Suche Bilder Maps Play YouTube News Gmail Drive Mehr »
Anmelden
Nutzer von Screenreadern: Klicke auf diesen Link, um die Bedienungshilfen zu aktivieren. Dieser Modus bietet die gleichen Grundfunktionen, funktioniert aber besser mit deinem Reader.

Patentsuche

  1. Erweiterte Patentsuche
VeröffentlichungsnummerUSRE46346 E1
PublikationstypErteilung
AnmeldenummerUS 14/226,413
Veröffentlichungsdatum21. März 2017
Eingetragen26. März 2014
Prioritätsdatum30. Okt. 2006
Auch veröffentlicht unterUS7975192, US8145984, US20100165730, US20110225472, WO2008053472A2, WO2008053472A3
Veröffentlichungsnummer14226413, 226413, US RE46346 E1, US RE46346E1, US-E1-RE46346, USRE46346 E1, USRE46346E1
ErfinderNaftali Sommer, Ofir Shalvi, Dotan Sokolov
Ursprünglich BevollmächtigterApple Inc.
Zitat exportierenBiBTeX, EndNote, RefMan
Externe Links: USPTO, USPTO-Zuordnung, Espacenet
Reading memory cells using multiple thresholds
US RE46346 E1
Zusammenfassung
A method for operating a memory (28) includes storing data, which is encoded with an Error Correction Code (ECC), in analog memory cells (32) of the memory by writing respective analog input values selected from a set of nominal values to the analog memory cells. The stored data is read by performing multiple read operations that compare analog output values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells. At least two of the read thresholds are positioned between a pair of the nominal values that are adjacent to one another in the set of the nominal values. Soft metrics are computed responsively to the multiple comparison results. The ECC is decoded using the soft metrics, so as to extract the data stored in the analog memory cells.
Bilder(9)
Previous page
Next page
Ansprüche(42)
The invention claimed is:
1. A method for operating a memory, comprising:
storing data, which is encoded with an Error Correction Code (ECC), in analog memory cells of the memory by writing to the analog memory cells respective analog input values that program the analog memory cells to a set of memory states;
reading the stored data multiple times from each analog memory cell by performing multiple read operations that compare analog output values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells, wherein the analog output values associated with each memory state lie in a respective analog value region, wherein analog value regions are separated by one or more boundary regions, and wherein at least two of the read thresholds are positioned in a boundary region between a pair of adjacent ones of the analog value regions;
computing soft metrics responsively to the multiple comparison results; and
decoding the ECC using the soft metrics, so as to extract the data stored in the analog memory cells;
wherein a plurality of the memory cells stores two or more bits of the data, wherein reading the data comprises, for the plurality of the memory cells, reading the two or more data bits in respective two or more decoding stages, and wherein computing the soft metrics comprises modifying a soft metric of a first bit read in a first decoding stage responsively to a value of a second bit read in a second decoding stage that precedes the first decoding stage; and
wherein modifying the soft metric comprises conditionally inverting the soft metric of the first bit depending on the value of the second bit.
2. The method according to claim 1, wherein each of the analog memory cells stores one or more bits of the data, and wherein each of the soft metrics corresponds to one of the bits.
3. The method according to claim 2, wherein each of at least some of the analog memory cells stores two or more bits of the data, wherein reading the data comprises, for each of the at least some of the analog memory cells, reading the two or more data bits in respective two or more decoding stages, and wherein computing the soft metrics comprises modifying a soft metric of a first bit read in a first decoding stage responsively to a value of a second bit read in a second decoding stage that precedes the first decoding stage.
4. The method according to claim 3, wherein modifying the soft metric comprises conditionally inverting the soft metric of the first bit depending on the value of the second bit.
5. The method according to claim 1, and comprising making an initial attempt to decode the ECC using an initial set of the read thresholds, such that no more than one of the read thresholds in the initial set is positioned in any given boundary region, and comparing the analog output values to the multiple read thresholds upon a failure of the initial attempt.
6. The method according to claim 1, wherein each comparison result has one of first and second possible values, and wherein computing the soft metrics comprises determining respective first and second counts of the comparison results having the first and second possible values, and computing the soft metrics based on the first and second counts.
7. The method according to claim 1, and comprising, upon failing to decode the ECC, adding one or more additional read thresholds to the multiple read thresholds, re-computing the soft metrics responsively to the additional read thresholds, and decoding the ECC using the re-computed soft metrics.
8. The method according to claim 7, wherein adding the additional threshold comprises progressively increasing a number of the read thresholds until a predetermined condition is met.
9. The method according to claim 1, wherein reading the data from a first group of the analog memory cells further comprises estimating interference caused to the first group by a second group of the analog memory cells and canceling the estimated interference.
10. The method according to claim 9, wherein canceling the estimated interference comprises modifying the soft metrics associated with the first group responsively to the estimated interference.
11. The method according to claim 9, and comprising, upon failing to decode the ECC in the first group, selecting whether to perform one of:
re-reading the data in the second group, so as to re-estimate and cancel the interference;
re-estimating the interference by reading the data in a third group of the memory cells; and
adding one or more additional read thresholds and re-reading the data in the first group using the additional read thresholds.
12. The method according to claim 1, wherein computing the soft metrics comprises normalizing the soft metrics so as not to depend on a number of the read thresholds.
13. The method according to claim 1, wherein performing the multiple read operations comprises positioning the multiple read thresholds at non-uniform intervals with respect to one another.
14. The method according to claim 1, wherein the analog output values associated with each memory state are distributed in a respective statistical distribution, and wherein reading the stored data comprises positioning the at least two of the read thresholds about a midpoint between respective statistical distributions of the analog output values associated with the memory states represented by the adjacent analog value regions.
15. The method according to claim 1, wherein two or more of the comparison results for a given analog memory cell are inconsistent with one another.
16. A data storage apparatus, comprising:
an interface, which is operative to communicate with a memory that includes a plurality of analog memory cells; and
a memory signal processor (MSP), which is connected to the interface and is coupled configured to store data, which is encoded with an Error Correction Code (ECC), in the analog memory cells by writing respective input analog values that program the analog memory cells to a set of memory states, to read the stored data multiple times from each analog memory cell by performing multiple read operations that compare analog output values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells, wherein the analog output values associated with each memory state lie in a respective analog value region, wherein analog value regions are separated by one or more boundary regions, and wherein at least two of the read thresholds are positioned in a boundary region between a pair of adjacent ones of the analog value regions, to compute soft metrics responsively to the multiple comparison results, and to decode the ECC using the soft metrics, so as to extract the data stored in the analog memory cells;
wherein a plurality of the memory cells stores two or more bits of the data;
wherein the MSP is further configured to:
read the two or more data bits in respective two or more decoding stages, and
modify a soft metric of a first bit read in a first decoding stage dependent upon a value of a second bit read in a second decoding stage that precedes the first decoding stage; and
wherein to modify the soft metric, the MSP is further configured to conditionally invert the soft metric of the first bit depending on the value of the second bit.
17. The apparatus according to claim 16, wherein each of the analog memory cells stores one or more bits of the data, and wherein each of the soft metrics corresponds to one of the bits.
18. The apparatus according to claim 17, wherein each of at least some of the analog memory cells stores two or more bits of the data, and wherein the MSP is coupled to read the two or more data bits in respective two or more decoding stages, and to modify a soft metric of a first bit read in a first decoding stage responsively to a value of a second bit read in a second decoding stage that precedes the first decoding stage.
19. The apparatus according to claim 18, wherein the MSP is coupled to conditionally invert the soft metric of the first bit depending on the value of the second bit.
20. The apparatus according to claim 16, wherein the MSP is coupled configured to make an initial attempt to decode the ECC using an initial set of the read thresholds, such that no more than one of the read thresholds in the initial set is positioned in any given boundary region, and to compare the analog output values to the multiple read thresholds upon failure of the initial attempt.
21. The apparatus according to claim 16, wherein each comparison result has one of first and second possible values, and wherein the MSP is coupled configured to determine respective first and second counts of the comparison results having the first and second possible values, and to compute the soft metrics based on the first and second counts.
22. The apparatus according to claim 16, wherein, upon failing to decode the ECC, the MSP is coupled configured to add one or more additional read thresholds to the multiple read thresholds, to recompute the soft metrics responsively to the additional read thresholds and to decode the ECC using the recomputed soft metrics.
23. The apparatus according to claim 22, wherein the MSP is coupled configured to progressively increase a number of the read thresholds until a predetermined condition is met.
24. The apparatus according to claim 16, wherein the MSP is coupled configured to estimate interference caused to a first group of the analog memory cells by a second group of the analog memory cells, and to cancel the estimated interference.
25. The apparatus according to claim 24, wherein the MSP is coupled configured to modify the soft metrics associated with the first group responsively to the estimated interference.
26. The apparatus according to claim 24, wherein, upon failing to decode the ECC in the first group, the MSP is coupled configured to select whether to perform one of:
re-reading the data in the second group, so as to re-estimate and cancel the interference;
re-estimating the interference by reading the data in a third group of the memory cells; and
adding one or more additional read thresholds and re-reading the data in the first group using the additional read thresholds.
27. The apparatus according to claim 16, wherein the MSP is coupled configured to normalize the soft metrics so as not to depend on a number of read thresholds.
28. The apparatus according to claim 16, wherein the MSP is coupled configured to position the multiple read thresholds at non-uniform intervals with respect to one another.
29. The apparatus according to claim 16, wherein the analog output values associated with each memory state are distributed in a respective statistical distribution, and wherein the MSP is coupled configured to position the at least two of the read thresholds about a midpoint between respective statistical distributions of the analog output values associated with the memory states represented by the adjacent analog value regions.
30. The apparatus according to claim 16, wherein two or more of the comparison results for a given analog memory cell are inconsistent with one another.
31. A data storage apparatus, comprising:
a memory device, comprising:
a plurality of analog memory cells, which are configured to store data, which is encoded with an Error Correction Code (ECC) and written to the analog memory cells as respective analog input values that program the analog memory cells to a set of memory states; and
reading circuitry, which is coupled to read the stored data multiple times from each analog memory cell by performing multiple read operations that compare output analog values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells, wherein the analog output values associated with each memory state lie in a respective analog value region, wherein analog value regions are separated by one or more boundary regions, and wherein at least two of the read thresholds are positioned in a boundary region between a pair of adjacent ones of the analog value regions, to compute soft metrics responsively to the multiple comparison results, and to output the computed soft metrics; and
a Memory Signal Processor (MSP) device, which is connected to the memory device and is coupled to accept the soft metrics computed by the reading circuitry, and to decode the ECC using the soft metrics.
32. A method for operating a memory, comprising:
storing data, which is encoded with an Error Correction Code (ECC), in a group of analog memory cells of the memory by writing to the analog memory cells in the group respective analog input values;
reading the data from the analog memory cells in the group by comparing analog output values of the analog memory cells in the group to one or more read thresholds, and applying ECC decoding to the read data; and
upon a failure of the ECC decoding, canceling interference caused to the analog memory cells in the group by at least one other analog memory cell, and re-decoding the ECC.
33. A data storage apparatus, comprising:
an interface, which is operative to communicate with a memory that includes a plurality of analog memory cells; and
a memory signal processor (MSP), which is connected to the interface and is coupled to store data, which is encoded with an Error Correction Code (ECC), in a group of analog memory cells of the memory by writing to the analog memory cells in the group respective analog input values, to read the data from the analog memory cells in the group by comparing analog output values of the analog memory cells in the group to one or more read thresholds, and applying ECC decoding to the read data, and, upon a failure of the ECC decoding, to cancel interference caused to the analog memory cells in the group by at least one other analog memory cell, and to re-decode the ECC.
34. A method for reading a memory cell of a non-volatile memory, comprising:
performing a first read operation of a memory cell dependent upon a first read threshold, wherein data stored in the memory cell is encoded with an Error Correction Code (ECC);
performing a second read operation of the memory cell dependent upon a second read threshold, wherein the first read threshold and the second read threshold are positioned in a boundary region relative to two possible memory states; and
determining a soft metric using the results of the first read operation and the second read operation;
modifying the soft metric by conditionally inverting the soft metric depending on a value of a data bit read from another memory cell; and
decoding the ECC using the soft metric to extract the data from the memory cell.
35. The method of claim 34, wherein the boundary region comprises an area of overlap between distributions of two memory states.
36. The method of claim 34, further comprising performing an initial read operation of the non-volatile memory cell dependent upon an initial read threshold prior to the first and second read operations.
37. The method of claim 36, wherein the first read operation and the second read operation are performed responsive to a failure of an error correction process dependent upon results of the initial read operation.
38. The method of claim 36, wherein the first read threshold is less than the initial read threshold, and wherein the second threshold is greater than the initial read threshold.
39. The method of claim 34, further comprising determining an interference caused to the memory cell by at least one other memory cell and compensating for the interference.
40. A method for operating a non-volatile memory, wherein the non-volatile memory includes a plurality of memory cells, the method, comprising:
performing a first read on a first group of the plurality of memory cells dependent upon a first read threshold, wherein data stored in the first group of the plurality of memory cells is encoded with an Error Correction Code (ECC);
performing a second read on the first group of the plurality of memory cells dependent upon a second read threshold;
determining at least one soft metric dependent upon results of the first read operation and the second read operation;
decoding the ECC using the at least one soft metric;
responsive to a failure to decode the ECC, performing a third read on the first group of the plurality of memory cells dependent upon a third read threshold;
updating at least one soft metric dependent upon a result of the third read;
wherein a of the plurality memory cells stores two or more bits of the data;
wherein performing the first read on the first group includes reading the two or more data bits of a memory cell in the first group in respective two or more decoding stages;
wherein determining the soft metric includes modifying a soft metric of a first bit read in a first decoding stage dependent upon a value of a second bit read in a second decoding stage that precedes the first decoding stage; and
wherein modifying the at least one soft metric includes conditionally inverting the soft metric of the first bit dependent upon the value of the second bit.
41. The method of claim 40, wherein updating the at least one soft metric comprises determining a new soft metric dependent upon the results of the first read operation and the second read operation, and results of the third read operation.
42. The method of claim 40, wherein the ECC comprises a low-density parity-check (LDPC) code.
Beschreibung
CROSS-REFERENCE TO RELATED APPLICATIONS

This application is an application for reissue of U.S. Pat. No. 8,145,984 B2, which is a continuation of U.S. patent application Ser. No. 11/995,814 filed on Jan. 15, 2008, which is the national stage entry of PCT/IL2007/001315 filed on Oct. 30, 2007, which claims the benefit of U.S. Provisional Patent Application 60/863,506, filed Oct. 30, 2006, U.S. Provisional Patent Application 60/867,399, filed Nov. 28, 2006, U.S. Provisional Patent Application 60/888,828, filed Feb. 8, 2007, U.S. Provisional Patent Application 60/889,277, filed Feb. 11, 2007, U.S. Provisional Patent Application 60/892,869, filed Mar. 4, 2007, U.S. Provisional Patent Application 60/894,456, filed Mar. 13, 2007, U.S. Provisional Patent Application 60/917,653, filed May 12, 2007, U.S. Provisional Patent Application 60/950,884, filed Jul. 20, 2007, and U.S. Provisional Patent Application 60/951,215, filed Jul. 22, 2007. The disclosures of all these related applications are incorporated herein by reference.

FIELD OF THE INVENTION

The present invention relates generally to memory devices, and particularly to methods and systems for reading data from memory cells.

BACKGROUND OF THE INVENTION

Several types of memory devices, such as Flash memories, use arrays of analog memory cells for storing data. Each analog memory cell stores a quantity of an analog value, such as an electrical charge or voltage, which represents the information stored in the cell. In Flash memories, for example, each analog memory cell holds a certain amount of electrical charge. The range of possible analog values is typically divided into regions, each region corresponding to one or more data bit values. Data is written to an analog memory cell by writing a nominal analog value that corresponds to the desired bit or bits. The possible bit values that can be stored in an analog memory cell are also referred to as the memory states of the cell.

Some memory devices, commonly referred to as Single-Level Cell (SLC) devices, store a single bit of information in each memory cell, i.e., each memory cell can be programmed to assume two possible memory states. Higher-density devices, often referred to as Multi-Level Cell (MLC) devices, store two or more bits per memory cell, i.e., can be programmed to assume more than two possible memory states.

Flash memory devices are described, for example, by Bez et al., in “Introduction to Flash Memory,” Proceedings of the IEEE, volume 91, number 4, April, 2003, pages 489-502, which is incorporated herein by reference. Multi-level Flash cells and devices are described, for example, by Eitan et al., in “Multilevel Flash Cells and their Trade-Offs,” Proceedings of the 1996 IEEE International Electron Devices Meeting (IEDM), New York, N.Y., pages 169-172, which is incorporated herein by reference. The paper compares several kinds of multilevel Flash cells, such as common ground, DINOR, AND, NOR and NAND cells.

Eitan et al., describe another type of analog memory cell called Nitride Read Only Memory (NROM) in “Can NROM, a 2-bit, Trapping Storage NVM Cell, Give a Real Challenge to Floating Gate Cells?” Proceedings of the 1999 International Conference on Solid State Devices and Materials (SSDM), Tokyo, Japan, Sep. 21-24, 1999, pages 522-524, which is incorporated herein by reference. NROM cells are also described by Maayan et al., in “A 512 Mb NROM Flash Data Storage Memory with 8 MB/s Data Rate”, Proceedings of the 2002 IEEE International Solid-State Circuits Conference (ISSCC 2002), San Francisco, Calif., Feb. 3-7, 2002, pages 100-101, which is incorporated herein by reference. Other exemplary types of analog memory cells are Floating Gate (FG) cells, Ferroelectric RAM (FRAM) cells, magnetic RAM (MRAM) cells, Charge Trap Flash (CTF) and phase change RAM (PRAM, also referred to as Phase Change Memory—PCM) cells. FRAM, MRAM and PRAM cells are described, for example, by Kim and Koh in “Future Memory Technology including Emerging New Memories,” Proceedings of the 24th International Conference on Microelectronics (MIEL), Nis, Serbia and Montenegro, May 16-19, 2004, volume 1, pages 377-384, which is incorporated herein by reference.

The analog values read from analog memory cells are sometimes distorted. The distortion may be due to various reasons, such as electrical field coupling from neighboring memory cells, disturb noise caused by memory access operations on other cells in the array and threshold voltage drift caused by device aging. Some common distortion mechanisms are described in the article by Bez et al., cited above. Distortion effects are also described by Lee et al., in “Effects of Floating Gate Interference on NAND Flash Memory Cell Operation,” IEEE Electron Device Letters, (23:5), May, 2002, pages 264-266, which is incorporated herein by reference.

Reading data from analog memory cells often involves comparing the analog values stored in the cells to one or more thresholds, or reference levels. Several methods for determining the appropriate threshold values are known in the art. For example, U.S. Pat. No. 5,657,332, whose disclosure is incorporated herein by reference, describes methods for recovering from hard errors in a solid-state memory system. Hard errors may arise from cells whose threshold voltages drifted from their intended level to cause read errors. The memory system includes an array of memory cells, each cell capable of having its threshold voltage programmed or erased to an intended level. An error checking scheme is provided for each of a plurality of groups of cells for identifying read errors therein. A read reference level is adjusted before each read operation on the individual group of cells containing read errors, each time the read reference level being displaced a predetermined step from a reference level for normal read, until the error checking means no longer indicates read errors. The drifted threshold voltage of each cell associated with a read error is re-written to its intended level.

U.S. Pat. No. 7,023,735, whose disclosure is incorporated herein by reference, describes methods for reading Flash memory cells, which, in addition to comparing the threshold voltages of Flash cells to integral reference voltages, compare the threshold voltages to fractional reference voltages.

U.S. Patent Application Publication 2007/0091677, whose disclosure is incorporated herein by reference, describes methods, devices and computer readable code for reading data from one or more flash memory cells, and for recovering from read errors. In some embodiments, in the event of an error correction failure by an error detection and correction module, the flash memory cells are re-read at least once using one or more modified reference voltages, until successful error correction may be carried out. In some embodiments, after successful error correction, a subsequent read request is handled without re-writing data to the flash memory cells in the interim.

U.S. Pat. No. 6,963,505, whose disclosure is incorporated herein by reference, describes a method, circuit and system for determining a reference voltage. In some embodiments a set of operating reference cells is established to be used in operating cells in a Non-Volatile Memory (NVM) block or array. At least a subset of cells of the NVM block or array may be read using each of two or more sets of test reference cells, where each set of test reference cells may generate or otherwise provide reference voltages at least slightly offset from each other set of test reference cells. For each set of test reference cells used to read at least a subset of the NVM block, a read error rate may be calculated or otherwise determined. A set of test reference cells associated with a relatively low read error rate may be selected as the set of operating reference cells to be used in operating other cells, outside the subset of cells, in the NVM block or array.

U.S. Pat. No. 7,196,928 and U.S. patent Application Publications 2006/0221692, 2007/0103986, 2007/0109845 and 2007/0109849, whose disclosures are incorporated herein by reference, describe several processes for reading a memory cell, which take into account the programmed state of an adjacent memory cell.

Some known methods use information regarding the quality of stored data when reading the data from memory cells. For example, U.S. Pat. No. 6,751,766, whose disclosure is incorporated herein by reference, describes several methods for assessing the quality of data stored in a memory system, and for operating the memory system according to the assessed quality. The data quality is sometimes assessed during read operations. Subsequent use of an Error Correction Code (ECC) can utilize the quality indications to detect and reconstruct the data with improved effectiveness. Alternatively, a statistics of data quality can be constructed and digital data values can be associated in a modified manner to prevent data corruption.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a method for operating a memory, including:

storing data, which is encoded with an Error Correction Code (ECC), in analog memory cells of the memory by writing respective analog input values selected from a set of nominal values to the analog memory cells;

reading the stored data by performing multiple read operations that compare analog output values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells, wherein at least two of the read thresholds are positioned between a pair of the nominal values that are adjacent to one another in the set of the nominal values;

computing soft metrics responsively to the multiple comparison results; and

decoding the ECC using the soft metrics, so as to extract the data stored in the analog memory cells.

In some embodiments, each of the analog memory cells stores one or more bits of the data, and each of the soft metrics corresponds to one of the bits. In an embodiment, each of at least some of the analog memory cells stores two or more bits of the data, reading the data includes, for each of the at least some of the analog memory cells, reading the two or more data bits in respective two or more decoding stages, and computing the soft metrics includes modifying a soft metric of a first bit read in a first decoding stage responsively to a value of a second bit read in a second decoding stage that precedes the first decoding stage. Modifying the soft metric may include conditionally inverting the soft metric of the first bit depending on the value of the second bit.

In another embodiment, the method includes making an initial attempt to decode the ECC using an initial set of the read thresholds, such that no more than one of the read thresholds in the initial set is positioned between each pair of the nominal values that are adjacent to one another, and comparing the analog output values to the multiple read thresholds upon a failure of the initial attempt.

In yet another embodiment, each comparison result has one of first and second possible values, and computing the soft metrics includes determining respective first and second counts of the comparison results having the first and second possible values, and computing the soft metrics based on the first and second counts.

In still another embodiment, the method further includes, upon failing to decode the ECC, adding one or more additional read thresholds to the multiple read thresholds, re-computing the soft metrics responsively to the additional read thresholds, and decoding the ECC using the re-computed soft metrics. Adding the additional threshold may include progressively increasing a number of the read thresholds until a predetermined condition is met.

In a disclosed embodiment, reading the data from a first group of the analog memory cells further includes estimating interference caused to the first group by a second group of the analog memory cells and canceling the estimated interference. Canceling the estimated interference may include modifying the soft metrics associated with the first group responsively to the estimated interference. In some embodiment, the method includes, upon failing to decode the ECC in the first group, selecting whether to perform one of:

re-reading the data in the second group, so as to re-estimate and cancel the interference;

re-estimating the interference by reading the data in a third group of the memory cells; and

adding one or more additional read thresholds and re-reading the data in the first group using the additional read thresholds.

In an embodiment, computing the soft metrics includes normalizing the soft metrics so as not to depend on a number of the read thresholds. Performing the multiple read operations may include positioning the multiple read thresholds at non-uniform intervals with respect to one another.

There is additionally provided, in accordance with an embodiment of the present invention, a data storage apparatus, including:

an interface, which is operative to communicate with a memory that includes a plurality of analog memory cells; and

a memory signal processor (MSP), which is connected to the interface and is coupled to store data, which is encoded with an Error Correction Code (ECC), in the analog memory cells by writing respective input analog values selected from a set of nominal values to the analog memory cells, to read the stored data by performing multiple read operations that compare analog output values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells, wherein at least two of the read thresholds are positioned between a pair of the nominal values that are adjacent to one another in the set of the nominal values, to compute soft metrics responsively to the multiple comparison results, and to decode the ECC using the soft metrics, so as to extract the data stored in the analog memory cells.

There is also provided, in accordance with an embodiment of the present invention, a data storage apparatus, including:

a memory device, including:

a plurality of analog memory cells, which are configured to store data, which is encoded with an Error Correction Code (ECC) and written to the analog memory cells as respective analog input values selected from a set of nominal values; and

reading circuitry, which is coupled to read the stored data by performing multiple read operations that compare output analog values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells, wherein at least two of the read thresholds are positioned between a pair of the nominal values that are adjacent to one another in the set of the nominal values, to compute soft metrics responsively to the multiple comparison results, and to output the computed soft metrics; and

a Memory Signal Processor (MSP) device, which is connected to the memory device and is coupled to accept the soft metrics computed by the reading circuitry, and to decode the ECC using the soft metrics.

There is further provided, in accordance with an embodiment of the present invention, a method for operating a memory, including:

storing data, which is encoded with an Error Correction Code (ECC), in analog memory cells of the memory by writing respective analog input values to the analog memory cells;

reading the stored data by comparing analog output values of the analog memory cells to a set of read thresholds, so as to produce multiple comparison results for each of the analog memory cells;

computing soft metrics responsively to the multiple comparison results;

decoding the ECC using the soft metrics, so as to extract the data stored in the analog memory cells; and

upon a failure to successfully extract the data, extending the set of the read thresholds by adding one or more new read thresholds to the set, updating the multiple comparison results based on the extended set of the read thresholds, re-computing the soft metrics and re-decoding the ECC, so as to extract the data.

In an embodiment, extending the set of the read thresholds includes selecting the one or more new read thresholds responsively to the output analog values of the analog memory cells. Selecting the one or more new read thresholds may include determining at least one property selected from a group of properties consisting of a number of the new read thresholds and values of the new read thresholds.

The present invention will be more fully understood from the following detailed description of the embodiments thereof, taken together with the drawings in which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram that schematically illustrates a system for memory signal processing, in accordance with an embodiment of the present invention;

FIG. 2 is a diagram that schematically illustrates a memory cell array, in accordance with an embodiment of the present invention;

FIG. 3 is a diagram that schematically illustrates read thresholds in a Single-Level Cell (SLC), in accordance with an embodiment of the present invention;

FIG. 4 is a diagram that schematically illustrates read thresholds in a Multi-Level Cell (MLC), in accordance with an embodiment of the present invention;

FIG. 5 is a flow chart that schematically illustrates a method for reading data from analog memory cells, in accordance with an embodiment of the present invention;

FIG. 6 is a flow chart that schematically illustrates a method for computing soft metrics, in accordance with an embodiment of the present invention;

FIG. 7 is a block diagram that schematically illustrates a circuit for computing soft metrics, in accordance with an embodiment of the present invention;

FIG. 8 is a flow chart that schematically illustrates a method for reading data from analog memory cells, in accordance with another embodiment of the present invention;

FIG. 9 is a diagram that schematically illustrates a process for reading data from analog memory cells, in accordance with yet another embodiment of the present invention;

FIG. 10 is a flow chart that schematically illustrates a method for reading data from analog memory cells, in accordance with still another embodiment of the present invention; and

FIG. 11 is a block diagram that schematically illustrates a system for memory signal processing, in accordance with an alternative embodiment of the present invention.

DETAILED DESCRIPTION OF EMBODIMENTS Overview

Embodiments of the present invention provide improved methods and systems for reading data from analog memory cells, such as Flash memory cells. In some embodiments that are described hereinbelow, a Memory Signal Processor (MSP) stores data, which is encoded with an Error Correction Code (ECC), in an array of analog memory cells. The MSP stores the encoded data by writing respective analog values to the analog memory cells. The analog values are selected from a set of nominal analog values, which represent the data.

The MSP reads the data from the analog memory cells by performing multiple read operations, which compare the analog values written to the cells to multiple read thresholds. The read thresholds are set so that at least two of them are positioned between a pair of adjacent nominal analog values. The multiple threshold comparisons produce multiple comparison results for each of the analog memory cells. The MSP computes soft metrics based on the multiple comparison results. The soft metrics provide quantitative measures of the levels of confidence or certainty that are associated with the values read from the memory cells, or of individual bits within the memory cells. The MSP decodes the ECC using the soft metrics. In some embodiments, the MSP increases the number of read thresholds in an iterative manner, until successful decoding is achieved.

Some known reading methods differentiate between adjacent memory states using a single threshold at any given time. Unlike these known methods, the methods and systems described herein perform multiple read operations using multiple thresholds, which are positioned between adjacent memory states. Typically, multiple thresholds are positioned in boundary regions between adjacent nominal values, so that the multiple comparison results convey valuable information regarding the statistical distribution of the analog values in these regions. As a result, the soft metrics, which are based on this information, enable the ECC decoding process to correct a higher number of read errors and to provide an improved overall error probability.

Some known reading methods modify the threshold values in order to improve decoding performance. Unlike these known methods, the methods and systems described herein do not adapt the threshold values, but rather add new thresholds to the existing set, and improve the decoding performance by refining the accuracy of the soft metrics.

The improved decoding performance achieved by the disclosed methods and systems enables improving the data storage reliability, storage density and retention time of memory devices, and enables lowering the memory device cost and complexity for a given performance level.

System Description

FIG. 1 is a block diagram that schematically illustrates a system 20 for memory signal processing, in accordance with an embodiment of the present invention. System 20 can be used in various host systems and devices, such as in computing devices, cellular phones or other communication terminals, removable memory modules (“disk-on-key” devices), digital cameras, music and other media players and/or any other system or device in which data is stored and retrieved.

System 20 comprises a memory device 24, which stores data in a memory cell array 28. The memory array comprises multiple analog memory cells 32. In the context of the present patent application and in the claims, the term “analog memory cell” is used to describe any memory cell that holds a continuous, analog value of a physical parameter, such as an electrical voltage or charge. Array 28 may comprise analog memory cells of any kind, such as, for example, NAND, NOR and CTF Flash cells, PCM, NROM, FRAM, MRAM and DRAM cells. The charge levels stored in the cells and/or the analog voltages or currents written into and read out of the cells are referred to herein collectively as analog values.

System 20 stores data in the analog memory cells by programming the cells to assume respective memory states. The memory states are selected from a finite set of possible states, and each state corresponds to a certain nominal analog value. For example, a 2 bit/cell MLC can be programmed to assume one of four possible memory states by writing one of four possible nominal analog values into the cell.

Data for storage in memory device 24 is provided to the device and cached in data buffers 36. The data is then converted to analog voltages and written into memory cells 32 using a reading/writing (R/W) unit 40, whose functionality is described in greater detail below. When reading data out of array 28, R/W unit 40 converts the electrical charge, and thus the analog voltages of memory cells 32, into digital samples having a resolution of one or more bits. The samples are cached in buffers 36. The operation and timing of memory device 24 is managed by control logic 48.

The storage and retrieval of data in and out of memory device 24 is performed by a Memory Signal Processor (MSP) 52. MSP 52 comprises a signal processing unit 60, which processes the data that is written into and read from device 24. Unit 60 encodes the data to be written into the memory cells using an Error Correction Code (ECC), and decodes the ECC of the retrieved data.

In particular, MSP 52 reads data out of memory cells 32 by comparing the values read from the cells to multiple read thresholds. The ECC decoding scheme used by unit 60 operates on soft metrics, which are computed based on the multiple threshold comparisons. Exemplary methods for reading data and for computing soft metrics are described in detail below.

Many known ECC decoding schemes can accept soft metrics of the encoded bits or symbols as input. For example, unit 60 may use a block code such as the Bose-Chaudhuri-Hocquenghem (BCH) code, Low-Density Parity Check (LDPC) code or Reed-Solomon (RS) code, a trellis code, a turbo-code, or any other suitable ECC and decoding scheme, which is able to operate on soft metrics. The methods and systems described herein are not limited to block codes and can be used with convolutional codes, as well.

MSP 52 comprises a data buffer 72, which is used by unit 60 for storing data and for interfacing with memory device 24. MSP 52 also comprises an Input/Output (I/O) buffer 56, which forms an interface between the MSP and the host system. A controller 76 manages the operation and timing of MSP 52. Signal processing unit 60 and controller 76 may be implemented in hardware. Alternatively, unit 60 and/or controller 76 may comprise microprocessors that run suitable software, or a combination of hardware and software elements.

The configuration of FIG. 1 is an exemplary system configuration, which is shown purely for the sake of conceptual clarity. Any other suitable configuration can also be used. Elements that are not necessary for understanding the principles of the present invention, such as various interfaces, addressing circuits, timing and sequencing circuits and debugging circuits, have been omitted from the figure for clarity.

In the exemplary system configuration shown in FIG. 1, memory device 24 and MSP 52 are implemented as two separate Integrated Circuits (ICs). In alternative embodiments, however, the memory device and MSP may be integrated on separate semiconductor dies in a single Multi-Chip Package (MCP) or System on Chip (SoC). Further alternatively, some or all of the MSP circuitry may reside on the same die on which memory array 28 is disposed. An exemplary configuration of this sort is described in FIG. 11 below. Further alternatively, some or all of the functionality of MSP 52 can be implemented in software and carried out by a processor or other element of the host system. In some implementations, a single MSP 52 may be connected to multiple memory devices 24. Additional architectural aspects of certain embodiments of system 20 are described in greater detail in U.S. Provisional Patent Application 60/867,399, cited above.

In a typical writing operation, data to be written into memory device 24 is accepted from the host and cached in I/O buffer 56. The data is transferred, via data buffers 72, to memory device 24. The data may be pre-processed by MSP 52 before it is transferred to the memory device for programming. For example, unit 60 may encode the data using an ECC, add certain data for internal use, and/or scramble the data. In device 24 the data is temporarily stored in buffers 36. R/W unit 40 converts the data to nominal analog values and writes the nominal values into the appropriate cells 32 of array 28.

In a typical reading operation, R/W unit 40 reads analog values out of the appropriate memory cells 32 and converts them to soft digital samples. The samples are cached in buffers 36 and transferred to buffers 72 of MSP 52. In some embodiments, unit 60 of MSP 52 converts the samples to data bits. As noted above, the range of possible analog values is divided into two or more regions, with each region representing a certain combination of one or more data bits.

As will be described in greater detail further below, the memory cells are read by comparing their analog values to multiple sets of read thresholds. For each cell, the MSP computes a soft metric based on the multiple comparison results. The soft metrics are then used by the MSP when decoding the ECC. The decoded data is transferred via I/O buffer 56 to the host system.

Memory Array Structure

FIG. 2 is a diagram that schematically illustrates memory cell array 28, in accordance with an embodiment of the present invention. Although FIG. 2 refers to Flash memory cells that are connected in a particular array configuration, the principles of the present invention are applicable to other types of memory cells and other array configurations, as well. Some exemplary cell types and array configurations are described in the references cited in the Background section above.

Memory cells 32 of array 28 are arranged in a grid having multiple rows and columns. Each cell 32 comprises a floating gate Metal-Oxide Semiconductor (MOS) transistor. A certain amount of electrical charge (electrons or holes) can be stored in a particular cell by applying appropriate voltage levels to the transistor gate, source and drain. The value stored in the cell can be read by measuring the threshold voltage of the cell, which is defined as the minimal voltage that needs to be applied to the gate of the transistor in order to cause the transistor to conduct. The read threshold voltage is indicative of the charge stored in the cell.

In the exemplary configuration of FIG. 2, the gates of the transistors in each row are connected by word lines 80. The sources of the transistors in each column are connected by bit lines 84. In some embodiments, such as in some NOR cell devices, the sources are connected to the bit lines directly. In alternative embodiments, such as in some NAND cell devices, the bit lines are connected to strings of floating-gate cells.

Typically, R/W unit 40 reads the threshold voltage of a particular cell 32 by applying varying voltage levels to its gate (i.e., to the word line to which the cell is connected) and checking whether the drain current of the cell exceeds a certain threshold (i.e., whether the transistor conducts). Unit 40 usually applies a sequence of different voltage values to the word line to which the cell is connected, and determines the lowest gate voltage value for which the drain current exceeds the threshold. Typically, unit 40 reads a group of cells, referred to as a page, simultaneously. Alternatively, R/W unit may use any other technique or circuitry for reading and writing values to and from memory cells 32 of array 28.

The memory cell array is typically divided into multiple pages, i.e., groups of memory cells that are programmed and read simultaneously. In some embodiments, each page comprises an entire row of the array. In alternative embodiments, each row (word line) can be divided into two or more pages. For example, in some SLC devices each row is divided into two pages, one comprising the odd-order cells and the other comprising the even-order cells. Typically but not necessarily, a two-bit-per-cell memory device usually has four pages per row, a three-bit-per-cell memory device has six pages per row, and a four-bit-per-cell memory device has eight pages per row.

Erasing of cells is usually carried out in blocks that contain multiple pages. Typical memory devices may comprise several thousand erasure blocks. In a typical two-bit-per-cell MLC device, each erasure block is on the order of 32 word lines, each comprising several thousand cells. Each word line is often partitioned into four pages (odd/even order cells, least/most significant bit of the cells). Alternatively, other block sizes and configurations can also be used. Three-bit-per cell devices often have 192 pages per erasure block, and four-bit-per-cell devices often have 256 pages per block.

Some memory devices comprise two or more separate memory cell arrays, often referred to as planes. Since each plane has a certain “busy” period between successive write operations, data can be written alternately to the different planes in order to increase programming speed.

Memory Cell Distortion Mechanisms

The analog values (e.g., threshold voltages) stored in memory cells 32 may contain various types of distortion, which are caused by different distortion mechanisms in array 28. For example, electrical cross-coupling between nearby cells in the array may modify the threshold voltage of a particular cell. As another example, electrical charge may leak from the cells over time. As a result of this aging effect, the threshold voltage of the cells may drift over time from the initially-written value. Another type of distortion, commonly referred to as disturb noise, is caused by memory access operations (e.g., read, write or erase operations) on certain cells in the array, which cause unintended charge variations in other cells. As yet another example, the source-drain current of a particular cell can be affected by the charge in adjacent cells, e.g., other cells in the same NAND cell string, via an effect referred to as Back Pattern Dependency (BPD).

The distortion in memory cells 32 degrades the performance of the memory device, e.g., the error probability when reconstructing the data, the achievable storage capacity and/or the achievable data retention period. Performance degradation is particularly severe in MLC devices, in which the differences between the different voltage levels that represent the data are relatively small.

Reading Memory Cells Using Multiple Thresholds

Embodiments of the present invention provide improved methods and systems for reading data from analog memory cells 32 of array 28, by using multiple read thresholds. The methods described herein are suitable for both SLC devices (as illustrated, for example, in FIG. 3 below) and MLC devices (as illustrated, for example, in FIG. 4 below).

FIG. 3 is a diagram that schematically illustrates read thresholds in an SLC device, in accordance with an embodiment of the present invention. The figure shows two statistical distributions of the threshold voltages in a group (e.g., page) of analog memory cells. For a specific page, the diagram represents a histogram of the threshold voltages stored in the memory cells of the page. In the example of FIG. 3, each memory cell is programmed to one of two possible nominal levels, i.e., each cell stores a single data bit. Due to various distortion variations among the cells and various impairment mechanisms, the actual threshold voltages read from the memory cells may statistically vary from the nominal levels. In the present example, a curve 90A shows the distribution of threshold voltages of the cells, which are programmed to store a “1” value. A curve 90B shows the distribution of threshold voltages of the cells that are programmed to store “0”.

As can be seen in the figure, curves 90A and 90B overlap. In other words, there is a finite probability that a memory cell, which was programmed to a certain bit value, will be erroneously interpreted as being programmed to another bit value. The position of the read threshold or thresholds used to differentiate between “1” and “0” has a considerable effect on the probability of error. In some embodiments of the present invention, MSP 52 reconstructs the data stored in the memory cells by combining information, which is obtained using multiple read thresholds, in order to reduce the probability of error.

FIG. 3 shows five thresholds denoted T1 . . . T5. In some embodiments, MSP 52 reads each memory cell using each of the thresholds. Each read operation produces a comparison result, i.e., an indication of whether the read threshold voltage is greater or smaller than the threshold used in the operation. In the exemplary embodiment of FIG. 3, the MSP reads each memory cell five times, using thresholds T1 . . . T5, to produce five respective comparison results. The MSP may go through the different thresholds at any suitable order. For example, the MSP may begin with the threshold positioned in the middle of the desired region and gradually move away from this point by adding thresholds on either side of the initial threshold. Exemplary iterative methods that gradually increase the number of thresholds are described further below.

For each memory cell being read, the MSP computes a soft metric using the multiple comparison results. The soft metric indicates a confidence level or measure of certainty associated with the value read from the memory cell. In some embodiments, the soft metric indicates a likelihood that the read value corresponds to a certain data value (e.g., a very low metric value may indicate a high certainty that the read value corresponds to a “1”, a very high metric value indicates that the read value is likely to represent a “0”, and intermediate metric values indicate lower confidence). In other embodiments, the metric value indicates the reliability of the read value without indicating a particular bit value (e.g., low metric value indicates low confidence, high metric value represents high confidence).

In the context of the present patent application and in the claims, the term “soft metric” refers to any type of quantitative measure that conveys more than a single bit of information, i.e., more than two possible values. For example, the soft metric may comprise a fixed- or floating-point numerical value represented using two or more bits. Another exemplary type of soft metric, sometimes referred to as “erasure,” assigns each read memory cell one of three possible values—“0”, “1” or “uncertain.” Further alternatively, any other suitable type of soft metric can be used.

Note that when each cell stores multiple data bits, a soft metric value may be computed and assigned to each individual bit. For example, in a four-level MLC, one metric value is computed for the Least Significant Bit (LSB) and another metric value is computed for the Most Significant Bit (MSB). Detailed examples of metric computation methods for both SLC and MLC applications are described further below.

MSP 52 may use any suitable method for computing the soft metric value based on the multiple comparison results. In some embodiments, the MSP may use a table, which provides the metric values associated with different combinations of the comparison results. For example, the following table can be used with the five-threshold configuration of FIG. 3:

Comparison results Metric
T1 T2 T3 T4 T5 value
0 0 0 0 0 M1
0 0 0 0 1 M2
0 0 0 1 0 M3
0 0 0 1 1 M4
. . . . . . . . . . . . . . . . . .
1 1 1 1 0 M31
1 1 1 1 1 M32

The table above provides thirty-two soft metric values denoted M1 . . . M32, which correspond to the thirty-two possible combinations of five comparison results of thresholds T1 . . . T5. Following the notation of FIG. 3, a “0” comparison result means that the read value was higher than the threshold used, and a “1” comparison result means the read value was lower than the threshold.

Typically, M1 and M32 will indicate high confidence levels, since these metric values correspond to situations in which the read operations with all five thresholds produce the same comparison results. Other combinations of comparison results will usually be assigned metrics that indicate lower confidence levels.

Some sets of comparison results may be regarded as inconsistent or self-contradictory. For example, assume T1<T2<T3<T4<T5, and that the five comparison results produced by thresholds T1 . . . T5 are denoted C1 . . . C5, respectively. The result set ‘1,1,1,0,1’ for a certain memory cell is inconsistent because it indicates that the analog value is larger than T4 and smaller than T3, even though T4>T3. Such a result set may be caused, for example, when the cell has a high level of read noise in at least one of the read operations. Result sets such as ‘1,1,1,1,0’, ‘1,1,1,0,0’, or ‘1,0,0,0,0’, on the other hand, are consistent.

The MSP may treat inconsistent sets of comparison results in different manners, by assigning them different soft metric values. For example, the MSP may regard inconsistent result sets as uncertain and mark them as erasures to the ECC decoding process. Alternatively, the MSP may disregard or otherwise attempt to resolve some inconsistencies. For example, the MSP may regard a ‘1,1,0,1,1,’ result set similarly to a ‘1,1,1,1,1’ set, assuming that the “0” comparison result of T3 was caused by read noise.

Alternatively to using tables, MSP 52 may evaluate a function that operates on the multiple comparison results and produces the corresponding soft metric value. For example, the MSP may evaluate Log Likelihood Ratios (LLRs) of individual bits in each memory cell, which are defined as

LLR Λ ( X i ) = log [ p ( X i = 1 | r ) p ( X i = 0 | r ) ] [ 1 ]

wherein Xi denotes a particular data bit stored in the memory cell in question, and r denotes the analog value read from the cell. The use of LLRs as metrics that are provided to an ECC decoding process is described, for example, in PCT Patent Application PCT/IL2007/000580, entitled “Combined Distortion Estimation and Error Correction Coding For Memory Devices,” filed May 10, 2007, whose disclosure is incorporated herein by reference.

In order to calculate the LLR, the MSP may maintain two values for each memory cell: (1) the largest read threshold that was found to be below the analog value of the cell, denoted Va, and (2) the smallest read threshold that was found to be above the analog value of the cell, denoted Vb. The LLR of the cell can be shown to be approximated by

Λ ( X i ) log [ Q { V a - T 1 σ } - Q { V b - T 1 σ } ] - log [ Q { V a - T 0 σ } - Q { V b - T 0 σ } ] [ 2 ]
wherein T1 denotes the center analog value of the nearest distribution that has “1” as its data bit, and T0 denotes the center value of the nearest distribution having “0” as its data bit. The distribution of r is assumed Gaussian with variance σ2.

As the memory cell is read with an increasing number of read thresholds, the MSP updates Va and Vb. At each stage, the actual analog value of the cell is known to be within the interval [Va, Vb]. As the number of thresholds increases, the interval shrinks, the uncertainty becomes smaller and the estimated LLR becomes more accurate.

Further alternatively, the MSP may use any other suitable method or mechanism for computing the soft metric values based on the multiple comparison results.

MSP 52 uses the soft metrics when decoding the ECC. In a typical application, the data stored in a group of memory cells, such as in a certain memory page, forms a single codeword. When decoding a certain ECC codeword, signal processing unit 60 of the MSP uses the soft metric values of the memory cells in the group. As a result, memory cells that are considered to have a high confidence level are given more weight in the ECC decoding process, and vice versa.

FIG. 4 is a diagram that schematically illustrates read thresholds in an MLC device, in accordance with an embodiment of the present invention. In the example of FIG. 4, each memory cell is programmed to one of four possible nominal levels, thus storing two bits of data. Curves 94A . . . 94D show the threshold voltage distributions of the memory cells that are programmed to store “11”, “01”, “00” and “10” data, respectively. In the present example, MSP 52 reads the memory cells using five sets of thresholds. Each threshold set comprises three thresholds, which are typically positioned in the three boundary regions between pairs of adjacent distribution curves. The threshold sets are listed in the following table:

Threshold set Thresholds
1 T11, T21, T31
2 T12, T22, T32
3 T13, T23, T33
4 T14, T24, T34
5 T15, T25, T35

In some embodiments, MSP 52 reads the threshold voltage of the cell using each of the fifteen thresholds, and computes a soft metric based on the fifteen comparison results. The MSP may use any type of soft metric and any method of computing the metric value based on the multiple comparison results. The MSP uses the soft metric values as input to the ECC decoding process, as explained above.

In alternative embodiments, the memory cell is read in two stages, corresponding to the two bits stored in the cell. For example, in the configuration of FIG. 4, the R/W unit performs a first set of comparisons using the five thresholds T21, T22, T23, T24 and T25, i.e., the thresholds located in the middle of the voltage axis, between curves 94B and 94C. The MSP computes a first soft metric based on the five comparison results obtained using these five thresholds. Note that both nominal levels located above thresholds T21 . . . T25 have an LSB value of “0” and that both nominal levels located below thresholds T21 . . . T25 have an LSB value of “1”. Therefore, the first soft metric corresponds to the LSB. Once the LSB is decoded, the R/W unit performs a second set of comparisons. The R/W unit may use thresholds T11 . . . T15 or thresholds T31 . . . T35 in the second stage, depending on the decoded value of the LSB. If the LSB was determined to be “1”, i.e., the read value was determined to be in the lower part of the voltage range, the MSB will be decoded using thresholds T11 . . . T15 in the second stage. If the LSB was decoded as “0”, the MSB will be decoded using thresholds T31 . . . T35. The MSP computes a second metric, which corresponds to the MSB, based on the five comparison results obtained in the second comparison stage.

A similar multi-stage comparison process can be carried out in multi-level cells storing a higher number of bits. For example, in eight-level (3 bit/cell) cells, the MSP and R/W unit may perform a three-stage comparison process to decode the individual bits. Apart from the first stage, the selection of the thresholds used in each stage typically depends on the decoded values of the previous bits.

In alternative multi-stage reading processes, each bit is read independently of the other bits. For example, referring to FIG. 4, the LSB can be read using thresholds T21 . . . T25. The MSB is read by sequentially reading the cell using both thresholds T11 . . . T15 and T31 . . . T35. If the comparison results indicate that the analog value is between T11 . . . T15 and T31 . . . T35, the bit is determined to be “0”. If, on the other hand, the comparison results indicate that the analog value is larger than T31 . . . T35 or smaller than T11 . . . T15, the bit is determined to be “1”. In this example, the comparison results to thresholds T21 . . . T25, which were used for reading the LSB, are not used for reading the MSB. Similar processes may be performed for other types of MLC, such as eight-level cells storing three bits per cell.

The threshold configurations shown in FIGS. 3 and 4 above are exemplary configurations, which were chosen purely for the sake of conceptual clarity. In alternative embodiments, system 20 may use any desired number of nominal levels, any other mapping of bit values to nominal levels and any desired number of threshold sets. Although FIGS. 3 and 4 show thresholds that are spaced at regular increments, the methods and systems described herein may use irregularly-spaced thresholds, as well. In MLC devices, the threshold spacing may vary from one voltage region to another. For example, in FIG. 4, thresholds T11 . . . T15 may be spaced differently than thresholds T21 . . . T25. Different threshold spacing may be used, for example, when different analog value distributions have different shapes or different spacing with respect to one another. The MSP may modify the threshold spacing, or otherwise select the threshold values to use, such as based on estimation of the analog value distributions.

FIG. 5 is a flow chart that schematically illustrates a method for reading data from analog memory cells 32, in accordance with an embodiment of the present invention. For a certain memory cell, the method begins with system 20 performing multiple read operations using respective multiple thresholds, at a reading step 100. The multiple read operations produce respective multiple comparison results, i.e., indications of whether the threshold voltage of the cell is smaller or greater than the different thresholds. MSP 52 computes a soft metric of the memory cell based on the multiple comparison results, at a metric computation step 104.

The MSP typically repeats the process of steps 100 and 104 above over a group of memory cells, whose data forms a single ECC codeword. In a typical implementation, R/W unit 40 reads the cells of an entire page of the memory device, using a particular threshold value, simultaneously. Once the soft metrics of the cells that store a certain codeword are computed, the MSP decodes the codeword using the metrics, at a decoding step 108. The MSP extracts the decoded data, at a data extraction step 112. The decoded data is typically output to the host system.

The multiple-threshold reading methods described herein can also be viewed as an efficient means for obtaining accurate information regarding the stored analog values using a relatively small number of read operations. Theoretically, if the exact analog values stored in the memory cells were known to the MSP (e.g., by employing high-resolution analog-to-digital conversion), this information could be used to extract probability measures on the stored data. However, the basic read operation of analog memory devices, such as Flash memories, usually comprises comparison operations, which compare the analog value stored in a cell to a single threshold. In order to obtain the analog value with a given resolution, the entire possible voltage range would have to be searched or scanned with the desired resolution. For example, if the range of possible analog values is 0-4V, and the desired resolution is 10 mV, 400 read operations would be needed. In practice, however, much of the valuable statistical information can be obtained by performing a much smaller number of read operations, for example by positioning the read thresholds in a region around the midpoint between distributions. The methods and systems described herein thus provide efficient means of gaining insight to such analog value statistics using a relatively small number of read operations.

In many practical cases, performing a large number of read operations on a certain memory cell is a computationally-intensive task, which complicates and slows down the data retrieval process. Moreover, the ECC is usually strong enough to successfully decode the vast majority of codewords, even when the memory cells are read using a single set of thresholds. Therefore, in some embodiments, the MSP initially reads the memory cells using a single set of thresholds. The MSP reverts to read the memory cells that correspond to a certain codeword using the multiple-threshold methods described herein only when the ECC decoding process fails.

The methods of FIGS. 3-5 above can be applied iteratively, gradually increasing the number of thresholds used. For example, the MSP may attempt to reconstruct the data using soft metrics that are based on two sets of thresholds. If the data cannot be reconstructed (i.e., if the ECC fails), the MSP can re-read the memory cells using a third threshold set. The iterations may continue until ECC decoding succeeds, or until reaching a predetermined maximum number of threshold sets. Note that at each stage of the iterative process, the MSP computes the soft metrics based on the multiple comparison results that are available so far. In some cases, the MSP may use information, such as metric values, which was calculated in previous iterations. The iterative process enables a gradual increase in the number of computations, only as needed to carry out successful decoding.

Soft Metrics Based on Counting Computation Results

In some embodiments, the MSP computes the soft metric value based on the number of computation results falling on either side of the thresholds. (In the description that follows, a “0” comparison result means the read value was higher than the threshold, and vice versa. This convention, however, is chosen purely for the sake of convenience, and the opposite convention can also be used.) Consider, for example, the exemplary SLC embodiment of FIG. 3 above. In this embodiment, out of the five comparison results, if the number of “0” comparison results is considerably higher than the number of “1” results, it is likely that the cell was programmed to “0”. Similarly, if the number of “1” comparison results is considerably higher than the number of “0” results, the programmed bit is likely to be “1”. Similar logic can also be used within each stage of the multi-stage comparison process that decodes the individual bits of a MLC cell, which was described in FIG. 4 above.

FIG. 6 is a flow chart that schematically illustrates an exemplary method for computing soft metrics, in accordance with an embodiment of the present invention. The method description refers to an SLC application and makes reference to FIG. 3 above. This choice, however, is made purely for the sake of simplicity of explanation. The method can similarly be used in MLC applications, as well.

The method begins with the MSP defining multiple thresholds, at a threshold definition step 116. Typically but not necessarily, the thresholds are defined within the boundary region between the voltage distribution. In FIG. 3, five thresholds are defined in the region in which curves 90A and 90B overlap. For example, assuming the mid-point between curves 90A and 90B is at 1 volt, and that the thresholds can be represented at a resolution of 20 mV, a set of thresholds can be defined to cover the voltage range of 1V±40 mV at 20 mV intervals.

The MSP reads the memory cells using the multiple thresholds, at a reading step 120. The MSP counts the number of comparison results falling on either side of the thresholds, at a counting step 124. In other words, the MSP determines the number of “0” comparison results and/or the number of “1” results out of the total number of threshold comparisons.

The MSP computes a soft metric associated with the cell (or with an individual bit within the cell) based on the count of comparison results, at a metric computation step 128. For example, assuming the five-threshold configuration of FIG. 3 above and a four-bit metric value, the MSP may compute the soft metric according to the following table:

Number of “0” Number of “1”
computation results computation results Soft metric value
0 5  “1111” = 15
1 4  “1100” = 12
2 3 “1001” = 9
3 2 “0110” = 6
4 1 “0011” = 3
5 0 “0000” = 0

In the table above, if all five computation results are “1”, the stored data bit is “1” with high likelihood, therefore the maximum metric value of “1111” is assigned. At the other extreme, if all five comparison results are “0”, the stored bit is likely to be “0”, and the minimum metric value of “0000” is assigned. If some comparison results are “0” and others are “1”, the metric value is set to an intermediate value, which grows monotonously with the number of “1” results out of the total.

Alternatively, any other suitable method for determining the soft metric value based on the count of comparison results can be used. The metric computation may be implemented by querying a table that holds the metric values and is indexed by the count of comparison results, evaluating a function that operates on the count of the comparison results, or using any other suitable mechanism.

Exemplary Hardware Implementation for MLC Metric Computation

As noted above, when computing the soft metrics of individual bits in a multi-level cell, the selection of thresholds may depend on the values of previously-decoded bits. Moreover, the values of previously-decoded bits may in some cases affect the metric value itself.

Consider, for example, the four-level cell configuration of FIG. 4 above. When reading the data out of such a cell in a two-stage process, the LSB is first decoded by determining whether the value read from the cell falls on the left- or right-hand-side of thresholds T21 . . . T25. Note that in the example of FIG. 4, the two nominal levels located below these thresholds have an LSB value of “1”, and the two nominal levels located above the thresholds have an LSB value of “0”.

The second decoding stage (decoding of the MSB) depends on the results of the first stage. When the LSB is “0”, decoding the MSB comprises determining whether the read value is likely to belong to curve 94C or to curve 94D. When the LSB is “1”, decoding the MSB comprises determining whether the read value is likely to belong to curve 94A or to curve 94B.

Note, however, that when comparing curves 94A and 94B (i.e., when LSB=“1”), high threshold voltages correspond to MSB=“0” and low threshold voltages correspond to MSB=“1”. When comparing curves 94C and 94D (i.e., when LSB=“0”) the situation is reversed, with high threshold voltages corresponding to MSB=“1” MSB and low threshold voltages corresponding to MSB=“0”. In such a situation, the soft metric value that depends on the count of comparison results should sometimes be inverted, so as to maintain the convention that a high metric value corresponds to “0” data. The decision whether or not to invert the metric value depends on the value of the previous bit. Equivalently, the value of the currently-read bit can be inverted instead of inverting the metric value. The conditional operation of inverting a value only if a previous value is equal to “1” can be implemented by performing an eXclusive-OR (XOR) operation between the current and previous bit values.

FIG. 7 is a block diagram that schematically illustrates an exemplary circuit for computing soft metrics in a multi-level cell, in accordance with an embodiment of the present invention. Although the description that follows refers to a hardware or firmware implementation, a similar mechanism can be implemented in software, or as a combination of software and hardware elements.

The circuit of FIG. 7 computes soft metrics of the LSBs and MSBs of a group of four-level cells, assuming the LSBs represent a certain memory page and the MSBs represent another page. The LSB page is read first and is referred to as the previous page. The MSB page is read second and is referred to as the current page. The computation process of the metrics of the current page makes conditional inversion (XOR) operations depending on the bit values of the previous page.

The circuit comprises a XOR circuit 134, which performs a bit-wise XOR operation between a byte 130 of hard bit decisions from the current page (MSBs) and a byte 132 of previously-decoded data bits (LSBs) from the previous page. Thus, for a particular cell, when the previously-decoded LSB is “1”, the currently-read MSB is inverted. An adder 136 sums the results of the XOR operations. The adder output is accumulated as a soft metric 140 of the MSB. A vector 138 holds the accumulated metrics of the MSBs of the different cells. The same circuit can also be used to compute the soft metrics of the LSBs, which do not depend on any previous values. In order to compute the LSB soft metrics, byte 132 is filled with zeros so that the XOR operation is bypassed and byte 130 is provided to adder 136 unchanged.

The circuit of FIG. 7 refers to four-level, 2 bit/cell MLC. Similar circuits can be used, however, to compute soft metrics for other types of multi-level cells, such as eight-level, 3 bits/cell MLC.

In alternative embodiments, the soft metrics of individual bits of a multi-level cell can be calculated independently for different bits. These methods may be of particular benefit when the read data values of previous bits are not available when reading a certain bit. Referring to the 2 bit/cell example of FIG. 4 above, the soft metrics of the MSB may be computed without knowledge of the LSB. As noted above, the MSB value can be assumed to be “0” if the analog value falls between thresholds T11 . . . T15 and T31 . . . T35, and “1” otherwise. In order to compute a soft metric for such a reading process, the MSP may group the thresholds in pairs that move progressively inwards into the region in which MSB=“0”. The MSP counts the comparison results falling inside and/or outside the MSR=“0” interval using the different threshold pairs.

In the present example, the MSP forms the pairs (T14, T35), (T12, T33), (T11, T31), (T13, T32) and (T15, T34). For each pair, the MSP performs two read operations and checks whether the read value falls in the interval between the thresholds, or outside the interval. The MSP counts the number of threshold pairs in which the analog value falls between the two thresholds (indicating MSB=“0”) and/or the number of threshold pairs in which the analog value falls outside the interval between the two thresholds (indicating MSB=“1”). The MSP computes a soft metric based on the counts.

A similar method can be applied to eight-level, 3 bit/cell MLC. Assume, for example, an eight-level MLC device whose eight levels are denoted L1 . . . L8 and are mapped to the bit triplets ‘111’, ‘011’, ‘001’, ‘101’, ‘100’, ‘000’, ‘010’, ‘110’, respectively. The MSP can compute the soft metric of the MSB (leftmost bit in the triplet) of such a cell independently of the other bits by performing comparisons using four sets of multiple thresholds. Each threshold set is positioned between adjacent levels having different MSB values. In the present example, one set is positioned between levels L1 and L2, another set between L3 and L4, a third set between L5 and L6 and a fourth set between L7 and L8. The four threshold sets divide the analog value axis into five intervals denoted I1 . . . I5, such that the MSB has the same value within each interval.

Using this division, the MSP determines that the MSB is “0” if the read analog value read from the cell falls within interval I2 or I4, and “1” if the analog value falls within interval I1, I3 or I5. In order to compute the soft metric of the MSB, the MSP forms groups of four thresholds, with each group containing one threshold from each set. Moving from group to group, each threshold is moved in the direction in which the MSB value transition is from “1” to “0”. For each threshold group, the MSP performs four read operations and checks whether the read value falls in intervals corresponding to “1” or in intervals corresponding to “0”. The MSP counts the number of threshold groups in which the analog value falls in intervals that correspond to MSB=“0” and/or the number of groups in which the analog value falls in intervals that correspond to MSB=“1”. The MSP computes a soft metric based on the counts.

Typically but not necessarily, soft metrics that are based on counting comparison results of a given type assume that the read thresholds are positioned symmetrically around the midpoint between distributions.

Gradually Increasing the Number of Thresholds

The comparison and metric computation operations described above consume both time and computation resources, which grow with the number of thresholds. Therefore, it is sometimes advantageous to use only as many thresholds as needed to successfully reconstruct the data. In some embodiments, the MSP initially attempts to compute the soft metrics and decode the data with a relatively small number of thresholds, and increase their number only when needed.

For example, the MSP may make an initial attempt to decode the ECC using an initial set of read thresholds in which only a single threshold is positioned between each pair of adjacent nominal values (memory states). In these embodiments, the MSP reverts to multiple-threshold decoding upon failure of the initial decoding attempt.

FIG. 8 is a flow chart that schematically illustrates an exemplary method for reading data from analog memory cells by gradually increasing the number of thresholds, in accordance with another embodiment of the present invention. Initially, it is assumed that the MSP attempted to decode a particular codeword stored in a group of memory cells using soft metrics that were obtained using a certain number of thresholds, and that ECC decoding has failed. The metrics are assumed to be based on the count of comparison results, as explained above.

The method begins with the MSP adding one or more additional thresholds to the set of thresholds used, and reading the group of memory cells using the added thresholds, at a threshold addition step 142.

The MSP updates the count of comparison results (i.e., the number of “0” and/or “1” results out of the total), at a count updating step 144. The updated count reflects the comparison results of the previous thresholds as well as of the newly-added thresholds. The MSP then computes the soft metrics based on the updated accumulated count of comparison results, and a metric updating step 146.

In some cases, the MSP may compute the metrics from scratch at each iteration. Alternatively, the MSP may store the metric values and/or comparison result counts from previous iterations, and update them to account for the newly-added comparison results. Generally, the soft metric computed at a given iteration may depend on the current comparison result count, on previous counts and on previous metric values.

The MSP computes soft metrics based on the accumulated count of comparison results, at a metric computation step 146. Any suitable metric computation method can be used, such as the exemplary methods described above. The MSP attempts to decode the codeword using the soft metrics, at an ECC decoding step 148. The MSP checks whether the ECC decoding was successful, at an ECC checking step 150. If successful, the MSP extracts and outputs the data, at a data extraction step 152, and the method terminates.

If, on the other hand, ECC decoding fails, the method loops back to threshold addition step 142 above. The MSP adds one or more additional thresholds to the set of thresholds, computes soft metrics based on the extended set, and attempts to decode the ECC again.

The method of FIG. 8 enables the MSP to use only as many thresholds as needed to successfully decode the ECC. When distortion is not severe, most codewords can be decoded using a small number of thresholds, enabling a high overall or average reading speed.

Alternatively to continuing the iterations until successful decoding of the ECC, the MSP may evaluate any other suitable condition, and stop the iterative process when the condition is met. For example, the MSP may continue to add thresholds until reaching a maximum number of thresholds, or a maximum number of iterations.

In some embodiments, the ECC decoding process may comprise an iterative process. Iterative decoding processes are commonly used to decode codes such as LDPC and turbo codes. In these embodiments, the iterative decoding process is provided with increasingly-improving metrics, which are based on an increasing number of read thresholds. In other words, the iterative decoding process starts decoding using metrics, which are based on a certain initial number of thresholds. Subsequent iterations of the iterative decoding process are provided with metrics that are based on an increasing number of read thresholds, until the iterative decoding process converges to a valid codeword.

Additionally or alternatively to using an ECC in the method of FIG. 8, the MSP may use an error detection code, such as a Cyclic Redundancy Check (CRC) or checksum. In such embodiments, the MSP iteratively adds read thresholds until the error detection code detects no errors. Thus, in the context of the present patent application and in the claims, the term “ECC” is used to address various types of error detection codes, as well. In some embodiments, the MSP may use an error detection code to determine when to stop adding new thresholds, even though the data is encoded using an ECC. This scheme may be advantageous, for example, when the ECC does not provide a reliable indication of decoding success or failure.

The MSP may use various methods and criteria for selecting how many thresholds to add at each iteration, and in which order. For example, thresholds can be added two at a time, gradually moving away from the initial threshold position in both directions. In other words, assuming the MSP initially attempts to use a threshold denoted T and that the thresholds are spaced at regular intervals of Δ, the threshold sets in the first four iterations are:

{T, T+Δ, T−Δ}

{T, T+Δ, T−Δ, T+2Δ, T−2Δ}

{T, T+Δ, T−Δ, T+2Δ, T−2Δ, T+3Δ, T−3Δ}

{T, T+A, T−Δ, T+2Δ, T−2Δ, T+3Δ, T−3Δ, T+4Δ, T−4Δ}

Metric Normalization and Interference Cancellation

When computing soft metrics that depend on varying numbers of thresholds, such as in the method of FIG. 8 above, the possible range of metric values may vary with the number of thresholds used. For example, when the soft metric comprises the count of comparison results, the metric values based on three thresholds will be in the range [0 . . . 3], whereas metric values based on five thresholds will be in the range [0 . . . 5]. This effect is generally undesirable. In other words, it is generally desired to provide the ECC decoder with metrics, which use the same dynamic range for quantifying confidence or certainty, regardless of the number of threshold comparisons on which the metric is based.

In some embodiments, the MSP normalizes the soft values read from the memory cells based on the number of thresholds. For example, the MSP may apply bit extension to the values to reach a certain constant number of bits, e.g., five-bits. For example, the bit-extended value may be given by

ExtendedValue = { 0 Val = 0 MaxVal Val = N Val + MaxVal - N 2 0 < Val < N } [ 3 ]
wherein Val denotes the input soft value and N denotes the number of thresholds used to evaluate Val. Max Val denotes the maximum value of the bit-extended soft value, e.g., 31 for five-bit representation. Alternatively, the MSP may apply any other suitable data scaling mechanism.

In some embodiments, the MSP has information regarding the level of distortion or interference in the memory cells being read. Various methods can be used to estimate interference levels in memory cells. Exemplary methods are described in PCT Patent Application PCT/IL2007/000580, cited above and in PCT Patent Application PCT/IL2007/000576, entitled “Distortion Estimation and Cancellation in Memory Devices,” filed May 10, 2007, and PCT Patent Application PCT/IL2007/001059, entitled “Estimation of Non-Linear Distortion in Memory Devices,” filed Aug. 27, 2007, whose disclosures are incorporated herein by reference.

When an estimate of the interference is available to the MSP, the MSP may add the effect of the interference to the soft values, or otherwise modify the soft values based on the estimated interference, before these values are provided to the ECC decoder.

FIG. 9 is a diagram that schematically illustrates a process for reading data from analog memory cells, which involves data scaling and interference cancellation, in accordance with yet another embodiment of the present invention. Although the configuration of FIG. 9 is used to demonstrate both interference cancellation and scaling, each mechanism may be carried out with or without the other.

In the process of FIG. 9, a scaling module 154 accepts the conditionally-inverted soft values read from the memory cells (e.g., the outputs of XOR circuit 134 of FIG. 7 above). Module 154 also accepts an indication of the iteration number and/or the number of thresholds that are currently used. Module 154 applies bit extension or other scaling to the input soft values. The amount of scaling depends on the input iteration number.

The scaled soft values are provided to an interference cancellation module 156, which also accepts estimates of the interference level in the respective memory cells. Module 156 subtracts or otherwise cancels out the interference estimates from the corresponding soft values, to produce soft values that are properly scaled and contain reduced levels of interference. The soft values are provided to a metric computation module 158, which computes the soft metrics and provides them to the ECC decoder.

Trading-Off Threshold Comparisons and Interference Estimation

Re-reading cells with additional thresholds and estimating the interference from neighboring memory cells are two operations that on one hand improve the reading performance, and on the other hand consume time and computational power. In some embodiments, the MSP may combine the two operations and trade-off one operation for another. For example, the MSP may determine at each iteration whether it is preferable to refine the decoding accuracy by re-reading the current page using an additional threshold, or to refine the interference estimation by reading (or re-reading) a group of interfering cells.

FIG. 10 is a flow chart that schematically illustrates a method for reading data from analog memory cells that involves trading-off re-reading and interference estimation, in accordance with another embodiment of the present invention.

The method begins with the MSP reading a page of memory cells, at a reading step 160. At each cycle of the process, the MSP may select to either (1) re-read the desired page using an additional threshold, or (2) read a page of interfering cells. The MSP may apply various policies or heuristics in determining which of the two actions to take at each cycle. The MSP may read different groups of interfering cells at different cycles.

For example, the MSP may alternate between the two operations, thus adding a threshold every two cycles and estimating interference every two cycles. Alternatively, the MSP may choose which action to take based on the estimated level of the distortion. For example, if recent interference estimations indicate that the level of interference is low, the MSP may give precedence to adding threshold comparisons, and refine the interference estimation at larger intervals. Further alternatively, the decision may depend on the type of page being read. For example, even- and odd-order pages may experience different interference levels, and the MSP may apply different decision logic for different page types. Pages located on the last word line in a block may also experience different interference levels and may be treated differently. Since the interference may depend on the order in which the pages were written, different trade-offs may apply to higher- and lower-number pages within a word line.

In some cases, memory cells within the desired page may cause interference to one another. Thus, the group of interfered cells and the group of interfering cells may sometimes overlap.

Based on the updated information, the MSP subtracts the interference estimation from the read soft values, at an interference cancellation step 162, and computes the soft metrics, at a metric calculation step 164. The MSP then decodes the ECC, at a decoding step 166, and checks whether ECC decoding was successful, at a success checking step 168.

If the ECC was decoded successfully, the method terminates, at a success termination step 170, and the MSP typically extracts and outputs the data. Otherwise, the MSP checks whether the number of iterations (cycles) exceeds a predetermined maximum number, at an iteration number checking step 172. If the maximum number of iterations was exceeded, the method terminates without successfully reading the data, at an error termination step 174. Otherwise, the method loops back to reading step 160 above, and the MSP again determines whether to add another threshold or refine the interference estimation in the next cycle.

In both iterative methods of FIGS. 8 and 10 above, the MSP may select the number of new thresholds that are added at a particular iteration based on the values read from the cells or the data represented by these values. For example, when the MSP detects severe ECC failure or an exceptionally high level of interference, it may decide to add a high number of thresholds. The MSP may also determine the values (i.e., positions) of the new thresholds based on the read values or read data. For example, the values of new thresholds that are added in response to ECC failure may be different from the values of thresholds added in response to high interference.

Performing Multiple Read Operations Internally to the Memory Device

When using the methods and systems described above, the multiple comparison results associated with the multiple thresholds are typically communicated from memory device 24 to MSP 52. The resulting communication bandwidth between the memory device and the MSP may become prohibitive, especially when using a large number of threshold sets and/or when the number of nominal levels per memory cell is high. In some practical cases, the communication bandwidth over the interface between the MSP and the memory device may become the limiting factor that determines the memory access speed of system 20. This effect becomes even more severe when a single MSP 52 is connected to multiple memory devices 24.

In alternative embodiments of the present invention, some of the re-reading functions are carried out internally to the memory device, so as to reduce the communication bandwidth between the memory device and the MSP.

FIG. 11 is a block diagram that schematically illustrates a system 200 for memory signal processing, in accordance with an alternative embodiment of the present invention. In the exemplary embodiment of FIG. 11, multiple memory devices 204 are connected to an MSP 208 over an external bus 212. Each memory device 204 comprises a memory cell array 216 and a R/W unit 220, which are similar to array 28 and R/W unit 40 of FIG. 1 above, respectively.

Unlike the embodiment shown in FIG. 1 above, each memory device 204 comprises a threshold setting and metric calculation unit 224, also referred to as “metric calculation unit” for brevity. Unit 224 is connected to R/W unit 220 by an internal bus 228. When the memory device accepts a request to retrieve data from a group of memory cells (e.g., a page), unit 224 controls R/W unit 220 to set the appropriate threshold values and read the memory cells using the thresholds, such as using any of the methods described above. The R/W unit carries out the multiple comparison operations and sends the corresponding comparison results to unit 224. Unit 224 computes the soft metrics based on the comparison results, and sends the metric values over external bus 212 to MSP 208.

MSP 208 comprises an ECC decoder 232. The ECC decoder accepts the soft metrics sent from unit 224 of memory device 204 and decodes the ECC based on the metrics. The MSP typically outputs the decoded data to the host system. When MSP 208 controls multiple memory devices 204, a single ECC decoder may decodes the data sent from all the memory devices. Alternatively, multiple ECC decoders may be used.

When using the configuration of FIG. 11, the communication bandwidth between the MSP and the memory device is significantly reduced in comparison with the configuration of FIG. 1 above, since individual comparison results are not communicated to the MSP. Instead, unit 224 sends the soft metric values, typically comprising a single value for each read memory cell. The large communication bandwidth needed to communicate the multiple comparison results is confined to internal bus 228, i.e., internally to the memory device. A high bandwidth bus of this sort is considerably simpler to implement internally to the memory device than between separate devices. Moreover, the traffic over the internal bus comprises only the traffic generated by the particular memory device, regardless of the number of memory devices controlled by the MSP.

The functional partitioning between R/W unit 220 and metric calculation unit 224 is an exemplary partitioning, which is chosen purely for the sake of conceptual clarity. In alternative embodiments, the reading, threshold comparison, threshold setting and metric computation functions can be partitioned in any other way, as desired. Thus, R/W unit 220, internal bus 228 and metric calculation unit 224 are collectively regarded as a reading circuit, which reads the analog memory cell and produces soft metrics.

As noted above, the soft metric computation sometimes takes into account estimation and cancellation of the interference in the read memory cells. In some embodiments, the interference estimation and cancellation functionality can also be carried out by the reading circuit internally to memory device 204, e.g., by unit 224. In these embodiments, unit 224 sends to the MSP soft metrics, in which the interference is already taken into account. Some aspects of carrying out signal processing functions internally to the memory device are described in U.S. Provisional Patent Application 60/917,653, cited above.

Although the embodiments described herein mainly address retrieving data from solid-state memory devices, the principles of the present invention can also be used for storing and retrieving data in Hard Disk Drives (HDD) and other data storage media and devices.

It will thus be appreciated that the embodiments described above are cited by way of example, and that the present invention is not limited to what has been particularly shown and described hereinabove. Rather, the scope of the present invention includes both combinations and sub-combinations of the various features described hereinabove, as well as variations and modifications thereof which would occur to persons skilled in the art upon reading the foregoing description and which are not disclosed in the prior art.

Patentzitate
Zitiertes PatentEingetragen Veröffentlichungsdatum Antragsteller Titel
US366863113. Febr. 19696. Juni 1972IbmError detection and correction system with statistically optimized data recovery
US366863213. Febr. 19696. Juni 1972IbmFast decode character error detection and correction system
US405885118. Okt. 197615. Nov. 1977Sperry Rand CorporationConditional bypass of error correction for dual memory access time selection
US411250218. Juli 19775. Sept. 1978Sperry Rand CorporationConditional bypass of error correction for dual memory access time selection
US439476329. Aug. 198019. Juli 1983Fujitsu LimitedError-correcting system
US441333924. Juni 19811. Nov. 1983Digital Equipment CorporationMultiple error detecting and correcting system employing Reed-Solomon codes
US455696119. Mai 19823. Dez. 1985Tokyo Shibaura Denki Kabushiki KaishaSemiconductor memory with delay means to reduce peak currents
US455843112. Dez. 198310. Dez. 1985Nec CorporationMemory system for storing analog information
US460868713. Sept. 198326. Aug. 1986International Business Machines CorporationBit steering apparatus and method for correcting errors in stored data, storing the address of the corrected data and using the address to maintain a correct data condition
US465484728. Dez. 198431. März 1987International Business MachinesApparatus for automatically correcting erroneous data and for storing the corrected data in a common pool alternate memory array
US466192924. Dez. 198428. Apr. 1987Hitachi, Ltd.Semiconductor memory having multiple level storage structure
US476817120. Juni 198530. Aug. 1988Nec CorporationMemory circuit having a plurality of cell arrays
US481128511. Jan. 19887. März 1989The Board Of Trustees Of The Leland Stanford Junior UniversityAnalog storage integrated circuit
US48993421. Febr. 19886. Febr. 1990Thinking Machines CorporationMethod and apparatus for operating multi-unit array of memories
US49107064. Aug. 198320. März 1990Hyatt Gilbert PAnalog memory for storing digital information
US499302913. März 198912. Febr. 1991International Business Machines CorporationMethod and apparatus for randomizing data in a direct access storage device
US50560896. Febr. 19898. Okt. 1991Mitsubishi Denki Kabushiki KaishaMemory device
US50777226. Jan. 198931. Dez. 1991Convex Computer CorporationDisk drive insertion and removal interlock
US512680823. Okt. 198930. Juni 1992Advanced Micro Devices, Inc.Flash EEPROM array with paged erase architecture
US516302122. Juli 199110. Nov. 1992Sundisk CorporationMulti-state EEprom read and write circuits and techniques
US517233811. Apr. 199015. Dez. 1992Sundisk CorporationMulti-state EEprom read and write circuits and techniques
US518255825. Okt. 199126. Jan. 1993Halliburton Geophysical Services, Inc.System for generating correction signals for use in forming low distortion analog signals
US518275229. Juni 199026. Jan. 1993Digital Equipment CorporationMethod and apparatus for transferring data between a data bus and a data storage device
US519158420. Febr. 19912. März 1993Micropolis CorporationMass storage array with efficient parity calculation
US520095917. Okt. 19896. Apr. 1993Sundisk CorporationDevice and method for defect handling in semi-conductor memory
US52375359. Okt. 199117. Aug. 1993Intel CorporationMethod of repairing overerased cells in a flash memory
US527266920. Febr. 199121. Dez. 1993Sundisk CorporationMethod and structure for programming floating gate memory cells
US527664912. Aug. 19914. Jan. 1994Mitsubishi Denki Kabushiki KaishaDynamic-type semiconductor memory device having staggered activation of column groups
US528746927. Dez. 198915. Febr. 1994Nec CorporationElectrically erasable and programmable non-volatile memory (EEPROM), wherein write pulses can be interrupted by subsequently received read requests
US536548423. Aug. 199315. Nov. 1994Advanced Micro Devices, Inc.Independent array grounds for flash EEPROM array with paged erase architechture
US538806426. Nov. 19917. Febr. 1995Information Storage Devices, Inc.Programmable non-volatile analog voltage source devices and methods
US541664612. Nov. 199316. Mai 1995Kabushiki Kaisha ToshibaData recording and reproducing apparatus having read/write circuit with programmable parameter means
US541678230. Okt. 199216. Mai 1995Intel CorporationMethod and apparatus for improving data failure rate testing for memory arrays
US544685420. Okt. 199329. Aug. 1995Sun Microsystems, Inc.Virtual memory computer apparatus and address translation mechanism employing hashing scheme and page frame descriptor that support multiple page sizes
US545042424. Mai 199312. Sept. 1995Mitsubishi Denki Kabushiki KaishaSemiconductor memory device with error checking and correcting function
US546944416. Nov. 199421. Nov. 1995Kabushiki Kaisha ToshibaElectrically erasable and programmable non-volatile memory system with write-verify controller using two reference levels
US547375330. Okt. 19925. Dez. 1995Intel CorporationMethod of managing defects in flash disk memories
US54791702. Nov. 199326. Dez. 1995California Institute Of TechnologyMethod and apparatus for long-term multi-valued storage in dynamic analog memory
US550895829. Sept. 199416. Apr. 1996Intel CorporationMethod and apparatus for sensing the state of floating gate memory cells by applying a variable gate voltage
US55198311. Juni 199421. Mai 1996Intel CorporationNon-volatile disk cache
US553296221. März 19952. Juli 1996Sandisk CorporationSoft errors handling in EEPROM devices
US554188627. Dez. 199430. Juli 1996Intel CorporationMethod and apparatus for storing control information in multi-bit non-volatile memory arrays
US560067716. Jan. 19964. Febr. 1997Zenith Electronics CorporationTrellis coded modulation system for digital television signal
US563832011. Jan. 199610. Juni 1997Invoice Technology, Inc.High resolution analog storage EPROM and flash EPROM
US5657332 *20. März 199512. Aug. 1997Sandisk CorporationSoft errors handling in EEPROM devices
US567554022. Jan. 19967. Okt. 1997Micron Quantum Devices, Inc.Non-volatile memory system having internal data verification test mode
US56823528. Febr. 199628. Okt. 1997Invoice Technology, Inc.Digital testing of analog memory devices
US56871146. Okt. 199511. Nov. 1997Agate Semiconductor, Inc.Integrated circuit for storage and retrieval of multiple digital bits per nonvolatile memory cell
US569671715. Okt. 19969. Dez. 1997Samsung Electronics Co., Ltd.Nonvolatile integrated circuit memory devices having adjustable erase/program threshold voltage verification capability
US572664929. Aug. 199510. März 1998Yamaha CorporationControl device suitable for use in an apparatus for reproducing video, audio and accompanying characters
US57269349. Apr. 199610. März 1998Information Storage Devices, Inc.Method and apparatus for analog reading values stored in floating gate structures
US574275229. Dez. 199521. Apr. 1998Symbios Logic Inc.Method for performing a RAID stripe write operation using a drive XOR command set
US574853326. März 19965. Mai 1998Invoice Technology, Inc.Read circuit which uses a coarse-to-fine search when reading the threshold voltage of a memory cell
US574853426. März 19965. Mai 1998Invox TechnologyFeedback loop for reading threshold voltage
US575163712. März 199612. Mai 1998Macronix International Co., Ltd.Automatic programming algorithm for page mode flash memory with variable programming pulse height and pulse width
US576140227. Sept. 19942. Juni 1998Hitachi, Ltd.Array type disk system updating redundant data asynchronously with data access
US579896631. März 199725. Aug. 1998Intel CorporationFlash memory VDS compensation techiques to reduce programming variability
US579920028. Sept. 199525. Aug. 1998Emc CorporationPower failure responsive apparatus and method having a shadow dram, a flash ROM, an auxiliary battery, and a controller
US580198528. Juli 19951. Sept. 1998Micron Technology, Inc.Memory system having programmable control parameters
US58388323. Apr. 199517. Nov. 1998Iterated Systems, Inc.Method and system for representing a data set with a data transforming function and data mask
US586010613. Juli 199512. Jan. 1999Intel CorporationMethod and apparatus for dynamically adjusting power/performance characteristics of a memory subsystem
US586711429. Febr. 19962. Febr. 1999Mitel CorporationMethod and apparatus for performing data compression
US58674288. Juli 19972. Febr. 1999Hitachi, Ltd.Nonvolatile memory system semiconductor memory and writing method
US586742919. Nov. 19972. Febr. 1999Sandisk CorporationHigh density non-volatile flash memory without adverse effects of electric field coupling between adjacent floating gates
US587798629. Dez. 19972. März 1999Sandisk CorporationMulti-state Flash EEprom system on a card that includes defective cell substitution
US588993726. Juni 199730. März 1999Nec CorporationHard disk apparatus capable of transforming logical addresses of apparatus diagnosis cylinders to HDD-by-HDD physical addresses
US59010892. Apr. 19984. Mai 1999Agate Semiconductor, Inc.Stabilization circuits and techniques for storage and retrieval of single or multiple digital bits per memory cell
US59094498. Sept. 19971. Juni 1999Invox TechnologyMultibit-per-cell non-volatile memory with error detection and correction
US591290623. Juni 199715. Juni 1999Sun Microsystems, Inc.Method and apparatus for recovering from correctable ECC errors
US593016730. Juli 199727. Juli 1999Sandisk CorporationMulti-state non-volatile flash memory capable of being its own two state write cache
US593742427. Febr. 199710. Aug. 1999Intel CorporationMethod and apparatus for suspending the writing of a nonvolatile semiconductor memory with program suspend command
US594200431. Okt. 199524. Aug. 1999Stmicroelectronics, S.R.L.Device and a method for storing data and corresponding error-correction information
US594671630. Mai 199631. Aug. 1999Hewlett-Packard CompanySectored virtual memory management system and translation look-aside buffer (TLB) for the same
US596998623. Juni 199819. Okt. 1999Invox TechnologyHigh-bandwidth read and write architectures for non-volatile memories
US598266818. Aug. 19989. Nov. 1999Hitachi, Ltd.Nonvolatile memory system, semiconductor memory and writing method
US599151720. Dez. 199623. Nov. 1999Sandisk CorporationFlash EEprom system with cell by cell programming verification
US599541720. Okt. 199830. Nov. 1999Advanced Micro Devices, Inc.Scheme for page erase and erase verify in a non-volatile memory array
US60090143. Juni 199828. Dez. 1999Advanced Micro Devices, Inc.Erase verify scheme for NAND flash
US60090162. Juni 199928. Dez. 1999Hitachi, Ltd.Nonvolatile memory system semiconductor memory and writing method
US602342529. Juni 19998. Febr. 2000Hitachi, Ltd.Nonvolatile memory system, semiconductor memory, and writing method
US60348911. Dez. 19977. März 2000Micron Technology, Inc.Multi-state flash memory defect management
US604099323. Febr. 199821. März 2000Macronix International Co., Ltd.Method for programming an analog/multi-level flash EEPROM
US60414303. Nov. 199721. März 2000Sun Microsystems, Inc.Error detection and correction code for data and check code fields
US607320423. Apr. 19976. Juni 2000Micron Technology, Inc.Memory system having flexible architecture and method
US610161421. Okt. 19978. Aug. 2000Intel CorporationMethod and apparatus for automatically scrubbing ECC errors in memory via hardware
US61282376. Dez. 19993. Okt. 2000Micron Technology, Inc.Method and apparatus for enhancing the performance of semiconductor memory devices
US613414014. Mai 199817. Okt. 2000Kabushiki Kaisha ToshibaNonvolatile semiconductor memory device with soft-programming to adjust erased state of memory cells
US613414319. Nov. 199917. Okt. 2000Micron Technology, Inc.Multi-state flash memory defect management
US613463131. Okt. 199617. Okt. 2000Hyundai Electronics America, Inc.Non-volatile memory with embedded programmable controller
US614126131. Dez. 199931. Okt. 2000Patti; RobertDRAM that stores multiple bits per storage cell
US615124625. Nov. 199821. Nov. 2000Sandisk CorporationMulti-bit-per-cell flash EEPROM memory with refresh
US615757321. Dez. 19995. Dez. 2000Hitachi, Ltd.Nonvolatile memory system, semiconductor memory, and writing method
US616696224. Juni 199926. Dez. 2000Amic Technology, Inc.Circuit and method for conditioning flash memory array
US616969115. Sept. 19992. Jan. 2001Stmicroelectronics S.R.L.Method for maintaining the memory content of non-volatile memory cells
US617846612. Juni 199823. Jan. 2001Unisys CorporationSystem for maximizing bandpass on an interface directly coupling two units where the interface has independently operative data and address interconnections, and computer sysem employing same.
US61851348. Dez. 19996. Febr. 2001Kabushiki Kaisha ToshibaFlash memory control method, flash memory system using the control method and flash memory device using the control method
US62091131. Sept. 199827. März 2001Micron Technology, Inc.Method and apparatus for performing error correction on data read from a multistate memory
US621265422. Juli 19973. Apr. 2001Lucent Technologies Inc.Coded modulation for digital storage in analog memory devices
US621927625. Febr. 200017. Apr. 2001Advanced Micro Devices, Inc.Multilevel cell programming
US621944720. Febr. 199817. Apr. 2001Samsung Electronics Co., Ltd.Method and circuit for extracting histogram and cumulative distribution function for image enhancement apparatus
US62227627. Aug. 199724. Apr. 2001Sandisk CorporationMulti-state memory
US623023313. Sept. 19918. Mai 2001Sandisk CorporationWear leveling techniques for flash EEPROM systems
US624045822. Dez. 199829. Mai 2001Unisys CorporationSystem and method for programmably controlling data transfer request rates between data sources and destinations in a data processing system
US625962727. Jan. 200010. Juli 2001Multi Level Memory TechnologyRead and write operations using constant row line voltage and variable column line load
US627541913. Okt. 200014. Aug. 2001Sandisk CorporationMulti-state memory
US627863212. Dez. 199721. Aug. 2001Micron Technology, Inc.Method and circuitry for performing analog over-program and under-program detection for a multistate memory cell
US627906926. Dez. 199621. Aug. 2001Intel CorporationInterface for flash EEPROM memory arrays
US628894423. Mai 200011. Sept. 2001Fujitsu LimitedNAND type nonvolatile memory with improved erase-verify operations
US629239429. Juni 200018. Sept. 2001Saifun Semiconductors Ltd.Method for programming of a semiconductor memory cell
US63011519. Aug. 20009. Okt. 2001Information Storage Devices, Inc.Adaptive programming method and apparatus for flash memory analog storage
US630448620. Dez. 199916. Okt. 2001Fujitsu LimitedSensing time control device and method
US63077766. Okt. 200023. Okt. 2001Sandisk CorporationMulti-bit-per-cell flash EEPROM memory with refresh
US631736313. Okt. 200013. Nov. 2001Sandisk CorporationMulti-state memory
US631736413. Okt. 200013. Nov. 2001Sandisk CorporationMulti-state memory
US634500426. Dez. 20005. Febr. 2002Mitsubishi Denki Kabushiki KaishaRepair analysis circuit for redundancy, redundant repairing method, and semiconductor device
US636034626. Aug. 199819. März 2002Sony CorporationStorage unit, method of checking storage unit, reading and writing method
US636300817. Febr. 200026. März 2002Multi Level Memory TechnologyMulti-bit-cell non-volatile memory with maximized data capacity
US636345418. Apr. 200026. März 2002Micron Technology, Inc.Memory system having flexible architecture and method
US63664962. Aug. 20002. Apr. 2002Stmicroelectronics S.R.L.Method for programming multi-level non-volatile memories by controlling the gate voltage
US638509225. Sept. 20017. Mai 2002Hitachi, Ltd.Nonvolatile memory system, semiconductor memory, and writing method
US639293217. Nov. 200021. Mai 2002Tatsuya IshiiNonvolatile memory system, semiconductor memory, and writing method
US639674228. Juli 200028. Mai 2002Silicon Storage Technology, Inc.Testing of multilevel semiconductor memory
US639736410. Nov. 199828. Mai 2002Mordecai BarkanDigital data representation for multi-bit data storage and transmission
US640532330. März 199911. Juni 2002Silicon Storage Technology, Inc.Defect management for interface to electrically-erasable programmable read-only memory
US640534210. Sept. 199911. Juni 2002Western Digital Technologies, Inc.Disk drive employing a multiple-input sequence detector responsive to reliability metrics to improve a retry operation
US64180603. Jan. 20029. Juli 2002Ememory Technology Inc.Method of programming and erasing non-volatile memory cells
US644258526. Nov. 199727. Aug. 2002Compaq Computer CorporationMethod for scheduling contexts based on statistics of memory system interactions in a computer system
US644560226. Okt. 19993. Sept. 2002Nucore Technology Inc.Structure and method for correction of defective analog data in a nonvolatile semiconductor memory
US645283827. Nov. 200117. Sept. 2002Hitachi, Ltd.Nonvolatile memory system, semiconductor memory and writing method
US645652817. Sept. 200124. Sept. 2002Sandisk CorporationSelective operation of a multi-state non-volatile memory system in a binary mode
US646647618. Jan. 200115. Okt. 2002Multi Level Memory TechnologyData coding for multi-bit-per-cell memories having variable numbers of bits per memory cell
US646706210. Nov. 199815. Okt. 2002Mordecai BarkanDigital data (multi-bit) storage with discrete analog memory cells
US64699312. Juli 200122. Okt. 2002M-Systems Flash Disk Pioneers Ltd.Method for increasing information content in a computer memory
US64902361. Sept. 20003. Dez. 2002Hitachi, Ltd.Method and apparatus for optical disk device
US652258027. Juni 200118. Febr. 2003Sandisk CorporationOperating techniques for reducing effects of coupling between storage elements of a non-volatile memory operated in multiple data states
US65259528. Apr. 200225. Febr. 2003Sony CorporationRecording system, data recording apparatus, memory apparatus, and data recording method
US653255627. Jan. 200011. März 2003Multi Level Memory TechnologyData management for multi-bit-per-cell memories
US653892227. Sept. 200025. März 2003Sandisk CorporationWritable tracking cells
US65494642. Juli 200215. Apr. 2003Kabushiki Kaisha ToshibaNonvolatile semiconductor memory device
US65535102. Sept. 199922. Apr. 2003Micron Technology, Inc.Memory device including redundancy routine for correcting random errors
US655896731. Jan. 20016. Mai 2003Multi Level Memory TechnologyMulti-bit-per-cell memory system with numbers of bits per cell set by testing of memory units
US65601522. Nov. 20016. Mai 2003Sandisk CorporationNon-volatile memory with temperature-compensated data read
US656731113. März 200220. Mai 2003Hitachi, Ltd.Nonvolatile memory system, semiconductor memory, and writing method
US657753911. Okt. 200110. Juni 2003Kabushiki Kaisha ToshibaNon-volatile semiconductor memory device with programming voltage generating system and data programming method
US65840124. Juni 200224. Juni 2003Btg International Inc.Electrically alterable non-volatile memory with N-bits per cell
US661530710. Mai 20002. Sept. 2003Micron Technology, Inc.Flash with consistent latency for read operations
US662173918. Jan. 200216. Sept. 2003Sandisk CorporationReducing the effects of noise in non-volatile memories through multiple reads
US664032630. Mai 200028. Okt. 2003Hewlett-Packard Development Company, L.C.Data set recovery by codeword overlay
US664316918. Sept. 20014. Nov. 2003Intel CorporationVariable level memory
US664691311. Okt. 200111. Nov. 2003Stmicroelectronics S.R.L.Method for storing and reading data in a multilevel nonvolatile memory
US66781922. Nov. 200113. Jan. 2004Sandisk CorporationError management for writable tracking storage units
US668381117. März 200327. Jan. 2004Hitachi, Ltd.Nonvolatile memory system, semiconductor memory, and writing method
US66871559. Mai 20033. Febr. 2004Oki Electric Industry Co., Ltd.Analog storage semiconductor memory that uses plural write voltages and plural read voltages having different voltage levels
US670774831. Dez. 200216. März 2004Ritek CorporationBack up power embodied non-volatile memory device
US670825712. Juli 199916. März 2004Koninklijke Philips Electronics N.V.Buffering system bus for external-memory access
US671444924. Jan. 200330. März 2004Sandisk CorporationSense amplifier suitable for analogue voltage levels
US671784726. Aug. 20026. Apr. 2004Sandisk CorporationSelective operation of a multi-state non-volatile memory system in a binary mode
US673155720. Juni 20024. Mai 2004Stmicroelectronics S.R.L.Method of refreshing an electrically erasable and programmable non-volatile memory
US673829315. Apr. 200318. Mai 2004Kabushiki Kaisha ToshibaNon-volatile semiconductor memory device and data programming method
US675176620. Mai 200215. Juni 2004Sandisk CorporationIncreasing the effectiveness of error correction codes and operating multi-level memory systems by using information about the quality of the stored data
US67571933. Apr. 200229. Juni 2004Macronix International Co., Ltd.Coding method of multi-level memory cell
US677480817. Mai 200010. Aug. 2004Dell Products L.P.Electromechanical lock for components
US67818776. Sept. 200224. Aug. 2004Sandisk CorporationTechniques for reducing effects of coupling between storage elements of adjacent rows of memory cells
US680480526. Juni 200112. Okt. 2004Seagate Technology LlcMethod and apparatus for encoding with unequal protection in magnetic recording channels having concatenated error correction codes
US680709518. Dez. 200219. Okt. 2004Sandisk CorporationMulti-state nonvolatile memory capable of reducing effects of coupling between storage elements
US680710123. Dez. 200219. Okt. 2004Renesas Technology Corp.Semiconductor memory device
US680996430. Aug. 200126. Okt. 2004Micron Technology, Inc.Nonvolatile semiconductor memory device capable of transferring data internally without using an external bus
US681959229. März 200216. Nov. 2004Kabushiki Kaisha ToshibaSemiconductor memory
US682916712. Dez. 20027. Dez. 2004Sandisk CorporationError recovery for nonvolatile memory
US684505230. Mai 200318. Jan. 2005Macronix International Co., Ltd.Dual reference cell sensing scheme for non-volatile memory
US685101827. März 20021. Febr. 2005Hewlett-Packard Development Company, L.P.Exchanging operation parameters between a data storage device and a controller
US685108124. Juli 20011. Febr. 2005Nec Electronics CorporationSemiconductor memory device having ECC type error recovery circuit
US685654613. Nov. 200115. Febr. 2005Sandisk CorporationMulti-state memory
US686221825. Febr. 20041. März 2005Sandisk CorporationMulti-state memory
US687076718. Sept. 200322. März 2005Intel CorporationVariable level memory
US687077310. März 200422. März 2005Kabushiki Kaisha ToshibaData writing method for semiconductor memory device and semiconductor memory device
US68735529. Okt. 200329. März 2005Renesas Technology Corp.Nonvolatile memory system, semiconductor memory, and writing method
US687952020. Juni 200312. Apr. 2005Kabushiki Kaisha ToshibaNon-volatile semiconductor memory device and electric device with the same
US68825676. Dez. 200219. Apr. 2005Multi Level Memory TechnologyParallel programming of multiple-bit-per-cell memory cells on a continuous word line
US689492630. Apr. 200317. Mai 2005Sandisk CorporationMulti-state memory
US690749720. Dez. 200214. Juni 2005Kabushiki Kaisha ToshibaNon-volatile semiconductor memory device
US692500918. Aug. 20042. Aug. 2005Kabushiki Kaisha ToshibaSemiconductor memory
US693092514. Okt. 200316. Aug. 2005Atmel CorporationSuspend-resume programming method for flash memory
US693418818. Juni 200423. Aug. 2005Micron Technology, Inc.Non-volatile memory erase circuitry
US693751127. Jan. 200430. Aug. 2005Macronix International Co., Ltd.Circuit and method for programming charge storage memory cells
US69589389. Dez. 200425. Okt. 2005Kabushiki Kaisha ToshibaData writing method for semiconductor memory device and semiconductor memory device
US696350529. Okt. 20038. Nov. 2005Aifun Semiconductors Ltd.Method circuit and system for determining a reference voltage
US69729937. Febr. 20036. Dez. 2005Sandisk CorporationMethod and structure for efficient data verification operation for non-volatile memories
US698817530. Juni 200317. Jan. 2006M-Systems Flash Disk Pioneers Ltd.Flash memory management method that is resistant to data corruption by power loss
US699293229. Okt. 200331. Jan. 2006Saifun Semiconductors LtdMethod circuit and system for read error detection in a non-volatile memory array
US699934411. März 200514. Febr. 2006Kabushiki Kaisha ToshibaNon-volatile semiconductor memory device and electric device with the same
US700284327. Jan. 200421. Febr. 2006Sandisk CorporationVariable current sinking for coarse/fine programming of non-volatile memory
US70063791. März 200528. Febr. 2006Kabushiki Kaisha ToshibaSemiconductor memory
US70128353. Okt. 200314. März 2006Sandisk CorporationFlash memory data correction and scrub techniques
US70200176. Apr. 200428. März 2006Sandisk CorporationVariable programming of non-volatile memory
US7023735 *16. Juni 20044. Apr. 2006Ramot At Tel-Aviv University Ltd.Methods of increasing the reliability of a flash memory
US703121029. Juni 200418. Apr. 2006Hynix Semiconductor Inc.Method of measuring threshold voltage for a NAND flash memory device
US703121411. Dez. 200218. Apr. 2006Silicon Storage Technology, Inc.Digital multilevel memory system having multistage autozero sensing
US703121622. Juni 200418. Apr. 2006Hynix Semiconductor Inc.Refresh controller with low peak current
US7039846 *28. Mai 20032. Mai 2006Comtech Telecommunications Corp.Enhanced turbo product code decoder system utilizing a codeword organization method
US704276622. Juli 20049. Mai 2006Spansion, LlcMethod of programming a flash memory device using multilevel charge storage
US70541935. Nov. 200430. Mai 2006Samsung Electronics Co., Ltd.Non-uniform programming pulse width for writing of multi-bit-per-cell memories
US705419922. Dez. 200430. Mai 2006Samsung Electronics Co., Ltd.Multi level flash memory device and program method
US705795830. Sept. 20036. Juni 2006Sandisk CorporationMethod and system for temperature compensation for memory cells with temperature-dependent behavior
US706514722. Juni 200120. Juni 2006Texas Instruments IncorporatedSystem and method of data communication using turbo trellis coded modulation combined with constellation shaping with or without precoding
US706853927. Jan. 200427. Juni 2006Sandisk CorporationCharge packet metering for coarse/fine programming of non-volatile memory
US707184931. März 20054. Juli 2006Guobiao ZhangFractional-Bit Systems
US707222210. März 20054. Juli 2006Renesas Technology Corp.Nonvolatile memory system, semiconductor memory and writing method
US70795555. Nov. 200118. Juli 2006Pulse Communications, Inc.Integrated digital loop carrier system with virtual tributary mapper circuit
US708861527. Juli 20048. Aug. 2006Sandisk CorporationMulti-state memory
US70991943. Dez. 200429. Aug. 2006Sandisk CorporationError recovery for nonvolatile memory
US71029243. Jan. 20055. Sept. 2006Sandisk CorporationTechniques of recovering data from memory cells affected by field coupling with adjacent memory cells
US711343219. Nov. 200326. Sept. 2006Sandisk CorporationCompressed event counting technique and application to a flash memory system
US713021013. Jan. 200531. Okt. 2006Spansion LlcMulti-level ONO flash program algorithm for threshold width control
US71391926. Febr. 200421. Nov. 2006Samsung Electronics Co., Ltd.Programming of multi-level memory cells on a continuous word line
US713919827. Jan. 200421. Nov. 2006Sandisk CorporationEfficient verification for coarse/fine programming of non-volatile memory
US714580531. Jan. 20065. Dez. 2006Renesas Technology Corp.Nonvolatile memory system, semiconductor memory, and writing method
US715169220. Mai 200519. Dez. 2006Macronix International Co., Ltd.Operation scheme for programming charge trapping non-volatile memory
US71707817. Apr. 200530. Jan. 2007Sandisk CorporationMulti-bit-per-cell flash EEPROM memory with refresh
US717080231. Dez. 200330. Jan. 2007Sandisk CorporationFlexible and area efficient column redundancy for non-volatile memories
US71738595. Apr. 20056. Febr. 2007Sandisk CorporationFaster programming of higher level states in multi-level cell flash memory
US71771845. Apr. 200413. Febr. 2007Sandisk CorporationSelective operation of a multi-state non-volatile memory system in a binary mode
US717719527. Juli 200513. Febr. 2007Sandisk CorporationReducing the effects of noise in non-volatile memories through multiple reads
US717719920. Okt. 200313. Febr. 2007Sandisk CorporationBehavior based programming of non-volatile memory
US71772002. Aug. 200413. Febr. 2007Msystems Ltd.Two-phase programming of a flash memory
US718433830. Aug. 200527. Febr. 2007Spansion LlcSemiconductor device, semiconductor device testing method, and programming method
US718719510. Dez. 20046. März 2007Hynix Semiconductor Inc.Parallel compression test circuit of memory device
US718759218. Aug. 20046. März 2007Sandisk CorporationMulti-state memory
US719061420. Mai 200513. März 2007Macronix International Co., Ltd.Operation scheme for programming charge trapping non-volatile memory
US719389820. Juni 200520. März 2007Sandisk CorporationCompensation currents in non-volatile memory read operations
US719392111. Apr. 200520. März 2007Samsung Electronics, Co., Ltd.Cascade wake-up circuit preventing power noise in memory device
US71966441. Juli 200427. März 2007Seagate Technology LlcDecoupling of analog input and digital output
US71969285. Apr. 200527. März 2007Sandisk CorporationCompensating for coupling during read operations of non-volatile memory
US71969339. Dez. 200527. März 2007Kabushiki Kaisha ToshibaHigh-speed verifiable semiconductor memory device
US719759423. Sept. 200327. März 2007Infineon Technologies Flash Gmbh & Co. KgCircuit, system and method for encoding data to be stored on a non-volatile memory array
US720006231. Aug. 20043. Apr. 2007Micron Technology, Inc.Method and system for reducing the peak current in refreshing dynamic random access memory devices
US721007729. Jan. 200424. Apr. 2007Hewlett-Packard Development Company, L.P.System and method for configuring a solid-state storage device with error correction coding
US722159225. Febr. 200522. Mai 2007Micron Technology, Inc.Multiple level programming in a non-volatile memory device
US722461316. Aug. 200529. Mai 2007Sandisk CorporationOperating techniques for reducing effects of coupling between storage elements of a non-volatile memory operated in multiple data states
US723147431. Mai 200512. Juni 2007Advanced Micro Devices, Inc.Serial interface having a read temperature command
US72315629. Jan. 200412. Juni 2007Infineon Technologies AgMemory module, test system and method for testing one or a plurality of memory modules
US724327514. Dez. 200510. Juli 2007Sandisk CorporationSmart verify for multi-state memories
US725469020. Mai 20047. Aug. 2007S. Aqua Semiconductor LlcPipelined semiconductor memories and systems
US72547631. Sept. 20047. Aug. 2007Agere Systems Inc.Built-in self test for memory arrays using error correction coding
US72570276. Dez. 200514. Aug. 2007Hynix Semiconductor Inc.Non-volatile memory device, and multi-page program, read and copyback program method thereof
US725998721. Dez. 200521. Aug. 2007Sandisk CorporationSystems for variable programming of non-volatile memory
US72660269. Aug. 20054. Sept. 2007Sandisk CorporationSymbol frequency leveling in a storage system
US72660696. Febr. 20044. Sept. 2007Via Technologies, Inc.,Method and apparatus for retry calculation in an optical disk device
US726906611. Mai 200511. Sept. 2007Micron Technology, Inc.Programming memory devices
US727275728. Apr. 200518. Sept. 2007Infineon Technologies AgMethod for testing a memory chip and test arrangement
US727461128. Juli 200625. Sept. 2007Micron Technology, Inc.Method and architecture to calibrate read operations in synchronous flash memory
US727735529. Aug. 20052. Okt. 2007Micron Technology, Inc.Method and apparatus for generating temperature-compensated read and verify operations in flash memories
US728039831. Aug. 20069. Okt. 2007Micron Technology, Inc.System and memory for sequential multi-plane page memory operations
US728040917. März 20069. Okt. 2007Matsushita Electric Industrial Co., Ltd.Non-volatile memory device with threshold voltage control function
US728041528. Apr. 20069. Okt. 2007Samsung Electronics Co., Ltd.Flash memory device and method of repairing defects and trimming voltages
US72833993. Aug. 200616. Okt. 2007Renesas Technology Corp.Nonvolatile memory system, semiconductor memory, and writing method
US728934410. Nov. 200530. Okt. 2007Sandisk CorporationReverse coupling effect with timing information for non-volatile memory
US730180722. Febr. 200527. Nov. 2007Sandisk CorporationWritable tracking cells
US730181727. Okt. 200527. Nov. 2007Sandisk CorporationMethod for programming of multi-state non-volatile memory using smart verify
US730852528. März 200511. Dez. 2007Sandisk Il Ltd.Method of managing a multi-bit cell flash memory with improved reliablility and performance
US731025529. Dez. 200518. Dez. 2007Sandisk CorporationNon-volatile memory with improved program-verify operations
US73102696. März 200718. Dez. 2007Kabushiki Kaisha ToshibaHigh-speed verifiable semiconductor memory device
US73102717. Apr. 200618. Dez. 2007Hynix Semiconductor Inc.Program-verify method of non-volatile memory device
US73102722. Juni 200618. Dez. 2007Sandisk CorporationSystem for performing data pattern sensitivity compensation using different voltage
US731034718. Jan. 200518. Dez. 2007Sandisk, Il Ltd.States encoding in multi-bit flash cells
US732150916. Febr. 200722. Jan. 2008Sandisk CorporationCompensating for coupling in non-volatile storage
US732838429. Sept. 20055. Febr. 2008Xilinx, Inc.Method and apparatus using device defects as an identifier
US734283116. Juni 200611. März 2008Sandisk CorporationSystem for operating non-volatile memory using temperature compensation of voltages of unselected word lines and select gates
US734333026. Juni 200111. März 2008Boesjes Eimar MSystems and methods for acquisition, evaluation, inventory, distribution, and/or re-sale of pre-owned recorded data products
US734592411. Okt. 200618. März 2008Micron Technology, Inc.Programming memory devices
US734592814. Dez. 200518. März 2008Sandisk CorporationData recovery methods in multi-state memory after program fail
US73492632. Mai 200625. März 2008Samsung Electronics Co., Ltd.Circuit and method for adaptive incremental step-pulse programming in a flash memory device
US735675516. Okt. 20038. Apr. 2008Intel CorporationError correction for multi-level cell memory with overwrite capability
US736342015. Dez. 200322. Apr. 2008Nortel Networks LimitedMethod and file structures for managing data on a flash disk
US736567110. Okt. 200629. Apr. 2008Seagate Technology LlcCommunication channel with undersampled interpolative timing recovery
US73887812. Okt. 200617. Juni 2008Sandisk Il Ltd.Multi-bit-per-cell flash memory device with non-bijective mapping
US73976975. Jan. 20078. Juli 2008Sandisk CorporationMulti-bit-per-cell flash EEPROM memory with refresh
US740597419. Mai 200429. Juli 2008Sharp Kabushiki KaishaSemiconductor memory device, page buffer resource assigning method and circuit therefor, computer system and mobile electronic device
US74059797. Sept. 200729. Juli 2008Renesas Technology Corp.Nonvolatile memory system, semiconductor memory, and writing method
US74088046. Dez. 20055. Aug. 2008Sandisk CorporationSystems for soft programming non-volatile memory utilizing individual verification and additional soft programming of subsets of memory cells
US740881022. Febr. 20065. Aug. 2008Micron Technology, Inc.Minimizing effects of program disturb in a memory device
US740947321. Dez. 20045. Aug. 2008Sandisk CorporationOff-chip data relocation
US74096234. Nov. 20045. Aug. 2008Sigmatel, Inc.System and method of reading non-volatile computer memory
US742084714. Dez. 20052. Sept. 2008Sandisk CorporationMulti-state memory having data recovery after program fail
US743323126. Apr. 20067. Okt. 2008Micron Technology, Inc.Multiple select gates with non-volatile memory cells
US743369722. Juni 20047. Okt. 2008Broadcom CorporationSynchronized UWB piconets for Simultaneously Operating Piconet performance
US74341114. Apr. 20057. Okt. 2008Kabushiki Kaisha ToshibaNon-volatile memory system having a pseudo pass function
US743749826. Juli 200414. Okt. 2008San Disk Il, LtdMethod and system for optimizing reliability and performance of programming data in non-volatile memory devices
US744032429. Dez. 200621. Okt. 2008Sandisk CorporationApparatus with alternating read mode
US74403311. Juni 200621. Okt. 2008Sandisk CorporationVerify operation for non-volatile storage using different voltages
US744106715. Nov. 200421. Okt. 2008Sandisk CorporationCyclic flash memory wear leveling
US744797016. Juni 20044. Nov. 2008Seagate Technology, Inc.Soft-decision decoding using selective bit flipping
US74504212. Juni 200611. Nov. 2008Sandisk CorporationData pattern sensitivity compensation using different voltage
US74537377. Juni 200718. Nov. 2008Micron Technology, Inc.Program method with optimized voltage level for flash memory
US74571631. Juni 200625. Nov. 2008Sandisk CorporationSystem for verifying non-volatile storage using different voltages
US745789717. März 200425. Nov. 2008Suoer Talent Electronics, Inc.PCI express-compatible controller and interface for flash memory
US746041011. Aug. 20062. Dez. 2008Panasonic CorporationNonvolatile semiconductor memory device and programming or erasing method therefor
US746041229. Dez. 20062. Dez. 2008Samsung Electronics Co., Ltd.Flash memory device and erasing method thereof
US746659220. Aug. 200716. Dez. 2008Renesas Technology Corp.Semiconductor memory device
US746890718. Sept. 200623. Dez. 2008Samsung Electronics Co., Ltd.Programming method for flash memory capable of compensating reduction of read margin between states due to hot temperature stress
US74689112. Nov. 200623. Dez. 2008Sandisk CorporationNon-volatile memory using multiple boosting modes for reduced program disturb
US747158122. März 200730. Dez. 2008Silicon Storage Technology, Inc.Wide dynamic range and high speed voltage mode sensing for a multilevel digital non-volatile memory
US748331927. Mai 200827. Jan. 2009Brown Terry CMethod and system for reducing volatile memory DRAM power budget
US74873295. Febr. 20083. Febr. 2009International Business Machines CorporationPage replacement policy for systems having multiple page sizes
US749264120. Juni 200617. Febr. 2009Kabushiki Kaisha ToshibaNon-volatile semiconductor memory device
US750871013. Nov. 200624. März 2009Sandisk CorporationOperating non-volatile memory with boost structures
US752671113. Okt. 200528. Apr. 2009Nec Electronics CorporationDecoding method and device for repeatedly decoding first and second encoded data
US75390617. Nov. 200626. Mai 2009Hynix Semiconductor Inc.Method of programming flash memory device
US753906220. Dez. 200626. Mai 2009Micron Technology, Inc.Interleaved memory program and verify method, device and system
US75514928. März 200723. Juni 2009Mosaid Technologies, Inc.Non-volatile semiconductor memory with page erase
US75581093. Nov. 20067. Juli 2009Sandisk CorporationNonvolatile memory with variable read threshold
US755883914. Sept. 20057. Juli 2009Netapp, Inc.Read-after-write verification for improved write-once-read-many data storage
US756813530. März 200728. Juli 2009Apple Inc.Use of alternative value in cell detection
US757052027. Dez. 20064. Aug. 2009Sandisk CorporationNon-volatile storage system with initial programming voltage based on trial
US759000229. Dez. 200615. Sept. 2009Sandisk CorporationResistance sensing and compensation for non-volatile storage
US759325913. Juni 200722. Sept. 2009Mosaid Technologies IncorporatedFlash multi-level threshold distribution scheme
US759409312. Okt. 200622. Sept. 2009Foundry Networks, Inc.Virtual memory mapping for efficient memory usage
US75967076. Mai 200529. Sept. 2009Sun Microsystems, Inc.System and method for efficient power throttling in multiprocessor chip
US76097871. Okt. 200327. Okt. 2009TdfReception of a signal modulated according to a multilevel coding technique
US761304330. März 20073. Nov. 2009Apple Inc.Shifting reference values to account for voltage sag
US761649829. Dez. 200610. Nov. 2009Sandisk CorporationNon-volatile storage system with resistance sensing and compensation
US761991829. Dez. 200617. Nov. 2009Intel CorporationApparatus, method, and system for flash memory
US76312456. Jan. 20068. Dez. 2009Sandisk Il Ltd.NAND flash memory controller exporting a NAND interface
US763379821. Nov. 200715. Dez. 2009Micron Technology, Inc.M+N bit programming and M+L bit read for M bit memory cells
US763380229. Dez. 200815. Dez. 2009Sandisk CorporationNon-real time reprogramming of non-volatile memory to achieve tighter distribution of threshold voltages
US763953210. Okt. 200729. Dez. 2009Micron Technology, Inc.Non-equal threshold voltage ranges in MLC NAND
US764434730. Sept. 20055. Jan. 2010Intel CorporationSilent data corruption mitigation using error correction code with embedded signaling fault detection
US765673429. Juni 20072. Febr. 2010Sandisk 3D LlcMethods and apparatus for extending the effective thermal operating range of a memory
US766015819. Juni 20089. Febr. 2010Micron Technology, Inc.Programming method to reduce gate coupling interference for non-volatile memory
US76601831. Aug. 20059. Febr. 2010Rambus Inc.Low power memory device
US766105430. Sept. 20059. Febr. 2010Intel CorporationMethods and arrangements to remap degraded storage blocks
US766500730. Juni 200416. Febr. 2010Seagate Technology, LlcRetrial and reread methods and apparatus for recording channels
US768098729. März 200616. März 2010Emc CorporationSub-page-granular cache coherency using shared virtual memory mechanism
US773371220. Mai 20088. Juni 2010Siliconsystems, Inc.Storage subsystem with embedded circuit for protecting against anomalies in power signal from host
US774235113. Juni 200722. Juni 2010Semiconductor Energy Laboratory Co., Ltd.Semiconductor device and electronic device
US776162430. Aug. 200720. Juli 2010Virident Systems, Inc.Systems and apparatus for main memory with non-volatile type memory modules, and related technologies
US781001720. März 20065. Okt. 2010Micron Technology, Inc.Variable sector-count ECC
US784814912. Febr. 20077. Dez. 2010Sandisk CorporationReducing the effects of noise in non-volatile memories through multiple reads
US78692734. Sept. 200711. Jan. 2011Sandisk CorporationReducing the impact of interference during programming
US788511920. Juli 20068. Febr. 2011Sandisk CorporationCompensating for coupling during programming
US79246136. Juli 200912. Apr. 2011Anobit Technologies Ltd.Data storage in analog memory cells with protection against programming interruption
US792593611. Juli 200812. Apr. 2011Anobit Technologies Ltd.Memory device with non-uniform programming levels
US792849713. Sept. 200719. Apr. 2011Kabushiki Kaisha ToshibaNonvolatile semiconductor memory and manufacturing method thereof
US793051529. Juli 200819. Apr. 2011International Business Machines CorporationVirtual memory management
US794582525. Nov. 200817. Mai 2011Spansion Isreal, LtdRecovery while programming non-volatile memory (NVM)
US79785168. Apr. 200812. Juli 2011Pliant Technology, Inc.Flash memory controller having reduced pinout
US801409431. Aug. 20096. Sept. 2011Western Digital Technologies, Inc.Disk drive expediting defect scan when quality metric exceeds a more stringent threshold
US80373808. Juli 200811. Okt. 2011International Business Machines CorporationVerifying data integrity of a non-volatile memory system during data caching process
US80407445. Jan. 200918. Okt. 2011Sandisk Technologies Inc.Spare block management of non-volatile memories
US8156403 *10. Mai 200710. Apr. 2012Anobit Technologies Ltd.Combined distortion estimation and error correction coding for memory devices
US2001000217223. Jan. 200131. Mai 2001Tomoharu TanakaNonvolatile semiconductor memory device
US2001000647926. Dez. 20005. Juli 2001Tamio IkehashiRead circuit of nonvolatile semiconductor memory
US2002003844010. Nov. 199828. März 2002Moti BarkanDigital data representation for multi-bit data storage and transmission
US2002005606414. Nov. 20019. Mai 2002Tycom (Us) Inc.Method and apparatus for enhanced forward error correction in network
US2002011857426. Febr. 200129. Aug. 2002Geoffrey GongwerNon-volatile memory with improved programming and method therefor
US2002013368417. Jan. 200219. Sept. 2002Honeywell International, Inc.Enhanced memory module architecture
US2002016609125. März 20027. Nov. 2002Kidorf Howard D.Concatenated forward error correction decoder
US2002017429529. Jan. 200221. Nov. 2002Ulrich Thomas R.Enhanced file system failure tolerance
US2002019651028. März 200226. Dez. 2002Hietala Vincent MarkMethod and system for decoding multilevel signals
US2003000234827. Juni 20012. Jan. 2003Sandisk CorporationOperating techniques for reducing effects of coupling between storage elements of a non-volatile memory operated in multiple data states
US2003010340011. Dez. 20025. Juni 2003Hieu Van TranMultistage autozero sensing for a multilevel non-volatile memory integrated circuit system
US2003016118311. Dez. 200228. Aug. 2003Tran Hieu VanDigital multilevel non-volatile memory system
US2003018985619. Nov. 20029. Okt. 2003Samsung Electronics Co., Ltd.Multi-level flash memory with temperature compensation
US200400572659. Juli 200325. März 2004Stmicroelectronics SaMethod of controlling an electronic non-volatile memory and associated device
US2004005728524. Sept. 200225. März 2004Sandisk CorporationNon-volatile memory and method with reduced neighboring field errors
US200400833332. Okt. 200329. Apr. 2004Sandisk CorporationHybrid implementation for error correction codes within a non-volatile memory system
US200400833342. Okt. 200329. Apr. 2004Sandisk CorporationMethod and apparatus for managing the integrity of data in non-volatile memory system
US2004010531112. Nov. 20033. Juni 2004Raul-Adrian CerneaNon-volatile memory with improved sensing and method therefor
US2004011443713. Dez. 200217. Juni 2004Li Chi Nan BrianCompaction scheme in NVM
US200401608425. Sept. 200319. Aug. 2004Renesas Technology Corp.Semiconductor memory device suppressing peak current
US200402233718. Mai 200311. Nov. 2004Micron Technology, Inc.NAND flash memory with improved read and verification threshold uniformity
US2005000780212. Apr. 200413. Jan. 2005Gerpheide George E.Zero drift analog memory cell, array and method of operation
US2005001316516. Juni 200420. Jan. 2005Amir BanFlash memories with adaptive reference voltages
US2005002494116. Dez. 20033. Febr. 2005M-Systems Flash Disk Pioneers, Ltd.Method of archiving data
US2005002497826. Juli 20043. Febr. 2005M-Systems Flash Disk Pioneers Ltd.Method and system for optimizing reliability and performance of programming data in non-volatile memory devices
US200500307884. Aug. 200310. Febr. 2005Parkinson Ward D.Analog phase change memory
US2005008657416. Okt. 200321. Apr. 2005Fackenthal Richard E.Error correction for multi-level cell memory with overwrite capability
US200501214363. Dez. 20049. Juni 2005Fuji Photo Film Co., Ltd.Heating method for a band-shaped body and heating apparatus for a band-shaped body
US2005015755521. Jan. 200421. Juli 2005Tsuyoshi OnoNonvolatile semiconductor memory device
US2005016291326. Jan. 200428. Juli 2005Jian ChenMethod of reading NAND memory to compensate for coupling between storage elements
US2005016905122. Febr. 20054. Aug. 2005Khalid Shahzad B.Writable tracking cells
US2005018964925. Apr. 20051. Sept. 2005Fujitsu LimitedLSI package, LSI element testing method, and semiconductor device manufacturing method
US2005021339322. Febr. 200529. Sept. 2005M-Systems Flash Disk Pioneers, Ltd.States encoding in multi-bit flash cells for optimizing error rate
US2005022485319. März 200313. Okt. 2005Narumi OhkawaSemiconductor device, manufacturing process thereof and imaging device
US2005024074517. Dez. 200427. Okt. 2005Sundar IyerHigh speed memory control and I/O processor system
US2005024362627. Okt. 20043. Nov. 2005M-Systems Flash Disk Pioneers Ltd.Refreshing data stored in a flash memory
US200600049528. Aug. 20055. Jan. 2006M-Systems Flash Disk Pioneers, Ltd.Method of managing a multi-bit-cell flash memory
US2006002887518. Aug. 20059. Febr. 2006M-Systems Flash Disk Pioneers, Ltd.Drift compensation in a flash memory
US2006002887719. Jan. 20059. Febr. 2006M-Systems Flash Disk Pioneers Ltd.Different numbers of bits per cell in non-volatile memory devices
US2006010119314. März 200511. Mai 2006M-Systems Flash Disk Pioneers, Ltd.States encoding in multi-bit flash cells for optimizing error rate
US2006010697215. Nov. 200418. Mai 2006Gorobets Sergey ACyclic flash memory wear leveling
US2006010713614. Dez. 200518. Mai 2006Gongwer Geoffrey SSmart verify for multi-state memories
US2006012975015. Dez. 200515. Juni 2006Samsung Electronics Co., Ltd.Method and apparatus for storing multimedia data in nonvolatile storage device in units of blocks
US2006013314122. Dez. 200422. Juni 2006Gorobets Sergey AErased sector detection mechanisms
US2006015618921. Dez. 200413. Juli 2006Andrew TomlinMethod for copying data in reprogrammable non-volatile memory
US200601793349. Febr. 200510. Aug. 2006International Business Machines CorporationDynamic power management via DIMM read operation limiter
US2006019069930. Dez. 200524. Aug. 2006Samsung Electronics Co., Ltd.Method, medium, and apparatus transforming addresses of discs in a disc drive
US2006020354629. Nov. 200514. Sept. 2006M-Systems Flash Disk Pioneers, Ltd.Method of achieving wear leveling in flash memory using relative grades
US2006021835922. März 200528. Sept. 2006Sigmatel, Inc.Method and system for managing multi-plane memory devices
US200602216925. Apr. 20055. Okt. 2006Jian ChenCompensating for coupling during read operations on non-volatile memory
US200602217056. Dez. 20055. Okt. 2006Hemink Gerrit JSoft programming non-volatile memory utilizing individual verification and additional soft programming of subsets of memory cells
US200602217145. Apr. 20055. Okt. 2006Yan LiRead operation for non-volatile storage that includes compensation for coupling
US2006023907713. Apr. 200626. Okt. 2006Park Ki-TaeNAND Flash Memory Device Having Dummy Memory cells and Methods of Operating Same
US200602390811. Mai 200626. Okt. 2006Micron Technology, Inc.NAND flash memory with read and verification threshold uniformity
US2006025662011. Mai 200516. Nov. 2006Micron Technology, Inc.Programming memory devices
US2006025662627. Juli 200616. Nov. 2006Werner Carl WIntegrated Circuit with Analog or Multilevel Storage Cells and User-Selectable Sampling Frequency
US2006025689117. Juli 200616. Nov. 2006Applied Microcircuits CorporationSystem for five-level non-causal channel equalization
US2006027174831. Mai 200530. Nov. 2006Intel CorporationPartial page scheme for memory technologies
US200602853921. Mai 200621. Dez. 2006Micron Technology, Inc.Selective slow programming convergence in a flash memory device
US2006028539617. Juni 200521. Dez. 2006Micron Technology, Inc.Program method with optimized voltage level for flash memory
US2007000601326. Juli 20064. Jan. 2007Simpletech, Inc.Solid-state memory device with protection against power failure
US2007001948120. Apr. 200625. Jan. 2007Park Chul WSemiconductor memories with block-dedicated programmable latency register
US200700335812. Aug. 20058. Febr. 2007Andrew TomlinSituation sensitive memory performance
US2007004731431. Aug. 20051. März 2007Micron Technology, Inc.Programming method for NAND EEPROM
US2007004732611. Okt. 20061. März 2007Micron Technology, Inc.Programming memory devices
US2007005053630. Aug. 20061. März 2007Cypress Semiconductor CorporationFlash drive fast wear leveling
US2007005844621. Dez. 200515. März 2007Hynix Semiconductor Inc.Erase and Program Method of Flash Memory Device for Increasing Program Speed of Flash Memory Device
US2007006150228. Dez. 200515. März 2007M-Systems Flash Disk Pioneers Ltd.Flash memory storage system and method
US2007006766717. Febr. 200622. März 2007Fujitsu LimitedWrite back method for RAID apparatus
US200700740936. Jan. 200629. März 2007M-Systems Flash Disk Pioneers Ltd.NAND flash memory controller exporting and NAND interface
US2007008623926. Jan. 200619. Apr. 2007M-Systems Flash Disk Pioneers, Ltd.Probabilistic error correction in multi-bit-per-cell flash memory
US2007008626013. Okt. 200519. Apr. 2007Sinclair Alan WMethod of storing transformed units of data in a memory system having fixed sized storage blocks
US2007008903411. Jan. 200619. Apr. 2007M-Systems Flash Disk Pioneers, Ltd.Method of error correction in MBC flash memory
US200700916775. Apr. 200626. Apr. 2007M-Systems Flash Disk Pioneers Ltd.Method for recovering from errors in flash memory
US2007009169416. Juni 200626. Apr. 2007Samsung Electronics Co., Ltd.Flash memory device capable of improving reliability
US200701039788. Nov. 200510. Mai 2007Conley Kevin MMemory with retargetable memory cell redundancy
US2007010398627. Dez. 200610. Mai 2007Jian ChenCompensating for coupling during read operations of non-volatile memory
US2007010984527. Dez. 200617. Mai 2007Jian ChenCompensating for coupling during read operations of non-volatile memory
US2007010984927. Dez. 200617. Mai 2007Jian ChenCompensating for coupling during read operations of non-volatile memory
US2007011572617. Okt. 200624. Mai 2007Guy CohenMethod, system and circuit for programing a non-volatile memory array
US2007011871322. Nov. 200524. Mai 2007Guterman Daniel CMemory system for legacy hosts
US2007014337821. Dez. 200521. Juni 2007Gorobets Sergey ANon-volatile memories with adaptive file handling in a directly mapped file storage system
US2007014353115. Dez. 200521. Juni 2007Atri Sunil RPower loss recovery for bit alterable memory
US2007015988918. Sept. 200612. Juli 2007Dong-Ku KangProgramming method for flash memory capable of compensating reduction of read margin between states due to hot temperature stress
US2007015989231. Aug. 200612. Juli 2007Dong-Ku KangProgramming method for flash memory capable of compensating reduction of read margin between states due to high temperature stress
US200701599077. Nov. 200612. Juli 2007Pan-Suk KwakMulti-chip package reducing peak power-up current
US200701688373. Apr. 200619. Juli 2007M-Systems Flash Disk Pioneers Ltd.Method for implementing error-correction codes in flash memory
US200701717148. Nov. 200626. Juli 2007Marvell International Ltd.Flash memory with coding and signal processing
US200701832101. Febr. 20079. Aug. 2007Choi Ki-HwanProgram method of flash memory capable of compensating read margin reduced due to charge loss
US2007018907316. Febr. 200616. Aug. 2007Micron Technology, Inc.Programming method to reduce gate coupling interference for non-volatile memory
US2007019560213. Apr. 200723. Aug. 2007Yupin FongReducing floating gate to floating gate coupling effect
US2007020642617. März 20066. Sept. 2007Nima MokhlesiSystem for performing read operation on non-volatile storage with compensation for coupling
US200702089043. März 20066. Sept. 2007Wu-Han HsiehWear leveling method and apparatus for nonvolatile memory
US2007022659922. Mai 200627. Sept. 2007Broadcom Corporation, A California CorporationSymbol by symbol map detection for signals corrupted by colored and/or signal dependent noise
US2007023699028. März 200611. Okt. 2007Micron Technology, Inc.Programming method to reduce word line to word line breakdown for NAND flash
US200702532494. Dez. 20061. Nov. 2007Sang-Gu KangMulti-bit nonvolatile memory device and related programming method
US2007025662010. Apr. 20078. Nov. 2007Autotether, Inc.Sailboat safety system for a person falling overboard
US20070263455 *30. März 200715. Nov. 2007Apple Inc.Iterative Memory Cell Charging Based on Reference Cell Value
US2007026623210. Nov. 200615. Nov. 2007Stephane RodgersMethod and System For Command Interface Protection To Achieve a Secure Interface
US2007027142422. März 200722. Nov. 2007Samsung Electronics Co., Ltd.Memory module, a memory system including a memory controller and a memory module and methods thereof
US200702800002. Aug. 20076. Dez. 2007Kabushiki Kaisha ToshibaMethod of writing data to a semiconductor memory device
US200702915718. Juni 200620. Dez. 2007Intel CorporationIncreasing the battery life of a mobile computing system in a reduced power state through memory compression
US2007029723431. Aug. 200727. Dez. 2007Raul-Adrian CerneaNon-Volatile Memory And Method With Bit Line To Bit Line Coupled Compensation
US200800103956. Juli 200610. Jan. 2008Nokia CorporationPerformance optimization in solid-state media
US200800251212. Okt. 200731. Jan. 2008Micron Technology, Inc.Method and apparatus for generating temperature-compensated read and verify operations in flash memories
US2008004353518. Aug. 200621. Febr. 2008Micron Technology, Inc.Method and system for minimizing number of programming pulses used to program rows of non-volatile memory cells
US2008004950412. Mai 200528. Febr. 2008Matsushita Electric Industrial Co., Ltd.Memory Control Circuit, Nonvolatile Storage Apparatus, and Memory Control Method
US2008004950631. Okt. 200728. Febr. 2008Guterman Daniel CAlternate Row-Based Reading and Writing for Non-Volatile Memory
US200800524469. Febr. 200728. Febr. 2008Sandisk Il Ltd.Logical super block mapping for NAND flash memory
US2008005599321. Sept. 20076. März 2008Micron Technology, Inc.System and memory for sequential multi-plane page memory operations
US2008008024328. Sept. 20073. Apr. 2008Kabushiki Kaisha ToshibaSemiconductor memory device
US2008008273028. Dez. 20063. Apr. 2008Samsung Electronics Co., Ltd.Multi-chip flash memory device and copy-back method thereof
US2008008912330. Aug. 200717. Apr. 2008Dong-Hyuk ChaeMethod for Programming a Multi-Level Non-Volatile Memory Device
US200801043096. Febr. 20071. Mai 2008Cheon Won-MoonFlash memory device with multi-level cells and method of writing data therein
US2008010431225. Okt. 20071. Mai 2008Sandisk Il Ltd.States encoding in multi-bit flash cells for optimizing error rate
US2008010959028. Dez. 20068. Mai 2008Samsung Electronics Co., Ltd.Flash memory system and garbage collection method thereof
US2008011501731. Okt. 200615. Mai 2008Jacobson Michael BDetection and correction of block-level data corruption in fault-tolerant data-storage systems
US200801234203. Nov. 200629. Mai 2008Yigal BrandmanNonvolatile Memory With Variable Read Threshold
US2008012668627. Nov. 200729. Mai 2008Anobit Technologies Ltd.Memory power and performance management
US200801303413. Dez. 20075. Juni 2008Anobit Technologies Ltd.Adaptive programming of analog memory
US2008014811517. Dez. 200719. Juni 2008Anobit Technologies Ltd.High-speed programming of memory devices
US2008015161813. Juni 200726. Juni 2008Sandisk Il Ltd.Flash memory device and system with randomizing for suppressing errors
US2008015166719. Okt. 200726. Juni 2008Semiconductor Manufacturing International ( Shanghai) CorporationMethod for Decreasing Program Disturb in Memory Cells
US2008015895817. Dez. 20073. Juli 2008Anobit Technologies Ltd.Memory device with reduced reading
US2008018100124. Jan. 200831. Juli 2008Anobit TechnologiesMemory device with negative thresholds
US2008019865010. Mai 200721. Aug. 2008Anobit Technologies Ltd.Distortion Estimation And Cancellation In Memory Devices
US2008019865421. Apr. 200821. Aug. 2008Kabushiki Kaisha ToshibaSemiconductor memory device
US200802091161. Mai 200828. Aug. 2008Jason CaulkinsMulti-Processor Flash Memory Storage Device and Management System
US200802093045. Mai 200828. Aug. 2008International Business Machines CorporationConvolution-Encoded RAID with Trellis-Decode-Rebuild
US2008021579813. Juni 20074. Sept. 2008Sandisk Il Ltd.Randomizing for suppressing errors in a flash memory
US2008021905026. Febr. 200811. Sept. 2008Anobit Technologies Ltd.Reduction of back pattern dependency effects in memory devices
US200802390936. Mai 20082. Okt. 2008Lightsurf Technologies, Inc.Digital Camera Device Providing Improved Methodology for Rapidly Taking Successive Pictures
US2008023981228. März 20082. Okt. 2008Kabushiki Kaisha ToshibaNonvolatile semiconductor memory system
US2008025318819. Juni 200816. Okt. 2008Micron Technology, Inc.Programming method to reduce gate coupling interference for non-volatile memory
US2008026326210. März 200823. Okt. 2008Anobit Technologies Ltd.Command interface for memory devices
US2008026367622. Jan. 200823. Okt. 2008Samsung Electronics Co., Ltd.System and method for protecting data information stored in storage
US2008027073014. Apr. 200830. Okt. 2008Sandisk Il Ltd.Method for efficient storage of metadata in flash memory
US2008028210612. Mai 200813. Nov. 2008Anobit Technologies LtdData storage with incremental redundancy
US2008028871425. Sept. 200720. Nov. 2008Sandisk Il LtdFile storage in a computer system with diverse storage media
US200900132336. Juli 20078. Jan. 2009Micron Technology, Inc.Error recovery storage along a nand-flash string
US2009002490510. Mai 200722. Jan. 2009Anobit Technologies Ltd.Combined distortion estimation and error correction coding for memory devices
US200900343371. Aug. 20075. Febr. 2009Seiichi AritomeMethod, apparatus, and system for improved read operation in memory
US2009004383129. Febr. 200812. Febr. 2009Mcm Portfolio LlcSmart Solid State Drive And Method For Handling Critical Files
US200900439516. Aug. 200812. Febr. 2009Anobit Technologies Ltd.Programming schemes for multi-level analog memory cells
US2009004923417. Jan. 200819. Febr. 2009Samsung Electronics Co., Ltd.Solid state memory (ssm), computer system including an ssm, and method of operating an ssm
US2009007376229. Aug. 200819. März 2009Samsung Electronics Co., Ltd.Methods of operating multi-bit flash memory devices and related systems
US2009008654228. Sept. 20072. Apr. 2009Dana LeeHigh Voltage Generation and Control in Source-Side Injection Programming of Non-Volatile Memory
US2009008948416. Jan. 20082. Apr. 2009Phison Electronics Corp.Data protection method for power failure and controller using the same
US200900919795. Okt. 20089. Apr. 2009Anobit TechnologiesReliable data storage in analog memory cells in the presence of temperature variations
US2009009493017. Apr. 200716. Apr. 2009Artur SchwoererWooden Lattice Girder for Construction
US2009010648523. Juli 200823. Apr. 2009Anobit Technologies Ltd.Reading analog memory cells using built-in multi-threshold commands
US2009011294931. Okt. 200730. Apr. 2009Microsoft CorporationCompressed storage management
US2009013275521. Nov. 200721. Mai 2009Micron Technology, Inc.Fault-tolerant non-volatile integrated circuit memory
US2009014460026. Nov. 20084. Juni 2009Anobit Technologies LtdEfficient re-read operations from memory devices
US2009015089410. Dez. 200711. Juni 2009Ming HuangNonvolatile memory (NVM) based solid-state disk (SSD) system for scaling and quality of service (QoS) by parallelizing command execution
US2009015795014. Dez. 200718. Juni 2009Robert David SelingerNAND flash module replacement for DRAM module
US2009015796411. Dez. 200818. Juni 2009Anobit Technologies Ltd.Efficient data storage in multi-plane memory devices
US2009015812611. Dez. 200818. Juni 2009Anobit Technologies LtdEfficient interference cancellation in analog memory cell arrays
US2009016852425. Dez. 20082. Juli 2009Anobit Technologies Ltd.Wear level estimation in analog memory cells
US200901722578. Apr. 20082. Juli 2009Pliant Technology, IncSystem and method for performing host initiated mass storage commands using a hierarchy of data structures
US200901722618. Apr. 20082. Juli 2009Pliant Technology, Inc.Multiprocessor storage controller
US200901931843. Apr. 200930. Juli 2009Super Talent Electronics Inc.Hybrid 2-Level Mapping Tables for Hybrid Block- and Page-Mode Flash-Memory System
US200901990743. Febr. 20096. Aug. 2009Anobit Technologies Ltd.Parameter estimation based on error correction code parity check equations
US2009020482430. Dez. 200813. Aug. 2009Lin Jason TSystem, method and memory device providing data scrambling compatible with on-chip copy operation
US2009020487221. Apr. 200913. Aug. 2009Super Talent Electronics Inc.Command Queuing Smart Storage Transfer Manager for Striping Data to Raw-NAND Flash Modules
US2009021365319. Febr. 200927. Aug. 2009Anobit Technologies LtdProgramming of analog memory cells using a single programming pulse per state transition
US2009021365423. Febr. 200927. Aug. 2009Anobit Technologies LtdProgramming analog memory cells for reduced variance after retention
US2009022559511. Mai 200910. Sept. 2009Mosaid Technologies IncorporatedFlash multi-level threshold distribution scheme
US200902287614. März 200910. Sept. 2009Anobit Technologies LtdEfficient readout from analog memory cells using data compression
US2009024087217. März 200924. Sept. 2009Anobit Technologies LtdMemory device with multiple-accuracy read commands
US200902655092. Juli 200922. Okt. 2009Micron Technology, Inc.Memory system and method having volatile and non-volatile memory devices at same hierarchical level
US2009030022727. Mai 20083. Dez. 2009Sandisk Il Ltd.Method of monitoring host activity
US2009032341230. Juni 200831. Dez. 2009Nima MokhlesiRead disturb mitigation in non-volatile memory
US2009032760826. Juni 200831. Dez. 2009Eschmann Michael KAccelerated resume from hibernation in a cached disk system
US2010001765025. Mai 200921. Jan. 2010Nanostar Corporation, U.S.ANon-volatile memory data storage system with reliability management
US201000340228. Aug. 200811. Febr. 2010Deepanshu DuttaCompensating for coupling during read operations in non-volatile storage
US2010005797626. Aug. 20084. März 2010Menahem LasserMultiple performance mode memory system
US2010006115129. Dez. 200811. März 2010Toru MiwaMulti-pass programming for memory with reduced data storage requirement
US2010008288326. Febr. 20091. Apr. 2010Ming-Dar ChenHybrid density memory system and control method thereof
US2010008324726. Sept. 20081. Apr. 2010Netapp, Inc.System And Method Of Providing Multiple Virtual Machines With Shared Access To Non-Volatile Solid-State Memory Using RDMA
US201001105804. Sept. 20096. Mai 2010Kabushiki Kaisha ToshibaMagnetic disk device
US2010012408815. Nov. 200920. Mai 2010Anobit Technologies LtdStorage at m bits/cell density in n bits/cell analog memory cell devices, m>n
US2010013169723. Nov. 200827. Mai 2010Sandisk Il Ltd.Methods for tag-grouping of blocks in storage devices
US2010013182716. Apr. 200827. Mai 2010Anobit Technologies LtdMemory device with internal signap processing unit
US201001422689. Febr. 201010. Juni 2010Micron Technology, Inc.Programming method to reduce gate coupling interference for non-volatile memory
US2010014227729. Juni 200910. Juni 2010Yang Chang WonPage buffer circuit, nonvolatile memory device including the page buffer circuit, and method of operating the nonvolatile memory device
US2010015767511. März 201024. Juni 2010Anobit Technologies LtdProgramming orders for reducing distortion in arrays of multi-level analog memory cells
US2010016568930. Dez. 20091. Juli 2010Anobit Technologies LtdRejuvenation of analog memory cells
US2010016954715. März 20091. Juli 2010Silicon Motion, Inc.Method for preventing data loss during solder reflow process and memory device using the same
US2010016974331. Dez. 20081. Juli 2010Andrew Wayne VoganError correction in a solid state disk
US201001748475. Jan. 20098. Juli 2010Alexander PaleyNon-Volatile Memory and Method With Write Cache Partition Management Methods
US2010019539012. Apr. 20105. Aug. 2010Anobit Technologies LtdMemory device with negative thresholds
US2010019915012. Okt. 20085. Aug. 2010Anobit Technologies LtdData Storage In Analog Memory Cell Arrays Having Erase Failures
US2010022050928. Febr. 20102. Sept. 2010Anobit Technologies LtdSelective Activation of Programming Schemes in Analog Memory Cell Arrays
US201002205104. Nov. 20082. Sept. 2010Anobit Technologies LtdOptimized Selection of Memory Chips in Multi-Chips Memory Devices
US2010025083622. März 201030. Sept. 2010Anobit Technologies LtdUse of Host System Resources by Memory Controller
US2011006679314. Sept. 201017. März 2011Gregory BurdImplementing RAID In Solid State Memory
US2011007548229. Sept. 200931. März 2011Zac ShepardMaintaining integrity of preloaded content in non-volatile memory during surface mounting
US2011010704930. Juli 20105. Mai 2011Samsung Electronics Co., Ltd.Method and apparatus adapted to prevent code data from being lost in solder reflow
US201101998235. Okt. 200918. Aug. 2011Sandisk Il Ltd.Preloading data into a flash storage device
US201103023542. Juni 20108. Dez. 2011Conexant Systems, Inc.Systems and methods for reliable multi-level cell flash storage
EP0783754B122. Sept. 199511. Juli 2001Intel CorporationSensing state of a memory by variable gate voltage
EP1434236B13. März 200028. Dez. 2005Oki Electric Industry Co. Ltd., Legal &amp; Intellectual Property DivisionSemiconductor memory device generating accurate internal reference voltage
EP1605509A119. März 200314. Dez. 2005Fujitsu LimitedSemiconductor device, process for producing the same and imaging device
WO1996010256A122. Sept. 19954. Apr. 1996Intel CorporationSensing state of a memory by variable gate voltage
WO1998028745A123. Okt. 19972. Juli 1998Intel CorporationNonvolatile writeable memory with fast programming capability
WO2002100112A13. Juni 200112. Dez. 2002Seelive Ltd.System and method for rapid video compression
WO2003100791A112. Mai 20034. Dez. 2003Sandisk CorporationIncreasing the effectiveness of error correction codes and operating multi-level memory systems by using information about the quality of the stored data
WO2007046084A215. Okt. 200626. Apr. 2007Ramot At Tel-Aviv University Ltd.Probabilistic error correction in multi-bit-per-cell flash memory
WO2007132452A210. Mai 200722. Nov. 2007Anobit TechnologiesReducing programming error in memory devices
WO2007132453A210. Mai 200722. Nov. 2007Anobit Technologies Ltd.Distortion estimation and cancellation in memory devices
WO2007132456A210. Mai 200722. Nov. 2007Anobit Technologies Ltd.Memory device with adaptive capacity
WO2007132457A210. Mai 200722. Nov. 2007Anobit Technologies Ltd.Combined distortion estimation and error correction coding for memory devices
WO2007132458A210. Mai 200722. Nov. 2007Anobit Technologies Ltd.Memory device programming using combined shaping and linear spreading
WO2007146010A26. Juni 200721. Dez. 2007Micron Technology, Inc.Programming a non-volatile memory device
WO2008026203A227. Aug. 20076. März 2008Anobit TechnologiesEstimation of non-linear distortion in memory devices
WO2008053472A230. Okt. 20078. Mai 2008Anobit Technologies Ltd.Reading memory cells using multiple thresholds
WO2008053473A230. Okt. 20078. Mai 2008Anobit Technologies Ltd.Memory cell readout using successive approximation
WO2008068747A23. Dez. 200712. Juni 2008Anobit Technologies Ltd.Automatic defect management in memory devices
WO2008077284A127. Dez. 20063. Juli 2008Intel CorporationInitiative wear leveling for non-volatile memory
WO2008083131A224. Dez. 200710. Juli 2008Sandisk CorporationMethod for programming with initial programming voltage based on trial
WO2008099958A113. Febr. 200821. Aug. 2008Kabushiki Kaisha ToshibaMethod of writing data into semiconductor memory and memory controller
WO2008111058A211. März 200818. Sept. 2008Anobit Technologies Ltd.Adaptive estimation of memory cell read thresholds
WO2008124760A29. Apr. 200816. Okt. 2008Sandisk CorporationNon-volatile memory and method for predictive programming
WO2008139441A216. Apr. 200820. Nov. 2008Anobit Technologies Ltd.Memory device with internal signal processing unit
WO2009037691A23. Sept. 200826. März 2009Anobit TechnologiesProgramming orders for reducing distortion in arrays of multi-level analog memory cells
WO2009037697A217. Sept. 200826. März 2009Densbits Technologies Ltd.Improved systems and methods for determining logical values of coupled flash memory cells
WO2009038961A22. Sept. 200826. März 2009Sandisk CorporationNonvolatile memory and method for compensating during programming for perturbing charges of neighboring cells
WO2009050703A212. Okt. 200823. Apr. 2009Anobit TechnologiesData storage in analog memory cell arrays having erase failures
WO2009053961A217. Sept. 200830. Apr. 2009Densbits Technologies Ltd.Systems and methods for multiple coding rates in flash devices
WO2009053962A217. Sept. 200830. Apr. 2009Densbits Technologies Ltd.Systems and methods for averaging error rates in non-volatile devices and storage systems
WO2009053963A217. Sept. 200830. Apr. 2009Densbits Technologies Ltd.Methods for adaptively programming flash memory devices and flash memory systems incorporating same
WO2009063450A24. Nov. 200822. Mai 2009Anobit TechnologiesOptimized selection of memory units in multi-unit memory devices
WO2009072100A217. Sept. 200811. Juni 2009Densbits Technologies Ltd.Systems and methods for temporarily retiring memory portions
WO2009072101A217. Sept. 200811. Juni 2009Densbits Technologies Ltd.Apparatus and methods for generating row-specific reading thresholds in flash memory
WO2009072102A217. Sept. 200811. Juni 2009Densbits Technologies Ltd.System and methods employing mock thresholds to generate actual reading thresholds in flash memory devices
WO2009072103A217. Sept. 200811. Juni 2009Densbits Technologies Ltd.Flash memory apparatus and methods using a plurality of decoding stages including optional use of concatenated bch codes
WO2009072104A217. Sept. 200811. Juni 2009Densbits Technologies Ltd.Flash memory device with physical cell value deterioration accommodation and methods useful in conjunction therewith
WO2009072105A217. Sept. 200811. Juni 2009Densbits Technologies Ltd.A low power chien-search based bch/rs decoding system for flash memory
WO2009074978A217. Sept. 200818. Juni 2009Densbits Technologies Ltd.Systems and methods for error correction and decoding on multi-level physical media
WO2009074979A217. Sept. 200818. Juni 2009Densbits Technologies Ltd.Chien-search system employing a clock-gating scheme to save power for error correction decoder and other applications
WO2009078006A217. Sept. 200825. Juni 2009Densbits Technologies Ltd.Apparatus for coding at a plurality of rates in multi-level flash memory systems, and methods useful in conjunction therewith
WO2009095902A217. Sept. 20086. Aug. 2009Densbits Technologies Ltd.Systems and methods for handling immediate data errors in flash memory
WO2011024015A125. Aug. 20093. März 2011Sandisk Il Ltd.Restoring data into a flash storage device
Nichtpatentzitate
Referenz
1Agrell et al., "Closest Point Search in Lattices", IEEE Transactions on Information Theory, vol. 48, No. 8, pp. 2201-2214, Aug. 2002.
2Ankolekar et al., "Multibit Error-Correction Methods for Latency-Constrained Flash Memory Systems", IEEE Transactions on Device and Materials Reliability, vol. 10, No. 1, pp. 33-39, Mar. 2010.
3Berman et al., "Mitigating Inter-Cell Coupling Effects in MLC NAND Flash via Constrained Coding", Flash Memory Summit, Santa Clara, USA, Aug. 19, 2010.
4Bez et al., "Introduction to Flash memory", Proceedings of the IEEE, vol. 91, No. 4, pp. 489-502, Apr. 2003.
5Blahut, R.E., "Theory and Practice of Error Control Codes," Addison-Wesley, May 1984, section 3.2, pp. 47-48.
6Chang, L., "Hybrid Solid State Disks: Combining Heterogeneous NAND Flash in Large SSDs", ASPDAC, Jan. 2008.
7Chinese Application # 200780026181.3 Official Action dated Apr. 8, 2011.
8Chinese Patent Application # 200780040493.X Official Action dated Jun. 15, 2011.
9Cho et al., "Multi-Level NAND Flash Memory with Non-Uniform Threshold Voltage Distribution," IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, Feb. 5-7, 2001, pp. 28-29 and 424.
10Compaq et al., "Universal Serial Bus Specification", revision 2.0, Apr. 27, 2000.
11Conway et al., "Sphere Packings, Lattices and Groups", 3rd edition, chapter 4, pp. 94-135, Springer, New York, USA 1998.
12Databahn(TM), "Flash memory controller IP", Denali Software, Inc., 1994 https://www.denali.com/en/products/databahn-flash.jsp.
13Databahn™, "Flash memory controller IP", Denali Software, Inc., 1994 https://www.denali.com/en/products/databahn-flash.jsp.
14Datalight, Inc., "FlashFX Pro 3.1 High Performance Flash Manager for Rapid Development of Reliable Products", Nov. 16, 2006.
15Duann, N., Silicon Motion Presentation "SLC & MLC Hybrid", Flash Memory Summit, Santa Clara, USA, Aug. 2008.
16Eitan et al., "Can NROM, a 2-bit, Trapping Storage NVM Cell, Give a Real Challenge to Flating Gate Cells?", Proceedings of the 1999 International Conference on Solid State Devices and Materials (SSDM), pp. 522-524, Tokyo, Japan 1999.
17Eitan et al., "Multilevel Flash Cells and their Trade-Offs", Proceedings of the 1996 IEEE International Electron Devices Meeting (IEDM), pp. 169-172, New York, USA 1996.
18Engh et al., "A self adaptive programming method with 5 mV accuracy for multi-level storage in Flash", pp. 115-118, Proceedings of the IEEE 2002 Custom Integrated Circuits Conference, May 12-15, 2002.
19Engineering Windows 7, "Support and Q&A for Solid-State Drives", e7blog, May 5, 2009.
20Goodman et al., "On-Chip ECC for Multi-Level Random Access Memories," Proceedings of the IEEE/CAM Information Theory Workshop, Ithaca, USA, Jun. 25-29, 1989.
21Gotou, H., "An Experimental Confirmation of Automatic Threshold Voltage Convergence in a Flash Memory Using Alernating Word-Line Voltage Pulses", IEEE Electron Device Letters, vol. 18, No. 10, pp. 503-505, Oct. 1997.
22Han et al., "An Intelligent Garbage Collection Algorithm for Flash Memory Storages", Computational Science and Its Applications-ICCSA 2006, vol. 3980/2006, pp. 1019-1027, Springer Berlin / Heidelberg, Germany, May 11, 2006.
23Han et al., "CATA: A Garbage Collection Scheme for Flash Memory File Systems", Ubiquitous Intelligence and Computing, vol. 4159/2006, p. 103-112, Springer Berlin / Heidelberg, Aug. 25, 2006.
24Hong et al., "NAND Flash-based Disk Cache Using SLC/MLC Combined Flash Memory", 2010 International Workshop on Storage Network Architecture and Parallel I/Os, pp. 21-30, USA, May 3, 2010.
25Horstein, "On the Design of Signals for Sequential and Nonsequential Detection Systems with Feedback," IEEE Transactions on Information Theory IT-12:4 (Oct. 1966), pp. 448-455.
26How to Resolve Bad Super Block: Magic Number Wrong"in BSD", Free Online Articles Director Article Base, posted Sep. 5, 2009.
27Huffman, A., "Non-Volatile Memory Host Controller Interface (NVMHCI)", Specification 1.0, Apr. 14, 2008.
28International Application PCT/IL2007/000575 Search Report dated May 30, 2008.
29International Application PCT/IL2007/000576 Search Report dated Jul. 7, 2008.
30International Application PCT/IL2007/000579 Search report dated Jul. 3, 2008.
31International Application PCT/IL2007/000580 Search Report dated Sep. 11, 2008.
32International Application PCT/IL2007/000581 Search Report dated Aug. 25, 2008.
33International Application PCT/IL2007/001059 Search report dated Aug. 7, 2008.
34International Application PCT/IL2007/001315 search report dated Aug. 7, 2008.
35International Application PCT/IL2007/001316 Search report dated Jul. 22, 2008.
36International Application PCT/IL2007/001488 Search report dated Jun. 20, 2008.
37International Application PCT/IL2008/000329 Search report dated Nov. 25, 2008.
38International Application PCT/IL2008/000519 Search report dated Nov. 20, 2008.
39International Application PCT/IL2008/001188 Search Report dated Jan. 28, 2009.
40International Application PCT/IL2008/001356 Search Report dated Feb. 3, 2009.
41International Application PCT/IL2008/001446 Search report dated Feb. 20, 2009.
42Jedec Standard JESD84-C44, "Embedded MultiMediaCard (eMMC) Mechanical Standard, with Optional Reset Signal", Jedec Solid State Technology Association, USA, Jul. 2009.
43Jedec, "UFS Specification", version 0.1, Nov. 11, 2009.
44Jung et al., in "A 117 mm.sup.2 3.3V Only 128 Mb Multilevel NAND Flash Memory for Mass Storage Applications," IEEE Journal of Solid State Circuits, (11:31), Nov. 1996, pp. 1575-1583.
45Kang et al., "A Superblock-based Flash Translation Layer for NAND Flash Memory", Proceedings of the 6th ACM & IEEE International Conference on Embedded Software, pp. 161-170, Seoul, Korea, Oct. 22-26, 2006.
46Kawaguchi et al. 1995. A flash-memory based file system. In Proceedings of the USENIX 1995 Technical Conference , New Orleans, Louisiana. 155-164.
47Kim et al., "Future Memory Technology including Emerging New Memories", Proceedings of the 24th International Conference on Microelectronics (MIEL), vol. 1, pp. 377-384, Nis, Serbia and Montenegro, May 16-19, 2004.
48Kim et al., "Multi-bit Error Tolerant Caches Using Two-Dimensional Error Coding", Proceedings of the 40th Annual ACM/IEEE International Symposium on Microarchitecture (MICRO-40), Chicago, USA, Dec. 1-5, 2007.
49Lee et al., "Effects of Floating Gate Interference on NAND Flash Memory Cell Operation", IEEE Electron Device Letters, vol. 23, No. 5, pp. 264-266, May 2002.
50Maayan et al., "A 512 Mb NROM Flash Data Storage Memory with 8 MB/s Data Rate", Proceedings of the 2002 IEEE International Solid-State circuits Conference (ISSCC 2002), pp. 100-101, San Francisco, USA, Feb. 3-7, 2002.
51Micron Technology Inc., "Memory Management in NAND Flash Arrays", Technical Note, year 2005.
52Mielke et al., "Recovery Effects in the Distributed Cycling of Flash Memories", IEEE 44th Annual International Reliability Physics Symposium, pp. 29-35, San Jose, USA, Mar. 2006.
53Numonyx, "M25PE16: 16-Mbit, page-erasable serial flash memory with byte-alterability, 75 MHz SPI bus, standard pinout", Apr. 2008.
54Onfi, "Open NAND Flash Interface Specification," revision 1.0, Dec. 28, 2006.
55Panchbhai et al., "Improving Reliability of NAND Based Flash Memory Using Hybrid SLC/MLC Device", Project Proposal for CSci 8980-Advanced Storage Systems, University of Minnesota, USA, Spring 2009.
56Park et al., "Sub-Grouped Superblock Management for High-Performance Flash Storages", IEICE Electronics Express, vol. 6, No. 6, pp. 297-303, Mar. 25, 2009.
57Phison Electronics Corporation, "PS8000 Controller Specification (for SD Card)", revision 1.2, Document No. S-07018, Mar. 28, 2007.
58SD Group and SD Card Association, "SD Specifications Part 1 Physical Layer Specification", version 3.01, draft 1.00, Nov. 9, 2009.
59Serial ATA International Organization, "Serial ATA Revision 3.0 Specification", Jun. 2, 2009.
60Shalvi, et al., "Signal Codes," Proceedings of the 2003 IEEE Information Theory Workshop (ITW'2003), Paris, France, Mar. 31-Apr. 4, 2003.
61Shiozaki, A., "Adaptive Type-II Hybrid Broadcast ARQ System", IEEE Transactions on Communications, vol. 44, Issue 4, pp. 420-422, Apr. 1996.
62ST Microelectronics, "Bad Block Management in NAND Flash Memories", Application note AN-1819, Geneva, Switzerland, May 2004.
63ST Microelectronics, "Wear Leveling in Single Level Cell NAND Flash Memories," Application note AN-1822 Geneva, Switzerland, Feb. 2007.
64Suh et al., "A 3.3V 32Mb NAND Flash Memory with Incremental Step Pulse Programming Scheme", IEEE Journal of Solid-State Circuits, vol. 30, No. 11, pp. 1149-1156, Nov. 1995.
65Super User Forums, "SD Card Failure, can't read superblock", posted Aug. 8, 2010.
66Takeuchi et al., "A Double Level VTH Select Gate Array Architecture for Multi-Level NAND Flash Memories", Digest of Technical Papers, 1995 Symposium on VLSI Circuits, pp. 69-70, Jun. 8-10, 1995.
67Takeuchi et al., "A Multipage Cell Architecture for High-Speed Programming Multilevel NAND Flash Memories", IEEE Journal of Solid State Circuits, vol. 33, No. 8, Aug. 1998.
68U.S. Appl. No. 11/945,575 Official Action dated Aug. 24, 2010.
69U.S. Appl. No. 11/949,135 Official Action dated Oct. 2, 2009.
70U.S. Appl. No. 11/957,970 Official Action dated May 20, 2010.
71U.S. Appl. No. 11/995,801 Official Action dated Oct. 15, 2010.
72U.S. Appl. No. 11/995,812 Office Action dated Oct. 28, 2011.
73U.S. Appl. No. 11/995,813 Official Action dated Jun. 16, 2011.
74U.S. Appl. No. 11/995,814 Official Action dated Dec. 17, 2010.
75U.S. Appl. No. 12/019,011 Official Action dated Nov. 20, 2009.
76U.S. Appl. No. 12/037,487 Office Action dated Jan. 3, 2012.
77U.S. Appl. No. 12/037,487 Official Action dated Oct. 3, 2011.
78U.S. Appl. No. 12/045,520 Official Action dated Nov. 16, 2010.
79U.S. Appl. No. 12/063,544 Office Action dated Dec. 14, 2011.
80U.S. Appl. No. 12/119,069 Office Action dated Nov. 14, 2011.
81U.S. Appl. No. 12/171,797 Official Action dated Aug. 25, 2010.
82U.S. Appl. No. 12/178,318 Official Action dated May 31, 2011.
83U.S. Appl. No. 12/186,867 Office Action dated Jan. 17, 2012.
84U.S. Appl. No. 12/251,471 Official Action dated Jan. 3, 2011.
85U.S. Appl. No. 12/251,471, filed Oct. 15, 2008.
86U.S. Appl. No. 12/323,544 Office Action dated Dec. 13, 2011.
87U.S. Appl. No. 12/332,368 Office Action dated Nov. 10, 2011.
88U.S. Appl. No. 12/344,233 Official Action dated Jun. 24, 2011.
89U.S. Appl. No. 12/388,528 Official Action dated Nov. 29, 2010.
90U.S. Appl. No. 12/405,275 Official Action dated Jul. 29, 2011.
91U.S. Appl. No. 12/497,707 Official Action dated Sep. 15, 2010.
92U.S. Appl. No. 12/534,893, filed Aug. 4, 2009.
93U.S. Appl. No. 12/534,898 Official Action dated Mar. 23, 2011.
94U.S. Appl. No. 12/534,898, filed Aug. 4, 2009.
95U.S. Appl. No. 12/551,567 Office Action dated Oct. 27, 2011.
96U.S. Appl. No. 12/551,567, filed Sep. 1, 2009.
97U.S. Appl. No. 12/551,583, filed Sep. 1, 2009.
98U.S. Appl. No. 12/558,528, filed Sep. 13, 2009.
99U.S. Appl. No. 12/579,430, filed Oct. 15, 2009.
100U.S. Appl. No. 12/579,432, filed Oct. 15, 2009.
101U.S. Appl. No. 12/607,078, filed Oct. 28, 2009.
102U.S. Appl. No. 12/607,085, filed Oct. 28, 2009.
103U.S. Appl. No. 12/618,732 Office Action dated Nov. 4, 2011.
104U.S. Appl. No. 12/649,358, filed Dec. 30, 2009.
105U.S. Appl. No. 12/649,360 Official Action dated Aug. 9, 2011.
106U.S. Appl. No. 12/649,360, filed Dec. 30, 2009.
107U.S. Appl. No. 12/649,382 Office Action dated Jan. 6, 2012.
108U.S. Appl. No. 12/688,883, filed Jan. 17, 2010.
109U.S. Appl. No. 12/728,296, filed Mar. 22, 2010.
110U.S. Appl. No. 12/758,003, filed Apr. 11, 2010.
111U.S. Appl. No. 12/822,207, filed Jun. 24, 2010.
112U.S. Appl. No. 12/880,101, filed Sep. 12, 2010.
113U.S. Appl. No. 12/890,724, filed Sep. 27, 2010.
114U.S. Appl. No. 12/963,649, filed Dec. 9, 2010.
115U.S. Appl. No. 12/987,174, filed Jan. 10, 2011.
116U.S. Appl. No. 12/987,175, filed Jan. 10, 2011.
117U.S. Appl. No. 13/021,754, filed Feb. 6, 2011.
118U.S. Appl. No. 13/047,822, filed Mar. 15, 2011.
119U.S. Appl. No. 13/069,406, filed Mar. 23, 2011.
120U.S. Appl. No. 13/088,361, filed Apr. 17, 2011.
121U.S. Appl. No. 13/170,202, filed Jun. 28, 2011.
122U.S. Appl. No. 13/171,467, filed Jun. 29, 2011.
123U.S. Appl. No. 13/176,761, filed Jul. 6, 2011.
124U.S. Appl. No. 13/192,495, filed Jul. 28, 2011.
125U.S. Appl. No. 13/192,501, filed Jul. 28, 2011.
126U.S. Appl. No. 13/192,504, filed Jul. 28, 2011.
127U.S. Appl. No. 13/192,852, filed Aug. 2, 2011.
128U.S. Appl. No. 13/214,257, filed Aug. 22, 2011.
129U.S. Appl. No. 13/231,963, filed Sep. 14, 2011.
130U.S. Appl. No. 13/239,408, filed Sep. 22, 2011.
131U.S. Appl. No. 13/239,411, filed Sep. 22, 2011.
132U.S. Appl. No. 13/284,909, filed on Oct. 30, 2011.
133U.S. Appl. No. 13/284,913, filed on Oct. 30, 2011.
134U.S. Appl. No. 13/338,335, filed on Dec. 28, 2011.
135U.S. Appl. No. 13/355,536, filed on Jan. 22, 2012.
136Ubuntu Forums, "Memory Stick Failed IO Superblock", posted Nov. 11, 2009.
137US 7,161,836, 01/2007, Wan et al. (withdrawn).
138Wei, L., "Trellis-Coded Modulation With Multidimensional Constellations", IEEE Transactions on Information Theory, vol. IT-33, No. 4, pp. 483-501, Jul. 1987.
139Wu et al., "eNVy: A non-Volatile, Main Memory Storage System", Proceedings of the 6th International Conference on Architectural support for programming languages and operating systems, pp. 86-87, San Jose, USA, 1994.
Klassifizierungen
Internationale KlassifikationG11C8/20, G11C11/56, G11C27/00, G06F11/30, G06F11/10
UnternehmensklassifikationG11C11/5642, G11C27/005, G11C8/20, G11C11/56, G06F11/1068