WO2001049002A1

WO2001049002A1 - Memoryless non-linear predistortion of digital amplitude modulation

Info

Publication number: WO2001049002A1
Application number: PCT/US2000/030285
Authority: WO
Inventors: Richard Steven Griph; Albert Howard Higashi
Original assignee: Motorola, Inc.
Priority date: 1999-12-28
Filing date: 2000-11-02
Publication date: 2001-07-05
Also published as: AU1582101A

Abstract

A transmission system (10) which removes memoryless non-linear predistortion of a digital amplitude modulation signal introduced by an amplifier (26) in the transmission system (10). A portion of the output from the amplifier is passed through a predistorter update module (15). Predistorter update module (15) determines an error between an actual and an ideal signal. The error is input to a predistorter (20) in order to vary a modulation signal prior to input to the amplifier (26).

Description

MEMORYLESS NON-LINEAR PREDISTORTION

OF DIGITAL AMPLITUDE MODULATION

BACKGROUND OF THE INVENTION

1. Technical Field

The present invention relates to a system for predistorting a signal prior to input to an amplifier in order to cancel out non-linear components introduced by the amplifier and, more particularly, to a memoryless non-linear predistortion of a digital amplitude modulation signal.

2. Discussion

It is usually necessary to employ high power amplifiers (HPAs), such as traveling wavetube amplifiers (TWTA) and solid state power amplifiers (SSPA), in transmitters used in high data rate communication links, such as in certain satellite communications systems. These types of high speed communication systems typically require a relatively high output power so that the signal being transmitted can travel greater distances before being significantly attenuated. In these types of communication systems, a low frequency digital baseband signal comprising a stream of digital data bits is transmitted after being modulated onto a high frequency carrier wave.

Different modulation schemes in the art distinguish the digital bits. Example digital modulation schemes for different applications include amplitude-shift keying (ASK), binary phase-shift keying (BPSK), quadrature-phase shift keying (QPSK), and quadrature amplitude modulation (QAM). Also, the digital baseband signals may be multilevel (M-ary) signals requiring multilevel modulation methods. Quadrature modulation schemes provide both amplitude and phase modulation of the carrier because both complex and imaginary representations of the signal are used.

In quadrature modulation schemes, such as QAM, each bit is converted to a bit symbol representing a complex value having an in-phase (real) component and a quadrature-phase (imaginary) component. A constellation pattern represents a group of symbols positioned within a circle around the origin of an imaginary axis and a real axis. The distance from the origin represents the amount of power being transmitted. For example, a group of four bits transmitted at a particular time is represented as sixteen (2⁴) symbols in the circle. Each symbol in the pattern identifies a complex voltage value having an in-phase component and a quadrature-phase component and represents the voltage value for a particular symbol period, which is the time during which each symbol is transmitted. The analog voltage value for each symbol is used to modulate a carrier wave. The symbols in the constellation pattern are geometrically spread so that they are equally spaced apart to more readily distinguish the symbols and reduce bit errors and may be positioned on one or more circles centered about the origin of the constellation pattern. Preferably, the constellation patterns get processed through the transmitter without being distorted so that the bits are readily distinguishable from each other at the receiver end.

HPAs are desirable in high speed communication applications because they provide high gain over wide bandwidths. However, the input signals to a HPA must be controlled because the HPA exhibits non-linear transfer characteristics. At lower input powers, the output-input power relationship of the HPA is approximately linear. However, at peak power output, the HPA saturates and further increases in the input power beyond the saturation point actually decrease the output power of the amplifier. The non-linearity of the HPA affects the position of the symbols in the constellation pattern by moving them away from the origin. Therefore, it is known to provide amplifier predistortion techniques in the transmitter when the amplifier is being operated in its non-linear range near peak output power. This predistortion approach typically includes using a memoryless mapping function that employs look- up tables that preset the constellation pattern symbols closer to the origin, so that when the signal passes through the amplifier, the symbols are moved towards locations representative of a linear transfer function. Such a look-up table approach is preset and non-adaptive and as such does not track changes in the HPA response that can occur over time due to aging or thermal changes in the HPA. High power amplifiers also include filtering distortions that cause the amplifier to have memory of previous constellation symbols already transmitted. Amplifier memory is the effect that the transmission of one symbol or group of symbols has on the transmission of the following symbol or groups of bits. High gain amplifiers introduce AM-AM (amplitude modulation) and AM-PM (phase modulation) distortion as a result of the non-constant envelope nature of the signals that are provided as inputs to the amplifier. Because the data is digitally encoded on a waveform, the pulse shape of the waveform creates artifact portions, where preceding pulses combine to interfere with the particular pulse being sampled. This is known as intersymbol interference (ISI), and requires that the signal pulses be shaped to reduce the memory of the amplifier.

Multiple possible transmission paths of a signal through a transmitter exist for an input signal. A typical input signal into a high power amplifier, such as a TWTA or SSPA, undergoes a filtering effect by the transmitter hardware before the amplifier. The input signal also experiences filtering effects of the HPA as a result of its memory. Because the amplifier has memory, a symbol can follow different paths, depending on what symbols were transmitted before the current symbol period.

The non-linearity of the amplifier distorts the filtered input signal due to its non- constant envelope. By applying memory predistortion techniques, the ISI of the amplifier can be reduced, thus limiting the distortion.

It is thus desirable to do additional predistortion beyond memoryless mapping to correct for the filtering effect of the amplifier memory and the filtering effect of the transmission hardware. Known techniques of memory predistortion have employed inverse filters that invert the filtering of the amplifier for predistortion purposes. Locally-adapted linear predistorters typically intend to invert the filtering prior to the HPA (pre-HPA filtering) and ignore filtering after the HPA (post-HPA filtering). The presence of filtering after the non-linearity of HPA will provide a linear signal at the receiver. The receiver equalizer typically suitably removes most filtering with minimal distortion so long as only linear distortion exists. The above techniques can prove very effective in minimizing memory and linear-related predistortion in a transmission system. Even where the memory has been properly predistorted, the inherent non-linearity of the HPA, whether a TWTA or SSPA, needs to be addressed, and, more particularly, can be better addressed than by using the memoryless mapping functions presently employed.

BRIEF DESCRIPTION OF THE DRAWINGS

The various advantages of the present invention will become apparent to one skilled in the art by reading the following specification and appended claims, and by referencing the following drawings in which:

FIG. 1 illustrates a schematic block diagram of a transmitter and a receiver system arranged in accordance with the principles of the present invention;

FIG. 2 illustrates an expanded block diagram of the predistorter update block of FIG. 1 ; and

FIG. 3 illustrates an example of a constellation diagram demonstrating the principles of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

FIG. 1 depicts a block diagram of a communication system 10 for exchanging modulated data signals between a transmitter 12 and a receiver 14 via a communication link 16, such as a air link or a hard-wired interconnection, arranged in accordance with the principles of the present invention. Transmitter 12 includes a modulator 18 which receives a digital data stream at baseband frequency. Modulator 18 modulates the data stream, utilizing a quadrature amplitude modulation (QAM) format. Modulator 18 modulates the bits onto an analog carrier wave. During modulation, modulator 18 identifies for each bit pattern a symbol that includes an in-phase and quadrature phase component, and maps the symbols into a constellation pattern, for transmission. The modulated signal has an analog voltage for each symbol to be transmitted. The modulator 18 can be any suitable quadrature modulator for the purpose described herein, as will be apparent to those skilled in the art.

The modulated signal is input into a predistorter 20. Predistorter 20 is a non- linear predistorter and is embodied as a programmable filter, as will be described in greater detail herein. Predistorter 20 adds a predistortion signal to the modulated signal, which is an inverse of the non-linear distortion introduced by transmitter 12 and, in particular, HPA filter 28. The distortion signal is later cancelled by distortion intentionally introduced by other components of transmitter 12. As will be discussed in more detail below, predistorter 20 receives voltage signals from predistorter update module 15 which receives the amplified signal that has been distorted by amplifier system 26.

Predistorter 20 is an inverse filter that compensates for changes the nonlinear distortion caused by the AM-AM/AM-PM nature of non-linear devices of various points in the constellation pattern by varying the complex voltage value output by modulator 18. The predistorter 20 implements a zero-forcing function algorithm. For example, predistorter 20 can employ a simple multi-tap delay-line digital or analog filter to provide filtering. This inverse filtering adjustment predistorts the constellation pattern representing the complex signal so that when non-linear distortion from amplifier system 26 occurs, the signal actually returns to desirable undistorted state for transmission. In this embodiment, predistorter 20 is positioned after modulator 18 and acts as an analog-type predistorter. However, as will be appreciated by those skilled in the art, predistorter 20 can be a digital predistorter. For example, modulator 18 can output digital symbols that have been modulated, where predistorter 20 operates on digital symbols and a digital-to- analog converter (not shown) after predistorter 20 can provide digital-to-analog conversion. Predistorter 20 outputs a predistorted, modulated signal which is shown as being input into a pre-HPA filter 22. Pre-HPA filter 22 represents the filtering effect of the physical circuitry in transmitter 12. Such filtering may be removed by other predistortion techniques known in the art.

The radio frequency (RF) signal from modulator 18 and predistorted by predistorter 20 is at a baseband frequency and must be upconverted to a high frequency for transmission. A mixer 24 upconverts the baseband frequency with a high frequency signal, such as cos(ω_ct).. Mixer 24 converts the in-phase and quadrature-phase representations of the complex voltage from the modulation process to a single high frequency RF signal. The predistortion technique of the present invention can also be done at RF frequency, where predistorter 20 would be located after mixer 24. The upconverted RF signal is then applied to the amplifier system 26 that significantly increases the gain for transmission. The operation of the mixing step and amplification step for a transmitter of this type is well understood to those skilled in the art.

The upconverted, amplified signal from amplifier system 26, has been distorted back to its desirable pattern and is applied to a RF filter 32 for subsequent RF filtering for conforming with Federal Communications Commission (FCC) requirements and then to an antenna (not shown) for transmission. The amplified signal from amplifier system 26 is also applied to predistorter update module 15 from a testpoint 48, as will be described herein, following amplifier system 26. A suitable power coupler (not shown) is provided at testpoint 48 to remove a small portion of the high power signal from amplifier system 26. Any type of suitable power splitter can be used to split the signal at testpoint 48 to send a portion of the signal to predistorter update module 15. According to the invention, predistorter update module 15 continually provides a voltage signal to predistorter 20 to make adaptive changes to the arrangement of the constellation pattern to invert the nonlinear distortion caused by amplifier system 26, which changes over time.

Amplifier system 26 includes a HPA 30, such as a TWTA or SSPA, and also includes a filter 28 which represents a memory filtering effect which is a natural byproduct of operation of amplifier system 26 and, in particular, HPA 30. Such filtering may be removed by other predistortion techniques (not shown) know in the art. In addition to the filtering effect represented by filter 28, HPA 30 also introduces a memoryless non-linearity into the RF signal output by amplifier system 26 and input to RF filter 32.

The signal output by RF filter 32 is broadcast across a channel 34 via communication link 16. The signal is received at receiver 14 by an antenna (not shown) that applies a signal to a receiver filter 36. The receiver filter 36 provides initial filtering of the received signal, for filtering channel noise and the like, and is typically closely matched to the transmitted signal. Receiver filter 36 rejects thermal noise and allows optimal reception. A mixer 38 downconverts the RF signal to an intermediate frequency signal by mixing the signal with a high frequency signal cos(ω_ct). The downconverted signal from mixer 38 includes baseband in-phase and quadrature-phase components. The downconverted signal is applied to low-pass filter 40 to provide filtering at baseband frequencies. Thus, receiver filter 36 typically acts as a course filter, and low-pass filter 40 typically acts as a fine filter. The filtered baseband signal from low-pass filter 40 is applied to a linear equalizer 42 that removes the ISI from transmission of the signal through channel 34. Receiver filter 36 and low-pass filter 40 may also generate the ISI. Linear equalizer 42 typically includes a tapped delay line filter, which is known in the art, where the taps are adjusted by a data estimator 44. Data estimator 44 takes the voltage represented by the in-phase and quadrature-phase values and converts it back to bits. Data estimator 44 can use any suitable algorithm to perform this function, such as a zero-forcing algorithm. Data estimator 44 measures the symbol locations, and generates an estimate between the actual symbol locations and the desired symbol locations. Thus, data estimator 44 provides an error correction between the constellation pattern actually received verses the expected constellation pattern. The equalizer update signal sent from data estimator 44 to linear equalizer 42 provides a filter correction to achieve the desired constellation pattern based on the error calculation. FIG. 2 illustrates an expanded block diagram of the predistorter update block of FIG. 1. The signal received from testpoint 48 at the output of amplifier system 26 (FIG.1) is input to attenuator 50. Attenuator 50 reduces the amplified signal power in preparation for applying the attenuated signal to demodulator 52. Attenuating the power level may also be accomplished by a power coupler which inherently attenuates the signal level output to testpoint 48. Demodulator 52 receives the attenuated signal and demodulates the attenuated signal to baseband in order to recover the baseband in-phase and quadrature-phase signal components as indicated in the constellation diagram of FIG. 3. The demodulated signal is output to an error detector 56. Error detector 56 determines an error signal based on the difference between the output of demodulator 52 and the output of modulator 18 (FIG. 1). The output from modulator 18 passes through a delay element 54 to compensate for the delay introduced by amplifier system 26. Prior to discussing operation of error detector 56, reference should be made to the constellation diagram of FIG. 3. As can be seen in the constellation diagram of FIG. 3, a 12/4 QAM system can be depicted on Cartesian coordinates defined by an in-phase axis 60 and a quadrature-phase axis 62. The 12/4 QAM constellation has an inner amplitude level 64 and outer amplitude level 66. The inner and outer amplitude levels represent different power levels that when introduced to a non-linearity such as HPA 30 (FIG. 1), will produce AM-AM and AM-PM distortions. Because the outer amplitude level 66 represents the peak power that can be transmitted by HPA 30, outer amplitude level 66 is selected as the reference for displacing the other points in the constellation. One skilled in the art will recognize that varying the power level of these points is of little value. The typical saturation curves of HPA 30 are such that a multiple decibel change of the power level at the input of HPA 30 would not significantly vary the output power. The relative flatness of this saturation curve results in these points being substantially at the correct power level regardless of small errors in their input power level, and adaptive control of this power level provides minimal incremental benefit. The phase of the outer points, however, is not similarly insensitive to errors, and the subject invention adaptively controls the phase of the outer points. The symbols on inner amplitude level 64, however, are amplitude distorted as well as phase distorted. Inner symbols 70 represent an undistorted configuration of the symbols on inner circle amplitude level 64 of the constellation diagram. Outer symbols 72 represent the distorted constellation points. The vector 74 represents the error or the difference between the undistorted constellation symbols 70 and the distorted constellation symbols 72.

As referred to above with reference to FIG. 2, error detector 56 outputs an error signal. The error signal is used for adaptation and is determined in accordance with the difference between the output of testpoint 48, passed through demodulator 52, and the ideal locations output by modulator 18 (FIG. 1), passed through delay element 54. The error signal contains phase and amplitude information for each of the transmitted signal locations. Correlator 58 subtracts the errors for each of the constellation points from the positions utilized during transmission. Correlator 58 multiplies the error by a percentage, such as 5-10%. When applied iteratively, this approach generates an adaptive value for predistorter 20. Correlator 58 outputs locations for predistorting the constellation points, which are mapped by predistorter 20 to the modulated signals. Predistorter 20 receives the ideal constellation points output by modulator 18 (FIG. 1 ) and modifies their location to the locations output by correlator 58. If the modulation format has, for example, sixteen points, then correlator 58 outputs sixteen locations to predistorter 20. Predistorter 20 essentially chooses from these sixteen predistorted locations or points based on the symbol to be transmitted.

During any iteration, the phase of the constellation point output by modulator 18 for transmission may be significantly different than the phase of the signal at testpoint 48 due to the AM-PM distortion of amplifier system 26. If the AM-PM distortion of amplifier system 26 is relatively small, then this phase difference will minimally impact predistortion. If the phase difference exceeds, for example, forty- five degrees, the phase difference will impact predistortion. There can also be an arbitrary phase shift in the system if a completely blind adaptation is desired. To compensate for an arbitrary phase shift, the error signal which is summed by correlator 58 can be corrected or rotated by the phase difference. The phase difference can be determined because predistorter 20 rotates the signals in order to cancel this rotation. This correction or rotation is performed by predistorter 20 prior to applying the signal input from correlator 58 for HPAs which have significant AM- PM distortion.

Because the phase of the outer constellation points is varied, but the amplitude is not, the outer points are rotated slightly differently than the inner points. The outer points are updated similarly to the inner points except that the amplitude of the outer points is normalized following phase correction. This maintains the power level of the outer points constant while allowing a phase shift of the outer points. Because the outer points provide a reference, the impact on the normalization process is preferably considered. If correlator 58 continuously reduces the power of the outer points when normalizing the outer points, this implies that the power of the reference signal from modulator 18 is low. If not corrected, the inner points would be corrected to an undesirably low power level. To compensate for this, the power level of the reference signal is adjusted based on what occurs during normalization. If normalization reduces the power of the outer points, then the reference signal is increased slightly. If normalization increases the power of the outer points, the reference signal is decreased slightly.

From the foregoing, one skilled in the art will recognize that the communication system 10 provides a novel method for memoryless non-linear predistortion of digital amplitude modulation. While discussed herein as being applicable to a quadrature amplitude modulation system, the invention described herein can be applied to any multi-amplitude modulation system. This invention simplifies the non-linear predistortion process by detecting different amplitude levels and providing a separate error for the corresponding amplitude. Thus, when employing predistortion, a single-tap equivalent filter can be applied to the modulated signal. Thus, a one-tap filter can be used for each particular amplitude level.

While specific embodiments have been shown and described in detail to illustrate the principles of the present invention, it will be understood that the invention may be embodied otherwise without departing from such principles. For example, one skilled in the art will readily recognize from such discussion and from the accompanying drawings and claims that various changes, modifications and variations can be made therein without departing from the spirit and scope of the invention as described in the following claims.

Claims

-10- CLAIMSWhat is claimed is:

1. A transmission system, comprising: a modulator for modulating a digital data stream onto a carrier wave to generate a modulated signal, the modulator converting data on the data stream into bit symbols for transmission by the transmission system of the modulated signal being in proximity to one of a plurality of predetermined amplitudes for a given bit symbol; an amplifier for amplifying the modulated signal prior to transmission to generate an amplified signal, the amplifier having a non-linear characteristic that generates a non-linear distortion in the modulated signal; a predistorter, the predistorter receiving the modulated signal prior to amplification by the amplifier, the predistorter generating an inverse signal of the non-linear distortion introduced by the amplifier, the inverse signal varying the modulated signal to define a predistorted signal; and a predistorter update controller, the predistorter update controller receiving a portion of the amplified signal, the predistorter update controller being operable to vary generation of the inverse signal in accordance with an amplitude of the amplified signal.

2. The transmission system of claim 1 wherein the predistorter update controller further comprises a demodulator, the demodulator demodulating the portion of the amplified signal from the carrier wave to output a demodulated signal, the demodulated signal including the non-linear distortion introduced by the amplifier.

3. The transmission system of claim 1 wherein the predistorter update controller distinguishes between the plurality of predetermined amplitudes of the modulated signal.

4. The transmission system of claim 1 wherein the predistorter update controller further comprises an attenuator, the attenuator reducing the amplification of the output signal.

5. The transmission system of claim 1 further comprising an error detector, the error detector comparing the modulated signal with the portion of the amplified signal to generate an error signal.

6. The transmission system of claim 5 further comprising a correlator, the correlator correlating the error signal with an ideal signal to generate a control signal, the control signal varying operation of the predistorter.

7. The transmission system of claim 1 wherein the amplifier is one of a traveling wave tube amplifier (TWTA) or a solid state power amplifier (SSPA).

8. The transmission system of claim 1 wherein the predistorter update controller receives the amplified signal from a sampling point immediately following the amplifier.

9. A transmission system, comprising: a modulator for modulating a digital data stream onto a carrier wave to generate a modulated signal, the modulator converting data on the data stream into bit symbols for transmission by the transmission system, the modulated signal being in proximity to one of a plurality of predetermined amplitudes for a given bit symbol; an amplifier for amplifying the modulated signal prior to transmission to generate an amplified signal, the amplifier having a non-linear characteristic that generates a non-linear distortion in the modulated signal; a predistorter, the predistorter receiving the modulated signal prior to amplification by the amplifier, the predistorter generating an inverse signal of the non-linear distortion introduced by the amplifier, the inverse signal varying the modulated signal to define a predistorted signal; an error detector, the error detector comparing the modulated signal with a portion of the amplified signal to generate an error signal, the error signal indicating the non-linear distortion introduced by the amplifier; and a correlator, the correlator correlating the error signal with an ideal signal to generate a control signal, the control signal varying operation of the predistorter.

10. The transmission system of claim 9 further comprising an attenuator, the attenuator reducing amplification of the output signal to generate an attenuated signal.

11. The transmission system of claim 10 further comprising a demodulator, the demodulator demodulating the attenuated signal from the carrier wave to output a demodulated signal, the demodulated signal including the nonlinear distortion introduced by the amplifier.

12. The transmission system of claim 11 wherein the attenuator receives the amplified signal from a sampling point immediately following the amplifier.

13. The transmission system of claim 12 wherein the amplifier is one of a traveling wave tube amplifier (TWTA) or a solid state power amplifier (SSPA).

14. A method for signal transmission, comprising the steps of: modulating a digital data stream onto a carrier wave to generate a modulated signal, by converting data in the digital data stream into bit symbols for transmission, the modulated signal being in proximity to one of a plurality of predetermined amplitudes for a given bit symbol; amplifying the modulated signal prior to transmission to generate an amplified signal, said amplifying including a non-linear characteristic that generates a non-linear distortion in the modulated signal; predistorting the modulated signal prior to amplification by generating an inverse signal of the non-linear distortion introduced by the amplifying step, the inverse signal varying the modulated signal to define a predistorted signal; and varying generation of the inverse signal in accordance with a selected one of predetermined amplitudes of the amplified signal and a proximity of the amplified signal to one of the plurality of predetermined amplitudes in order to generate the inverse signal and remove the non-linear distortion.

15. The method of claim 14 further comprising the step of demodulating a portion of the amplified signal from the carrier wave to generate a demodulated output signal, the demodulated output signal including the non-linear distortion introduced by the amplifying step.

16. The method of claim 15 further comprising the step of reducing the amplification of the output signal prior to demodulating the portion of the amplified signal.

17. The method of claim 16 further comprising the step of distinguishing between the plurality of predetermined amplitudes of the modulated signal.

18. The method of claim 17 further comprising the step of comparing the modulated signal with the portion of the amplified signal to generate an error signal.

19. The method of claim 18 further comprising the step of correlating the error signal with an ideal signal to generate a control signal, the control signal varying the step of predistorting.

20. The method of claim 14 further comprising the step of sampling the amplified signal from a sampling point immediately following the step of amplifying.