CA2479977C - Data communication protocol in an automatic meter reading system - Google Patents

Abstract

An automatic meter reading (AMR) system includes a fixed or mobile reader and an endpoint. The endpoint is interfaced to a utility meter and the fixed or mobile reader is capable of communicating with the endpoint via RF communication. In this system the fixed or mobile reader sends a message to the endpoint that includes a response mode direction; the response mode direction from the reader tells the endpoint to respond to the reader either in a mobile network mode or a fixed network mode.

Description

DATA COMMUNICATION PROTOCOL IN AN
AUTOMATIC METER READING SYSTEM
FiF.I.~ of THE uwENT'ION
The present invention relates to automatic metes reading systems ate, morn particularly, to the communication protocol used far the endpoint to reader hop in the automatic meter reading system.

Current au~c metex reading (AMR) systems are significantly limited in the information that can be obtained from the meter. Generally the AMR system comprises a reader and an endpoint that is interfaced to a meter. In a typical system, the endpoint obtains the consumption reading from the meter and then bubbles up every few seconds to send that consumption reading, via RF signal, to the reader. Alternatively, the endpoint receives a wake-up tone from the reader that prompts the endpoint to send the consumption reading to the header.
All that is obtained from this configuration is a single consumption reading from the meter and that reading is based on what meter register the endpoint was progranutted with initially at the factory.
As such, there is a need for an AMR system that enables the user of the system to have mane access to and itmre control over the information that die meter and endpoint can provide.
SUMMARY OF THE INVENTION
The present invention is a data communication protocol used between an endpoint and a reader in an automatic meter reading (AMR) system. The comraunication protocol enables the reader to have a conversation with the endpoint in that the reader can tell the meter what to do, it can reconfigure the meter, it can tell the endpoint to reconfigure the meter, it can request a specific response, it can request the endpoint to reprogram certain values in both the endpoint and the meter, it can request that the end point get specific information from the meter, return it to the end point, which returns it to the reader, eta In a preferred embodiment of the present invention., an automatic miter reading (AMR) system includes a fixed or mobile reader and an endpoint. The erulpoint is interfaced to a utility meter and the fixed or mobile reader is capable of communicating with the endpoim via RF
communication. 1n this system the fixed or mobile reader sends a message to the endpoint that includes a response mode direction; the response mode direction from the reader tells the endpoint to isspond to the reader either in a mobile network mode ~ a fixed network mode.
1n another erabodiment of the invention, the reads uses a single RF
communication ZO protocol in communicating with the endpoint whether one-way communication or two-way communication is used In still another embodiment of the invention, the RF
communication between the endpoint and reader occurs through the use of a communication protocol that uses a data link layer that directs all outboutrd transmissions from the reader to the endpoint to be either Manchester encoded or transmitted as non-return to zero data. In still another embodiment of the invention, the communication protocol includes a transport layer that provides slotting control for all data transferred between the reads and the endpoint. In yet another embodirae~ of the present invention, the communication protocol utilizes a comanand and control frame for communication between the reader and endpoint. In still another embodiment of the present invention, a plurality of readaa and et~points an provided. The i,eadas are qaasi-synchronized in time to provide a control frame to at least one of the rte.
BRIEF DESCRIPTION OF ~ DRAWINGS
Fig. 1 depicts a radio-based auoomatic meter reading system that utilizes the data communication ~xotocol of the present invention.
Fig. 2 is a table containing the physical layer specifications of the reader.

Fig. 3A is a table containing the physical Iayer specifications of the endpoint at data note 1.
Fig. 3B is a table contaitung the physical layer specification of the endpoint at data rats 2.
1~ig. 4 is a table containing the physical lays spoc~cations of the endpoint in a one-way AMR system Fig. 5 is a diagram of a Manchester encoding structure.
Fig. 6 is an example of a Sequence Inversion Keyed Countdown Timer.
Fg. 7 diagrams the data packet srnxx~u~e.
Fig. 8 diagrams a high power pulse data packet structure.
Fig. 9A diagrams a two-way command and control txante.
Fig. 9H diagrams a one-way commsad and control frame.
Fig. 10 is a table containing universal command types for the data communication protocol of the present invention. .
Fig. 11 is a table containing type spxific contmsnds for the data communication protocol of the present invention Fig. 12 diagrams crommand 48 of the data communication protocol, Multiple Ungtouped Endpoint Command.
Fig. 13 diagrams conunand 49 of the data communication protocol, Vector and Listen Fig. 14 diagrams contntand 50 of the data communication pnxocoi, Multiple Commands to Individual Endpoint Fig. 15 is a diagram of the channel spectrum of the system.
Fig. 16 is example of a timing diagram for a staged wakeup sequence for a three cell reuse pattern.
Fig. 17 is an ezample of a three-cell cellular reuse pattern.
Fig. 18 is an example of a four-cell cellular reuse pattern.

Fig. 19 is an example of a five-cell cellular reuse pattern.
Fig. 20 depicts mobile operation of the system over five chancels.
Fig. 2I depicts coverage rings.
DETAIL» DESCRIPTION OF THE PREFERRED EMBODIMENTS
'ibe present invention is a data communication protocol for automatic meter reading (AMR) systems. The protocol is designed to be flexible and expandable enabling both one-way and two-way meter reading in both fixed and mobile meter reading systems.
I. ~~gnn Components 1n an AMR system 100, as depicted in Fig. 1, that is utilized with the present invention, the components generally include a plurality of telemetry devices including, but not limited to, electric esters 102, gas meters 104 and water meters 106. Each of the meters may be either electrically or battery powered. The sysoem further includes a plurality of endpoints 108, wherein each co~rraesponds and interfaces to a meter. Each of the endpoints 108 preferably incorporates a radio ieceiver/transmitter, e.g., the Itron, Ins. ERT. The system additionally includes one or more readers that tray be fixed or mobile, Fig. 1 depicts: ( 1) a mobile hand-held reader 110, such as that used in the Itron Off sibs meter reading system; (2) a mobile vehick-equipped reader 112, such as that used in the Itron Mobile AM<t system; (3) a fated radio communication network 114, such as the Itron Fixed Network AMIL system that utilizes the additional components of cell central control units (CCUs) and network control nodes tNCNs);
and (4) a fixed micro-network system, such as the Itmn MicroNetwork AMR system that utilizes both radio communication through concentrators and telephone communications through PSTN.
Of course other types of racers may be used without. departing from the spirit or scope of the invention. Further izscluded in AMR system 100 is a head-end, host processor 118. The host processor incorporates software that manages ttx collection of metering data and facilita0es the transfer of that data to a utility or supplier billing system 120.
The .4h9t system 100 and the data protocol is usable in both ~e-way meter reading and in two-way meter reading. The one-way meter reading system enables the reader to listen to messages sent asynchronously from the endpoint while the two-way meter reading system enables the reader to communicate with and command the endpoint while also enabling the endpoint to respond to the reader.
~y'~(~~ol ZO The present communication protocol will be described with reference to the 1430 MHz band that may be utilized within North America, however, it should be understood that any other radio frequency band may be used, as suitable, without departing from the spirit or scope of the invention. The present communication protocol will also be described with reference to the Open Systems Interconnection (OSI) protocol stack of the International Starslards Organization is which includes: (1) the physical layer; (2) the data link layer; (3) the network layer; (4) the transport layer, (5) the session layer; (5) the presattation layer, and (7) the application layer.
ILA. Svstem Protoap~ - Physics 1 gyer The physical layer describes the physical characteristics of the communication. This 20 layer conveys the bit smcam through the network at the electrical and mechanical level. It provides the hardware means of sending and receiving data on a cartier. The physical layer specifications for the reader may be found in l ig. 2 wherein: (1) the operational modes; (2) the frequency band; (3) the channel bandwidth; (4) the modulation scheme; (5) the deviation; (5) the encoding; (7) the bit rate; ($) the frequency stability; (9) the minimum reception sensitivity; ( 10) 25 the transmission power, (11) the preamble length; and (12) the transmission modes are Ixovided.
s The physical layer specefrcation for the endpoint in a two-way AMR system, at a first data rate and a second data rate, are found in the tables of Fig. 3A and Fg.
3B, respectively. The specifications provided include: ( 1 ) the operational modes; (2) tlu frequency band; (3) the channel bandwidth; (4) the modulation scheme; (5) the deviation; (6) the encoding; f n the bit rate: (8) the fr~xr~y stability: (4) the minimum reception sensitivity: ( 10) the minimum preamble length; and (11) the factory default freqixncy. The physical layer specification for the endpoint in a one-way AMR system is provided, similarly, in the table of Fig.
4. However, it should be understood that any other physical layer specifications may be used, as suitable, without departing from the spirit or scope of the invention.
ILB. System atocol - Data Iank Laver The data link layer specifies how packets are transported over the physical layer, including the framing, i.e., the bit patterns that mark the start and end of packets. This layer provides syrsyhronizatioa for the physical level. It furnishes transmission protocol knowledge and management. In the present data communication protocol. all outbound data transmissiods, ie., all communications frown the remder's central radio to endpoint, are Manchester encoded with the guaraatad transition mid-bit and each data bit encoded as aesa(bar)-(Sx Fg. 5 fnr the Manchester Encoding Structure). Inbound tranamissiorts from the endpoint are either transmitted as Manchester encoded data, idlenacal to otransmissions. or are transmitted as NRZ
(non-return to zero) data. Selection is b~nsed on the value of the MCH flag in the command and control frame.
The data link layer provides a countdown timer. The countdown timer uses Sequence Inversion Keying to represent timer bits. F.aclt system is assigned a 10~bit pseudo noise (PIE
sequence (for valid sequences, see Table 1 below). That sequence in the data stream represents a timer bit value 0 and the inverse of that sequence in the data stream represents a tuner bit value 1. Timer values are composed of 10 timer bits, or 100 data bits. The countdown timer begins at 1023, or 11111 I 1 I 11 binary, and counts sequentially to zero, encoding all timer bits as either the system PN sequence or its inverse. The total counter time, in seconds, is 102400Jr, where r is the bit rate, in bits per second. Fig. 6 provides an example of a Sequence Inversion Keyed Countdown Timer.
Table 1. PN Sequences Sequence Usage Seqaenc~0 inverted Number Se eoce=1 0 Facto Default0000000010 1111111101 1 Electric 0000000110 1111111001 Devices 2 Electric 0000001010 11 I 1 I 10101 Devices 3 Electric 0000001110 1111110001 Devices 4 Electric 0000011010 1111100101 Devices 5 Electric 0000010110 1111101001 Devices 6 Electric 0000111010 11 i 1000101 Devices 7 Ba Devices 0000101110 1111010001 8 Ba Devices 0001110110 1110001001 9 Ba Devices 0001101110 1110010001 Ba Devices 0000011110 1111100001 11 Bane Devices0001011110 1110100001 12 Ba Devices 0001111010 1110000101 10 All inbound lmcket transmissions are prxeded by a 24-bit or ZS bit prramble and appended with a lb-bit CRC code, which is i~lusive of all header information, but not the preamble, length, or lengthbar bytes. The CRC polynomial is 0x1021. The CRC initialization value is 0x0000.
CRC processing is perfornued most significant byte (MSH) first, and the final checksum is not inverted.
ILC. Stem Protocol -Network Laver 'I'l~ network layer specifies how packets get from the source network to the destination network. This layer handles the roetting of the data (sending it in the right direction to the right destination an outgoing transmissions and recxiving incoming transmissi~s at the packet level).
The network layer does routing and forwarding. In the present data communication protocol, the network layer functionality is only implemented in electric endpoints, ix., it is not used for battery-powered endpoints, or in any endpoint that acts as translator or repeater. This layer controls the hopping functions that need to occur between a reader and any erulpoint in order to transfer data. This hopping protocol is currently used within the Itron AMR
systems and is S therefore not described in detail herein.
~~~,, Svstem Protocol-Trarts~Ort I,~lrer The transport layer is used to solve prnblems like reliability ("did the data reach the destination?") and ensure that data arrives in the contct order. This layer manages the end-to end control (for example, determining whether all packets have arrived) and error-checking. It ensures complete data txansfer. 1n the present data communication protocol, slotting control is handled in the transport layer. This inclu~s slot assignments, timing, and any necessary packetixation. Fig. 7 details the packet stxucture. The message, message type, and flags are received from the presentation layer, and broken into appropriately sized packets. Each packet is prefaced with the endpoint >D, flags, message type, endpoint type, and packet length. The packet length reflects the number of bytes in the message itself, exclusive of header information.
In the case whetre more than 254 bytes are required in a packet, the value of the length field is set to OxFP, and the actual length of the message structure is placed in bytes 14 (high byte) and 15 (Iow byte), with the message bytes to follow. All packets must have a whole number of bytes in the message.
The packet number byte, when used as part of the tixssagc, is configured as below in Table 2, wherein the first four bits comprise the total number of packets in this message and the last four bits oomprisc the packet number.
Ts~bte 2. Puget Number T T T ~T N ~N N N
MSH LSB

The flags byte is configured as below in Table 3. The first two bits are rZSaved while the xcond two bits provides the encoder number (for niulti-encoder units), whet~ein 00 = encoder 0, Ol = encoder 1, 10 = encoder 2, and 11 = encoder 3. The fifth bit signifies the stenos of a pending event, wherein 0 = no pending event and 1 = a pending event. The sixth bit comprises the security bit, wherein 0 = security disabled and 1 = security enable. The xventh bit comprises the relay bit, wherein 0 = message from originating endpoint and 1 = message via relay. The eighth bit comprises the rexnd bit, wherein 0 = first attempt at packet transmission and 1 =
resend attempt.

Table 3. Flab R R ENC ENC EVT SEC RLY RSD
MSH LSB
Some endpoints in the system have the option of sending out an infrequent (several times a day) fixed forn~at message at a higher power level, for use in 1~way fined network applications.
Ttu; message has its own structure., as defured in Fig. 8. The enstom packet is then 1BCH (255, 139, 15) encoded, prior to transmission. The encoding polynomial is 0x461407132060175561570722730?~7453567445s . For multi-encoder endpoints this packet is generamd a~ xnt for each individual encoder. The flags for the high power pulse data packet structure are configured as shown in Table 4 below. The first four bits are re$erved while the fifth and bits provide the encoder number, wlrenein 00 s encoder 0, Ol =
enood~ 1, 10 =
Decoder 2, and 11 = encoder 3. The seventh bit comprises the relay bit, wherein 0 = messs~ge from originating endpoint and 1 = message via relay. The eighth bit comprises the error code indicating that a critical endpoint error has occutmed.
Table 4. Flags R ~ R INC ENC RLY ERR
MSB
The endpoints may also be set to send out any preprogrammed message type in place of the fined format message described above.

ILE. Sy~~;m Protocol - Session Layer The session layer sets up, coordinates, and terminates conversations, exchanges, and dialogs between the applications at each end It deals with session and connection coordination.
In the present data communication protocol, the sesgion-layet generally-comprises the command and control frame that is sent from the reader to the endpoint.
IIl E i~S~tem~ptocol - Sossion Laycrll'wo-Wa7~Command and Control The command and control frame is used to issue corrunand to two-way endpoints either individually or in groups. It also serves to realign the endpoint real-time clock. Fig. 9A
diagrams the two-way communication command and control frame. As shown, the command and control frame transmission is preceded by a 24-bit preamble, as indicatod by the three "P"
fields within the fimne. The first 16 bits are preferably an alternating pattern, AAAAh, and are used for clock recovery. The last 8 bits are used for frame and timing synchronization.
Field "0" of the command and control frame comprises the system i~ntification (iD).
Each system is issued an 8-bit 1D value, which is stored in the endpoint, to distinguish different systems within geographic proximity. The endpoints arc designed to respond to commands from their own system or to commands that address them specifically by lm number, proper security password, and have a 0x00 in field "0". The system m functions nearly identically to the cell 1D, described below. However, the system 1D is universal, while the cell m is local, i.e., a single system will have multiple cells each having the same system 1D but a different ceil 1D.
Field "1" of the command control frame comprises the frame >D. Each reader within the system is assigned a frame )D to use based on its position in the wake-up sequence. The position in the wakeup sequence is directly related to the frequency reuse pattern that is used in a given system. Table 1, described earlier, correlates the frame ID to the channel, which is correlated to the cell reuse ratio.
)~ield "2" of the command and control frame comprises the cell 1D. Each cell is issued an 8-bit 1D value, which is stored in the endpoint, to distinguish different systems within geogaphic prozimity.
Fieids "3" through "6" of the command and control frame is the RTC, which is defined as UTC time (coordinat!ed universal time), which is a 32-bit value representing the number of seconds since midnight (00:00;00) on January 1, 1970 GMT.
Field "7" is the command flags I field, wherein the first three bits define a slot length according to Tabk 5.
TABLE 5. Slot Lengths Valve of L Bite Nominal Len in Nominal L in ms Tlclrs*

000 819 24.99390 001 1638 49.98779 010 3277 100.00610 Oll 6553 199.98169 100 9830 299.98780 101 16384 500.00000 i 10 32768 1000.00000 111- __~ _ 5000.00000 .163840 ~Laerinea as ttcJCS of an ldenl 32,768 Hz Clock.
The fourth bit is the forward error ion bit, wherein 0 = no forward correction en~or and 1=
forward error correct all responses. The fifth bit provides the slot mode, wherein 0 = respond to command in pseudo-random slot (Slotted Aloha) and 1 = respond to command in the defined slot. The sixth bit of field "T' defines the data type, wherein 0 = NRZ
response from endpoint and 1 = Manchester encoded from the endpoira. The seventh and eighth bits of field "T' comprise the command target, wherein 00 =- the entire cell, Ol = the group defined in FPm~HI
(field "12'x, 10 = the goup defuud in EPID~.O (field "15"), and 11 = the endpoint defuiod by I..PID (including HI/i.0), fields "12" through "15". It should be noted that in single endpoint communications the command target (TGT) is set to 11 and tl~ endpoint responds immediately after command processing with a minimum of 25 milliseconds between this frame and the endpointresponse.
F'~eld "8" of the command and control fiante is the command flags 2 field, wherein the first four bits are reserved. The fifth and sixth bits defined the encoder number, wherein 00 =
Encoder 0, Ol = Encoder 1, 10 = Encoder 2, and 11= Encoder 3. The final seventh and eighth bits define the transmit mode, wherein 00 = transmit mode 1, e.g., mobile response rcquin~, Ol = transmit mode 2, e.g., fixed network response required, and 10lI1 arc reserved. Also see section V below.
Field "9" of the conunand and control frame comprises the slot offset. Slot offset defines the number of slots between packets in mufti-packet messages. For example, if the endpoint has an initial slot number of 50, and the slot offset is 120, a three-packet message would be transmitted in slots 50, 170, and 290.
Fields "10" and "11" of the command and control frame define the first unsolicited message. Specifically, they define the slot number where the unsolicited messages (LJMs) are to l5 begin. Any UMs genuatEd during tire cell nad would be reported in a pseudo-randomly selectaed slot after the slot defined here. If the value of this field is 0x0000, no UMs arc sent fra~m the endpoir>t.
Fields "12" through "15" of the command and control frame provide the endpoint fi4s far those endpoints that the r~eada is desiring to co~muoicate with.
Fields "16" and "17" am the security fields and are described fmd~cr in rotation to the pt~sentation layer.
Field "18", defines tire commsrtd set. The comtttands are divided into two groups: (t) universal and (2) type-spxifrc. Universal commands ara: nurabered 0-63 and are applicable W
all the system endpoint4. Type specifrc commands are numbertd tS4-255 and vary depending on the lower nibble of the commend set field in accorda~nx with Table 6 below.

Table 6. Command Sets Comrrwnd Set CDS Value Usaee 0 (default) 0000 , UW' ity Meterin End oints 1 0001 tens and Translators 2 0010 Teteme Devices 3-14 0011-1110 Reserved 15 1111 Reserved En 'neerin Use Unl >

Fields "19" through "21" of the command and control frame define the command and command body. Specifically, the eight command bits of field "19" indicate the command type, wherein the numbers 0-63 are universal corcunands and b4-255 are the type specific commands.
Fields "20" and "21" provide sixteen bits wherein any data needed to carry out the command type is provided. The tables in Figs. 10 and 11 indicate the comrnaud types and command bodies that are possible with the system of the present invention. lZefernng to tine universal commands Wig. 10), it can be seen that the present system is capable of but not limited to: (1) reporting a status; (2) changing a system number to a new system number; (3) changing a group number to a new group number, (4) changing a system slot number to a new system slot number; (5) changing the cell m to a new cell 1D; (5) reporting slot numbers; (7) resending identified packets of data; (8) setting the receiver bubble-up period; (9) setting the bubble-up charu~el; (10) setting the bubble-up time; (11) configuring the transmission power; (12) setting the channel frequency;
etc.
Refe~ning to the type specific commands (Fig. 11), nunxnous other commands are available including but not limited to: (1) reporting consumption data; (2) reporting time of use (TOU) data; (3) reporting logged data; (4) reporryng temperature; (5) reporting tamper data: (6) setting configuration flags: (7) initializing consumption; (8) reporting an event summary; (9) performing an endpoint diagnostic check; ( 10) reporting memory contents; etc.
Fields "22" and "23" of the command and control frame designate the response frequency for the endpoint. The response frequency is configured as 16 bit flags, identifying valid response frequencies for the endpoint. For example, if the response frequency has a value of Ox00C 1 (bits, 7, 6, and 0 are set), the endpoint may respond on channel, 7, channel 6, or channel 0.
Fxld "24" is reserved for later use.
Field "25" indithe length of the extcndcd control frame in bytes. A value of 0 indicates that no extended frame is presenk Fields "26" and "27" of the command and control frame provides the cyclic redundancy check (CRC). Specifically, fields "26" and "ZT' provide a I6-bit CRC. The CRC
is preferably a polynomial defined as 0x1021. The CRC initialization value is 0x0000. CRC
processing is performed most significant bit (MSB) first, and the final checksum is not inverted.
ILE.ii. System Protocol - Ses3~on LayerlOne-War Command and Control Far simplicity one-way devices may opt to use the programming frame shown in Fig. 9B.
The command and command body bytes are similar to that described above with reference to the two-way devices. The byte for number of commands provides the total number of commands to follow in this frame, with a maximum value of 8. The command flags are diagratnrncd in Table 7 below. The first two bits indicate the transmit mode, wherein 00 = transmit mode 0, 01 =
transmit mode I, and IQII 1 are reserved. The third bit designates the data logging, wttenein 0 =
data logging is disabled and 1 = data logging is enabled. The fourth bit designates the forward error correction, wherein 0 = disable forward error correction on response and 1 = enable forward error correction on response. The fifth and sixth bits designate the mode set, wherein 00 = stock mode, 01 = test mode, 10 = reserved mode, and 11 = normal mode. The seventh and eighth bits are reserved Table 7. d Flags TXM T'XM DLG FEC MDE 11~E R R
MSB LSB

lLE.iii. System Protocol - Session Lower ISnecial Command-Channel Freouencx Certain of the commands provided in the command and control frame are described in detail below. For instance, Command 33, which is the set channel frequency.
Each of the system endpoints support up to 16 charnels, which are set individually. They may or may not be contiguous channels. The channel numbering differs based on frequency band.
For example, in the present innplttnentation of the invention, the 1427 -1432 MHz band is divided inoo 6.25 kHz frequency channels, with frequency channel 0 centered at 1427.000 MHz, freque~xy channel 1 centered at 1427.00625 MHz, etc. If endpoint channel 15 is programmed to a value of 480, that endpoint tnceiver will always operate at 142?.000 + (.00625 * 480) = 1430.000 MHz. This tray be extended to other frequency bands. For example, the 43335 MHz band is divided into 25 KHz frequency channels with the frequency channel 0 centered at 433.000 MHz, frequency chapel 1 centered at 433.025 MHz a~ so on.
The command body of the set channel frequency command is detailed below in Table 8:
TABLE 8. Command Body/Channel Frequency car CE~1 Cl'trT Get FR - FR FR FR FR FR FR FR FR PR
MSB LSB
Individual frcqaen~cies are pmogrammed iirto the endpoint by selecting the channel being programmed (1-15) wig the top nibble, and the frequency number in the lower 12 bits.
ZO Endpoint channel 0 is preferably the manufactuaing default frequency, and may not be odited.
Endpoint chsmiel 15 is the roceiver frequency. 1t is initialimed to the same fioquency as chaa~l 0 at manufacture, and is preferably programmed prior to or at installation.
The endpoint channel uses are defined in Table 9 below:
Table 9. Endpoint Channel Use End t C6anacl~ Charred Use 0 Facto~fauh. This channel is not re io ranunable.

1 General use Tx/Rx (Transtnission/Rece lion 2 General use Tx/Rx 3 General use TxlRx 4 G_enarat use Tx/Rx Genera( use Tx/Rx 6 General use Tx/Rx 7 General use Tx/Rx 8 General use TxIRx 9 General use Tx/R.x General use Tx/Rx 11 General use Tx/Rx 12 General use Tx/Rx 13 General use Tx/Rx 14 Default UM Channel (unsolicited mesas IS Default Rx Channel The configuration flag commands, i.e., commands 90, 91, and 92 are used for setting individual flags in the endpoints. Fach flag commend includes an 8-bit flag mask and an &bit flag as shown below (the configuration flags 1 command body):
5 Table Z0. Flag Mask MSK MSK MSK MSK MSK MSK MSK MSK
MSB I-SB
Table 11. Flags R R R TxB UMC FN FEC MMI
The flag mask field determines which flags are to be modified by this command A "I" in any bit position means the associated value in the flags field should be modified.
For example, A
value of 0x1? (bits 4,2,1 and 0 are high) means that the values in the Flags field, bits 4, 2, 1, and 0 must be written to the associated flags in the endpoint. With regard oo the flags field of Table i 5 7, the first thnde bits m~e reserved for future growth while the fourth bit, TxB, determines if the endpoint is in transmit bubble up mode, the fifth bit, ITMC, defuxs the unsolicited message channel, i.e., UMC = 0 then transmit UMs on Channel 14, and UMC = 1 then transmit UHis on channel 15. The sixth bit of the flags field defines the fixed network mode, wherein 0 = this endpoint operates in MobilelHandbeld mode only and I= this endpoint operates in mobileJhandheld/fixed network mode. The seventh bit of the flags field d~efu~s the forward error correction, wherein 0 = no forward error correction applied to the high power pulse and 1 =

forward error correction is applied to the high power pulse. The eighth bit of the flags feeld defines the multiple message integration, wherein 0 = no multiple message integration applied to high power pulse arxi I = multiple message integration applied to high power pulse.
ILE.iv. Sy~m Protocol - Session La~erl5necial Commands - Test Commands The present data communication protocol provides at least two commands for use in system testing and analysis. The fu~st command is command 210, i.e., Generate UM (unsolicited message). This command automatically generates an unsolicited message in all endpoirns addressed by the command and control frame. It generates the lowest numbered LJM supported by the endpoint. The second command is cornrnand 211, i.e., Enter Screaming Viking Mode.
Screaming Viking Mode is a constant transmission mode, to be used for test only. When this command is received, the endpoint repetitively transmits its m for the number of minutes declared in the command. If a value of0 is sent, the mode is active for 15 seconds.
ILE.v. S~ loco - Session Layy(~~ial Commands - Extension Commands Commands 48, 49, 50 and 51 of the data communication protocol are implemented as extensions roe the command end control frame. The extension commands immediately follow the command and control frame in the same transmit session. Command 48 is the multiple ungcouped endpoint command. In the case where the system needs to command a group of specific endpoints and vector them to specific slots, command 48 is issued The central radio then issues commands to these endpoints, as shown in Fig. 12. This carnmand can be used to address a maximum of 16 distinct endpoints. The packet length reflects the number of endpoints addressed by the message. Note that the conunand 48 may not be used for any command that requires the security password. The structure of command 48 provides for an 8-byte preamble having the value of OxAAAA AAAA AAAA AA9b, the length, the endpoint lDs, and the command bodies for tech of the endpoints and a response byte for each of the endpoints. The response byte is dia~d in Table 12 below:
Table 12. Respom9e Byte R. ~ R
MSB GSB
The response byte reserves the fast four bits and utilizes the last four bits to define the response frequency nibble. Specifically, the four bit flags define which of the pm-programnKd channels the endpoint may respond on. if CHN = 0000, then use the response frequency byte from the lU original command and control frame. The structure of the command 48 also includes the CRC as described earlier:
Command 49, i.e., the vector and listen frame, is issued in the instance where the central radio or reader need to download an arbitrary block of data to the endpoint.
The endpoint, upon receiving this command raxives a data frame, as defined in Fig. 13. This carnmand is valid only when the endpoints are individually addressed (i.e., TGT = 11). The data is endpoint-type specific. Note that the vector and lisOen frame has an $-byte preamble with a value of OxAAAA
AAAA AAAA AA96. Further. note that the packet length icfkcts the number of bytes in the message itself. exclusive of header information, and that the CRCs a~mputed over all bytes in the message body.
Command 50, the multiple caartmand to individual endpoint command, is used in the case where the central radio or rtader need to download a series of canunaads to one specific endpoint. The endpoint, upon receiving this command, receives a data frame as defined in Fig.
14. This command is only valid when the endpoints are individually addressed (i.e.,1'GT =11).
Up to 24 commands may be issued on as endpoint using this sttuctune. Note that the packet length reflects the number of commands to be issued within this structure.

Command 51, the Extended Frame Mobile Read command, uses the multiple ungrouped endpoint comma, structure, with a Slotted-ALOHA period between the extended fray and the queried response slots. All endpoints which recognize the command respond. If the endpoint is among those addressed by the extended frame, it responds as commanded, being offset by 16 slots. If the e~point is not specifically addressed it responds in the Slotted-ALOHA section with its programmed default message.
ILF. System Protocol - Presentation Layer The presentation layer, which is usually part of an operating system, converts incoming and outgoing data from one presentation format to another and it is sometimes called the syntax layer. In the present data communication protocol, the presentation layer handles data security and any necessary data compression and decompression.
The data security is preferably a simple two-level protocol, which may be enabled or disabled by the customer. Level 1 provides simple encryption for the transfer of normal data while level 2 provides write security to the endpoint to prevent unauthorized users from changing endpoint parameters.
Level 1 is intended for use on ordinary data being transmitted from the endpoint to the head end. All data is encrypted with a simple 8-bit XOR mask. The level 1 socurity enables flag and encryption mask and are editable by a level 2 parameter write. The factory default for the XOR mask is the bottom 8 bits of the serial number. Level 1 security is applied only to the message itself and not to the EPA, flags, or message type. Level 1 security may be disabled by setting the mask value to 0.
Level 2 security is intended for use on any head end conunands to change endpoint parameters. It includes modification of operational, security and reprogramming parameters.
Levet 2 functionality is independent and can be applied with or without Level 1 functions enabled. Each endpoint has a 16-bit password. This password is originally defir~d at install, and can be edited by a valid Level 2 command. Any write command must include the current password to be considered valid by the endpoint. For added security, the Level 1 erxryption mask may be applied to the password, if Level 1 functionality is active. There is no compression performed on packet data.
ILG. System Protocol - Application Laver The application layer is the layer at which communication partners arc identified, quality of service is ident~ed, user authentication and privacy are considered, and any constraints on data syntax are identified. (T'his layer is not the application itself, alth~gh some applications may perform application layer functions.) 1n the present data communication protocol, an endpoint application layer is used in conjunction with the application progranuning interface (APn. When data is requested by the presentation layer, via the API, the application layer performs its processing and returns the requested message as a single block, along with one 8-bit value. The value represents the message type.
BI. S~ste~ation The two-way AMR system of the present invention, at 1430 MHz, is designed to operate most effxiently in five contiguous RF channels. This allows the use of a cheaper (wider) receiver section in the endpoint while still maintaining the FCC mandated SO
KHz maximum transmit spectmm. The transmit spectrum in all devices, endpoints, arrd readers, raust maintain a 50 KHz or less occupied bandwidth during transmit. The receiver in the wader must also have a good selectivity on the channel of interest. The endpoint receiver is allowed to acecpt a wider receive bandwidth primarily to redtxe the cost of the endpoint.
Refer to Fig. 15 to observe the 250 KHz of spectmm allocated to the system. As shown, the spectrum is divided in to five 50 KHz channels. The center channel, i.e., channel 3, is designated as the control channel for the system 100. All endpoints 106 listen on this channel.

As such, if the readers art quasi-sy~hronized in their outbound transmissions the cancer channel approach allows the endpoints to use a wider r~eive bandwidth while avoiding the interference that would normally be a problem (synchronization is described in further detail below). 'fhe diagram of Fig. I5, illustrates the bandwidth differences graphically. Since the reader has good selectivity the endpoints can respond on a different channel in each cell simultaneously allowing the maximum data throughput in the system (cell re-use is described in further detail below).
By utilizing an appropriate RF AS1C, the architecture can be reduced to three contiguous channels with the reaming two or more channels scattered throughout the band to ease specwm allocation requirements. With a reduction in the interference protection to the end point, a completely xparated channel model could be used in an alternative configuration. I3owcver, in the separaoe channel model, the endpoint requires additional base band filtering and is still slightly more susceptible to adjacent channel interference on the control channel especially if operating in the high power portion of the band. The separate channel option also allows multiple control channels in the system when mobile operation is used with multiple outbound channels. When using the separate cltanntl model, channels 2 and 4, of a 5-channel block, are used for control signals.
To alleviate cell-to-cell interference in a system with a single control channel the readers must be synchmnizod in time so that the control frames, which are descn'bed in further detail below, do not overlap. The addition of "dead time" in between sequernial control frames allow for the receivers to be quasi-synchronized instead of in perfect lock step. 1n the preferred embodiment, quasi-synchronized means that the receivers are within 0.5 saccade of each other, which can easily by achieved via protocols such as NT'P (network time protocol). Other quasi-synchronization times may be used without departing from the spirit or scope of the invention.
As such, a GPS or other high accuracy time base is not required within the readers.
Within the AMR system, each reader is assigned a frame 1D to use based on its position in a wakeup sequence. The position in the wakeup sequence is directly related to the frequency reuse pattern used in a given system. The timings in the diagram of Pig. 1 fi are provided as an example of a staged walceup sequence for three cell reuse. As shown, the timings are for an endpoint ~ endpoint clock accuracy of +/ 0.5 seconds, if the value obtainable is only +/- l second then the dead time must be increased to 5 and the nominal frame time to 22.5 seconds.
All other titninga remain the same. If GPS is available in the reader, the dead time can be reduced and the time frame timing can be shortened. In any case, the minimum dead time is preferably 0.5 xconds.
As shown in Fig. 16, the first wake-up xquer~ce is initiated at time T=0. For the first 18.5 secoc~ds, get wakeup (SIK countdown timer), next 0.25 sands (command and control, frame 2), and last 2S second is dead time. The remaining time in the timeline is the hold off time for response sluts, which is the frame number ~ the nominal frame time, or 2*20 = 40 seconds of hold off time. At T=20, the second wake-up sequence is initiated.
Similarly, the first 18.5 seconds, get wakeup (SILK cwmtdown timer), next f? 25 seconds (command and control, frame 1 ), and the last 2.5 seconds is dead ame. The hold off time for response slots in this instance is, again, the frame number * flue raminal frame time, which is 1 *20 = 20 seconds off hold off time. At T~iO, the third wakaup xqucnce is initiated For the first 18.5 seconds, get wakeup (SIK countdown timed the next 0.25 seconds (commtd and c~atnol, frame 0), and the Last 2.5 sacoads is dead time. The hold off time for response is cakulabed as follows, frame number * nominal frame lane, or 0~20 = 0 seconds hold off time meaning the endpoi~ have 25 seconds bef~e the begi~intg of slot 0 is this cell.
As mentioned, the example of Fig. 16 is for a three cell reuse pattern.
However, the example can be easily extended to higher celhtler reuse ratios by adding mode frames as appropriate. In the 1430 MHz system, the maximum recommended cxllular reuse is 5. This leads to a hold off time of 100 seconds in the first cell transmitoed which is short enough for the endpoint to maintain accurate timing with regard to slot timings.

unless otherwise specif'~ed by the system, the frame m is preferably tied to the cellular frequency used based on Table 13 below:
Table 13. Frame ID
Cell Reuse Ratio Channel to Frame ll) in 3 Cell Channel 1= Frame ID 0 Channel 3 = Frame 1D 1 Channel 5 = Frame 1D 2 4 Ceti Channel 1= Frame ID 0 Channel 2 = Frame 1D 1 Channel 4 = Frame ID 2 Channel 5 = Frame ID 3 Cell Channel 1= Frame 1D 0 Channel 2 = Frame n71 Channel 3 = Frame ID 2 Channel 4 = Frame >D 3 Channel 5 = Frame 1D 4 To maximize throughput in the system 100, a cellular reuse scheme is employed in the 1430 MHz bad. The rouse ratio is preferably a 3, 4. 5, 7, ar 9 cell pattern.
Smauer patterns are preferred from a delay perspective, however, the final choicae is preferably made during the RF
planning and installation of actual systems in the field. The 7 and 9 patterns are preferably used in die virtual cell model. The reuse patterns are provided in Figs. 17, 18, and 19 depicting three_ oell (ABC), four-cell (ABCD), and five-cell reuse patterns (ABCDE~
n~pectively.
When operating in the mobile or head-held mode, the 2.5 sa~osrda of "dead time" does 1S not apply. Rather slot "4" occurs at the end of the command and control frame plus 25 milliseconds. Note, that due to time required to read the attached meter andlor bring the charge pump to fall operation the endp~ir~t may or may not respond in slot "Or' even if told to respond immediately.
In prog<~amming mode, the hand-held control may reduce its sensitivity by as much as 30 dB to avoid overload conditions at close programming distances. The hand-held and endpoint must work with programming distances as close as 0.5 meters and as far as 300 meters when in the mobile mode of operation with a tine of site propagation path In mobile operation the wake-up sequence, the command ~ control data, and the coceive portions of a standard read cycle are continuously repeated as the mobile moves through the system. The timing is preferably in the range of a one to five second cycle.
The diagram depicted in Fg. 20 gives a general over view of the mobile operation over the five channels.
The command & control fi~ame preferably contains a gmup call read that solicits a consumptive type reading from all of the endpoints that can hear the mobile and that have the carrect system m. The endpoint responds to the group call in a random slot, on a random channel. The randora channel is chosen from the list of available channels that is provided in the command & conaol frame. The random slot is one of the 50 ms slots in the Slotted-ALOHA
portion of the frame. (Slotted ALOHA is a random access scheme just like regular ALOHA
except that the transmissions ace required to begin and end within the defined timeslot. 'The timeslots are marked from the end of the command d~t control frame just like in the fued netarark).
When the reader hears a response from a given endpoint, it knows that it is within range and can reqaeat a specific response from the endpoint in the next command dt control frame.
The command Bt control frame is expected to contain both a standard command frame and sa exOended control frame to allow for the mobile to access the most endpoints possible in a single pass. Wben the mobik requests a tr;aponse fram the end point it will tell It the chscmel and time slot that it is supposed to respond on. This is to minimize the chances of a collision on the longer messages that can be delivered in the 11~P type of responses. During the mobile cycle, battery endpoints may be required to bubble up their receivers up at a higher rate than normal or synchronize to the fu~st command dt control frame to improve mobile performanac.
If the van is moving at a maximum of 30 miles per hour it will travel 440 feet in 10 seconds. The van will also have a communications radius of approximately S00 feet give a 1400 MHz system operating at a data rate of 22.6 I~chipslsecond, with the expected power levels and receiver sensitivities (e.g., +14 dHm endpoint TX power, -110 dBM RX
sensitivity in the van, 20 dB margin, endpoint at 5'). The margin is included because the MDP data packet is much longer than the current SCM type messages and is not repeated unless an eaar occurs.
To achieve a low re-try rate, it is desirable to bring the HER down to 0.0196. To do this under norms!
situations would require an additional 20 dB of margin, however, a diversity setup on the van receivers can be used to achieve the same results. This requires two antennas on the van placed five to six feet apart along with an additional receiver demodulator chain per channel. For SCM
data that is repeated multiple times, the system can operate at a much lower margin and still achieve excellent read reliability in the van. A coverage radius of about 1200 feet is obtained for the system when collecting standard consumptive data.
The diagram of Fig. 2I, shows the coverage rings for low margin SCM messages and for the 20 dB margin )DR messages for the present system in comparison with the current 0 dB
margin SCM messages from the ERT.
With the current mobile protocol each endpoint is, on average, in the range of the van for approximately 12 to 25 seconds. This is an appropriate amount of time to wake up the endpoint, identify who it is, raNest an MDP (mobile data packet = 250 bytes of raw data maximum) to be sent, receive the MDP and potentially retry the request and receive portions of the process if necessary.
In the basic system, there are five channels at a maximum 7596 utilization for MDP
responses. This gives an effective data rate of 42375 BPS or 5296 bytes per second or 21 blocks per second. Sinct the system is looking at a single block per meter, the system can support 21 new meters per second. The mobile then has a nominal range of 500 feet. This gives tt~ system of about 175 meters in range at any given time, even in the densest specifxd.
systems. If the van is moving at 30 mph, the system gets 44 feet of new meters per second. In performing a geometric approximation, the result is about 12 ~w meters per second. So, the system can handle 21 new meters per second but can only get in the range of 10 to 12 meters per second.
This allows for a full set of tactics in a dense system. (This assumes the low 11.3b363 KBPS
data rate and the full 250 byte 11~P, for smaller packers and with the higl~r data rate option, the situation is even better.
V. ResDOnse O~mi~g 'bon f~ obile ~ Fixgd ~jetwork Ooeratio~
In order w optimize the batter efficiency, range, and overall system robustness far endpoints that must operate in both a mobile and fixed network scenario without repmgraa~ming, the following methodology is preferably used. The outbound transmission from the Trader includes a flag that states the response mode of the endpoint. When the response mode flag is set to "mobile" the endpoint responds at a lower power (e.g., +l4dBm) and in a dynamically tatndoraized slot determined as described above. When the endpoint sees the "fixed network"
flag set it responds in its assigned slot at high power (e.g., +30 dHm). The advantage provided by this scenario is that in the nwbile case the reader is not burdented wilt slot dynamic alloeati~
of multiple, which can be computationally intensive and consume additional air time to successfully communicate to all the in-range endpoints. It also allows the endpoint to conserver power and reduce interference. This leads to the ability to transmit more data with leas retries.
In the fixed network case, the high power mode enables the system to get maximum range from tlm device (reducing in5~sfmettn~e costs) while interfaenoe is mitigated by assigned slots. The slots are efficiently assigned in the fixed network case because of the pseudo-static nature of the system. Note that prior art systems enabled only static programmi~tg of the endpoint to operate is one mode or the ot>mr. As such, the previous mettadology did riot allow for mixed mode operation without reprogramming the endpoint. Thus, the present invention presents the combination of low power operation and dynamic slot assigmnent for mobile operation with the high power slotted operation for the fixed network all controlled by a flag in the outbound wakeup data. Refer to field "8," bits 7 and 8, of the command 8t control frame that define the transmitlresponse mode.
The present invention may be embodied in other specifx forms without departing from the spirit of the essential attributes thereof; therefore, the illustrated embodiment should be considered in all respects as illustrative and not restrictive, reference being made tv the appended claims rather than to the foregoing description to indicate the scope of the invention.

Claims (6)

1. An automatic meter reading (AMR) system, comprising:
a reader;

an endpoint interfaced to a utility meter, wherein said reader is capable of communicating with a plurality of endpoint devices via RF communication; and wherein said reader uses a single RF communication protocol in communicating with all of said plurality of endpoint devices and said single RF
communication protocol supports both one-way communication and two-way communication, such that said reader uses said single RF communication protocol to permit communications with at least some of said plurality of endpoint devices via one-way communication and at least some of said endpoint devices via two-way communication; and wherein said single RF communication protocol includes:

a physical layer that defines a common frequency band, channel bandwidth, modulation scheme, and preamble length for endpoint devices operating in one-way and two-way communication modes; and a transport layer that defines timing and packetization for all data transferred between said reader and endpoint, wherein said packetization provides packets originating from endpoints operating in one-way and two-way communication modes to include at least a common endpoint ID field, a common endpoint type field, and a message field following the endpoint type field.

2. The AMR system of claim 1, wherein one-way communication enables said endpoint to communicate with said reader and deliver a specified message type.

3. The AMR system of claim 1, wherein two-way communication enables said reader to communicate with said endpoint, command said endpoint, and enable said endpoint to respond to said reader.

4. The AMR system of claim 2, wherein two-way communication enables said reader to communicate with said endpoint, command said endpoint, and enable said endpoint to respond to said reader.

5. An automatic meter reading (AMR) system, comprising a reader;

an endpoint interfaced to a utility meter, wherein said reader is capable of communicating with said endpoint via RF communication;

wherein said RF communication occurs through the use of a communication protocol, and wherein said communication protocol includes a data link layer directing all outbound data transmissions from said reader to said endpoint to be Manchester encoded and directing inbound transmissions from said endpoint to said reader to be in a selectable encoding format selected from the set consisting of: Manchester encoding, and non-return to zero data encoding.

6. The AMR system of claim 5, wherein said data link layer provides a sequence inversion keyed (SIK) countdown timer.