# US 2017/0257251 A1 · Predistortion Processing Apparatus and Method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2017/0257251 A1
- Document ID: US20170257251A1
- Status: Published application
- Kind: A1
- Application number: 15/599,956
- Filed: 2017-05-19
- Published: 2017-09-07
- Earliest claimed: 2014-11-19
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Yanqing Zhao; Chunlei Yu; Tingjian TIAN
- Web page: https://patentgrep.com/patent/US20170257251A1

## CPC

- H04L 27/368: {adaptive predistortion}
- H03F 1/3247: {using feedback acting on predistortion circuits (H03F1/3264 takes precedence)}
- H03F 1/3258: {based on polynomial terms}
- H03F 1/3294: {Acting on the real and imaginary components of the input signal}
- H03F 3/189: High-frequency amplifiers, e.g. radio frequency amplifiers
- H03F 3/245: {with semiconductor devices only}
- H04L 25/03343: {Arrangements at the transmitter end}
- H03F 2200/102: A non-specified detector of a signal envelope being used in an amplifying circuit
- H03F 2201/32: Indexing scheme relating to modifications of amplifiers to reduce non-linear distortion
- H03F 2201/3233: Adaptive predistortion using lookup table, e.g. memory, RAM, ROM, LUT, to generate the predistortion
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- H03F: AMPLIFIERS

## Abstract

A predistortion processing apparatus: an auxiliary feedback module, configured to: extract a nonlinear distortion signal from an analog signal, and input an obtained feedback signal corresponding to the nonlinear distortion signal into an auxiliary model coefficient training module; the auxiliary model coefficient training module, configured to: train an auxiliary coefficient according to the feedback signal and a predistortion signal, and transmit a first auxiliary coefficient obtained through training to a predistortion processing module; a radio frequency signal feedback module, configured to extract a fundamental wave feedback signal; a predistortion model coefficient training module, configured to: train a predistortion coefficient according to the predistortion signal and the fundamental wave feedback signal, and transmit an obtained predistortion coefficient to the predistortion processing module; the predistortion processing module, configured to: perform predistortion processing on an input intermediate frequency signal by performing nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient.

## Claims

### Claim 1

A predistortion processing apparatus comprising: an auxiliary feedback module, configured to:
receive an amplified analog signal,
extract a nonlinear distortion signal from the amplified analog signal,
obtain a feedback signal corresponding to the nonlinear distortion signal, and
input the feedback signal into an auxiliary model coefficient training module;
the auxiliary model coefficient training module being configured to:
receive the feedback signal and a predistortion signal that is output by a predistortion processing module,
train an auxiliary coefficient of the auxiliary model coefficient training module according to the feedback signal and the predistortion signal that is output by the predistortion processing module, and
transmit a first auxiliary coefficient obtained through training to the predistortion processing module;
a radio frequency signal feedback module configured to:
receive the amplified analog signal, extract a fundamental wave signal from the amplified analog signal,
obtain a fundamental wave feedback signal corresponding to the fundamental wave signal, and
input the fundamental wave feedback signal into a predistortion model coefficient training module;
the predistortion model coefficient training module configured to:
receive the fundamental wave feedback signal and the predistortion signal that is output by the predistortion processing module,
train a predistortion coefficient of the predistortion model coefficient training module according to the fundamental wave feedback signal and the predistortion signal that is output by the predistortion processing module, and
transmit a predistortion coefficient obtained through training to the predistortion processing module; and
the predistortion processing module configured to perform predistortion processing on an input intermediate frequency signal by performing nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, and output a predistortion signal obtained through predistortion processing.

### Claim 2 (depends on 1)

The apparatus according to claim 1, wherein the nonlinear distortion signal comprises an envelope signal, and the auxiliary feedback module is configured to: receive the amplified analog signal, extract the envelope signal from the amplified analog signal, perform sampling processing on the envelope signal, obtain an envelope feedback signal corresponding to the envelope signal, and input the envelope feedback signal into the auxiliary model coefficient training module.

### Claim 3 (depends on 2)

The apparatus according to claim 2, wherein the predistortion processing module comprises a first auxiliary filtering submodule and a first predistortion processing submodule, and the predistortion model coefficient training module comprises a second auxiliary filtering submodule and a first predistortion coefficient processing submodule; and
the auxiliary model coefficient training module is configured to:
train the auxiliary coefficient of the auxiliary model coefficient training module by performing modeling by using the predistortion signal and the envelope feedback signal that are input into the auxiliary model coefficient training module,
obtain a second auxiliary coefficient after the training, and
transmit the second auxiliary coefficient to the first auxiliary filtering submodule and the second auxiliary filtering submodule.

### Claim 4 (depends on 3)

The apparatus according to claim 3, wherein the first auxiliary filtering submodule is configured to:
construct, according to the intermediate frequency signal input into the predistortion processing module, an envelope reference signal corresponding to the intermediate frequency signal,
perform filtering processing on the envelope reference signal by using the second auxiliary coefficient, and
input an obtained first output signal into the first predistortion processing submodule;
the first predistortion processing submodule is configured to:
generate a predistortion vector according to a preset predistortion model by using the first output signal output by the first auxiliary filtering submodule and a clipping-processed baseband signal that is input into the first predistortion processing submodule,
multiply the predistortion vector by the predistortion coefficient to obtain the predistortion signal, and
output the predistortion signal;
the second auxiliary filtering submodule is configured to:
perform, by using the second auxiliary coefficient,
filtering processing on the fundamental wave feedback signal output by the radio frequency signal feedback module, and
input an obtained second output signal into the first predistortion coefficient processing submodule; and
the first predistortion coefficient processing submodule is configured to:
perform nonlinear modeling according to the second output signal, the fundamental wave feedback signal input into the first predistortion coefficient processing submodule, and the predistortion signal,
train the predistortion coefficient of the predistortion model coefficient training module, and
obtain the predistortion coefficient.

### Claim 5 (depends on 1)

The apparatus according to claim 1, wherein the nonlinear distortion signal comprises an envelope signal and a harmonic signal, and the auxiliary feedback module is configured to:
receive the amplified analog signal, extract the envelope signal and the harmonic signal from the amplified analog signal,
perform sampling processing on the envelope signal and the harmonic signal,
obtain an envelope feedback signal corresponding to the envelope signal and a harmonic feedback signal corresponding to the harmonic signal, and
input the envelope feedback signal and the harmonic feedback signal into the auxiliary model coefficient training module.

### Claim 6 (depends on 5)

The apparatus according to claim 5, wherein the predistortion processing module comprises a third auxiliary filtering submodule and a second predistortion processing submodule, and the predistortion model coefficient training module comprises a fourth auxiliary filtering submodule and a second predistortion coefficient processing submodule; and
the auxiliary model coefficient training module is configured to:
train the auxiliary coefficient of the auxiliary model coefficient training module by performing modeling by using the predistortion signal, the envelope feedback signal, and the harmonic feedback signal that are input into the auxiliary model coefficient training module,
obtain a third auxiliary coefficient after training, and transmit the third auxiliary coefficient to the third auxiliary filtering submodule and the fourth auxiliary filtering submodule.

### Claim 7 (depends on 6)

The apparatus according to claim 6, wherein the third auxiliary filtering submodule is configured to:
construct, according to the intermediate frequency signal input into the predistortion processing module, an envelope reference signal and a harmonic reference signal that correspond to the intermediate frequency signal,
perform filtering processing on the envelope reference signal and the harmonic reference signal by using the third auxiliary coefficient, and
input an obtained third output signal into the second predistortion processing submodule;
the second predistortion processing submodule is configured to:
generate a predistortion vector according to a preset predistortion model by using the third output signal output by the third auxiliary filtering submodule and a clipping-processed baseband signal that is input into the predistortion processing submodule,
multiply the predistortion vector by the predistortion coefficient to obtain the predistortion signal, and output the predistortion signal;
the fourth auxiliary filtering submodule is configured to:
perform, by using the third auxiliary coefficient, filtering processing on the fundamental wave feedback signal output by the radio frequency signal feedback module, and
input an obtained fourth output signal into the second predistortion coefficient processing submodule; and
the second predistortion coefficient processing submodule is configured to:
perform nonlinear modeling according to the fourth output signal, a fundamental wave feedback signal input into the second predistortion coefficient processing submodule, and the predistortion signal,
train the predistortion coefficient of the predistortion model coefficient training module, and
obtain the predistortion coefficient.

### Claim 8 (depends on 1)

The apparatus according to claim 1, wherein the nonlinear distortion signal comprises a harmonic signal, and the auxiliary feedback module is configured to: extract the harmonic signal from the amplified analog signal, perform sampling processing on the harmonic signal, obtain a harmonic feedback signal corresponding to the harmonic signal, and input the harmonic feedback signal into the auxiliary model coefficient training module.

### Claim 9 (depends on 8)

The apparatus according to claim 8, wherein the predistortion processing module comprises a fifth auxiliary filtering submodule and a third predistortion processing submodule, and the predistortion model coefficient training module comprises a sixth auxiliary filtering submodule and a third predistortion coefficient processing submodule, and
the auxiliary model coefficient training module is configured to:
train the auxiliary coefficient of the auxiliary model coefficient training module by performing modeling by using the predistortion signal and the harmonic feedback signal that are input into the auxiliary model coefficient training module,
obtain a fourth auxiliary coefficient after training, and
transmit the fourth auxiliary coefficient to the fifth auxiliary filtering submodule and the sixth auxiliary filtering submodule.

### Claim 10 (depends on 9)

The apparatus according to claim 9, wherein the fifth auxiliary filtering submodule is configured to:
construct, according to the intermediate frequency signal input into the predistortion processing module, a harmonic reference signal corresponding to the intermediate frequency signal,
perform filtering processing on the harmonic reference signal, and
input a fifth output signal obtained by using the fourth auxiliary coefficient into the third predistortion processing submodule;
the third predistortion processing submodule is configured to:
generate a predistortion vector according to a preset predistortion model by using the fifth output signal output by the fifth auxiliary filtering submodule and a clipping-processed baseband signal that is input into the third predistortion processing submodule,
multiply the predistortion vector by the predistortion coefficient to obtain the predistortion signal, and
output the predistortion signal;
the sixth auxiliary filtering submodule is configured to:
perform, by using the fourth auxiliary coefficient, filtering processing on the fundamental wave feedback signal output by the radio frequency signal feedback module, and
input an obtained sixth output signal into a predistortion model processing submodule; and
the third predistortion coefficient processing submodule is configured to:
perform nonlinear modeling according to the sixth output signal, a fundamental wave feedback signal input into the third predistortion coefficient processing submodule, and the predistortion signal,
train the predistortion coefficient of the predistortion model coefficient training module, and
obtain the predistortion coefficient.

### Claim 11

A predistortion processing method comprising:
receiving, by an auxiliary feedback module, an amplified analog signal, extracting a nonlinear distortion signal from the amplified analog signal, obtaining a feedback signal corresponding to the nonlinear distortion signal, and inputting the feedback signal into an auxiliary model coefficient training module;
receiving, by the auxiliary model coefficient training module, the feedback signal and a predistortion signal that is output by a predistortion processing module, training an auxiliary coefficient of the auxiliary model coefficient training module according to the feedback signal and the predistortion signal that is output by the predistortion processing module, and transmitting a first auxiliary coefficient obtained through training to the predistortion processing module;
receiving, by a radio frequency signal feedback module, the amplified analog signal, extracting a fundamental wave signal from the amplified analog signal, obtaining a fundamental wave feedback signal corresponding to the fundamental wave signal, and inputting the fundamental wave feedback signal into a predistortion model coefficient training module;
receiving, by the predistortion model coefficient training module, the fundamental wave feedback signal and the predistortion signal that is output by the predistortion processing module, training a predistortion coefficient of the predistortion model coefficient training module according to the fundamental wave feedback signal and the predistortion signal that is output by the predistortion processing module, and transmitting a predistortion coefficient obtained through training to the predistortion processing module; and
performing, by the predistortion processing module, predistortion processing on the input intermediate frequency signal by performing nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, and outputting a predistortion signal obtained through predistortion processing.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the nonlinear distortion signal comprises an envelope signal, and the receiving, by an auxiliary feedback module, an amplified analog signal, extracting a nonlinear distortion signal from the amplified analog signal, obtaining a feedback signal corresponding to the nonlinear distortion signal, and inputting the feedback signal into an auxiliary model coefficient training module specifically comprises:
receiving, by the auxiliary feedback module, the amplified analog signal, extracting the envelope signal from the amplified analog signal, performing sampling processing on the envelope signal, obtaining an envelope feedback signal corresponding to the envelope signal, and inputting the envelope feedback signal into the auxiliary model coefficient training module.

### Claim 13 (depends on 12)

The method according to claim 12, wherein the predistortion processing module comprises a first auxiliary filtering submodule and a first predistortion processing submodule, the predistortion model coefficient training module comprises a second auxiliary filtering submodule and a first predistortion coefficient processing submodule, and the method further comprises:
training, by the auxiliary model coefficient training module, the auxiliary coefficient of the auxiliary model coefficient training module by performing modeling by using the predistortion signal and the envelope feedback signal that are input into the auxiliary model coefficient training module, obtaining a second auxiliary coefficient after training, and transmitting the second auxiliary coefficient to the first auxiliary filtering submodule and the second auxiliary filtering submodule.

### Claim 14 (depends on 13)

The method according to claim 13, wherein the method further comprises:
constructing, by the first auxiliary filtering submodule according to the intermediate frequency signal input into the predistortion processing module, an envelope reference signal corresponding to the intermediate frequency signal, performing filtering processing on the envelope reference signal by using the second auxiliary coefficient, and inputting an obtained first output signal into the first predistortion processing submodule;
generating, by the first predistortion processing submodule, a predistortion vector according to a preset predistortion model by using the first output signal output by the first auxiliary filtering submodule and a clipping-processed baseband signal that is input into the first predistortion processing submodule, multiplying the predistortion vector by the predistortion coefficient to obtain the predistortion signal, and outputting the predistortion signal;
performing, by the second auxiliary filtering submodule by using the second auxiliary coefficient, filtering processing on the fundamental wave feedback signal output by the radio frequency signal feedback module, and inputting an obtained second output signal into a predistortion model processing submodule; and
performing, by the first predistortion coefficient processing submodule, nonlinear modeling according to the second output signal, the fundamental wave feedback signal input into the first predistortion coefficient processing submodule, and the predistortion signal, and training the predistortion coefficient of the predistortion model coefficient training module to obtain the predistortion coefficient.

### Claim 15 (depends on 11)

The method according to claim 11, wherein the nonlinear distortion signal comprises an envelope signal and a harmonic signal, and the method further comprises:
extracting, by the auxiliary feedback module, the envelope signal and the harmonic signal from the analog signal, performing sampling processing on the envelope signal and the harmonic signal, obtaining an envelope feedback signal corresponding to the envelope signal and a harmonic feedback signal corresponding to the harmonic signal, and inputting the envelope feedback signal and the harmonic feedback signal into the auxiliary model coefficient training module.

### Claim 16 (depends on 15)

The method according to claim 15, wherein the predistortion processing module comprises a third auxiliary filtering submodule and a second predistortion processing submodule, the predistortion model coefficient training module comprises a fourth auxiliary filtering submodule and a second predistortion coefficient processing submodule, and the method further comprises:
training, by the auxiliary model coefficient training module, the auxiliary coefficient by using the predistortion signal, the envelope feedback signal, and the harmonic feedback signal that are input into the auxiliary model coefficient training module, obtaining a third auxiliary coefficient after training, and transmitting the third auxiliary coefficient to the third auxiliary filtering submodule and the fourth auxiliary filtering submodule.

### Claim 17 (depends on 16)

The method according to claim 16, wherein the method further comprises: constructing, by the third auxiliary filtering submodule according to the intermediate frequency signal input into the predistortion processing module, an envelope reference signal and a harmonic reference signal that correspond to the intermediate frequency signal, performing filtering processing on the envelope reference signal and the harmonic reference signal by using the third auxiliary coefficient, and inputting an obtained third output signal into the second predistortion processing submodule;
generating, by the second predistortion processing submodule, a predistortion vector according to a preset predistortion model by using the third output signal output by the third auxiliary filtering submodule and a clipping-processed baseband signal that is input into the second predistortion processing submodule, multiplying the predistortion vector by the predistortion coefficient to obtain the predistortion signal, and outputting the predistortion signal;
performing, by the fourth auxiliary filtering submodule by using the third auxiliary coefficient, filtering processing on the fundamental wave feedback signal output by the radio frequency signal feedback module, and inputting an obtained fourth output signal into the second predistortion coefficient processing submodule; and
performing, by the second predistortion coefficient processing submodule, nonlinear modeling according to the fourth output signal, a fundamental wave feedback signal input into the second predistortion coefficient processing submodule, and the predistortion signal, and training the predistortion coefficient of the predistortion model coefficient training module to obtain the predistortion coefficient.

### Claim 18 (depends on 11)

The method according to claim 11, wherein the nonlinear distortion signal comprises a harmonic signal, and the method further comprises:
extracting, by the auxiliary feedback module, the harmonic signal from the analog signal, performing sampling processing on the harmonic signal, obtaining a harmonic feedback signal corresponding to the harmonic signal, and inputting the harmonic feedback signal into the auxiliary model coefficient training module.

### Claim 19 (depends on 18)

The method according to claim 18, wherein the predistortion processing module comprises a fifth auxiliary filtering submodule and a third predistortion processing submodule, the predistortion model coefficient training module comprises a sixth auxiliary filtering submodule and a third predistortion coefficient processing submodule, and the method further comprises:
training, by the auxiliary model coefficient training module, the auxiliary coefficient by using the predistortion signal and the harmonic feedback signal that are input into the auxiliary model coefficient training module, obtaining a fourth auxiliary coefficient after training, and transmitting the fourth auxiliary coefficient to the fifth auxiliary filtering submodule and the sixth auxiliary filtering submodule.

### Claim 20 (depends on 19)

The method according to claim 19, wherein the method further comprises:
constructing, by the fifth auxiliary filtering submodule according to the intermediate frequency signal input into the predistortion processing module, a harmonic reference signal corresponding to the intermediate frequency signal, performing filtering processing on the harmonic reference signal, and inputting a fifth output signal obtained by using the fourth auxiliary coefficient into the third predistortion processing submodule;
generating, by the third predistortion processing submodule, a predistortion vector according to a preset predistortion model by using the fifth output signal output by the fifth auxiliary filtering submodule and a clipping-processed baseband signal that is input into the third predistortion processing submodule, multiplying the predistortion vector by the predistortion coefficient to obtain the predistortion signal, and outputting the predistortion signal;
performing, by the sixth auxiliary filtering submodule by using the fourth auxiliary coefficient, filtering processing on the fundamental wave feedback signal output by the radio frequency signal feedback module, and inputting an obtained sixth output signal into a predistortion model processing submodule; and
performing, by the third predistortion coefficient processing submodule, nonlinear modeling according to the sixth output signal, a fundamental wave feedback signal input into the third predistortion coefficient processing submodule, and the predistortion signal, and training the predistortion coefficient of the predistortion model coefficient training module to obtain the predistortion coefficient.
