WO2019085555A1 - 超宽带数字预失真方法、装置和*** - Google Patents

超宽带数字预失真方法、装置和*** Download PDF

Info

Publication number
WO2019085555A1
WO2019085555A1 PCT/CN2018/097584 CN2018097584W WO2019085555A1 WO 2019085555 A1 WO2019085555 A1 WO 2019085555A1 CN 2018097584 W CN2018097584 W CN 2018097584W WO 2019085555 A1 WO2019085555 A1 WO 2019085555A1
Authority
WO
WIPO (PCT)
Prior art keywords
sub
signal
power amplifier
frequency domain
effective bandwidth
Prior art date
Application number
PCT/CN2018/097584
Other languages
English (en)
French (fr)
Inventor
刁穗东
Original Assignee
京信通信***（中国）有限公司
京信通信***（广州）有限公司
京信通信技术(广州)有限公司
天津京信通信***有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 京信通信***（中国）有限公司, 京信通信***（广州）有限公司, 京信通信技术(广州)有限公司, 天津京信通信***有限公司 filed Critical 京信通信***（中国）有限公司
Publication of WO2019085555A1 publication Critical patent/WO2019085555A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
    • H03F1/32—Modifications of amplifiers to reduce non-linear distortion
    • H03F1/3241—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits
    • H03F1/3247—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits using feedback acting on predistortion circuits
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
    • H03F1/32—Modifications of amplifiers to reduce non-linear distortion
    • H03F1/3241—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits
    • H03F1/3258—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits based on polynomial terms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69—Spread spectrum techniques
    • H04B1/7163—Spread spectrum techniques using impulse radio
    • H04B1/71635—Transmitter aspects
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/38—Synchronous or start-stop systems, e.g. for Baudot code
    • H04L25/40—Transmitting circuits; Receiving circuits
    • H04L25/49—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems

Definitions

  • the present invention relates to the field of signal processing technologies for wireless communications, and in particular, to an ultra-wideband digital predistortion method, apparatus and system.
  • wireless communication equipment With the increasing traffic of data services, wireless communication equipment is required to operate in a wider working frequency band to ensure a higher wireless communication rate.
  • the latest wireless communication equipment has an instantaneous working bandwidth exceeding 200 MHz, which places high demands on power amplifiers, but Existing power amplifiers are difficult to guarantee good linearity in the ultra-wideband frequency range, while the traditional digital predistortion method has a poor nonlinear correction effect on ultra-wideband power amplifiers.
  • An ultra-wideband digital predistortion method comprising the following steps:
  • the frequency domain reconstruction is performed on each sub-band in turn, and the input signals of the power amplifier after frequency domain reconstruction are respectively corresponding to the input reconstructed signals in each sub-band, and the output feedback signals of the power amplifier after frequency domain reconstruction are respectively calculated.
  • the baseband sub-signals of the corresponding sub-bands are pre-distorted according to the pre-distortion coefficients of the respective sub-bands.
  • An ultra-wideband digital predistortion device comprising:
  • a subband division module configured to divide the effective bandwidth into a plurality of subbands according to an amplitude frequency response within an effective bandwidth of the power amplifier
  • the predistortion coefficient calculation module is configured to perform frequency domain reconstruction on each subband in turn, respectively calculate an input signal of the power amplifier corresponding to the input reconstructed signal in each subband after frequency domain reconstruction, and respectively calculate frequency domain reconstruction
  • the output feedback signal of the post power amplifier corresponds to the output feedback reconstructed signal in each subband, and the predistortion coefficients of the respective subbands are respectively calculated according to the input reconstructed signal and the output feedback reconstructed signal;
  • the predistortion processing module is configured to perform predistortion processing on the baseband sub-signals of the corresponding sub-bands according to the pre-distortion coefficients of the respective sub-bands.
  • An ultra-wideband digital predistortion system comprising:
  • a predistortion processing unit a data acquisition unit, a subband division unit, and a predistortion coefficient calculation unit;
  • the data acquisition unit acquires an input signal of each sub-band of the power amplifier and an output feedback signal of each sub-band;
  • the subband dividing unit sequentially performs frequency domain reconstruction on each subband, and respectively calculates an input signal of the power amplifier corresponding to the input reconstructed signal in each subband after frequency domain reconstruction, and separately calculates power in the frequency domain reconstruction.
  • the output feedback signal of the amplifier corresponds to the output feedback reconstruction signal in each sub-band;
  • the pre-distortion coefficient calculation unit respectively calculates pre-distortion coefficients of the respective sub-bands according to the input reconstruction signal and the output feedback reconstruction signal;
  • the predistortion processing unit performs predistortion processing on the baseband sub-signals of the corresponding sub-bands according to the pre-distortion coefficients of the respective sub-bands.
  • the above ultra-wideband digital predistortion method, device and system divide subbands according to frequency response, separate subbands and predistort each subband separately, which can effectively improve the nonlinear correction effect on the ultra-wideband power amplifier, thereby improving the super Wideband digital predistortion performance.
  • 1 is a flow chart of an ultra-wideband digital pre-distortion method of an embodiment
  • FIG. 2 is a schematic diagram of dividing a subband within an effective bandwidth according to a frequency response characteristic value according to an embodiment
  • FIG. 3 is a block diagram showing the structure of an ultra-wideband digital predistortion apparatus according to an embodiment
  • FIG. 4 is a schematic diagram of the connection of an ultra-wideband digital pre-distortion system of one embodiment in one embodiment.
  • the present invention provides an ultra-wideband digital predistortion method, which may include the following steps:
  • the effective bandwidth of the power amplifier is the working frequency band of the power amplifier.
  • the power amplifier only processes the data in the working frequency band to ensure the accuracy of the predistortion processing. Assuming that the effective bandwidth of a power amplifier is 1.8 GHz to 2 GHz, signals outside this range cannot be amplified normally.
  • N is the number of points at which the signal is acquired, is a constant, and n is a variable.
  • the meaning of such expression in the following embodiments is similar, and will not be further described in the following embodiments.
  • the amplitude-frequency response within the effective bandwidth of the power amplifier can then be calculated from the effective bandwidth of X(k), Y(k), and the power amplifier. Further, the frequency domain sampling point offset k_off and the frequency domain sampling point number K_in may be calculated according to the effective bandwidth; the frequency domain sampling point offset k_off is taken as the starting point, and the frequency domain sampling point number K_in is the length, and the effective bandwidth is obtained.
  • Frequency domain index k calculating an amplitude frequency within an effective bandwidth of the power amplifier according to a frequency domain form X(k) of the input signal, a frequency domain form Y(k) of the output feedback signal, and the frequency domain index k response.
  • the resulting amplitude-frequency response within the effective bandwidth can be written as:
  • the amplitude average of the amplitude frequency response in the effective bandwidth may be counted; the effective bandwidth is divided into several first level subbands according to the amplitude mean; The phase averages of the phase-frequency responses of the power amplifiers are separately calculated in the band; the corresponding first-order sub-bands are respectively divided into several sub-bands according to the phase mean values corresponding to the respective first-level sub-bands.
  • the amplitude of the amplitude-frequency response of each frequency sampling point in the effective bandwidth may be separately calculated; and the difference between the obtained amplitude and the first amplitude average is less than the preset difference a target frequency sampling point; dividing a first target frequency sampling point whose frequency interval is less than a preset interval into an area; determining a second target frequency sampling point closest to a center of gravity of the area; sampling points according to the second target frequency
  • the effective bandwidth is divided into a number of primary subbands.
  • the secondary sub-bands can be divided in a similar manner, and will not be described here. Firstly, according to the amplitude-frequency response division, and then according to the phase-frequency response division, more precise subdivision, to ensure that the sub-band power amplifier features are consistent.
  • the subband division process will be described below with a specific example.
  • calculate the amplitude mean A of H_in find the point of the mean line ⁇ A, and classify the point whose frequency interval is smaller than ⁇ F into one area, as shown in area A1 and area A2 in Fig. 2, respectively.
  • the points closest to the center of gravity of the region, as shown in Fig. 2 are points 58 and 165, denoted as f_A1 and f_A2, respectively, and H_in is divided into three first-level sub-bands by f_A1 and f_A2, respectively.
  • the second-order sub-band is divided according to the phase-frequency response in the same way.
  • the portion of the input signal other than the sub-band may be zeroed to obtain an input reconstructed signal corresponding to the input signal in each sub-band.
  • the output feedback signal when the output feedback signal is reconstructed, the output feedback signal may be zeroed in a portion other than the extended sub-band corresponding to the sub-band, so that the output feedback signal is corresponding to each sub-band.
  • the output feedback reconstructs the signal; wherein the extended subband is a frequency band in which the subband and its adjacent frequency bands are combined. Wherein the frequency band adjacent to the sub-band includes a higher frequency band and a lower frequency band adjacent to the sub-band.
  • the number of frequency domain points of the extended subband is three times that of the extended subband, and C1 is equal to the number of points of k_in1.
  • the time domain form of the input reconstructed signal and the time domain form of the output feedback reconstructed signal may be calculated. That is, performing an IFFT transform (Inverse Fast Fourier Transform) on the input reconstructed signal and the output feedback reconstructed signal, and then reconstructing the time domain of the signal according to the time domain form of the input reconstructed signal and the output feedback. The form calculates the predistortion coefficient.
  • IFFT transform Inverse Fast Fourier Transform
  • IFFT is converted to x_1 (X_1(k), Y_1(k), X_2(k), Y_2(k),..., X_6(k), Y_6(k), respectively.
  • n y_1(n), x_2(n), y_2(n), ..., x_6(n), y_6(n).
  • the predistortion coefficient c_i of the subband power amplifier model is fitted by y_i and x_i.
  • c_i) obtains the time domain form of the subband of the predistorted input signal x_i'.
  • each baseband sub-signal after the pre-distortion processing may be subjected to a summation process to obtain an input signal after pre-distortion processing;
  • the input signal after the predistortion processing is output to the signal input terminal of the power amplifier.
  • the previous example is continued by summing the different sub-band dimensions of x_1'(n),...,x_6'(n) to obtain x'(n), which is output to the power amplifier.
  • the above ultra-wideband digital predistortion method divides subbands according to frequency response, separates subbands and predistorts each subband separately, can effectively improve nonlinear correction effect on ultra-wideband power amplifier, thereby improving ultra-wideband digital predistortion Performance.
  • the present invention further provides an ultra-wideband digital predistortion apparatus, which may include:
  • the subband division module 110 is configured to divide the effective bandwidth into a plurality of subbands according to an amplitude frequency response within an effective bandwidth of the power amplifier;
  • N is the number of points at which the signal is acquired, is a constant, and n is a variable.
  • the meaning of such expression in the following embodiments is similar, and will not be further described in the following embodiments.
  • the amplitude-frequency response within the effective bandwidth of the power amplifier can then be calculated from the effective bandwidth of X(k), Y(k), and the power amplifier. Further, the frequency domain sampling point offset k_off and the frequency domain sampling point number K_in may be calculated according to the effective bandwidth; the frequency domain sampling point offset k_off is taken as the starting point, and the frequency domain sampling point number K_in is the length, and the effective bandwidth is obtained.
  • Frequency domain index k calculating an amplitude frequency within an effective bandwidth of the power amplifier according to a frequency domain form X(k) of the input signal, a frequency domain form Y(k) of the output feedback signal, and the frequency domain index k response.
  • the resulting amplitude-frequency response within the effective bandwidth can be written as:
  • the amplitude average of the amplitude frequency response in the effective bandwidth may be counted; the effective bandwidth is divided into several first level subbands according to the amplitude mean; The phase averages of the phase-frequency responses of the power amplifiers are separately calculated in the band; the corresponding first-order sub-bands are respectively divided into several sub-bands according to the phase mean values corresponding to the respective first-level sub-bands.
  • the amplitude of the amplitude-frequency response of each frequency sampling point in the effective bandwidth may be separately calculated; and the difference between the obtained amplitude and the first amplitude average is less than the preset difference a target frequency sampling point; dividing a first target frequency sampling point whose frequency interval is less than a preset interval into an area; determining a second target frequency sampling point closest to a center of gravity of the area; sampling points according to the second target frequency
  • the effective bandwidth is divided into a number of primary subbands.
  • the secondary sub-bands can be divided in a similar manner, and will not be described here. Firstly, according to the amplitude-frequency response division, and then according to the phase-frequency response division, more precise subdivision, to ensure that the sub-band power amplifier features are consistent.
  • the subband division process will be described below with a specific example.
  • calculate the amplitude mean A of H_in find the point of the mean line ⁇ A, and classify the point whose frequency interval is smaller than ⁇ F into one area, as shown in area A1 and area A2 in Fig. 2, respectively.
  • the points closest to the center of gravity of the region, as shown in Fig. 2 are points 58 and 165, denoted as f_A1 and f_A2, respectively, and H_in is divided into three first-level sub-bands by f_A1 and f_A2, respectively.
  • the second-order sub-band is divided according to the phase-frequency response in the same way.
  • the pre-distortion coefficient calculation module 120 is configured to perform frequency domain reconstruction on each sub-band in turn, respectively calculate an input signal of the power amplifier corresponding to the input reconstructed signal in each sub-band after frequency domain reconstruction, and calculate a frequency domain weight respectively.
  • the output feedback signal of the post-construction power amplifier corresponds to the output feedback reconstructed signal in each sub-band, and the pre-distortion coefficients of the respective sub-bands are respectively calculated according to the input reconstructed signal and the output feedback reconstructed signal;
  • the portion of the input signal other than the sub-band may be zeroed to obtain an input reconstructed signal corresponding to the input signal in each sub-band.
  • the output feedback signal when the output feedback signal is reconstructed, the output feedback signal may be zeroed in a portion other than the extended sub-band corresponding to the sub-band, so that the output feedback signal is corresponding to each sub-band.
  • the output feedback reconstructs the signal; wherein the extended subband is a frequency band in which the subband and its adjacent frequency bands are combined. Wherein the frequency band adjacent to the sub-band includes a higher frequency band and a lower frequency band adjacent to the sub-band.
  • the number of frequency domain points of the extended subband is three times that of the extended subband, and C1 is equal to the number of points of k_in1.
  • the time domain form of the input reconstructed signal and the time domain form of the output feedback reconstructed signal may be calculated. That is, performing an IFFT transform (Inverse Fast Fourier Transform) on the input reconstructed signal and the output feedback reconstructed signal, and then reconstructing the time domain of the signal according to the time domain form of the input reconstructed signal and the output feedback. The form calculates the predistortion coefficient.
  • IFFT transform Inverse Fast Fourier Transform
  • IFFT is converted to x_1 (X_1(k), Y_1(k), X_2(k), Y_2(k),..., X_6(k), Y_6(k), respectively.
  • n y_1(n), x_2(n), y_2(n), ..., x_6(n), y_6(n).
  • the predistortion coefficient c_i of the subband power amplifier model is fitted by y_i and x_i.
  • the predistortion processing module 130 is configured to perform predistortion processing on the baseband sub-signals of the corresponding sub-bands according to the pre-distortion coefficients of the respective sub-bands.
  • c_i) obtains the time domain form of the subband of the predistorted input signal x_i'.
  • each baseband sub-signal after the pre-distortion processing may be subjected to a summation process to obtain an input signal after pre-distortion processing;
  • the input signal after the predistortion processing is output to the signal input terminal of the power amplifier.
  • the previous example is continued by summing the different sub-band dimensions of x_1'(n),...,x_6'(n) to obtain x'(n), which is output to the power amplifier.
  • the ultra-wideband digital predistortion device divides the subband according to the frequency response, separates the subbands and separately predistorts the subbands, thereby effectively improving the nonlinear correction effect on the ultra-wideband power amplifier, thereby improving the ultra-wideband digital predistortion. Performance.
  • the ultra-wideband digital predistortion device of the present invention has a one-to-one correspondence with the ultra-wideband digital predistortion method of the present invention, and the technical features and beneficial effects described in the embodiments of the ultra-wideband digital predistortion method are applicable to ultra-wideband digital predistortion. In the embodiment of the device, it is hereby declared.
  • the present invention further provides an ultra-wideband digital predistortion system, which may include:
  • Predistortion processing unit 210 data acquisition unit 220, subband division unit 230 and predistortion coefficient calculation unit 240;
  • the data collection unit 220 acquires an input signal of each sub-band of the power amplifier and an output feedback signal of each sub-band;
  • the subband dividing unit 230 sequentially performs frequency domain reconstruction on each subband, and respectively calculates an input signal of the power amplifier corresponding to the input reconstructed signal in each subband after the frequency domain reconstruction, and respectively calculates the frequency domain reconstruction.
  • the output feedback signal of the power amplifier corresponds to the output feedback reconstruction signal in each sub-band;
  • the pre-distortion coefficient calculation unit 240 respectively calculates pre-distortion coefficients of the respective sub-bands according to the input reconstruction signal and the output feedback reconstruction signal;
  • the predistortion processing unit 210 performs predistortion processing on the baseband sub-signals of the corresponding sub-bands according to the pre-distortion coefficients of the respective sub-bands.
  • the effective bandwidth of the power amplifier is the working frequency band of the power amplifier.
  • the power amplifier only processes the data in the working frequency band to ensure the accuracy of the predistortion processing. Assuming that the effective bandwidth of a power amplifier is 1.8 GHz to 2 GHz, signals outside this range cannot be amplified normally.
  • N is the number of points at which the signal is acquired, is a constant, and n is a variable.
  • the amplitude-frequency response within the effective bandwidth of the power amplifier can then be calculated from the effective bandwidth of X(k), Y(k), and the power amplifier. Further, the frequency domain sampling point offset k_off and the frequency domain sampling point number K_in may be calculated according to the effective bandwidth; the frequency domain sampling point offset k_off is taken as the starting point, and the frequency domain sampling point number K_in is the length, and the effective bandwidth is obtained.
  • Frequency domain index k calculating an amplitude frequency within an effective bandwidth of the power amplifier according to a frequency domain form X(k) of the input signal, a frequency domain form Y(k) of the output feedback signal, and the frequency domain index k response.
  • the resulting amplitude-frequency response within the effective bandwidth can be written as:
  • the above operation may be performed in advance, and after obtaining the amplitude-frequency response in the effective bandwidth of the power amplifier, the amplitude-frequency response in the effective bandwidth may be pre-stored to the splitter 220 for performing subsequent operations.
  • the amplitude average of the amplitude frequency response in the effective bandwidth may be counted; the effective bandwidth is divided into several first level subbands according to the amplitude mean; The phase averages of the phase-frequency responses of the power amplifiers are separately calculated in the band; the corresponding first-order sub-bands are respectively divided into several sub-bands according to the phase mean values corresponding to the respective first-level sub-bands.
  • the amplitude of the amplitude-frequency response of each frequency sampling point in the effective bandwidth may be separately calculated; and the difference between the obtained amplitude and the first amplitude average is less than the preset difference a target frequency sampling point; dividing a first target frequency sampling point whose frequency interval is less than a preset interval into an area; determining a second target frequency sampling point closest to a center of gravity of the area; sampling points according to the second target frequency
  • the effective bandwidth is divided into a number of primary subbands.
  • the secondary sub-bands can be divided in a similar manner, and will not be described here. Firstly, according to the amplitude-frequency response division, and then according to the phase-frequency response division, more precise subdivision, to ensure that the sub-band power amplifier features are consistent.
  • the subband division process will be described below with a specific example.
  • calculate the amplitude mean A of H_in find the point of the mean line ⁇ A, and classify the point whose frequency interval is smaller than ⁇ F into one area, as shown in area A1 and area A2 in Fig. 2, respectively.
  • the points closest to the center of gravity of the region, as shown in Fig. 2 are points 58 and 165, denoted as f_A1 and f_A2, respectively, and H_in is divided into three first-level sub-bands by f_A1 and f_A2, respectively.
  • the second-order sub-band is divided according to the phase-frequency response in the same way.
  • the input signal when the data processing unit 240 performs the input signal reconstruction, the input signal may be zeroed in a portion other than the sub-band, and the input signal corresponding to the input reconstructed signal in each sub-band is obtained. .
  • the output feedback signal when the output feedback signal is reconstructed, the output feedback signal may be zeroed in a portion other than the extended sub-band corresponding to the sub-band, so that the output feedback signal is corresponding to each sub-band.
  • the output feedback reconstructs the signal; wherein the extended subband is a frequency band in which the subband and its adjacent frequency bands are combined. Wherein the frequency band adjacent to the sub-band includes a higher frequency band and a lower frequency band adjacent to the sub-band.
  • the number of frequency domain points of the extended subband is three times that of the extended subband, and C1 is equal to the number of points of k_in1.
  • the time domain form of the input reconstructed signal and the time domain form of the output feedback reconstructed signal may be calculated. That is, performing an IFFT transform (Inverse Fast Fourier Transform) on the input reconstructed signal and the output feedback reconstructed signal, and then reconstructing the time domain of the signal according to the time domain form of the input reconstructed signal and the output feedback. The form calculates the predistortion coefficient.
  • IFFT transform Inverse Fast Fourier Transform
  • IFFT is converted to x_1 (X_1(k), Y_1(k), X_2(k), Y_2(k),..., X_6(k), Y_6(k), respectively.
  • n y_1(n), x_2(n), y_2(n), ..., x_6(n), y_6(n).
  • the predistortion coefficient c_i of the subband power amplifier model is fitted by y_i and x_i.
  • c_i) obtains the time domain form of the subband of the predistorted input signal x_i'.
  • the ultra-wideband digital pre-distortion system of the present invention may further comprise a combiner; the combiner performs a summation process on each of the baseband sub-signals after the pre-distortion processing to obtain a pre-distorted input. And outputting the predistortion processed input signal to a signal input end of the power amplifier.
  • the previous example is continued by summing the different sub-band dimensions of x_1'(n),...,x_6'(n) to obtain x'(n), which is output to the power amplifier.
  • the above ultra-wideband digital predistortion system divides subbands according to frequency response, separates subbands and separately predistorts each subband, which can effectively improve the nonlinear correction effect on the ultra-wideband power amplifier, thereby improving the ultra-wideband digital predistortion. Performance.
  • a "computer-readable medium” can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with such an instruction execution system, apparatus, or device.
  • computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.
  • portions of the invention may be implemented in hardware, software, firmware or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: having logic gates for implementing logic functions on data signals. Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Power Engineering (AREA)
  • Nonlinear Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Algebra (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Amplifiers (AREA)

Abstract

本发明涉及一种超宽带数字预失真方法、装置和***，根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带，分离子带并对各个子带分别进行预失真，能够有效地提高对超宽带功率放大器的非线性纠正效果，从而提高超宽带数字预失真的性能。

Description

超宽带数字预失真方法、装置和*** 技术领域
本发明涉及无线通信的信号处理技术领域，特别是涉及一种超宽带数字预失真方法、装置和***。
背景技术
在现代无线通信***中，较高功率的发射机是整个无线通信***中能耗最高的部分，为了节省整个***的能耗，提高发射机的能效比是最佳选择，而提高静态工作点可以提高发射机能效比，却由于进入功率放大器的非线性区间，导致了发射信号的非线性失真。目前解决非线性失真的方法最常用的是数字预失真。
随着数据业务流量的日益增加，需要无线通信设备工作在更宽的工作频带来保证较高的无线通信速率，最新的无线通信设备瞬时工作带宽超过200MHz，对功率放大器提出很高的要求，但现有的功率放大器难以在超宽带频率范围内保证良好的线性指标，而传统的数字预失真方法对超宽带功率放大器的非线性纠正效果较差。
发明内容
基于此，有必要针对传统的数字预失真方法对超宽带功率放大器的非线性纠正效果较差的问题，提供一种超宽带数字预失真方法、装置和***。
一种超宽带数字预失真方法，包括以下步骤：
根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带；
依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
一种超宽带数字预失真装置，包括：
子带划分模块，用于根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带；
预失真系数计算模块，用于依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
预失真处理模块，用于分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
一种超宽带数字预失真***，包括：
预失真处理单元，数据采集单元，子带划分单元和预失真系数计算单元；
所述数据采集单元采集功率放大器各个子带的输入信号和各个子带的输出反馈信号；
所述子带划分单元依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号；
所述预失真系数计算单元根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
所述预失真处理单元分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
上述超宽带数字预失真方法、装置和***，根据频率响应划分子带，分离子带并对各个子带分别进行预失真，能够有效地提高对超宽带功率放大器的非线性纠正效果，从而提高超宽带数字预失真的性能。
附图说明
图1为一个实施例的超宽带数字预失真方法流程图；
图2为一个实施例的按照频率响应特征值划分有效带宽内的子带的示意图图；
图3为一个实施例的超宽带数字预失真装置的结构框图；
图4为一个实施例的超宽带数字预失真***在一个实施例中的连接示意图。
具体实施方式
下面将结合较佳实施例及附图对本发明的内容作进一步详细描述。显然，下文所描述的实施例仅用于解释本发明，而非对本发明的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与本发明相关的部分而非全部内容。
如图1所示，本发明提供一种超宽带数字预失真方法，可包括以下步骤：
S1，根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带；
在本步骤中，为了保证数字预失真处理结果的准确性，只对功率放大器有效带宽内的数据进行处理。有效带宽即功率放大器的工作频段，功率放大器只对工作频段内的数据做处理能保证预失真处理准确性。假设某功放的有效带宽为1.8GHz～2GHz，则超出这个范围的信号不能被正常的放大。在实际操作时，可以采集功率放大器的输入信号和输出反馈信号，分别设为x(n)和y(n)，n＝1：N。N是采集信号的点数，是常数，n是变量。其中，n＝1：N表示n的取值为大于或等于1，且小于或等于N的正整数，后续实施例中的此类表述含义与之类似，在后续实施例中不再赘述。可根据x(n)和y(n)来计算功率放大器的幅频响应。具体地，可以计算输入信号x(n)的频域形式和输出反馈信号y(n)的频域形式，即分别对x(n)和y(n)进行FFT变换(Fast Fourier Transformation，快速傅里叶变换)，得到X(k)和Y(k)，k＝1：K，其中K＝N。然后，可根据X(k)、Y(k)和功率放大器的有效带宽计算功率放大器有效带宽内的幅频响应。进一步地，可以根据有效带宽计算频域采样点偏置k_off和频域采样点个数K_in；以频域采样点偏置k_off为起始点，频域采样点个数K_in为长度，得到有效带宽的频域索引k；根据所述输入信号的频域形式X(k)、所述输出反馈信号的频域形式Y(k)和所述频域索 引k计算所述功率放大器有效带宽内的幅频响应。得到的有效带宽内的幅频响应可记为：
H_in(k_in)＝H(k)＝Y(k)/X(k)，k_in＝1：K_in，k＝k_in+k_off。
具体来说，假设时域采样点数为512，采样率为491.52MHz，有效带宽240MHz，起始点125.76MHz，则K_in＝512/491.52*240＝250，k_in＝1：250，k_off＝125.76/491.52*512＝131，k＝132：381。
在一个实施例中，在划分子带时，可以统计所述有效带宽内所述幅频响应的幅度均值；根据幅度均值将所述有效带宽划分为若干个一级子带；在各个一级子带内分别统计功率放大器的相频响应的相位均值；根据各个一级子带对应的相位均值分别将对应的一级子带划分为若干个子带。进一步地，在划分一级子带时，可以分别计算所述有效带宽内各个频率采样点的幅频响应的幅值；获取幅值与所述第一幅度均值之差小于预设差值的第一目标频率采样点；将频率间隔小于预设间隔的第一目标频率采样点划分为一个区域；确定最接近所述区域的重心的第二目标频率采样点；根据所述第二目标频率采样点将所述有效带宽划分为若干个一级子带。同理，可以采取类似的方式划分二级子带，此处不再赘述。先按照幅频响应划分，再按照相频响应划分，更加精确的细分，保证了子带内功放特征一致。
下面以一个具体的例子对子带划分过程进行说明。如图2所示，计算H_in的幅度均值A，找出距离均值线±ΔA的点，对频率间隔小于ΔF的点归类为一个区域，如图2中区域A1和区域A2，分别在区域找到最接近该区域重心的点，如图2中横坐标为58和165的点，分别记为f_A1和f_A2，分别以f_A1和f_A2为边界划分H_in为3个一级子带。然后在各个子带内以同样的方法按照相频响应划分二级子带，假设最终以边界f_B1，f_A1，f_B2，f_A2，f_B3划分了6个子带，得到H_in(k_in1)，H_in(k_in2)，...，H_in(k_in6)，其中，k_in1＝1：f_B1-1，k_in2＝f_B1：f_A1-1，k_in3＝f_A1：f_B2-1，k_in4＝f_B2：f_A2-1，k_in5＝f_A2：f_B3-1，k_in6＝B3：K_in。
S2，依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器 的输出反馈信号对应在各个子带内的输出反馈重构信号，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
在一个实施例中，进行输入信号重构时，可以将所述输入信号在所述子带以外的部分置零，得到所述输入信号对应在各个子带内的输入重构信号。在另一个实施例中，进行输出反馈信号重构时，可以将所述输出反馈信号在所述子带对应的扩展子带以外的部分置零，得到所述输出反馈信号对应在各个子带内的输出反馈重构信号；其中，所述扩展子带为所述子带及其相邻的频带合并而成的频带。其中，与子带相邻的频带包括与该子带相邻的更高频率的频带和更低频率的频带。
具体来说，继续前文例子，对第1子带重构过程中，对功率放大器的输入信号X(k)的重构的第1子带有X_1(k)＝X(k)，k＝k_in1+k_off；X_1(k)＝0，k≠k_in1+k_off，对功率放大器的输出反馈信号Y(k)的重构的第1子带有Y_1(k)＝Y(k)，k＝k_in1′+k_off；Y_1(k)＝0，k≠k_in1′+k_off，其中，k_in1′＝1-C1：f_B1-1+C1，C1为正整数。优选地，扩展子带的频域点数是扩展前子带的3倍，此时C1等于k_in1的点数。
同理，对第2子带重构过程中，对功率放大器的输入信号X(k)的重构的第2子带有X_2(k)＝X(k)，k＝k_in2+k_off；X_2(k)＝0，k≠k_in2+k_off，对功率放大器的输出反馈信号Y(k)的重构的第2子带有Y_2(k)＝Y(k)，k＝k_in2′+k_off；Y_2(k)＝0，k≠k_in2′+k_off，其中，k_in2′＝f_B1-C2：f_A1+C2，C2为正整数。
如此类推至第6子带重构。
在一个实施例中，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数时，可以计算输入重构信号的时域形式和输出反馈重构信号的时域形式，即对输入重构信号和输出反馈重构信号分别进行IFFT变换(Inverse Fast Fourier Transform，快速傅里叶反变换)，然后根据输入重构信号的时域形式和输出反馈重构信号的时域形式计算预失真系数。继续前文例子，具体来说，是对X_1(k)，Y_1(k)，X_2(k)，Y_2(k)，...，X_6(k)，Y_6(k)分别进行IFFT变换为x_1(n)，y_1(n)，x_2(n)，y_2(n)，...，x_6(n)，y_6(n)。
进一步地，在计算预失真系数时，可以将输入重构信号的时域形式和输出 反馈重构信号的时域形式代入预存的子带功放模型x_i＝f(y_i，c_i)，其中x_i为第i子带的输入信号x_i(n)的简写，y_i为第i子带的输出反馈信号y_i(n)的简写，c_i为预失真系数。通过y_i和x_i拟合出子带功放模型的预失真系数c_i。
S3，分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
在一个实施例中，可以对基带信号z(m)，按照前面的子带划分和重构方法处理后得到基带信号的子带形式z_i(m)，简写为z_i，代入x_i′＝f(z_i，c_i)得到预失真后的输入信号x_i′的子带的时域形式。
在进一步的实施例中，在对各个子带的基带子信号进行预失真处理之后，还可以对预失真处理后的各个基带子信号进行求和处理，得到预失真处理后的输入信号；将所述预失真处理后的输入信号输出至所述功率放大器的信号输入端。具体来说，继续前文例子，是对x_1′(n)，...，x_6′(n)的不同子带维度上求和，得到x′(n)，输出到功率放大器。
上述超宽带数字预失真方法，根据频率响应划分子带，分离子带并对各个子带分别进行预失真，能够有效地提高对超宽带功率放大器的非线性纠正效果，从而提高超宽带数字预失真的性能。
如图3所示，本发明还提供一种超宽带数字预失真装置，可包括：
子带划分模块110，用于根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带；
在本模块中，为了保证数字预失真处理结果的准确性，只对功率放大器有效带宽内的数据进行处理。有效带宽即功率放大器的工作频段，功率放大器只对工作频段内的数据做处理能保证预失真处理准确性。假设某功放的有效带宽为1.8GHz～2GHz，则超出这个范围的信号不能被正常的放大。在实际操作时，可以采集功率放大器的输入信号和输出反馈信号，分别设为x(n)和y(n)，n＝1：N。N是采集信号的点数，是常数，n是变量。其中，n＝1：N表示n的取值为大于或等于1，且小于或等于N的正整数，后续实施例中的此类表述含义与之类似，在后续实施例中不再赘述。可根据x(n)和y(n)来计算功率放大器的幅频响应。具体地，可以计算输入信号x(n)的频域形式和输出反馈信号y(n)的频域形式，即分别 对x(n)和y(n)进行FFT变换(Fast Fourier Transformation，快速傅里叶变换)，得到X(k)和Y(k)，k＝1：K，其中K＝N。然后，可根据X(k)、Y(k)和功率放大器的有效带宽计算功率放大器有效带宽内的幅频响应。进一步地，可以根据有效带宽计算频域采样点偏置k_off和频域采样点个数K_in；以频域采样点偏置k_off为起始点，频域采样点个数K_in为长度，得到有效带宽的频域索引k；根据所述输入信号的频域形式X(k)、所述输出反馈信号的频域形式Y(k)和所述频域索引k计算所述功率放大器有效带宽内的幅频响应。得到的有效带宽内的幅频响应可记为：
H_in(k_in)＝H(k)＝Y(k)/X(k)，k_in＝1：K_in，k＝k_in+k_off。
具体来说，假设时域采样点数为512，采样率为491.52MHz，有效带宽240MHz，起始点125.76MHz，则K_in＝512/491.52*240＝250，k_in＝1：250，k_off＝125.76/491.52*512＝131，k＝132：381。
在一个实施例中，在划分子带时，可以统计所述有效带宽内所述幅频响应的幅度均值；根据幅度均值将所述有效带宽划分为若干个一级子带；在各个一级子带内分别统计功率放大器的相频响应的相位均值；根据各个一级子带对应的相位均值分别将对应的一级子带划分为若干个子带。进一步地，在划分一级子带时，可以分别计算所述有效带宽内各个频率采样点的幅频响应的幅值；获取幅值与所述第一幅度均值之差小于预设差值的第一目标频率采样点；将频率间隔小于预设间隔的第一目标频率采样点划分为一个区域；确定最接近所述区域的重心的第二目标频率采样点；根据所述第二目标频率采样点将所述有效带宽划分为若干个一级子带。同理，可以采取类似的方式划分二级子带，此处不再赘述。先按照幅频响应划分，再按照相频响应划分，更加精确的细分，保证了子带内功放特征一致。
下面以一个具体的例子对子带划分过程进行说明。如图2所示，计算H_in的幅度均值A，找出距离均值线±ΔA的点，对频率间隔小于ΔF的点归类为一个区域，如图2中区域A1和区域A2，分别在区域找到最接近该区域重心的点，如图2中横坐标为58和165的点，分别记为f_A1和f_A2，分别以f_A1和f_A2为边界划分H_in为3个一级子带。然后在各个子带内以同样的方法按照相频响 应划分二级子带，假设最终以边界f_B1，f_A1，f_B2，f_A2，f_B3划分了6个子带，得到H_in(k_in1)，H_in(k_in2)，...，H_in(k_in6)，其中，k_in1＝1：f_B1-1，k_in2＝f_B1：f_A1-1，k_in3＝f_A1：f_B2-1，k_in4＝f_B2：f_A2-1，k_in5＝f_A2：f_B3-1，k_in6＝B3：K_in。
预失真系数计算模块120，用于依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
在一个实施例中，进行输入信号重构时，可以将所述输入信号在所述子带以外的部分置零，得到所述输入信号对应在各个子带内的输入重构信号。在另一个实施例中，进行输出反馈信号重构时，可以将所述输出反馈信号在所述子带对应的扩展子带以外的部分置零，得到所述输出反馈信号对应在各个子带内的输出反馈重构信号；其中，所述扩展子带为所述子带及其相邻的频带合并而成的频带。其中，与子带相邻的频带包括与该子带相邻的更高频率的频带和更低频率的频带。
具体来说，继续前文例子，对第1子带重构过程中，对功率放大器的输入信号X(k)的重构的第1子带有X_1(k)＝X(k)，k＝k_in1+k_off；X_1(k)＝0，k≠k_in1+k_off，对功率放大器的输出反馈信号Y(k)的重构的第1子带有Y_1(k)＝Y(k)，k＝k_in1′+k_off；Y_1(k)＝0，k≠k_in1′+k_off，其中，k_in1′＝1-C1：f_B1-1+C1，C1为正整数。优选地，扩展子带的频域点数是扩展前子带的3倍，此时C1等于k_in1的点数。
同理，对第2子带重构过程中，对功率放大器的输入信号X(k)的重构的第2子带有X_2(k)＝X(k)，k＝k_in2+k_off；X_2(k)＝0，k≠k_in2+k_off，对功率放大器的输出反馈信号Y(k)的重构的第2子带有Y_2(k)＝Y(k)，k＝k_in2′+k_off；Y_2(k)＝0，k≠k_in2′+k_off，其中，k_in2′＝f_B1-C2：f_A1+C2，C2为正整数。
如此类推至第6子带重构。
在一个实施例中，根据所述输入重构信号和输出反馈重构信号分别计算各 个子带的预失真系数时，可以计算输入重构信号的时域形式和输出反馈重构信号的时域形式，即对输入重构信号和输出反馈重构信号分别进行IFFT变换(Inverse Fast Fourier Transform，快速傅里叶反变换)，然后根据输入重构信号的时域形式和输出反馈重构信号的时域形式计算预失真系数。继续前文例子，具体来说，是对X_1(k)，Y_1(k)，X_2(k)，Y_2(k)，...，X_6(k)，Y_6(k)分别进行IFFT变换为x_1(n)，y_1(n)，x_2(n)，y_2(n)，...，x_6(n)，y_6(n)。
进一步地，在计算预失真系数时，可以将输入重构信号的时域形式和输出反馈重构信号的时域形式代入预存的子带功放模型x_i＝f(y_i，c_i)，其中x_i为第i子带的输入信号x_i(n)的简写，y_i为第i子带的输出反馈信号y_i(n)的简写，c_i为预失真系数。通过y_i和x_i拟合出子带功放模型的预失真系数c_i。
预失真处理模块130，用于分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
在一个实施例中，可以对基带信号z(m)，按照前面的子带划分和重构方法处理后得到基带信号的子带形式z_i(m)，简写为z_i，代入x_i′＝f(z_i，c_i)得到预失真后的输入信号x_i′的子带的时域形式。
在进一步的实施例中，在对各个子带的基带子信号进行预失真处理之后，还可以对预失真处理后的各个基带子信号进行求和处理，得到预失真处理后的输入信号；将所述预失真处理后的输入信号输出至所述功率放大器的信号输入端。具体来说，继续前文例子，是对x_1′(n)，...，x_6′(n)的不同子带维度上求和，得到x′(n)，输出到功率放大器。
上述超宽带数字预失真装置，根据频率响应划分子带，分离子带并对各个子带分别进行预失真，能够有效地提高对超宽带功率放大器的非线性纠正效果，从而提高超宽带数字预失真的性能。
本发明的超宽带数字预失真装置与本发明的超宽带数字预失真方法一一对应，在上述超宽带数字预失真方法的实施例阐述的技术特征及其有益效果均适用于超宽带数字预失真装置的实施例中，特此声明。
如图4所示，本发明还提供一种超宽带数字预失真***，可包括：
预失真处理单元210，数据采集单元220，子带划分单元230和预失真系数 计算单元240；
所述数据采集单元220采集功率放大器各个子带的输入信号和各个子带的输出反馈信号；
所述子带划分单元230依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号；
所述预失真系数计算单元240根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
所述预失真处理单元210分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
在工作时，为了保证数字预失真处理结果的准确性，可以只对功率放大器有效带宽内的数据进行处理。有效带宽即功率放大器的工作频段，功率放大器只对工作频段内的数据做处理能保证预失真处理准确性。假设某功放的有效带宽为1.8GHz～2GHz，则超出这个范围的信号不能被正常的放大。在实际操作时，可以采集功率放大器的输入信号和输出反馈信号，分别设为x(n)和y(n)，n＝1：N。N是采集信号的点数，是常数，n是变量。其中，n＝1：N表示n的取值为大于或等于1，且小于或等于N的正整数，后续实施例中的此类表述含义与之类似，在后续实施例中不再赘述。数据处理单元240可根据x(n)和y(n)来计算功率放大器的幅频响应。具体地，可以计算输入信号x(n)的频域形式和输出反馈信号y(n)的频域形式，即分别对x(n)和y(n)进行FFT变换(Fast Fourier Transformation，快速傅里叶变换)，得到X(k)和Y(k)，k＝1：K，其中K＝N。然后，可根据X(k)、Y(k)和功率放大器的有效带宽计算功率放大器有效带宽内的幅频响应。进一步地，可以根据有效带宽计算频域采样点偏置k_off和频域采样点个数K_in；以频域采样点偏置k_off为起始点，频域采样点个数K_in为长度，得到有效带宽的频域索引k；根据所述输入信号的频域形式X(k)、所述输出反馈信号的频域形式Y(k)和所述频域索引k计算所述功率放大器有效带宽内的幅频响应。得到的有效带宽内的幅频响应可记为：
H_in(k_in)＝H(k)＝Y(k)/X(k)，k_in＝1：K_in，k＝k_in+k_off。
具体来说，假设时域采样点数为512，采样率为491.52MHz，有效带宽240MHz，起始点125.76MHz，则K_in＝512/491.52*240＝250，k_in＝1：250，k_off＝125.76/491.52*512＝131，k＝132：381。
上述操作可以预先执行，得到功率放大器有效带宽内的幅频响应之后，可以将该有效带宽内的幅频响应预存至分路器220，以便执行后续操作。
在一个实施例中，在划分子带时，可以统计所述有效带宽内所述幅频响应的幅度均值；根据幅度均值将所述有效带宽划分为若干个一级子带；在各个一级子带内分别统计功率放大器的相频响应的相位均值；根据各个一级子带对应的相位均值分别将对应的一级子带划分为若干个子带。进一步地，在划分一级子带时，可以分别计算所述有效带宽内各个频率采样点的幅频响应的幅值；获取幅值与所述第一幅度均值之差小于预设差值的第一目标频率采样点；将频率间隔小于预设间隔的第一目标频率采样点划分为一个区域；确定最接近所述区域的重心的第二目标频率采样点；根据所述第二目标频率采样点将所述有效带宽划分为若干个一级子带。同理，可以采取类似的方式划分二级子带，此处不再赘述。先按照幅频响应划分，再按照相频响应划分，更加精确的细分，保证了子带内功放特征一致。
下面以一个具体的例子对子带划分过程进行说明。如图2所示，计算H_in的幅度均值A，找出距离均值线±ΔA的点，对频率间隔小于ΔF的点归类为一个区域，如图2中区域A1和区域A2，分别在区域找到最接近该区域重心的点，如图2中横坐标为58和165的点，分别记为f_A1和f_A2，分别以f_A1和f_A2为边界划分H_in为3个一级子带。然后在各个子带内以同样的方法按照相频响应划分二级子带，假设最终以边界f_B1，f_A1，f_B2，f_A2，f_B3划分了6个子带，得到H_in(k_in1)，H_in(k_in2)，...，H_in(k_in6)，其中，k_in1＝1：f_B1-1，k_in2＝f_B1：f_A1-1，k_in3＝f_A1：f_B2-1，k_in4＝f_B2：f_A2-1，k_in5＝f_A2：f_B3-1，k_in6＝B3：K_in。
在一个实施例中，数据处理单元240进行输入信号重构时，可以将所述输入信号在所述子带以外的部分置零，得到所述输入信号对应在各个子带内的输入重构信号。在另一个实施例中，进行输出反馈信号重构时，可以将所述输出 反馈信号在所述子带对应的扩展子带以外的部分置零，得到所述输出反馈信号对应在各个子带内的输出反馈重构信号；其中，所述扩展子带为所述子带及其相邻的频带合并而成的频带。其中，与子带相邻的频带包括与该子带相邻的更高频率的频带和更低频率的频带。
具体来说，继续前文例子，对第1子带重构过程中，对功率放大器的输入信号X(k)的重构的第1子带有X_1(k)＝X(k)，k＝k_in1+k_off；X_1(k)＝0，k≠k_in1+k_off，对功率放大器的输出反馈信号Y(k)的重构的第1子带有Y_1(k)＝Y(k)，k＝k_in1′+k_off；Y_1(k)＝0，k≠k_in1′+k_off，其中，k_in1′＝1-C1：f_B1-1+C1，C1为正整数。优选地，扩展子带的频域点数是扩展前子带的3倍，此时C1等于k_in1的点数。
同理，对第2子带重构过程中，对功率放大器的输入信号X(k)的重构的第2子带有X_2(k)＝X(k)，k＝k_in2+k_off；X_2(k)＝0，k≠k_in2+k_off，对功率放大器的输出反馈信号Y(k)的重构的第2子带有Y_2(k)＝Y(k)，k＝k_in2′+k_off；Y_2(k)＝0，k≠k_in2′+k_off，其中，k_in2′＝f_B1-C2：f_A1+C2，C2为正整数。
如此类推至第6子带重构。
在一个实施例中，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数时，可以计算输入重构信号的时域形式和输出反馈重构信号的时域形式，即对输入重构信号和输出反馈重构信号分别进行IFFT变换(Inverse Fast Fourier Transform，快速傅里叶反变换)，然后根据输入重构信号的时域形式和输出反馈重构信号的时域形式计算预失真系数。继续前文例子，具体来说，是对X_1(k)，Y_1(k)，X_2(k)，Y_2(k)，...，X_6(k)，Y_6(k)分别进行IFFT变换为x_1(n)，y_1(n)，x_2(n)，y_2(n)，...，x_6(n)，y_6(n)。
进一步地，在计算预失真系数时，可以将输入重构信号的时域形式和输出反馈重构信号的时域形式代入预存的子带功放模型x_i＝f(y_i，c_i)，其中x_i为第i子带的输入信号x_i(n)的简写，y_i为第i子带的输出反馈信号y_i(n)的简写，c_i为预失真系数。通过y_i和x_i拟合出子带功放模型的预失真系数c_i。
在一个实施例中，可以对基带信号z(m)，按照前面的子带划分和重构方法处理后得到基带信号的子带形式z_i(m)，简写为z_i，代入x_i′＝f(z_i，c_i)得到 预失真后的输入信号x_i′的子带的时域形式。
在进一步的实施例中，本发明的超宽带数字预失真***还可包括合路器；所述合路器对预失真处理后的各个基带子信号进行求和处理，得到预失真处理后的输入信号，并将所述预失真处理后的输入信号输出至所述功率放大器的信号输入端。具体来说，继续前文例子，是对x_1′(n)，...，x_6′(n)的不同子带维度上求和，得到x′(n)，输出到功率放大器。
上述超宽带数字预失真***，根据频率响应划分子带，分离子带并对各个子带分别进行预失真，能够有效地提高对超宽带功率放大器的非线性纠正效果，从而提高超宽带数字预失真的性能。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行***、装置或设备(如基于计算机的***、包括处理器的***或其他可以从指令执行***、装置或设备取指令并执行指令的***)使用，或结合这些指令执行***、装置或设备而使用。就本说明书而言，“计算机可读介质”可以是任何可以包含、存储、通信、传播或传输程序以供指令执行***、装置或设备或结合这些指令执行***、装置或设备而使用的装置。
计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本发明的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行***执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有 用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。

Claims (10)

  1. 一种超宽带数字预失真方法，其特征在于，包括以下步骤：
    根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带；
    依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
    分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
  2. 根据权利要求1所述的超宽带数字预失真方法，其特征在于，在分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理之后，还包括以下步骤：
    对预失真处理后的各个基带子信号进行求和处理，得到预失真处理后的输入信号；
    将所述预失真处理后的输入信号输出至所述功率放大器的信号输入端。
  3. 根据权利要求1所述的超宽带数字预失真方法，其特征在于，在根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带之前，还包括以下步骤：
    计算所述输入信号的频域形式和所述输出反馈信号的频域形式；
    根据所述输入信号的频域形式、所述输出反馈信号的频域形式和所述功率放大器的有效带宽计算所述功率放大器有效带宽内的幅频响应。
  4. 根据权利要求3所述的超宽带数字预失真方法，其特征在于，根据所述输入信号的频域形式、所述输出反馈信号的频域形式和所述功率放大器的有效带宽计算所述功率放大器有效带宽内的幅频响应的步骤包括：
    根据所述有效带宽计算频域采样点偏置和频域采样点个数；
    以频域采样点偏置为起始点，频域采样点个数为长度，得到有效带宽的频域索引；
    根据所述输入信号的频域形式、所述输出反馈信号的频域形式和所述频域索引计算所述功率放大器有效带宽内的幅频响应。
  5. 根据权利要求1所述的超宽带数字预失真方法，其特征在于，根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带的步骤包括：
    统计所述有效带宽内所述幅频响应的幅度均值；
    根据幅度均值将所述有效带宽划分为若干个一级子带；
    在各个一级子带内分别统计功率放大器的相频响应的相位均值；
    根据各个一级子带对应的相位均值分别将对应的一级子带划分为若干个子带。
  6. 根据权利要求5所述的超宽带数字预失真方法，其特征在于，根据第一幅度均值将所述有效带宽划分为若干个一级子带的步骤包括：
    分别计算所述有效带宽内各个频率采样点的幅频响应的幅值；
    获取幅值与所述第一幅度均值之差小于预设差值的第一目标频率采样点；
    将频率间隔小于预设间隔的第一目标频率采样点划分为一个区域；
    确定最接近所述区域的重心的第二目标频率采样点；
    根据所述第二目标频率采样点将所述有效带宽划分为若干个一级子带。
  7. 根据权利要求1所述的超宽带数字预失真方法，其特征在于，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号的步骤包括：
    将所述输入信号在所述子带以外的部分置零，得到所述输入信号对应在各个子带内的输入重构信号；
    和/或
    分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号的步骤包括：
    将所述输出反馈信号在所述子带对应的扩展子带以外的部分置零，得到所述输出反馈信号对应在各个子带内的输出反馈重构信号；
    其中，所述扩展子带为所述子带及其相邻的频带合并而成的频带。
  8. 一种超宽带数字预失真装置，其特征在于，包括：
    子带划分模块，用于根据功率放大器有效带宽内的幅频响应将所述有效带宽划分为若干个子带；
    预失真系数计算模块，用于依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号，根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
    预失真处理模块，用于分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
  9. 一种超宽带数字预失真***，其特征在于，包括：
    预失真处理单元，数据采集单元，子带划分单元和预失真系数计算单元；
    所述数据采集单元采集功率放大器各个子带的输入信号和各个子带的输出反馈信号；
    所述子带划分单元依次对各个子带进行频域重构，分别计算频域重构后功率放大器的输入信号对应在各个子带内的输入重构信号，并分别计算频域重构后功率放大器的输出反馈信号对应在各个子带内的输出反馈重构信号；
    所述预失真系数计算单元根据所述输入重构信号和输出反馈重构信号分别计算各个子带的预失真系数；
    所述预失真处理单元分别根据各个子带的预失真系数对对应子带的基带子信号进行预失真处理。
  10. 根据权利要求9所述的超宽带数字预失真***，其特征在于，所述预失真处理单元还对预失真处理后的各个基带子信号进行求和处理，得到预失真处理后的输入信号，并将所述预失真处理后的输入信号输出至所述功率放大器的信号输入端。
PCT/CN2018/097584 2017-11-03 2018-07-27 超宽带数字预失真方法、装置和*** WO2019085555A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711068293.7A CN107994923B (zh) 2017-11-03 2017-11-03 超宽带数字预失真方法、装置和***
CN201711068293.7 2017-11-03

Publications (1)

Publication Number Publication Date
WO2019085555A1 true WO2019085555A1 (zh) 2019-05-09

Family

ID=62030015

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/097584 WO2019085555A1 (zh) 2017-11-03 2018-07-27 超宽带数字预失真方法、装置和***

Country Status (2)

Country Link
CN (1) CN107994923B (zh)
WO (1) WO2019085555A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107994923B (zh) * 2017-11-03 2019-11-12 京信通信***（中国）有限公司 超宽带数字预失真方法、装置和***
CN109150240B (zh) * 2018-07-18 2020-07-14 中国电子科技集团公司第七研究所 一种用于无线跳频通信***的数字预失真在线训练方法
CN113162558B (zh) * 2021-03-15 2021-12-28 深圳市时代速信科技有限公司 一种数字预失真方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102014090A (zh) * 2010-12-13 2011-04-13 中兴通讯股份有限公司 数字预失真方法及装置
US20130094610A1 (en) * 2011-10-14 2013-04-18 Fadhel Ghannouchi Digital Multi-band Predistortion Linearizer with Nonlinear Subsampling Algorithm in the Feedback Loop
CN103491040A (zh) * 2013-09-30 2014-01-01 福州大学 一种数字基带自适应预失真***和方法
CN104639481A (zh) * 2015-03-05 2015-05-20 大唐移动通信设备有限公司 一种多频段信号处理方法及设备
CN105763495A (zh) * 2014-12-16 2016-07-13 中兴通讯股份有限公司 一种数字预失真的方法和装置
CN107994923A (zh) * 2017-11-03 2018-05-04 京信通信***（中国）有限公司 超宽带数字预失真方法、装置和***

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3805221B2 (ja) * 2001-09-18 2006-08-02 株式会社日立国際電気 歪み補償装置
JP5355508B2 (ja) * 2009-08-27 2013-11-27 株式会社エヌ・ティ・ティ・ドコモ べき級数型ディジタルプリディストータとその歪補償制御方法
CN101827054B (zh) * 2010-01-27 2012-10-17 国家广播电影电视总局无线电台管理局 非线性失真的预补偿方法与装置及发射机
EP3166223B1 (en) * 2015-10-13 2020-09-02 Analog Devices Global Unlimited Company Ultra wide band digital pre-distortion
CN107241070B (zh) * 2016-03-29 2020-07-10 大唐移动通信设备有限公司 一种确定dpd系数的方法、fpga及dpd处理***
CN106685868B (zh) * 2017-01-03 2020-09-08 电子科技大学 一种相邻多频带数字预失真***与方法
CN107124144B (zh) * 2017-04-20 2020-02-18 京信通信***（中国）有限公司 一种数字预失真处理方法和装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102014090A (zh) * 2010-12-13 2011-04-13 中兴通讯股份有限公司 数字预失真方法及装置
US20130094610A1 (en) * 2011-10-14 2013-04-18 Fadhel Ghannouchi Digital Multi-band Predistortion Linearizer with Nonlinear Subsampling Algorithm in the Feedback Loop
CN103491040A (zh) * 2013-09-30 2014-01-01 福州大学 一种数字基带自适应预失真***和方法
CN105763495A (zh) * 2014-12-16 2016-07-13 中兴通讯股份有限公司 一种数字预失真的方法和装置
CN104639481A (zh) * 2015-03-05 2015-05-20 大唐移动通信设备有限公司 一种多频段信号处理方法及设备
CN107994923A (zh) * 2017-11-03 2018-05-04 京信通信***（中国）有限公司 超宽带数字预失真方法、装置和***

Also Published As

Publication number Publication date
CN107994923B (zh) 2019-11-12
CN107994923A (zh) 2018-05-04

Similar Documents

Publication Publication Date Title
US20200366253A1 (en) Multi-band digital compensator for a non-linear system
WO2019085555A1 (zh) 超宽带数字预失真方法、装置和***
WO2019217811A1 (en) Digital compensator for a non-linear system
Guan et al. Simplified dynamic deviation reduction-based Volterra model for Doherty power amplifiers
US20160126903A1 (en) Weighted memory polynomial method and system for power amplifiers predistortion
GB2477986A (en) Method and apparatus for crest factor reduction
WO2016095528A1 (zh) 一种数字预失真的方法、装置和计算机存储介质
KR101664797B1 (ko) Ofdm신호의 papr를 감쇄하기 위한 방법 및 이를 이용한 ofdm 송신기
Enzinger et al. A survey of delay and gain correction methods for the indirect learning of digital predistorters
US10298366B2 (en) Circuit and method for predistortion
Tao et al. A random demodulation based reduced sampling rate method for wideband digital predistortion
CN113468814B (zh) 一种基于神经网络的数字预失真训练数据筛选方法及装置
JP6523287B2 (ja) デジタルプリディストーション装置のためのサンプル有効性の判別方法及び装置
Sezgin Different Digital Predistortion Techniques for Power Amplifier Linearization
TW201633752A (zh) 傳送器以及用來降低輸入訊號失真的方法
US10122560B2 (en) Tree search tone reservation for PAPR reduction in OFDM systems
Ghannouchi et al. A dual branch Hammerstein-Wiener architecture for behavior modeling of wideband RF transmitters
CN109842421B (zh) 频谱整形器装置和方法以及非暂时性计算机可读介质
Zhang et al. A new adaptive algorithm for digital predistortion using LS with singular value decomposition
Devi et al. Modeling broadband RF power amplifiers using a modified Hammerstein model
Jaber et al. Joint clipping and amplifying techniques for PAPR reduction in OFDM systems
WO2018129710A1 (zh) 一种信号传输方法、***及装置
Zeng et al. A crest factor reduction method in digital predistortion for improvement of power efficiency
Hoh et al. A nonlinearity predistortion technique for HPA with memory effects in OFDM systems
Masoumian et al. On joint compressed sensing based channel estimation and nonuniform PTS PAPR reduction without side information

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18872794

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18872794

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 16/09/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18872794

Country of ref document: EP

Kind code of ref document: A1