WO2016078038A1 - 一种预失真处理的装置及方法 - Google Patents

一种预失真处理的装置及方法 Download PDF

Info

Publication number
WO2016078038A1
WO2016078038A1 PCT/CN2014/091668 CN2014091668W WO2016078038A1 WO 2016078038 A1 WO2016078038 A1 WO 2016078038A1 CN 2014091668 W CN2014091668 W CN 2014091668W WO 2016078038 A1 WO2016078038 A1 WO 2016078038A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
module
predistortion
auxiliary
coefficient
Prior art date
Application number
PCT/CN2014/091668
Other languages
English (en)
French (fr)
Inventor
赵延青
余春蕾
田廷剑
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP14906288.7A priority Critical patent/EP3208938B1/en
Priority to CN201480082743.6A priority patent/CN107078702B/zh
Priority to PCT/CN2014/091668 priority patent/WO2016078038A1/zh
Publication of WO2016078038A1 publication Critical patent/WO2016078038A1/zh
Priority to US15/599,956 priority patent/US10075324B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/32—Carrier systems characterised by combinations of two or more of the types covered by groups H04L27/02, H04L27/10, H04L27/18 or H04L27/26
    • H04L27/34—Amplitude- and phase-modulated carrier systems, e.g. quadrature-amplitude modulated carrier systems
    • H04L27/36—Modulator circuits; Transmitter circuits
    • H04L27/366—Arrangements for compensating undesirable properties of the transmission path between the modulator and the demodulator
    • H04L27/367—Arrangements for compensating undesirable properties of the transmission path between the modulator and the demodulator using predistortion
    • H04L27/368—Arrangements for compensating undesirable properties of the transmission path between the modulator and the demodulator using predistortion adaptive predistortion
    • 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
    • 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
    • H03F1/3294—Acting on the real and imaginary components of the input signal
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
    • H03F3/189—High-frequency amplifiers, e.g. radio frequency amplifiers
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
    • H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
    • H03F3/24—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers of transmitter output stages
    • H03F3/245—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers of transmitter output stages with semiconductor devices only
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference
    • H04L25/03343—Arrangements at the transmitter end
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F2200/00—Indexing scheme relating to amplifiers
    • H03F2200/102—A non-specified detector of a signal envelope being used in an amplifying circuit
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F2201/00—Indexing scheme relating to details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements covered by H03F1/00
    • H03F2201/32—Indexing scheme relating to modifications of amplifiers to reduce non-linear distortion
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F2201/00—Indexing scheme relating to details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements covered by H03F1/00
    • H03F2201/32—Indexing scheme relating to modifications of amplifiers to reduce non-linear distortion
    • H03F2201/3233—Adaptive predistortion using lookup table, e.g. memory, RAM, ROM, LUT, to generate the predistortion

Definitions

  • the present invention relates to the field of linear modulation technologies, and in particular, to an apparatus and method for predistortion processing.
  • PA Power Amplify
  • RRU Radio Remote Unit
  • the nonlinearity and memory effect of the PA cause nonlinear distortion of the output signal of the power amplifier.
  • the distortion of the PA output signal and the out-of-band power leakage of the signal increase.
  • the receiver cannot correctly receive the signal and interferes with the communication system of the adjacent frequency band.
  • the ultra-wideband digital signal supporting multi-band and multi-standard is used.
  • Predistortion technology DPD (Digital Pre-Distortion) and RRU to reduce the nonlinear distortion of the PA output signal.
  • DPD Digital Pre-Distortion
  • RRU Radio Resource Unit
  • the DPD module is designed before the PA, and the digital IF signal generates nonlinear components and nonlinear components generated by the PA through the DPD module. Eliminate each other to achieve the effect of improving the linearity of the PA output.
  • the feedback signal and the predistortion circuit output signal are used as input signals of the predistortion coefficient learning circuit, respectively, performing predistortion model coefficient training for each frequency band, obtaining nonlinear model coefficients corresponding to N frequency bands, and training After convergence, the predistortion coefficient is copied to the predistortion circuit, and then the power amplifier is nonlinearly modeled by the predistortion circuit according to the predistortion coefficient.
  • the design of ultra-wideband power amplifiers is limited by device performance.
  • the low-frequency envelope impedance is difficult to optimize.
  • the lack of video bandwidth leads to a strong electrical memory effect of the power amplifier, and the harmonic frequency range will be wider.
  • Impedance is also difficult to optimize and can also cause an increase in electrical memory effects.
  • the traditional digital predistortion technology since the feedback channel only feeds back the fundamental signal, the fundamental signal and the harmonic signal cannot be fed back in real time, so that the predistortion processing circuit can only use the information of the fundamental signal to perform nonlinear modeling of the predistortion circuit.
  • the non-distortion circuit nonlinear modeling can not be performed in real time by using the information contained in the envelope signal and the harmonic signal output by the power amplifier, resulting in lower accuracy of the nonlinear modeling of the predistortion circuit and lowering the performance index of the predistortion circuit.
  • the invention provides a device and a method for predistortion processing, which can solve the problem that the accuracy of the predistortion processing circuit for predistortion modeling is not high under the broadband communication system in the prior art.
  • a first aspect of the present invention provides an apparatus for predistortion processing, including:
  • An auxiliary feedback module is configured to receive the 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 the auxiliary model coefficient Training module
  • the auxiliary model coefficient training module is configured to receive the feedback signal and a predistortion signal output from the predistortion processing module, and train the auxiliary model coefficient according to the feedback signal and the predistortion signal output from the predistortion processing module.
  • the auxiliary coefficient of the module is trained, the first auxiliary coefficient is obtained after convergence, and the first auxiliary coefficient obtained by the training is transmitted to the predistortion processing module;
  • the RF signal feedback module is configured to receive the amplified analog signal, and extract a fundamental wave signal from the amplified analog signal, obtain a fundamental wave feedback signal corresponding to the fundamental wave signal, and then feedback the fundamental wave Transmitting the signal into the predistortion model coefficient training module;
  • the predistortion model coefficient training module is configured to receive a fundamental wave feedback signal and a predistortion signal output from the predistortion processing module, and according to the fundamental wave feedback signal, a predistortion signal output from the predistortion processing module And training the predistortion coefficient of the predistortion model coefficient training module, and transmitting the trained predistortion coefficient to the predistortion processing module;
  • the predistortion processing module is configured to perform nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, perform predistortion processing on the input intermediate frequency signal, and output predistortion signal obtained by predistortion processing .
  • the non-linear distortion signal includes an envelope signal
  • the auxiliary feedback module is specifically configured to receive the amplified analog signal
  • the envelope signal is subjected to sampling processing to obtain an envelope feedback signal corresponding to the envelope signal
  • the envelope feedback signal is input to the auxiliary model coefficient training module.
  • the pre-distortion processing module includes a first auxiliary filtering sub-module and a first pre-distortion processing a submodule
  • the predistortion model coefficient training module includes a second auxiliary filtering submodule and a first predistortion coefficient processing submodule
  • the auxiliary model coefficient training module is specifically configured to use the auxiliary model coefficient training mode Modeling the predistortion signal and the envelope feedback signal of the block, training the auxiliary coefficients of the auxiliary model coefficient training module, obtaining a second auxiliary coefficient after convergence, and transmitting the second auxiliary coefficient to the first An auxiliary filtering sub-module and the second auxiliary filtering sub-module.
  • the first auxiliary filtering submodule is configured to be used according to the input of the predistortion processing module
  • the intermediate frequency signal constructs an envelope reference signal corresponding to the intermediate frequency signal, and performs filtering processing on the envelope reference signal by using the second auxiliary coefficient, and inputs the obtained first output signal into the first predistortion processing Submodule
  • the first pre-distortion processing sub-module is configured to generate, by using a preset pre-distortion model, a first output signal output by the first auxiliary filtering sub-module and a baseband signal input by the clipping process input to the pre-distortion processing module Predistortion vector, multiplying the predistortion vector and the predistortion coefficient to obtain the predistortion signal and outputting;
  • the second auxiliary filtering sub-module is configured to perform a filtering process on the fundamental feedback signal input from the RF feedback signal module, and use the second auxiliary coefficient to perform filtering processing, and input the obtained second output signal into the first pre-distortion coefficient Processing submodule;
  • the first pre-distortion coefficient processing sub-module is configured to perform nonlinear modeling according to the second output signal, a fundamental wave feedback signal input to the first pre-distortion coefficient processing sub-module, and the pre-distortion signal, and
  • the predistortion coefficient of the predistortion model coefficient training module is trained to obtain the predistortion coefficient.
  • the non-linear distortion signal includes an envelope signal and a harmonic signal
  • the auxiliary feedback module is specifically configured to receive the amplified analog signal. Extracting an envelope signal and a harmonic signal from the amplified analog signal, and performing sampling processing on the envelope signal and the harmonic signal to obtain an envelope feedback signal corresponding to the envelope signal, and The harmonic feedback signal corresponding to the harmonic signal is described, and the envelope feedback signal and the harmonic feedback signal are input to the auxiliary model coefficient training module.
  • the pre-distortion processing module includes a third auxiliary filtering sub-module and a second pre-distortion processing a submodule
  • the predistortion model coefficient training module includes a fourth auxiliary filtering submodule and a second predistortion coefficient processing submodule
  • the auxiliary model coefficient training module is specifically configured to use the auxiliary model coefficient training mode Modeling the predistortion signal, the envelope feedback signal and the harmonic feedback signal of the block, training the auxiliary coefficients of the auxiliary model coefficient training module, converging to obtain a third auxiliary coefficient, and transmitting the third auxiliary coefficient And the third auxiliary filtering sub-module and the fourth auxiliary filtering sub-module.
  • the third auxiliary filtering submodule is configured to be used according to the input of the predistortion processing module
  • the intermediate frequency signal constructs an envelope reference signal and a harmonic reference signal corresponding to the intermediate frequency signal, and performs filtering processing on the envelope reference signal and the harmonic reference signal by using the third auxiliary coefficient, and the obtained
  • the third output signal is input to the second predistortion processing submodule;
  • the second pre-distortion processing sub-module is configured to generate, by using a preset pre-distortion model, a third output signal output by the first auxiliary filtering sub-module and a baseband signal input by the clipping processing input to the pre-distortion processing module Predistortion vector, multiplying the predistortion vector and the predistortion coefficient to obtain the predistortion signal and outputting;
  • the fourth auxiliary filtering sub-module is configured to perform a filtering process on the fundamental feedback signal input from the RF feedback signal module, and use the third auxiliary coefficient to perform filtering processing, and input the obtained fourth output signal into the second pre-distortion coefficient Processing submodule;
  • the second pre-distortion coefficient processing sub-module is configured to perform nonlinear modeling according to the fourth output signal, a fundamental wave feedback signal input to the second pre-distortion coefficient processing sub-module, and the pre-distortion signal, and
  • the predistortion coefficient of the predistortion model coefficient training module is trained to obtain the predistortion coefficient.
  • the non-linear distortion signal includes a harmonic signal
  • the auxiliary feedback module is specifically configured to extract harmonics from the amplified analog signal. And sampling the harmonic signal to obtain a harmonic feedback signal corresponding to the harmonic signal, and inputting the harmonic feedback signal into the auxiliary model coefficient training module.
  • the pre-distortion processing module includes a fifth auxiliary filtering sub-module and a third pre-distortion processing a submodule
  • the predistortion model coefficient training module includes a sixth auxiliary filtering submodule and a third predistortion coefficient processing submodule
  • the auxiliary model coefficient training module is specifically configured to perform modeling by using a predistortion signal and a harmonic feedback signal input to the auxiliary model coefficient training module, and training the auxiliary coefficient of the auxiliary model coefficient training module, and obtaining the first after convergence Four auxiliary coefficients, and pass the fourth auxiliary coefficient to the The fifth auxiliary filtering sub-module and the sixth auxiliary filtering sub-module.
  • the fifth auxiliary filtering submodule is configured to input an intermediate frequency signal according to the predistortion processing module. Constructing 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 ;
  • the third pre-distortion processing sub-module is configured to generate, by using a preset pre-distortion model, a fifth output signal output by the fifth auxiliary filtering sub-module and a baseband signal input by the clipping processing input to the pre-distortion processing module Predistortion vector, multiplying the predistortion vector and the predistortion coefficient to obtain the predistortion signal and outputting;
  • the sixth auxiliary filtering sub-module is configured to perform filtering processing on the fundamental feedback signal input from the radio frequency feedback signal module, and use the fourth auxiliary coefficient to perform filtering processing, and input the obtained sixth output signal into the pre-distortion model processing sub-module;
  • the third pre-distortion coefficient processing sub-module is configured to perform nonlinear modeling according to the sixth output signal, a fundamental wave feedback signal input to the third pre-distortion coefficient processing sub-module, and the pre-distortion signal, and The predistortion coefficient of the predistortion model coefficient training module is trained to obtain the predistortion coefficient.
  • the apparatus further includes:
  • a pre-processing module configured to digitally upconvert a baseband signal input to the pre-processing module, combine all carriers in the same frequency band, perform clipping processing, and input the clipped baseband signal into the pre-distortion processing a module, the baseband signal comprising a carrier of multiple frequency bands;
  • a signal combining and transmitting module configured to receive a predistortion signal outputted from the predistortion processing module, and convert the predistortion signal outputted from the predistortion processing module into an analog signal, and output the signal to the power amplifying module;
  • the power amplification module is configured to amplify an analog signal input to the power amplification module, and output the amplified analog signal to the auxiliary feedback module and the radio frequency signal feedback module.
  • a second aspect of the present invention provides a method for predistortion processing, the method comprising:
  • the auxiliary feedback module receives the amplified analog signal and extracts the non-line from the amplified analog signal a distortion signal, obtaining a feedback signal corresponding to the nonlinear distortion signal, and inputting the feedback signal to the auxiliary model coefficient training module;
  • the auxiliary model coefficient training module receives the feedback signal and the predistortion signal outputted from the predistortion processing module, and assists the auxiliary model coefficient training module according to the feedback signal and the predistortion signal input from the predistortion processing module.
  • the coefficient is trained, and the first auxiliary coefficient obtained by the training is transmitted to the predistortion processing module;
  • the RF signal feedback module receives the amplified analog signal, and extracts a fundamental wave signal from the amplified analog signal, and obtains a fundamental wave feedback signal corresponding to the fundamental wave signal, and then inputs the fundamental wave feedback signal into the Pre-distortion model coefficient training module;
  • the predistortion model coefficient training module receives a fundamental wave feedback signal and a predistortion signal output from the predistortion processing module, and according to the fundamental wave feedback signal and a predistortion signal output from the predistortion processing module, Pre-distortion model coefficient training module pre-distortion coefficient is trained, and the trained pre-distortion coefficient is transmitted to the pre-distortion processing module;
  • the predistortion processing module performs nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, performs predistortion processing on the input intermediate frequency signal, and outputs a predistortion signal obtained by the predistortion processing.
  • the non-linear distortion signal includes an envelope signal
  • the auxiliary feedback module receives the amplified analog signal from the analog signal. Extracting a nonlinear distortion signal, and performing sampling processing on the nonlinear distortion signal to obtain a feedback signal corresponding to the nonlinear distortion signal, and inputting the feedback signal into the auxiliary model coefficient training module specifically includes:
  • the auxiliary feedback module receives the amplified analog signal, extracts the envelope signal from the analog signal, and performs sampling processing on the envelope signal to obtain a feedback signal corresponding to the envelope signal, and
  • the envelope feedback signal is input to an auxiliary model coefficient training module.
  • the pre-distortion processing module includes a first auxiliary filtering sub-module and a first pre-distortion processing sub-module
  • the pre-distortion model coefficient training module includes a second auxiliary filtering sub-module and a first pre-distortion coefficient processing sub-module, the method further comprising:
  • the auxiliary model coefficient training module performs modeling by inputting a predistortion signal and an envelope feedback signal of the auxiliary model coefficient training module, and training the auxiliary coefficient to obtain a second after convergence And a second auxiliary coefficient is transmitted to the first auxiliary filtering sub-module and the second auxiliary filtering sub-module.
  • the method further includes:
  • the first auxiliary filtering sub-module constructs an envelope reference signal corresponding to the intermediate frequency signal according to an intermediate frequency signal input to the pre-distortion processing module, performs filtering processing on the envelope reference signal, and uses the second
  • the first output signal obtained by the auxiliary coefficient is input to the predistortion processing submodule in the predistortion processing module;
  • the predistortion processing sub-module generates a predistortion vector by using a first output signal output by the first auxiliary filtering submodule and a baseband signal input by the clipping process of the predistortion processing submodule according to a preset predistortion model. And multiplying the predistortion vector and the predistortion coefficient to obtain the predistortion signal and outputting;
  • the second auxiliary filtering sub-module performs filtering processing on the fundamental feedback signal input from the radio frequency feedback signal module by using the second auxiliary coefficient, and inputs the obtained second output signal into the pre-distortion model processing sub-module;
  • the first pre-distortion coefficient processing sub-module performs nonlinear modeling according to the second output signal, a fundamental wave feedback signal input to the first pre-distortion coefficient processing sub-module, and the pre-distortion signal, and performs the The predistortion coefficient of the predistortion model coefficient training module is trained to obtain the predistortion coefficient.
  • the non-linear distortion signal includes an envelope signal and a harmonic signal
  • the method further includes:
  • the auxiliary feedback module extracts an envelope signal and a harmonic signal from the analog signal, and performs sampling processing on the envelope signal and the harmonic signal to obtain an envelope feedback signal corresponding to the envelope signal, and And a harmonic feedback signal corresponding to the harmonic signal, and inputting the envelope feedback signal and the harmonic feedback signal to the auxiliary model coefficient training module.
  • the pre-distortion processing module includes a third auxiliary filtering sub-module and a second pre-distortion processing sub-module
  • the pre-distortion model coefficient training module includes a fourth auxiliary filter sub-module and a second pre-distortion coefficient processing sub-module, the method further comprising:
  • the auxiliary model coefficient training module trains the auxiliary coefficient by using a predistortion signal, an envelope feedback signal, and a harmonic feedback signal input to the auxiliary model coefficient training module, and is condensed to obtain a third auxiliary coefficient, and the third auxiliary coefficient is transmitted to the third auxiliary filtering sub-module and the fourth auxiliary filtering sub-module.
  • the method further includes:
  • the third auxiliary filtering sub-module constructs an envelope reference signal and a harmonic reference signal corresponding to the intermediate frequency signal according to an intermediate frequency signal input to the predistortion processing module, and the envelope reference signal and the harmonic reference The signal is subjected to filtering processing, and the third output signal obtained by using the third auxiliary coefficient is input to the second predistortion processing submodule;
  • the second pre-distortion processing sub-module utilizes a third output signal output by the third auxiliary filtering sub-module and a clipping-processed baseband signal input to the second pre-distortion processing sub-module according to a preset pre-distortion model Generating a predistortion vector, multiplying the predistortion vector and the predistortion coefficient to obtain the predistortion signal and outputting the predistortion signal;
  • the fourth auxiliary filtering sub-module performs filtering processing on the fundamental feedback signal input from the radio frequency feedback signal module, and uses the third auxiliary coefficient to perform filtering processing, and inputs the obtained fourth output signal into the second pre-distortion coefficient processing unit.
  • the second pre-distortion coefficient processing sub-module performs nonlinear modeling according to the fourth output signal, a fundamental wave feedback signal input to the second pre-distortion coefficient processing sub-module, and the pre-distortion signal, and performs the The predistortion coefficient of the predistortion model coefficient training module is trained to obtain the predistortion coefficient.
  • the non-linear distortion signal includes a harmonic signal
  • the method further includes:
  • the auxiliary feedback module extracts a harmonic signal from the analog signal, performs sampling processing on the harmonic signal, obtains a harmonic feedback signal corresponding to the harmonic signal, and inputs the harmonic feedback signal into the Auxiliary model coefficient training module.
  • the auxiliary model coefficient training module trains the auxiliary coefficient by using a predistortion signal and a harmonic feedback signal input to the auxiliary model coefficient training module, and obtains a fourth auxiliary coefficient after convergence, and transmits the fourth auxiliary coefficient Giving the fifth auxiliary filtering sub-module and the sixth auxiliary filtering Submodule.
  • the method further includes:
  • the first auxiliary filtering sub-module constructs a harmonic reference signal corresponding to the intermediate frequency signal according to an intermediate frequency signal input to the pre-distortion processing module, performs filtering processing on the harmonic reference signal, and uses the fourth
  • the fifth output signal obtained by the auxiliary coefficient is input to the third predistortion processing submodule;
  • the third pre-distortion processing sub-module utilizes a first output signal output by the first auxiliary filtering sub-module and a clipping-processed baseband signal input to the third pre-distortion processing sub-module according to a preset pre-distortion model Generating a predistortion vector, multiplying the predistortion vector and the predistortion coefficient to obtain the predistortion signal and outputting the predistortion signal;
  • the sixth auxiliary filtering sub-module performs filtering processing on the fundamental feedback signal input from the radio frequency feedback signal module by using the fourth auxiliary coefficient, and inputs the obtained sixth output signal into the pre-distortion model processing sub-module;
  • the third pre-distortion coefficient processing sub-module performs nonlinear modeling according to the sixth output signal, a fundamental wave feedback signal input to the third pre-distortion coefficient processing sub-module, and the pre-distortion signal, and performs the The predistortion coefficient of the predistortion model coefficient training module is trained to obtain the predistortion coefficient.
  • the method further includes:
  • the pre-processing module digitally upconverts the baseband signal input to the pre-processing module, combines all carriers in the same frequency band, performs clipping processing, and inputs the clipped baseband signal into the pre-distortion processing module,
  • the baseband signal includes carriers of multiple frequency bands;
  • the signal combining and transmitting module will receive the predistortion signal outputted from the predistortion processing module, and convert the predistortion signal outputted from the predistortion processing module into an analog signal, and output the signal to the power amplifying module;
  • the power amplification module amplifies an analog signal input to the power amplification module, and outputs the amplified analog signal to the auxiliary feedback module and the radio frequency signal feedback module.
  • the nonlinear distortion signal is extracted from the analog signal by adding an auxiliary feedback module, and a feedback signal corresponding to the nonlinear distortion signal is obtained.
  • the nonlinear distortion signal is fed back, and the auxiliary model coefficient training module is used to train the auxiliary coefficient by using the feedback signal and the predistortion signal output by the auxiliary feedback module, and the obtained first auxiliary coefficient is transmitted to the predistortion processing module and the predistortion model.
  • the coefficient training module enables the predistortion processing module to perform nonlinear modeling by using the first auxiliary coefficient and the predistortion coefficient trained by the predistortion model coefficient training module, thereby effectively improving the accuracy of the nonlinear modeling and improving the digital preprocessing under the broadband communication system.
  • the performance of distortion technology is a technique that uses the first auxiliary coefficient to perform nonlinear modeling by using the first auxiliary coefficient and the predistortion coefficient trained by the predistortion model coefficient training module, thereby effectively improving the accuracy of the nonlinear modeling and improving the digital preprocessing under the broadband communication system.
  • FIG. 1 is a schematic structural diagram of a device for predistortion processing according to an embodiment of the present invention
  • FIG. 2 is another schematic structural diagram of an apparatus for predistortion processing according to an embodiment of the present invention.
  • FIG. 3 is another schematic structural diagram of an apparatus for predistortion processing according to an embodiment of the present invention.
  • FIG. 4 is another schematic structural diagram of an apparatus for predistortion processing according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of an internal operation process of a device for predistortion processing according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of an embodiment of a method for predistortion processing according to an embodiment of the present invention.
  • the terms “comprises” and “comprises” and “the” and “the” are intended to cover a non-exclusive inclusion, for example, a process, method, system, product, or device that comprises a series of steps or modules is not necessarily limited to Those steps or modules, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products or devices, the division of the modules presented herein is merely a logical division. In actual applications, there may be another way of dividing, for example, multiple modules can be combined or integrated into another In a system, or some features may be omitted or not performed.
  • the coupling or direct coupling or communication connection between the displayed or discussed may be through some interfaces, and the indirect coupling or communication connection between the modules may be Electrical or other similar forms are not limited herein.
  • the module or the sub-module described as the separate component may or may not be physically separated, may not be a physical module, or may not be divided into a plurality of circuit modules, and may select a part thereof according to actual needs or All modules are used to achieve the objectives of the embodiments of the present invention.
  • a device and a method for predistortion processing can solve the nonlinear distortion in the broadband communication system in the prior art, because the feedback circuit only feeds back the fundamental wave signal, and cannot feedback the nonlinear distortion such as the envelope signal and the harmonic signal in real time.
  • the signal causes the predistortion processing circuit to perform predistortion modeling only according to the fundamental feedback signal, and can not perform predistortion modeling according to the predistortion processing circuit such as envelope signal and harmonic signal in real time, resulting in the accuracy of nonlinear modeling. High, so that the problem of nonlinear characteristics of the power amplifier cannot be accurately analyzed.
  • the device and method can be used in the field of linear modulation, and is mainly applied to the RRU of a mobile communication system, and the main application scenario thereof is a broadband multi-band, multi-standard mobile communication system.
  • the DPD module is designed before the PA, and the nonlinear component generated by the digital intermediate frequency signal through the DPD module cancels out the nonlinear component generated by the PA, thereby achieving the effect of improving the linearity of the PA output, and the PA design is different, and the envelope impedance is different.
  • the harmonic impedance also has different effects on the linearity of the power amplifier and the DPD.
  • the PA since the PA is a nonlinear device, the PA output signal is a full spectrum in frequency.
  • the signal on the frequency band f is the fundamental signal, and the frequency band is 2f.
  • the signals of integer multiples such as 3f and 4f are harmonic signals, and the signals near the 0 frequency are envelope signals, wherein the envelope signal is the response of the fundamental signal envelope at low frequencies.
  • the fundamental signal is the signal that the circuit will transmit.
  • the predistortion signal appearing in the present invention includes an envelope signal and a harmonic signal, and the predistortion signal may be at least one of an envelope signal or a harmonic signal, or other similarities may cause nonlinear distortion or
  • the signal of the memory effect is acceptable, and the components of the signal in the predistortion signal fed back by the auxiliary feedback module are not limited herein;
  • the auxiliary feedback module feeds back at least one of the envelope signal or the harmonic signal
  • the corresponding auxiliary model coefficient training module, the auxiliary filtering sub-module and the pre-distortion model coefficient training module are correspondingly changed, for example, auxiliary
  • the auxiliary model coefficient training The training module, the auxiliary filtering sub-module, and the pre-distortion model coefficient training module respectively can only process the components related to the envelope feedback signal, which are not limited herein.
  • the auxiliary feedback module and the RF signal feedback module may be a zero intermediate frequency receiver architecture, or may be a digital intermediate frequency receiver architecture or a direct RF receiver architecture or other similar receiver architecture, as long as the implementation described herein is achieved.
  • the function can be, and the specific structure is not limited herein.
  • the apparatus for pre-distortion processing includes:
  • the auxiliary feedback module 101 is configured to receive the amplified analog signal, extract a nonlinear distortion signal from the amplified analog signal, acquire a feedback signal corresponding to the nonlinear distortion signal, and input the feedback signal to the auxiliary model coefficient training module 102. ;
  • the auxiliary model coefficient training module 102 is configured to receive the feedback signal and the predistortion signal output from the predistortion processing module, and perform the auxiliary coefficient of the auxiliary model coefficient training module 102 according to the feedback signal and the predistortion signal output from the predistortion processing module 105. Training, after convergence, the first auxiliary coefficient is obtained, and the first auxiliary coefficient obtained by the training is transmitted to the predistortion processing module 105;
  • the RF signal feedback module 103 is configured to receive the amplified analog signal, and 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 predistortion model coefficient training module 104;
  • the predistortion model coefficient training module 104 is configured to receive the fundamental wave feedback signal and the predistortion signal output from the predistortion processing module, and output the predistortion signal according to the predistortion signal output from the predistortion processing module 105, and output from the radio frequency signal feedback module 103.
  • the fundamental wave feedback signal, the predistortion coefficient of the predistortion model coefficient training module 103 is trained, and the predistortion coefficient obtained by the training is transmitted to the predistortion processing module 105;
  • the predistortion processing module 105 is configured to receive a fundamental wave feedback signal and a predistortion signal output from the predistortion processing module, perform nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, and perform predistortion processing on the input intermediate frequency signal. And outputting the predistortion signal obtained by the predistortion processing.
  • the auxiliary feedback module 101 extracts a nonlinear distortion signal from the analog signal, obtains a feedback signal corresponding to the nonlinear distortion signal, and implements a real-time feedback nonlinear distortion signal, and the auxiliary model coefficient training module 102 utilizes the auxiliary feedback module.
  • the feedback signal outputted by 101 and the predistortion signal output by the predistortion processing module 105 train the auxiliary coefficients, and the obtained first auxiliary coefficient
  • the pre-distortion processing module 105 and the pre-distortion model coefficient training module 104 are used, and the pre-distortion processing module 105 uses the first auxiliary coefficient and the pre-distortion coefficient trained by the pre-distortion model coefficient training module 104 to perform nonlinear modeling, thereby effectively improving nonlinearity.
  • the accuracy of the modeling, the predistortion processing of the intermediate frequency signal, the predistortion signal is obtained, so that the nonlinear quantity generated by the intermediate frequency signal input to the predistortion processing device after the predistortion processing is offset by the nonlinear quantity generated by the power amplifier Improve the performance of digital pre-distortion technology in the case of broadband.
  • the embodiment of the present invention is an example of a pre-distortion processing device in an embodiment of the present invention.
  • the auxiliary feedback module 201 is configured to receive the amplified analog signal, extract a nonlinear distortion signal from the amplified analog signal, extract a nonlinear distortion signal from the amplified analog signal, and obtain feedback corresponding to the nonlinear distortion signal. Signal and input the feedback signal to the auxiliary model coefficient training module 202;
  • the auxiliary model coefficient training module 202 is configured to train the auxiliary coefficient of the auxiliary model coefficient training module 202 according to the feedback signal and the predistortion signal output from the predistortion processing module 204, and obtain the first auxiliary coefficient after convergence, and the training result is obtained.
  • the first auxiliary coefficient is passed to the predistortion processing module 204;
  • the RF signal feedback module 203 is configured to receive the amplified analog signal, and 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 the predistortion model.
  • Coefficient training module ;
  • the predistortion model coefficient training module 204 is configured to receive the fundamental wave feedback signal and the predistortion signal output from the predistortion processing module, according to the predistortion signal output from the predistortion processing module 205, and the fundamental wave output from the radio frequency signal feedback module 203.
  • the feedback signal, the predistortion coefficient of the predistortion model coefficient training module 203 is trained, and the predistortion coefficient obtained by the training is transmitted to the predistortion processing module 205;
  • the predistortion processing module 205 is configured to receive the intermediate frequency signal, perform nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, perform predistortion processing on the input intermediate frequency signal, and output the predistortion signal obtained by the predistortion processing.
  • the nonlinear distortion signal may include an envelope signal
  • the predistortion processing module 205 includes a first auxiliary filtering submodule 2051 and a first predistortion processing submodule 2052
  • the distortion model coefficient training module 204 includes a second auxiliary filter sub-module 2041 and a first pre-distortion coefficient processing sub-module 2042;
  • the auxiliary feedback module 201 is specifically configured to receive the amplified analog signal, extract an envelope signal from the amplified analog signal, sample the envelope signal, obtain an envelope feedback signal corresponding to the envelope signal, and provide an envelope.
  • the auxiliary model coefficient training module 202 is specifically configured to perform modeling by using the predistortion signal and the envelope feedback signal of the input auxiliary model coefficient training module 202, and training the auxiliary coefficient of the auxiliary model coefficient training module 202, and obtaining the second auxiliary coefficient after convergence And transmitting the second auxiliary coefficient to the first auxiliary filtering sub-module 2051 and the second auxiliary filtering sub-module 2041.
  • the first auxiliary filtering sub-module 2041 is configured to construct an envelope reference signal corresponding to the intermediate frequency signal according to the intermediate frequency signal of the input predistortion processing module 204, and perform filtering processing on the envelope reference signal, and use the second The first output signal obtained by the auxiliary coefficient is input to the first predistortion processing submodule 2042;
  • the first pre-distortion processing sub-module 2042 is configured to generate pre-distortion by using the first output signal output by the first auxiliary filtering sub-module 2041 and the clipping processed baseband signal of the input pre-distortion processing module 204 according to the preset pre-distortion model.
  • Vector multiplying the predistortion vector and the predistortion coefficient to obtain a predistortion signal and outputting;
  • the second auxiliary filtering sub-module 2041 is configured to perform filtering processing on the fundamental feedback signal input from the RF feedback signal module 203 by using the second auxiliary coefficient, and input the obtained second output signal into the first pre-distortion coefficient.
  • the first pre-distortion coefficient processing sub-module 2042 is configured to perform nonlinear modeling according to the second output signal, the fundamental wave feedback signal and the pre-distortion signal input to the first pre-distortion coefficient processing sub-module 2042, and perform a pre-distortion model coefficient training module.
  • the pre-distortion coefficient of 204 is trained to obtain a pre-distortion coefficient.
  • the auxiliary feedback module 201 extracts an envelope signal from the analog signal, samples the envelope signal, obtains an envelope feedback signal, and implements a real-time feedback envelope signal, and the auxiliary model coefficient training module 202 utilizes the envelope.
  • the feedback signal and the predistortion signal train the auxiliary coefficient, and the obtained second auxiliary coefficient is transmitted to the predistortion processing module 205 and the predistortion model coefficient training module 204, and the predistortion processing module utilizes the second auxiliary coefficient and the predistortion model coefficient.
  • the pre-distortion coefficient trained by the training module 204 is nonlinearly modeled, the accuracy of the nonlinear modeling is effectively improved, and the intermediate frequency signal is pre-distorted to obtain a pre-distortion signal, so that the intermediate frequency signal input to the pre-distortion processing device is pre-distorted.
  • the generated nonlinear quantity cancels out the nonlinear quantity generated by the power amplifier, linearly amplifies, and improves the performance of the digital pre-distortion technology in the case of wideband.
  • the embodiment of the present invention is an example of a device for feeding back an envelope signal and a harmonic signal.
  • the auxiliary feedback module 301 is configured to receive the amplified analog signal, extract a nonlinear distortion signal from the amplified analog signal, sample the nonlinear distortion signal, and obtain a feedback signal corresponding to the nonlinear distortion signal, and The feedback signal is input to the auxiliary model coefficient training module 302;
  • the auxiliary model coefficient training module 302 is configured to receive the feedback signal and the predistortion signal output from the predistortion processing module 305, and the auxiliary coefficient of the auxiliary model coefficient training module 302 according to the feedback signal and the predistortion signal output from the predistortion processing module 304. Performing training, after convergence, obtaining a first auxiliary coefficient, and transmitting the first auxiliary coefficient obtained by training to the predistortion processing module 305;
  • the RF signal feedback module 303 is configured to receive the amplified analog signal, and extract a fundamental wave signal from the amplified analog signal, obtain a fundamental wave feedback signal corresponding to the fundamental wave signal, and then input the fundamental wave feedback signal into the predistortion model.
  • the predistortion model coefficient training module 304 is configured to receive the fundamental wave feedback signal and the predistortion signal output by the predistortion processing module 305, according to the predistortion signal output from the predistortion processing module 305, and the fundamental wave feedback output from the RF signal feedback module.
  • the signal, the predistortion coefficient of the predistortion model coefficient training module 303 is trained, the predistortion coefficient is obtained after convergence, and the predistortion coefficient obtained by the training is transmitted to the predistortion processing module 304;
  • the predistortion processing module 304 is configured to perform nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, perform predistortion processing on the input intermediate frequency signal, and output the predistortion signal obtained by the predistortion processing.
  • the nonlinear distortion signal may include an envelope signal and a harmonic signal.
  • the auxiliary feedback module 301 is specifically configured to receive the amplified analog signal, extract an envelope signal and a harmonic signal from the amplified analog signal, and sample and process the envelope signal and the harmonic signal to obtain a packet corresponding to the envelope signal.
  • the feedback signal is obtained, the harmonic feedback signal corresponding to the harmonic signal is obtained, and the envelope signal and the harmonic feedback signal are input to the auxiliary model coefficient training module 302;
  • the pre-distortion processing module 305 includes a third auxiliary filtering sub-module 3051 and a second pre-distortion processing sub-module 3052.
  • the pre-distortion model coefficient training module 304 includes a fourth auxiliary filtering sub-module 3041 and a second pre-distortion coefficient. Processing sub-module 3042,
  • the auxiliary model coefficient training module 302 is specifically configured to perform modeling by using the predistortion signal, the envelope feedback signal, and the harmonic feedback signal of the input auxiliary model coefficient training module 302, and perform auxiliary coefficients of the auxiliary model coefficient training module 302. Training, after convergence, the third auxiliary coefficient is obtained, and the third auxiliary coefficient is transmitted to the third auxiliary filtering sub-module 3051 and the fourth auxiliary filtering sub-module 3041;
  • the third auxiliary filtering sub-module 3051 is configured to construct an envelope reference signal and a harmonic reference signal corresponding to the intermediate frequency signal according to the intermediate frequency signal of the input predistortion processing module 305, and use the third auxiliary coefficient to the envelope reference signal. And the harmonic reference signal is filtered, and the obtained third output signal is input to the predistortion processing sub-module 3052;
  • the second pre-distortion processing sub-module 3052 is configured to generate a pre-distortion vector by using the third output signal output by the first auxiliary filtering sub-module 3051 and the clipped baseband signal input to the pre-distortion processing module 305 according to the preset pre-distortion model.
  • the predistortion vector and the predistortion coefficient are multiplied to obtain a predistortion signal and output.
  • the fourth auxiliary filtering sub-module 3041 is configured to perform filtering processing on the fundamental feedback signal input from the RF feedback signal module 307 by using the third auxiliary coefficient, and input the obtained fourth output signal into the pre-distortion model processing.
  • the second pre-distortion coefficient processing sub-module 3042 is configured to perform nonlinear modeling according to the fourth output signal, the fundamental wave feedback signal and the pre-distortion signal input to the second pre-distortion coefficient processing sub-module 3042, and train the pre-distortion model coefficients.
  • the pre-distortion coefficients of module 304 are trained to obtain pre-distortion coefficients.
  • first auxiliary filtering sub-module 3051 and the second auxiliary filtering sub-module 3041 have the same structure, and are all digital filters.
  • digital filters There are two main types of digital filters, one is an infinite-length unit impulse.
  • the response digital filter IIR, Infinite Impulse Response
  • the other is a finite-length unit impulse response digital filter (FIR, Finite Impulse Response)
  • the specific types of digital filters in this article are not limited, as long as The function of adaptive filtering can be achieved.
  • the auxiliary signal module and the harmonic signal are extracted from the analog signal by the auxiliary feedback module 301, and the envelope signal and the harmonic signal are sampled to obtain an envelope feedback signal and a harmonic feedback signal, thereby realizing real-time feedback.
  • the envelope signal and the harmonic signal, the auxiliary model coefficient training module 302 trains the auxiliary coefficient by using the envelope feedback signal, the harmonic feedback signal and the predistortion signal, and transmits the obtained third auxiliary coefficient to the predistortion processing module 305 and
  • the predistortion model coefficient training module 304 the predistortion processing module uses the third auxiliary coefficient and the predistortion coefficient trained by the predistortion model coefficient training module 304 to perform nonlinear modeling, thereby effectively improving the accuracy of the nonlinear modeling, and the intermediate frequency signal.
  • Pre-pre Distortion processing obtaining a predistortion signal, so that the OFDM signal input to the predistortion processing device undergoes predistortion processing, and the generated nonlinear quantity cancels out the nonlinear quantity generated by the power amplifier, thereby improving the performance of the digital predistortion technology under the broadband condition .
  • the embodiment of the present invention is an example of a device for pre-distortion processing in the embodiment of the present invention.
  • the auxiliary feedback module 401 is configured to receive the amplified analog signal, extract the nonlinear distortion signal from the amplified analog signal, obtain a feedback signal corresponding to the nonlinear distortion signal, and input the feedback signal into the auxiliary model coefficient training module 402;
  • the auxiliary model coefficient training module 402 is configured to receive the feedback signal and the predistortion signal output from the predistortion processing module 405, and the auxiliary coefficient of the auxiliary model coefficient training module 402 according to the feedback signal and the predistortion signal output from the predistortion processing module 405. Performing training, and transmitting the first auxiliary coefficient obtained by the training to the predistortion processing module 405;
  • the RF signal feedback module 403 is configured to receive the amplified analog signal, and extract a fundamental wave signal from the amplified analog signal, obtain a fundamental wave feedback signal corresponding to the fundamental wave signal, and then input the fundamental wave feedback signal into the predistortion model.
  • the predistortion model coefficient training module 404 is configured to receive the fundamental wave feedback signal and the predistortion signal output from the predistortion processing module, according to the predistortion signal output from the predistortion processing module 405, and the fundamental wave feedback output from the RF signal feedback module.
  • the signal, the predistortion coefficient of the predistortion model coefficient training module 404 is trained, and the predistortion coefficient obtained by the training is transmitted to the predistortion processing module 405;
  • the predistortion processing module 405 is configured to perform nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, perform predistortion processing on the input intermediate frequency signal, and output the predistortion signal obtained by the predistortion processing.
  • the nonlinear distortion signal may include a harmonic signal.
  • the auxiliary feedback module 401 is specifically configured to extract a harmonic signal from the amplified analog signal, sample the amplified harmonic signal, obtain a harmonic feedback signal corresponding to the harmonic signal, and input the harmonic feedback signal into the auxiliary signal.
  • the pre-distortion processing module 405 includes a fifth auxiliary filtering sub-module 4051 and a third pre-distortion processing sub-module 4052, and the pre-distortion model coefficient training module 404 includes a sixth auxiliary filter.
  • the auxiliary model coefficient training module 402 is specifically configured to perform modeling by using the predistortion signal and the harmonic feedback signal of the input auxiliary model coefficient training module 402, and training the auxiliary coefficient of the auxiliary model coefficient training module 402 to obtain the fourth auxiliary coefficient after convergence. And transmitting the fourth auxiliary coefficient to the fifth auxiliary filtering sub-module 4051 and the sixth auxiliary filtering sub-module 4041;
  • the fifth auxiliary filtering sub-module 4051 is further configured to construct a harmonic reference signal corresponding to the intermediate frequency signal according to the intermediate frequency signal of the input predistortion processing module 404, and filter the harmonic reference signal by using the fourth auxiliary coefficient, and obtain the obtained
  • the fifth output signal is input to the predistortion processing sub-module 4052;
  • the third pre-distortion processing sub-module 4052 is further configured to generate a pre-distortion by using the fifth output signal output by the fifth auxiliary filtering sub-module 4051 and the clipping-processed baseband signal of the input pre-distortion processing module 405 according to the preset pre-distortion model.
  • the vector multiplies the predistortion vector and the predistortion coefficient to obtain a predistortion signal and outputs it.
  • the sixth auxiliary filtering sub-module 4041 is configured to perform a filtering process on the fundamental feedback signal input from the RF feedback signal module 403, and input the obtained sixth output signal into the pre-distortion model processing sub-module 4032;
  • the third pre-distortion coefficient processing sub-module 4042 is configured to perform nonlinear modeling according to the sixth output signal, the fundamental feedback signal of the third pre-distortion coefficient processing sub-module 4042, and the pre-distortion signal, and train the pre-distortion model coefficients.
  • the predistortion coefficients of module 404 are trained to obtain predistortion coefficients.
  • the device further includes:
  • the pre-processing module 406 is configured to digitally upconvert the input baseband signal, combine all the carriers in the same frequency band, perform clipping processing, and input the clipped baseband signal into the predistortion processing module 405, where the baseband signal includes Carriers in multiple frequency bands.
  • the signal combining and transmitting module 407 is configured to receive the predistortion signal outputted from the predistortion processing module, and convert the predistortion signal outputted from the predistortion processing module into an analog signal, and output the signal to the power amplifying module 408;
  • the power amplification module 408 is configured to amplify the analog signal of the input power amplification module 406 and output the signal to the auxiliary feedback module 401 and the RF signal feedback module 407.
  • the auxiliary feedback module 401 extracts the harmonic signal from the analog signal, uses the harmonic signal to obtain the harmonic feedback signal, realizes the real-time feedback harmonic signal, and the auxiliary model coefficient training module 402 utilizes the envelope feedback signal and the pre- The distortion signal trains the auxiliary coefficients and will get the fourth
  • the auxiliary coefficient is transmitted to the fifth auxiliary filtering sub-module 4051 and the sixth auxiliary filtering sub-module 4041, and the pre-distortion processing module 405 performs nonlinear construction using the fourth auxiliary coefficient and the pre-distortion coefficient trained by the third pre-distortion coefficient processing sub-module 4042.
  • the module effectively improves the accuracy of the nonlinear modeling, and performs predistortion processing on the intermediate frequency signal to obtain a predistortion signal, so that the nonlinear frequency generated by the input intermediate frequency signal of the predistortion processing device is predistorted, and the generated power is generated by the power amplifier.
  • the non-linear quantities cancel each other out, improving the performance of digital predistortion techniques in the case of wideband.
  • the output signals of the pre-processing circuit are the intermediate frequency signals (signal X 1 , signal X 2 ; X 1 corresponds to band 1 and X 2 corresponds to band 2), and both band 1 and band 2 are
  • the internal operation flow of a predistortion processing device in the embodiment of the present invention is described in detail by taking a baseband signal of a carrier of a plurality of frequency bands and an auxiliary feedback module simultaneously feeding back the envelope signal and the harmonic signal.
  • the frequency range of the envelope signal is f 2 -f 1 -40MHz ⁇ f e ⁇ f 2 -f 1 +40MHz
  • the frequency range of the feedback signal of the second harmonic corresponding to the frequency band 1 and the frequency band 2 is 2f 1 -40MHz ⁇ f h1 ⁇ 2f 1 + 40MHz and 2f 2 -40MHz ⁇ f h2 ⁇ 2f 2 + 40MHz
  • a device specific application scenario of one embodiment of the predistortion process embodiment of the present invention is as follows:
  • the pre-processing module 406 digitally upconverts the input frequency band 1 and the frequency band 2, combines all the carriers in the same frequency band, performs clipping processing, obtains the signal X 1 and the signal X 2 , and inputs the signal X 1 and the signal X 2 .
  • the signal combining and transmitting module 407 converts the predistortion signal y 1 , y 2 outputted by the predistortion processing module 405 into an analog signal and outputs it to the power amplifying module 408;
  • the signal combining and transmitting module 407 converts y 1 and y 2 into analog signals, and the specific implementation manner is as follows:
  • the signals in each of the predistortion signals y 1 and y 2 are digitally upconverted, combined, and converted into analog signals.
  • the power amplification module 408 amplifies the analog signal of the input power amplification module 403, and outputs it to the auxiliary feedback module 401 and the RF signal feedback module 403;
  • the RF signal feedback module 403 receives the amplified analog signal input by the power amplification module 408, and extracts the fundamental wave signals Z 1 and Z 2 from the amplified analog signal, and samples and processes the fundamental wave signal Z to obtain a fundamental wave.
  • the fundamental feedback signal d f1 corresponding to the signal Z 1 and the fundamental feedback signal d f2 corresponding to the fundamental signal Z 2 are input to the predistortion model coefficient training module 407.
  • the fundamental feedback signals d f1 and d f2 are obtained by using the fundamental signals Z 1 and Z 2 as follows.
  • the two fundamental signals Z 1 , Z 2 and the predistortion signals y 1 and y 2 are respectively delayed. After phase alignment, the corresponding fundamental feedback signals d f1 , d f2 are obtained .
  • the auxiliary feedback module 401 extracts the envelope signal E and the harmonic signals F 1 and F 2 from the amplified analog signal, and separately samples the E, F 1 , and F 2 to obtain an envelope feedback corresponding to the envelope signal E.
  • a signal d e , an envelope feedback signal d f1 corresponding to F 1 , an envelope feedback signal d h2 corresponding to F 2 , and d e , d h1 and d h2 are input to the auxiliary model coefficient training module 407;
  • the digital intermediate frequency feedback signals of the three signals E, F 1 and F 2 are respectively corresponding to the reference signals.
  • the delay estimation, phase alignment and power adjustment are performed, and three adjusted envelope feedback signals and harmonic feedback signals are respectively obtained, which are respectively recorded as d e , d h1 and d h2 .
  • the auxiliary model coefficient training module 402 trains the auxiliary coefficients of the auxiliary model coefficient training module 402 according to d e , d h1 , d h1 and y 1 , y 2 output from the predistortion processing module 405, and obtains the target auxiliary coefficient after convergence, and Passing the target auxiliary coefficient to the predistortion processing module 405 and the predistortion model coefficient training module 404;
  • the specific implementation manner of training the auxiliary coefficients of the auxiliary model coefficient training module 402 by using d e , d h1 , d h2 , y 1 and y 2 may be:
  • the model is a linear model, the specific expression is as follows:
  • w e , w h1 , w h2 are the filter coefficients of the envelope signal, the second harmonic of band 1 and band 2, respectively, ie the target auxiliary coefficient, and three linear equations are constructed by using successive N sampling points, And using the least square method or an adaptive algorithm such as LMS, RLS, etc., the target auxiliary coefficients w e , w h1 , w h2 are calculated, and the target auxiliary coefficients are transmitted to the fifth auxiliary filtering submodule in the predistortion processing module 405. 4051 and a sixth auxiliary filtering sub-module 4041 in the pre-distortion model coefficient training module 404, the target auxiliary coefficient is used for auxiliary filtering.
  • model in this embodiment may also adopt nonlinear modeling, such as a memory polynomial model, and the specific modeling manner is not limited herein.
  • the predistortion model coefficient training module 404 trains the predistortion coefficients of the predistortion model coefficient training module 404 according to y 1 , y 2 , and the fundamental feedback signals d f1 , d f2 output from the RF signal feedback module, and obtains the target after convergence. Predistortion coefficient, and the trained predistortion coefficient is transmitted to the predistortion processing module 405;
  • d f1 , d f2 the use of d f1 , d f2 to train the predistortion coefficient is as follows:
  • the envelope reference signal of the envelope signal and the harmonic reference signal of the second harmonic signal corresponding to the frequency bands 1 and 2 are constructed by using the fundamental feedback signals d f1 and d f2 , respectively And inputting the three signals into the sixth auxiliary filtering sub-module 4041, and recording the obtained three output signals as p e , p f1 , p f2 , and respectively Correspondingly, the sixth filter sub-module in this article will not be described again.
  • the sixth auxiliary filtering sub-module 4041 completion function may be described by the following formula:
  • the third pre-distortion processing sub-module 4052 in the pre-distortion processing module 405 performs nonlinear construction by using the output signals p e , p f1 , p f2 of the sixth auxiliary filtering sub-module 4041 and the fundamental feedback signals d f1 , d f2 .
  • the modulo, fitting the predistortion circuit output signals y 1 , y 2 For the convenience of description, only the third-order nonlinear component is considered, and the specific nonlinear model is as follows:
  • i 1, 2, 3, 4, 5
  • the predistortion coefficients w 1 and w 2 of the band 1 band 2 are expressed. for:
  • Two linear equations are constructed by using successive N sampling points, and the predistortion coefficients w 1 and w 2 of the band 1 band 2 are solved by the least square method or other adaptive algorithms such as LMS, RLS, and the predistortion coefficient. Passed to the predistortion processing module 405 for predistortion processing.
  • the predistortion processing module 405 performs nonlinear modeling according to the target auxiliary coefficient and the predistortion coefficient, performs predistortion processing on the input intermediate frequency signal, and outputs the predistortion signal obtained by the predistortion processing.
  • the nonlinear modeling using the target auxiliary coefficients w e , w h1 , w h2 and the pre-distortion coefficient is as follows:
  • the predistortion processing sub-module 4052 constructs a reference signal of the corresponding envelope signal and a reference signal of the second harmonic signal by using the digital intermediate frequency signals x 1 and x 2 of the frequency band 1 and the frequency band 2, respectively.
  • the three signals are input into the fifth auxiliary filtering sub-module 4051, and three output signals are respectively recorded as q e , q f1 , q f2 , and the fifth auxiliary filtering sub-module 4051 completion function can be described by the following formula:
  • the third pre-distortion processing sub-module 4052 performs nonlinear modeling using the q e , q f1 , q f2 and the band 1 band 2 signals x 1 , x 2 output by the fifth auxiliary filtering sub-module 4051, and the pre-distortion coefficient w 1 And w 2 correspondingly multiplied, and summed to obtain the output signals y 1 , y 2 of the predistortion processing module 405, wherein the predistortion processing module 405 implementation function can be represented by the following two formulas:
  • the predistortion signals y 1 , y 2 are input to the signal combining and transmitting module 407 such that the signal combining and transmitting module 407 converts y 1 , y 2 into analog signals and outputs them to the power amplifying module 408.
  • the performance of the ACLR (Adjacent Channel Leakage Ratio) of the fundamental feedback signals d f1 , d f2 can be detected;
  • the coefficient training module 404 trains the auxiliary coefficients, the predistortion model coefficient training module 404 trains the predistortion coefficients and the predistortion processing module 405 follows the modeling steps until convergence.
  • the pre-distortion processing module herein may be a pre-distortion circuit
  • the fifth auxiliary filtering sub-module and the sixth auxiliary filtering sub-module may specifically be auxiliary filtering circuits
  • the power amplifying module may be a power amplifier and an auxiliary feedback module.
  • the auxiliary feedback circuit may be a predistortion model coefficient training module, which may be a predistortion model coefficient learning circuit
  • the auxiliary model coefficient training module may be an auxiliary model coefficient learning circuit
  • the signal combining and transmitting module may be a signal combining and transmitting circuit, and a radio frequency signal.
  • the feedback module may be a radio frequency signal feedback circuit.
  • the related circuit is selected to implement the solution of the embodiment of the present invention. For details, refer to the signal transmission flowchart shown in FIG. 5.
  • the envelope signal and the harmonic signal are extracted from the analog signal by the auxiliary feedback module 401, and the envelope signal and the harmonic signal are sampled, and the envelope feedback signal and the harmonic feedback signal are obtained to realize real-time feedback.
  • the envelope signal and the harmonic signal, the auxiliary model coefficient training module 402 trains the auxiliary coefficient by using the envelope feedback signal, the harmonic feedback signal and the predistortion signal, and transmits the obtained target auxiliary coefficient to the predistortion processing module 405 and the pre
  • the distortion model coefficient training module 404, the predistortion processing module 405 uses the target auxiliary coefficient and the predistortion coefficient trained by the predistortion model coefficient training module 403 to perform nonlinear modeling, thereby effectively improving the accuracy of the nonlinear modeling, and performing the IF signal.
  • the predistortion process obtains a predistortion signal, so that after the predistortion processing of the intermediate frequency signal input to the predistortion processing device, the generated nonlinear quantity cancels the nonlinear quantity generated by the power amplifying module 408, and the digital predistortion is improved under the broadband condition.
  • the device for predistortion processing in the embodiment of the present invention is described in detail.
  • the following describes a method for predistortion processing in the embodiment of the present invention. Referring to FIG. 7, a predistortion process in the embodiment of the present invention is described.
  • An embodiment of the method includes:
  • the auxiliary feedback module receives the amplified analog signal, and extracts a nonlinear distortion signal from the analog signal, obtains a feedback signal corresponding to the nonlinear distortion signal, and inputs the feedback signal into the auxiliary mode.
  • Type coefficient training module ;
  • the nonlinear distortion signal After extracting the nonlinear distortion signal from the analog signal, the nonlinear distortion signal is sampled to obtain a feedback signal corresponding to the nonlinear distortion signal, and the specific implementation manner of the sampling processing may be:
  • the nonlinear distortion signal is converted into a digital signal, and then the digital signal can be digitally down-converted and phase-aligned by an ADC (Digital to Analog Conversion) sampling method.
  • the signal is down-converted, and the digital signal is phase-aligned by using a phase shifter, so that a feedback signal corresponding to at least one frequency band can be obtained.
  • ADC Digital to Analog Conversion
  • the specific sampling processing manner is not limited herein, and the nonlinear distortion signal can be a single frequency band.
  • the dual frequency band may also include different frequency bands and the like, and specific components are not limited herein.
  • the auxiliary model coefficient training module receives the feedback signal and the predistortion signal output from the predistortion processing module, and trains the auxiliary coefficient of the auxiliary model coefficient training module according to the feedback signal and the predistortion signal input from the predistortion processing module, and The first auxiliary coefficient obtained by the training is transmitted to the predistortion processing module;
  • the RF signal feedback module receives the amplified analog signal, and extracts a fundamental wave signal from the amplified analog signal, obtains a fundamental wave feedback signal corresponding to the fundamental wave signal, and then inputs the fundamental wave feedback signal into the predistortion model coefficient training.
  • the predistortion model coefficient training module receives the fundamental wave feedback signal and the predistortion signal output from the predistortion processing module, and the predistortion model coefficient training module according to the fundamental wave feedback signal and the fundamental wave feedback signal input from the RF signal feedback module.
  • the pre-distortion coefficient is trained, and the pre-distortion coefficient obtained by the training is transmitted to the pre-distortion processing module;
  • the predistortion processing module performs nonlinear modeling according to the first auxiliary coefficient and the predistortion coefficient, performs predistortion processing on the input intermediate frequency signal, and outputs the predistortion signal obtained by the predistortion processing.
  • the auxiliary model coefficient training module transmits the auxiliary coefficient to the predistortion processing module and the predistortion model coefficient training module, and the predistortion model coefficient training module transmits the predistortion coefficient to the predistortion processing module. This can be done like this:
  • the data in the register corresponding to the storage coefficient of the circuit A is copied to the register corresponding to the storage coefficient of the circuit B, and only the data is transferred.
  • FPGA Field-Programmable Gate Array
  • the auxiliary distortion module extracts the nonlinear distortion signal from the analog signal, samples the nonlinear distortion signal, obtains the feedback signal corresponding to the nonlinear distortion signal, and realizes the real-time feedback nonlinear distortion signal, and the auxiliary model.
  • the coefficient training module trains the auxiliary coefficient by using the feedback signal output by the auxiliary feedback module and the predistortion signal output by the predistortion processing module, and transmits the obtained first auxiliary coefficient to the predistortion processing module and the predistortion model coefficient training module.
  • the distortion processing module uses the first auxiliary coefficient and the predistortion coefficient trained by the predistortion model coefficient training module to perform nonlinear modeling, effectively improves the accuracy of the nonlinear modeling, and performs predistortion processing on the intermediate frequency signal to obtain a predistortion signal.
  • the generated nonlinear quantity cancels out the nonlinear quantity generated by the power amplifier, linearly amplifies, and improves the performance of the digital predistortion technology under the broadband condition.
  • the nonlinear distortion signal includes an envelope signal
  • the predistortion processing module includes a second auxiliary filtering sub-module and a first pre-distortion coefficient processing sub-module
  • the auxiliary feedback module extracts the envelope signal from the analog signal, samples the envelope signal, obtains a feedback signal corresponding to the envelope signal, and inputs the envelope feedback signal into the auxiliary model coefficient training module;
  • the auxiliary model coefficient training module uses the pre-distortion signal and the envelope feedback signal of the input auxiliary model coefficient training module to model, trains the auxiliary coefficients, converges to obtain the second auxiliary coefficient, and transmits the second auxiliary coefficient to the first auxiliary.
  • a filtering submodule and a second auxiliary filtering submodule uses the pre-distortion signal and the envelope feedback signal of the input auxiliary model coefficient training module to model, trains the auxiliary coefficients, converges to obtain the second auxiliary coefficient, and transmits the second auxiliary coefficient to the first auxiliary.
  • the first auxiliary filtering sub-module constructs an envelope reference signal corresponding to the intermediate frequency signal according to the intermediate frequency signal of the input predistortion processing module, performs filtering processing on the envelope reference signal by using the second auxiliary coefficient, and obtains the first output.
  • the first pre-distortion processing sub-module generates a pre-distortion vector according to the preset pre-distortion model by using the first output signal output by the first auxiliary filtering sub-module and the clipping processed baseband signal input to the first pre-distortion processing sub-module,
  • the predistortion vector is multiplied by the predistortion coefficient to obtain a predistortion signal and output;
  • the second auxiliary filtering sub-module applies a second auxiliary coefficient to the fundamental feedback signal input from the RF feedback signal module, and inputs the obtained second output signal into the pre-distortion model processing sub-module;
  • the first pre-distortion coefficient processing sub-module performs nonlinear modeling according to the second output signal, the fundamental feedback signal of the first pre-distortion coefficient processing sub-module, and the pre-distortion signal, and pre-distortion coefficient of the pre-distortion model coefficient training module Train to get the predistortion coefficient.
  • the nonlinear distortion signal includes an envelope signal and a harmonic signal
  • the predistortion processing module includes a third auxiliary filtering submodule and a second predistortion processing submodule
  • the predistortion model coefficient training module includes a fourth auxiliary filtering submodule and a second pre Distortion coefficient processing sub-module
  • the auxiliary feedback module extracts the envelope signal and the harmonic signal from the analog signal, and samples and processes the envelope signal and the harmonic signal to obtain an envelope feedback signal corresponding to the envelope signal and a harmonic corresponding to the harmonic signal.
  • Feedback signal and input the envelope feedback signal and the harmonic feedback signal into the auxiliary model coefficient training module;
  • the auxiliary model coefficient training module trains the auxiliary coefficients by using the predistortion signal, the envelope feedback signal and the harmonic feedback signal of the input auxiliary model coefficient training module, and obtains the third auxiliary coefficient after convergence, and transmits the third auxiliary coefficient to the third.
  • the third auxiliary filtering sub-module constructs an envelope reference signal and a harmonic reference signal corresponding to the intermediate frequency signal according to the intermediate frequency signal of the input predistortion processing module, and filters the envelope reference signal and the harmonic reference signal by using the second auxiliary coefficient. Processing, and inputting the obtained third output signal into the second predistortion processing submodule;
  • the second pre-distortion processing sub-module generates a pre-distortion vector according to the preset pre-distortion model by using the third output signal output by the third auxiliary filtering sub-module and the baseband signal input by the clipping of the second pre-distortion processing sub-module.
  • the predistortion vector is multiplied by the predistortion coefficient to obtain a predistortion signal and output;
  • the fourth auxiliary filtering sub-module performs filtering processing on the fundamental wave feedback signal input from the RF feedback signal module, and uses the second auxiliary coefficient to perform filtering processing, and inputs the obtained fourth output signal into the second pre-distortion coefficient processing sub-module;
  • the second pre-distortion coefficient processing sub-module performs nonlinear modeling according to the fourth output signal, the fundamental feedback signal of the second pre-distortion coefficient processing sub-module, and the pre-distortion signal, and pre-distortion coefficient of the pre-distortion model coefficient training module Train to get the predistortion coefficient.
  • the nonlinear distortion signal is a harmonic signal
  • the predistortion processing module includes a fifth auxiliary filtering submodule and a third predistortion processing submodule
  • the predistortion model coefficient training module includes a sixth auxiliary filtering submodule and a third predistortion coefficient processor.
  • the auxiliary feedback module extracts the harmonic signal from the analog signal, samples the harmonic signal, obtains the harmonic feedback signal corresponding to the harmonic signal, and inputs the harmonic feedback signal into the auxiliary model coefficient training module;
  • the auxiliary model coefficient training module trains the auxiliary coefficients by using the predistortion signal and the harmonic feedback signal of the input auxiliary model coefficient training module, and obtains the fourth auxiliary coefficient after convergence, and transmits the fourth auxiliary coefficient to the fifth auxiliary filtering submodule and a sixth auxiliary filtering submodule;
  • the fifth auxiliary filtering sub-module constructs a harmonic reference signal corresponding to the intermediate frequency signal according to the intermediate frequency signal of the input predistortion processing module, performs filtering processing on the harmonic reference signal by using the fourth auxiliary coefficient, and obtains the fifth output signal.
  • a third predistortion processing submodule Inputting a third predistortion processing submodule;
  • the third pre-distortion processing sub-module generates a pre-distortion vector according to the preset pre-distortion model by using the fifth output signal output by the fifth auxiliary filtering sub-module and the baseband signal input by the clipping of the third pre-distortion processing sub-module.
  • the predistortion vector is multiplied by the predistortion coefficient to obtain a predistortion signal and output;
  • the sixth auxiliary filtering sub-module performs filtering processing on the fundamental feedback signal input from the RF feedback signal module, and uses the second auxiliary coefficient to perform filtering processing, and inputs the obtained sixth output signal into the pre-distortion model processing sub-module;
  • the third pre-distortion coefficient processing sub-module performs nonlinear modeling according to the sixth output signal, the fundamental feedback signal of the third pre-distortion coefficient processing sub-module, and the pre-distortion signal, and the pre-distortion coefficient of the pre-distortion model coefficient training module Train to get the predistortion coefficient.
  • the preprocessing module digitally upconverts the baseband signal input to the preprocessing module, combines all carriers in the same frequency band, performs clipping processing, and inputs the clipped baseband signal into the predistortion processing module, and the baseband signal includes multiple frequency bands.
  • the signal combining and transmitting module receives the predistortion signal outputted from the predistortion processing module, and converts the predistortion signal outputted from the predistortion processing module into an analog signal, and outputs the signal to the power amplifying module;
  • the power amplification module amplifies the analog signal of the input power amplification module, and outputs the amplified analog signal to the auxiliary feedback module and the RF signal feedback module;
  • At least one of the envelope signal or the harmonic signal may be fed back by the auxiliary feedback module, and the envelope signal is converted into an envelope feedback signal or the harmonic signal is converted into a harmonic feedback signal to assist the model coefficient.
  • the training module uses the envelope feedback signal or the harmonic feedback signal to perform nonlinear modeling, and transmits the trained auxiliary coefficients to the predistortion model coefficient training module and the predistortion processing module, so that the predistortion model coefficient training module utilizes the auxiliary coefficients and
  • the fundamental feedback signal is modeled and the trained predistortion coefficient is transmitted to the predistortion processing module to improve the accuracy of the predistortion modeling, so that the predistortion processing module uses the predistortion coefficient to predistort the intermediate frequency signal to make the intermediate frequency signal
  • the amount of nonlinearity generated is offset by the amount of nonlinearity produced by the power amplifier, thereby improving the performance of the digital predistortion technique in the case of wideband.
  • the predistortion processing module trained in the predistortion model coefficient training module herein can use the predistortion coefficient in the first preset threshold range, and the predistortion coefficient can be used without real-time update.
  • the distortion coefficient, and the auxiliary model coefficient training module can perform regular training on the auxiliary coefficient, or can perform training from time to time.
  • the auxiliary coefficient can be updated without being updated or preset as long as it is within the second preset threshold range.
  • the distortion model coefficient training module training pre-distortion coefficient and the auxiliary model coefficient training module training auxiliary coefficient period are determined according to circuit design and customer requirements, and the specific implementation manner is not limited herein.

Landscapes

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

Abstract

本发明提供一种预失真处理的装置及方法，装置包括：辅助反馈模块，用于从模拟信号中提取非线性失真信号，将获取的与非线性失真信号对应的反馈信号输入辅助模型系数训练模块；辅助模型系数训练模块，用于根据反馈信号、预失真信号对辅助系数进行训练，并将训练得到的第一辅助系数传递给预失真处理模块；射频信号反馈模块，用于提取基波反馈信号；预失真模型系数训练模块，用于根据预失真信号、基波反馈信号对预失真系数进行训练，将得到的预失真系数传递给预失真处理模块；预失真处理模块，用于根据第一辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。

Description

一种预失真处理的装置及方法 技术领域
本发明涉及线性调制技术领域，尤其涉及一种预失真处理的装置及方法。
背景技术
随着移动通信技术的发展和移动通信用户的需求越来越高，移动通信技术的应用场景越来越复杂，移动通信***在长时间内会有多种信号制式并存的局面，以适应更加复杂的移动通信应用场景，功率放大器PA(Power amplify)是移动通信***射频处理单元RRU(Radio Remote Unit)的重要组成部分，由于PA的非线性和记忆效应会引起功放输出信号的非线性失真，从而造成PA输出信号的畸变和信号带外功率泄露增加，严重时会导致接收机无法正确的接受信号，并对相邻频带的通信***造成干扰，通常利用支持多频带、多制式的超宽频的数字预失真技术DPD(Digital Pre-Distortion)和RRU来减小PA输出信号非线性失真，一般将DPD模块设计在PA之前，数字中频信号通过DPD模块所产生出非线性分量与PA产生的非线性分量相互抵消，达到改善PA输出线性度的效果。
现有技术中，将反馈信号和预失真电路输出信号作为预失真系数学习电路的输入信号，分别对每个频带进行预失真模型系数训练，得到N个频带对应的非线性模型系数，并将训练收敛后预失真系数复制给预失真电路，然后根据预失真系数对功率放大器进行预失真电路的非线性建模。
但是，在宽带通信***中，超宽带的功率放大器设计受到器件性能的制约，低频包络阻抗难以优化，视频带宽不足导致功放的电学记忆效应较强，同时谐波频率范围会更宽，谐波阻抗也难以优化，也会引起电学记忆效应增强。在传统的数字预失真技术中，由于反馈通道仅反馈基波信号，无法实时反馈基波信号和谐波信号，导致预失真处理电路只能利用基波信号的信息进行预失真电路非线性建模，无法实时利用功放输出的包络信号和谐波信号所包含的信息进行预失真电路非线性建模，导致预失真电路非线性建模准确度降低，以及降低预失真电路性能指标。
发明内容
本发明提供一种预失真处理的装置及方法，能够解决现有技术中宽带通信***下，预失真处理电路进行预失真建模的准确度不高的问题。
本发明第一方面提供一种预失真处理的装置，包括：
辅助反馈模块，用于接收放大后的模拟信号，并从放大后的模拟信号中提取非线性失真信号，获取与所述非线性失真信号对应的反馈信号，并将所述反馈信号输入辅助模型系数训练模块；
所述辅助模型系数训练模块，用于接收所述反馈信号和自预失真处理模块输出的预失真信号，根据所述反馈信号、自预失真处理模块输出的预失真信号对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第一辅助系数，并将训练得到的第一辅助系数传递给所述预失真处理模块；
射频信号反馈模块，用于接收放大后的模拟信号，并从所述放大后的模拟信号中提取基波信号，获取与所述基波信号对应的基波反馈信号后，将所述基波反馈信号输入所述预失真模型系数训练模块；
所述预失真模型系数训练模块，用于接收基波反馈信号和自所述预失真处理模块输出的预失真信号，根据所述基波反馈信号、自所述预失真处理模块输出的预失真信号，对所述预失真模型系数训练模块的预失真系数进行训练，并将训练得到的预失真系数传递给所述预失真处理模块；
所述预失真处理模块，用于根据所述第一辅助系数和所述预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
结合第一方面，本发明实施例中第一方面的第一种实现方式中，所述非线性失真信号包括包络信号，所述辅助反馈模块具体用于接收放大后的模拟信号，对所述包络信号进行采样处理，得到与所述包络信号对应的包络反馈信号，并将所述包络反馈信号输入辅助模型系数训练模块。
结合第一方面及第一方面的第一种实现方式，本发明实施例中第一方面的第二种实现方式中，所述预失真处理模块包括第一辅助滤波子模块和第一预失真处理子模块，所述预失真模型系数训练模块包括第二辅助滤波子模块和第一预失真系数处理子模块，
所述辅助模型系数训练模块具体用于利用输入所述辅助模型系数训练模 块的预失真信号、包络反馈信号进行建模，对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第二辅助系数，并将所述第二辅助系数传递给所述第一辅助滤波子模块和所述第二辅助滤波子模块。
结合第一方面及第一方面的第二种实现方式，本发明实施例中第一方面的第三种实现方式中，所述第一辅助滤波子模块用于根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号，并利用所述第二辅助系数对所述包络参考信号进行滤波处理，并将得到的第一输出信号输入所述第一预失真处理子模块；
所述第一预失真处理子模块用于根据预设预失真模型，利用所述第一辅助滤波子模块输出的第一输出信号和输入所述预失真处理模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
所述第二辅助滤波子模块用于对自射频反馈信号模块输入的基波反馈信号，利用所述第二辅助系数进行滤波处理，并将得到的第二输出信号输入所述第一预失真系数处理子模块；
所述第一预失真系数处理子模块用于根据所述第二输出信号、输入所述第一预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
结合第一方面，本发明实施例中第一方面的第四种实现方式中，所述非线性失真信号包括包络信号和谐波信号，所述辅助反馈模块具体用于接收放大后的模拟信号，从放大后的模拟信号中提取包络信号和谐波信号，对所述包络信号和所述谐波信号进行采样处理，得到与所述包络信号对应的包络反馈信号，及与所述谐波信号对应的谐波反馈信号，并将所述包络反馈信号、所述谐波反馈信号输入所述辅助模型系数训练模块。
结合第一方面及第一方面的第四种实现方式，本发明实施例中第一方面的第五种实现方式中，所述预失真处理模块包括第三辅助滤波子模块和第二预失真处理子模块，所述预失真模型系数训练模块包括第四辅助滤波子模块和第二预失真系数处理子模块，
所述辅助模型系数训练模块具体用于利用输入所述辅助模型系数训练模 块的预失真信号、包络反馈信号和谐波反馈信号进行建模，对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第三辅助系数，并将所述第三辅助系数传递给所述第三辅助滤波子模块和所述第四辅助滤波子模块。
结合第一方面及第一方面的第五种实现方式，本发明实施例中第一方面的第六种实现方式中，所述第三辅助滤波子模块用于根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号和谐波参考信号，并利用所述第三辅助系数对所述包络参考信号和所述谐波参考信号进行滤波处理，并将得到的第三输出信号输入所述第二预失真处理子模块；
所述第二预失真处理子模块用于根据预设预失真模型，利用所述第一辅助滤波子模块输出的第三输出信号和输入所述预失真处理模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
所述第四辅助滤波子模块用于对自射频反馈信号模块输入的基波反馈信号，利用所述第三辅助系数进行滤波处理，并将得到的第四输出信号输入所述第二预失真系数处理子模块；
所述第二预失真系数处理子模块用于根据所述第四输出信号、输入所述第二预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
结合第一方面，本发明实施例中第一方面的第七种实现方式中，所述非线性失真信号包括谐波信号，所述辅助反馈模块具体用于从放大后的模拟信号中提取谐波信号，并对所述谐波信号进行采样处理，得到与所述谐波信号对应的谐波反馈信号，并将所述谐波反馈信号输入辅助模型系数训练模块。
结合第一方面及第一方面的第七种实现方式，本发明实施例中第一方面的第八种实现方式中，所述预失真处理模块包括第五辅助滤波子模块和第三预失真处理子模块，所述预失真模型系数训练模块包括第六辅助滤波子模块和第三预失真系数处理子模块，
所述辅助模型系数训练模块具体用于利用输入所述辅助模型系数训练模块的预失真信号、谐波反馈信号进行建模，对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第四辅助系数，并将所述第四辅助系数传递给所 述第五辅助滤波子模块和所述第六辅助滤波子模块。
结合第一方面及第一方面的第八种实现方式，本发明实施例中第一方面的第九种实现方式中，第五辅助滤波子模块用于根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的谐波参考信号，并对所述谐波参考信号进行滤波处理，并将利用所述第四辅助系数得到的第五输出信号输入所述第三预失真处理子模块；
所述第三预失真处理子模块用于根据预设预失真模型，利用所述第五辅助滤波子模块输出的第五输出信号和输入所述预失真处理模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
所述第六辅助滤波子模块用于对自射频反馈信号模块输入的基波反馈信号，利用所述第四辅助系数进行滤波处理，并将得到的第六输出信号输入预失真模型处理子模块；
所述第三预失真系数处理子模块用于根据所述第六输出信号、输入所述第三预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
结合第一方面及第一方面的第一至第九种实现方式，本发明实施例中第一方面的第十种实现方式中，所述装置还包括：
预处理模块，用于将输入所述预处理模块的基带信号进行数字上变频，并将同一频段的所有载波合并后进行削波处理，并将削波处理后的基带信号输入所述预失真处理模块，所述基带信号包括多个频段的载波；
信号合并和传输模块，用于接收自所述预失真处理模块输出的预失真信号，并将自所述预失真处理模块输出的预失真信号转化为模拟信号后输出至功率放大模块；
所述功率放大模块，用于对输入所述功率放大模块的模拟信号进行放大，并将放大后的模拟信号输出至所述辅助反馈模块及所述射频信号反馈模块。
本发明第二方面提供一种预失真处理的方法，其特征在于，所述方法包括：
辅助反馈模块接收放大后的模拟信号，并从放大后的模拟信号中提取非线 性失真信号，获取与所述非线性失真信号对应的反馈信号，并将所述反馈信号输入辅助模型系数训练模块；
所述辅助模型系数训练模块接收所述反馈信号和自预失真处理模块输出的预失真信号，根据所述反馈信号、自预失真处理模块输入的预失真信号对所述辅助模型系数训练模块的辅助系数进行训练，并将训练得到的第一辅助系数传递给所述预失真处理模块；
射频信号反馈模块接收放大后的模拟信号，并从所述放大后的模拟信号中提取基波信号，获取与所述基波信号对应的基波反馈信号后，将所述基波反馈信号输入所述预失真模型系数训练模块；
所述预失真模型系数训练模块接收基波反馈信号和自所述预失真处理模块输出的预失真信号，根据所述基波反馈信号、自所述预失真处理模块输出的预失真信号，对所述预失真模型系数训练模块的预失真系数进行训练，并将训练得到的预失真系数传递给所述预失真处理模块；
所述预失真处理模块根据所述第一辅助系数和所述预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
结合第二方面，本发明实施例中第二方面的第一种实现方式中，所述非线性失真信号包括包络信号，所述辅助反馈模块接收放大后的模拟信号，从所述模拟信号中提取非线性失真信号，并对所述非线性失真信号进行采样处理，得到与所述非线性失真信号对应的反馈信号，并将所述反馈信号输入辅助模型系数训练模块具体包括：
所述辅助反馈模块接收放大后的模拟信号，从所述模拟信号中提取所述包络信号，并对所述包络信号进行采样处理，得到与所述包络信号对应的反馈信号，并将所述包络反馈信号输入辅助模型系数训练模块。
结合第二方面及第二方面的第一种实现方式，本发明第二方面的第二种实现方式中，所述预失真处理模块包括第一辅助滤波子模块和第一预失真处理子模块，所述预失真模型系数训练模块包括第二辅助滤波子模块和第一预失真系数处理子模块，所述方法还包括：
所述辅助模型系数训练模块利用输入所述辅助模型系数训练模块的预失真信号、包络反馈信号进行建模，对所述辅助系数进行训练，收敛后得到第二 辅助系数，并将所述第二辅助系数传递给所述第一辅助滤波子模块和所述第二辅助滤波子模块。
结合第二方面及第二方面的第二种实现方式，本发明第二方面的第三种实现方式中，所述方法还包括：
所述第一辅助滤波子模块根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号，对所述包络参考信号进行滤波处理，并将利用所述第二辅助系数得到的第一输出信号输入所述预失真处理模块中的预失真处理子模块；
所述预失真处理子模块根据预设预失真模型，利用所述第一辅助滤波子模块输出的第一输出信号和输入所述预失真处理子模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
所述第二辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用所述第二辅助系数进行滤波处理，并将得到的第二输出信号输入预失真模型处理子模块；
所述第一预失真系数处理子模块根据所述第二输出信号、输入所述第一预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
结合第二方面，本发明第二方面的第四种实现方式中，所述非线性失真信号包括包络信号和谐波信号，所述方法还包括：
所述辅助反馈模块从所述模拟信号中提取包络信号和谐波信号，并对所述包络信号和谐波信号进行采样处理，得到与所述包络信号对应的包络反馈信号，以及与所述谐波信号对应的谐波反馈信号，并将所述包络反馈信号和所述谐波反馈信号输入所述辅助模型系数训练模块。
结合第二方面及第二方面的第四种实现方式，本发明第二方面的第五种实现方式中，所述预失真处理模块包括第三辅助滤波子模块和第二预失真处理子模块，所述预失真模型系数训练模块包括第四辅助滤波子模块和第二预失真系数处理子模块，所述方法还包括：
所述辅助模型系数训练模块利用输入所述辅助模型系数训练模块的预失真信号、包络反馈信号和谐波反馈信号对所述辅助系数进行训练，收敛后得到 第三辅助系数，并将所述第三辅助系数传递给所述第三辅助滤波子模块和所述第四辅助滤波子模块。
结合第二方面及第二方面的第五种实现方式，本发明第二方面的第六种实现方式中，所述方法还包括：
所述第三辅助滤波子模块根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号和谐波参考信号，对所述包络参考信号和所述谐波参考信号进行滤波处理，并将利用所述第三辅助系数得到的第三输出信号输入所述第二预失真处理子模块；
所述第二预失真处理子模块根据预设预失真模型，利用所述第三辅助滤波子模块输出的第三输出信号和输入所述第二预失真处理子模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
所述第四辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用所述第三辅助系数进行滤波处理，并将得到的第四输出信号输入所述第二预失真系数处理子模块；
所述第二预失真系数处理子模块根据所述第四输出信号、输入所述第二预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
结合第二方面，本发明实施例中第二方面的第七种实现方式中，所述非线性失真信号包括谐波信号，所述方法还包括：
所述辅助反馈模块从所述模拟信号中提取谐波信号，并对谐波信号进行采样处理，得到与所述谐波信号对应的谐波反馈信号，并将所述谐波反馈信号输入所述辅助模型系数训练模块。
结合第二方面及第二方面的第七种实现方式，本发明实施例中第二方面的第八种实现方式中，所述预失真处理模块包括第五辅助滤波子模块和第三预失真处理子模块，所述预失真模型系数训练模块包括第六辅助滤波子模块和第三预失真系数处理子模块，所述方法还包括：
所述辅助模型系数训练模块利用输入所述辅助模型系数训练模块的预失真信号和谐波反馈信号对所述辅助系数进行训练，收敛后得到第四辅助系数，并将所述第四辅助系数传递给所述第五辅助滤波子模块和所述第六辅助滤波 子模块。
结合第二方面及第二方面的第八种实现方式，本发明实施例中第二方面的第九种实现方式中，所述方法还包括：
所述第一辅助滤波子模块根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的谐波参考信号，对所述谐波参考信号进行滤波处理，并将利用所述第四辅助系数得到的第五输出信号输入所述第三预失真处理子模块；
所述第三预失真处理子模块根据预设预失真模型，利用所述第一辅助滤波子模块输出的第一输出信号和输入所述第三预失真处理子模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
所述第六辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用所述第四辅助系数进行滤波处理，并将得到的第六输出信号输入预失真模型处理子模块；
所述第三预失真系数处理子模块根据所述第六输出信号、输入所述第三预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
结合第二方面及第二方面的第一至第九种实现方式，本发明实施例中第二方面的第十种实现方式中，所述方法还包括：
预处理模块将输入所述预处理模块的基带信号进行数字上变频，将同一频段的所有载波合并后进行削波处理，并将削波处理后的基带信号输入所述预失真处理模块，所述基带信号包括多个频段的载波；
所述信号合并和传输模块将接收自所述预失真处理模块输出的预失真信号，并将自所述预失真处理模块输出的预失真信号转化为模拟信号后输出至功率放大模块；
所述功率放大模块对输入所述功率放大模块的模拟信号进行放大，并将放大后的模拟信号输出至所述辅助反馈模块及所述射频信号反馈模块。
从以上技术方案中可以看出，本发明中，通过增加辅助反馈模块从模拟信号中提取非线性失真信号，并得到与所述非线性失真信号对应的反馈信号，实 现实时反馈非线性失真信号，增加辅助模型系数训练模块利用辅助反馈模块输出的反馈信号及预失真信号对辅助系数进行训练，并将得到的第一辅助系数传递给预失真处理模块和预失真模型系数训练模块，使得预失真处理模块利用第一辅助系数和预失真模型系数训练模块训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，提升宽带通信***下数字预失真技术的性能。
附图说明
图1为本发明实施例中一种预失真处理的装置一结构示意图；
图2为本发明实施例中一种预失真处理的装置另一结构示意图；
图3为本发明实施例中一种预失真处理的装置另一结构示意图；
图4为本发明实施例中一种预失真处理的装置另一结构示意图；
图5为本发明实施例中一种预失真处理的装置内部运作流程示意图；
图6为本发明实施例中一种预失真处理的方法一实施例示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例，基于本发明中的实施例，本领域技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”“第四”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或模块的过程、方法、***、产品或设备不必限于清楚地列出的那些步骤或模块，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或模块，本文中所出现的模块的划分，仅仅是一种逻辑上的划分，实际应用中实现时可以有另外的划分方式，例如多个模块可以结合成或集成在另 一个***中，或一些特征可以忽略，或不执行，另外，所显示的或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，模块之间的间接耦合或通信连接可以是电性或其他类似的形式，本文中均不作限定。并且，作为分离部件说明的模块或子模块可以是也可以不是物理上的分离，可以是也可以不是物理模块，或者可以分不到多个电路模块中，可以根据实际的需要选择其中的部分或全部模块来实现本发明实施例方案的目的。
本发明实施例中，一种预失真处理的装置及方法，能够解决现有技术中宽带通信***下，由于反馈电路仅反馈基波信号，无法实时反馈包络信号和谐波信号等非线性失真信号，导致预失真处理电路只能根据基波反馈信号进行预失真建模，无法实时根据包络信号、谐波信号等预失真处理电路进行预失真建模，导致非线性建模的准确度不高，从而无法准确分析功率放大器的非线性特性的问题。该装置和方法可以用于线性调制领域，主要应用于移动通信***的RRU,其主要的应用场景为宽带多频带，多制式的移动通信***。一般将DPD模块设计在PA之前，数字中频信号通过DPD模块所产生的非线性分量与PA产生的非线性分量相互抵消，从而达到改善PA输出线性度的效果，并且，PA设计不同，包络阻抗和谐波阻抗对功放线性度和DPD的校正效果也不同。另外，由于PA为非线性器件，PA输出信号在频率上为全频谱，如果想获取频段为f的信号，那么对应的PA输出的信号中，频段f上的信号为基波信号，频段为2f，3f，4f等整数倍的信号则为谐波信号，在靠近0频附近的信号则为包络信号，其中，包络信号为基波信号包络在低频上的响应。
所以，基波信号、包络信号及谐波信号都是从模拟信号中提取的，实际上基波信号就是电路将发射的信号。
需要说明的是，本文中所出现的预失真信号包括包络信号、谐波信号，预失真信号可以为包络信号或谐波信号中的至少一个，也可以是其他类似能够引起非线性失真或记忆效应的信号皆可，具体辅助反馈模块反馈的预失真信号中信号的成分，本文中均不作限定；
另外，本文中，当辅助反馈模块反馈包络信号或谐波信号中至少一个时，相对应的辅助模型系数训练模块、辅助滤波子模块以及预失真模型系数训练模块都会相应的改变，例如，辅助反馈模块仅反馈包络信号时，辅助模型系数训 练模块、辅助滤波子模块以及预失真模型系数训练模块相应的仅处理与包络反馈信号相关的成分即可，具体本文中均不作限定。
本文中，辅助反馈模块和射频信号反馈模块可以是零中频接收机架构，也可以是数字中频接收机架构或者直接射频接收机的架构或其他类似的接收机架构，只要能实现本文中所描述的功能即可，具体结构本文中均不作限定。
下面介绍本发明实施例中一种预失真处理的装置实施例，该装置应用与线性调制领域，请参阅图1，本发明实施例的预失真处理的装置包括：
辅助反馈模块101，用于接收放大后的模拟信号，并从放大后的模拟信号中提取非线性失真信号，获取与非线性失真信号对应的反馈信号，并将反馈信号输入辅助模型系数训练模块102；
辅助模型系数训练模块102，用于接收反馈信号和自预失真处理模块输出的预失真信号，根据反馈信号、自预失真处理模块105输出的预失真信号对辅助模型系数训练模块102的辅助系数进行训练，收敛后得到第一辅助系数，并将训练得到的第一辅助系数传递给预失真处理模块105；
射频信号反馈模块103，用于接收放大后的模拟信号，并从放大后的模拟信号中提取基波信号，获取与基波信号对应的基波反馈信号后输入预失真模型系数训练模块104；
预失真模型系数训练模块104，用于接收基波反馈信号和自预失真处理模块输出的预失真信号，根据自预失真处理模块105输出的预失真信号预失真信号、自射频信号反馈模块103输出的基波反馈信号，对预失真模型系数训练模块103的预失真系数进行训练，并将训练得到的预失真系数传递给预失真处理模块105；
预失真处理模块105，用于接收基波反馈信号和自预失真处理模块输出的预失真信号，根据第一辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
本发明实施例中，通过辅助反馈模块101从模拟信号中提取非线性失真信号，得到与非线性失真信号对应的反馈信号，实现实时反馈非线性失真信号，辅助模型系数训练模块102利用辅助反馈模块101输出的反馈信号及预失真处理模块105输出的预失真信号对辅助系数进行训练，并将得到的第一辅助系数 传递给预失真处理模块105和预失真模型系数训练模块104，预失真处理模块105利用第一辅助系数和预失真模型系数训练模块104训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，对中频信号进行预失真处理，得到预失真信号，使得输入预失真处理装置的中频信号经过预失真处理后，所产生的非线性量与功率放大器产生的非线性量相互抵消，提升宽带情况下数字预失真技术的性能。
请参阅图2，本发明实施以辅助反馈模块反馈包络信号为例，本发明实施例中一种预失真处理的装置另一实施例包括：
辅助反馈模块201，用于接收放大后的模拟信号，并从放大后的模拟信号中提取非线性失真信号，从放大后的模拟信号中提取非线性失真信号，获取与非线性失真信号对应的反馈信号，并将反馈信号输入辅助模型系数训练模块202；
辅助模型系数训练模块202，用于根据反馈信号、自预失真处理模块204输出的预失真信号对辅助模型系数训练模块202的辅助系数进行训练，收敛后得到第一辅助系数，并将训练得到的第一辅助系数传递给预失真处理模块204；
射频信号反馈模块203，用于接收放大后的模拟信号，并从放大后的模拟信号中提取基波信号，获取与基波信号对应的基波反馈信号后，将基波反馈信号输入预失真模型系数训练模块；
预失真模型系数训练模块204，用于接收基波反馈信号和自预失真处理模块输出的预失真信号，根据自预失真处理模块205输出的预失真信号、自射频信号反馈模块203输出的基波反馈信号，对预失真模型系数训练模块203的预失真系数进行训练，并将训练得到的预失真系数传递给预失真处理模块205；
预失真处理模块205，用于接收中频信号，根据第一辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
可选的，在上述图2对应的实施例的基础上，非线性失真信号可以包括包络信号，预失真处理模块205包括第一辅助滤波子模块2051和第一预失真处理子模块2052，预失真模型系数训练模块204包括第二辅助滤波子模块2041和第一预失真系数处理子模块2042；
辅助反馈模块201具体用于接收放大后的模拟信号，从放大后的模拟信号中提取包络信号，对包络信号进行采样处理，得到与包络信号对应的包络反馈信号，并将包络反馈信号输入辅助模型系数训练模块202；
辅助模型系数训练模块202具体用于利用输入辅助模型系数训练模块202的预失真信号、包络反馈信号进行建模，对辅助模型系数训练模块202的辅助系数进行训练，收敛后得到第二辅助系数，并将第二辅助系数传递给第一辅助滤波子模块2051和第二辅助滤波子模块2041。
可选的，第一辅助滤波子模块2041，用于根据输入预失真处理模块204的中频信号构造与中频信号对应的包络参考信号，并对包络参考信号进行滤波处理，并将利用第二辅助系数得到的第一输出信号输入第一预失真处理子模块2042；
第一预失真处理子模块2042，用于根据预设预失真模型，利用第一辅助滤波子模块2041输出的第一输出信号和输入预失真处理模块204的削波处理后的基带信号生成预失真向量，将预失真向量和预失真系数相乘，得到预失真信号并输出；
可选的，第二辅助滤波子模块2041用于对自射频反馈信号模块203输入的基波反馈信号，利用第二辅助系数进行滤波处理，并将得到的第二输出信号输入第一预失真系数处理子模块2042；
第一预失真系数处理子模块2042用于根据第二输出信号、输入第一预失真系数处理子模块2042的基波反馈信号、预失真信号进行非线性建模，并对预失真模型系数训练模块204的预失真系数进行训练，得到预失真系数。
本发明实施例中，通过辅助反馈模块201从模拟信号中提取包络信号，对包络信号进行采样处理，得到包络反馈信号，实现实时反馈包络信号，辅助模型系数训练模块202利用包络反馈信号及预失真信号对辅助系数进行训练，并将得到的第二辅助系数传递给预失真处理模块205和预失真模型系数训练模块204，预失真处理模块利用第二辅助系数和预失真模型系数训练模块204训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，对中频信号进行预失真处理，得到预失真信号，使得输入预失真处理装置的中频信号经过预失真处理后，所产生的非线性量与功率放大器产生的非线性量相互抵消，线性放大，提升宽带情况下数字预失真技术的性能。
请参阅图3，本发明实施以辅助反馈模块反馈包络信号和谐波信号为例，本发明实施例中一种预失真处理的装置另一实施例包括：
辅助反馈模块301，用于接收放大后的模拟信号，并从放大后的模拟信号中提取非线性失真信号，对非线性失真信号进行采样处理，得到与非线性失真信号对应的反馈信号，并将反馈信号输入辅助模型系数训练模块302；
辅助模型系数训练模块302，用于接收反馈信号和自预失真处理模块305输出的预失真信号，根据反馈信号、自预失真处理模块304输出的预失真信号对辅助模型系数训练模块302的辅助系数进行训练，收敛后得到第一辅助系数，并将训练得到的第一辅助系数传递给预失真处理模块305；
射频信号反馈模块303，用于接收放大后的模拟信号，并从放大后的模拟信号中提取基波信号，获取与基波信号对应的基波反馈信号后，将基波反馈信号输入预失真模型系数训练模块304；
预失真模型系数训练模块304，用于接收基波反馈信号和预失真处理模块305输出的预失真信号，根据自预失真处理模块305输出的预失真信号、自射频信号反馈模块输出的基波反馈信号，对预失真模型系数训练模块303的预失真系数进行训练，收敛后得到预失真系数，并将训练得到的预失真系数传递给预失真处理模块304；
预失真处理模块304，用于根据第一辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
可选的，在上述图3对应的实施例的基础上，非线性失真信号可以包括包络信号和谐波信号，
辅助反馈模块301具体用于接收放大后的模拟信号，从放大后的模拟信号中提取包络信号和谐波信号，对包络信号和谐波信号进行采样处理，得到与包络信号对应的包络反馈信号，得到与谐波信号对应的谐波反馈信号，并将包络信号和谐波反馈信号输入辅助模型系数训练模块302；
本发明实施例中，预失真处理模块305包括第三辅助滤波子模块3051和第二预失真处理子模块3052，预失真模型系数训练模块304包括第四辅助滤波子模块3041和第二预失真系数处理子模块3042，
可选的，辅助模型系数训练模块302具体用于利用输入辅助模型系数训练模块302的预失真信号、包络反馈信号和谐波反馈信号进行建模，对辅助模型系数训练模块302的辅助系数进行训练，收敛后得到第三辅助系数，并将第三辅助系数传递给第三辅助滤波子模块3051和第四辅助滤波子模块3041；
可选的，第三辅助滤波子模块3051用于根据输入预失真处理模块305的中频信号构造与中频信号对应的包络参考信号和谐波参考信号，并利用第三辅助系数对包络参考信号和谐波参考信号进行滤波处理，并将得到的第三输出信号输入预失真处理子模块3052；
第二预失真处理子模块3052用于根据预设预失真模型，利用第一辅助滤波子模块3051输出的第三输出信号和输入预失真处理模块305的削波处理后的基带信号生成预失真向量，将预失真向量和预失真系数相乘，得到预失真信号并输出。
可选的，第四辅助滤波子模块3041，用于对自射频反馈信号模块307输入的基波反馈信号，利用第三辅助系数进行滤波处理，并将得到的第四输出信号输入预失真模型处理子模块3042；
第二预失真系数处理子模块3042，用于根据第四输出信号、输入第二预失真系数处理子模块3042的基波反馈信号、预失真信号进行非线性建模，并对预失真模型系数训练模块304的预失真系数进行训练，得到预失真系数。
需要说明的是，本文中的第一辅助滤波子模块3051和第二辅助滤波子模块3041的结构相同，都为数字滤波器，该数字滤波器主要有两种，一种是无限长单位冲激响应数字滤波器(IIR,Infinite Impulse Response)，另一种是有限长单位冲激响应数字滤波器(FIR,Finite Impulse Response)，具体本文中的数字滤波器的类型均不作限定性说明，只要可以实现自适应滤波的功能即可。
本发明实施例中，通过辅助反馈模块301从模拟信号中提取包络信号和谐波信号，对包络信号和谐波信号进行采样处理，得到包络反馈信号和谐波反馈信号，实现实时反馈包络信号和谐波信号，辅助模型系数训练模块302利用包络反馈信号、谐波反馈信号及预失真信号对辅助系数进行训练，并将得到的第三辅助系数传递给预失真处理模块305和预失真模型系数训练模块304，预失真处理模块利用第三辅助系数和预失真模型系数训练模块304训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，对中频信号进行预 失真处理，得到预失真信号，使得输入预失真处理装置的中频信号经过预失真处理后，所产生的非线性量与功率放大器产生的非线性量相互抵消，提升宽带情况下数字预失真技术的性能。
请参阅图4，本发明实施以辅助反馈模块反馈谐波信号为例，本发明实施例中一种预失真处理的装置另一实施例包括：
辅助反馈模块401，用于接收放大后的模拟信号，从放大后的模拟信号中提取非线性失真信号，获取与非线性失真信号对应的反馈信号，并将反馈信号输入辅助模型系数训练模块402；
辅助模型系数训练模块402，用于接收反馈信号和自预失真处理模块405输出的预失真信号，根据反馈信号、自预失真处理模块405输出的预失真信号对辅助模型系数训练模块402的辅助系数进行训练，并将训练得到的第一辅助系数传递给预失真处理模块405；
射频信号反馈模块403，用于接收放大后的模拟信号，并从放大后的模拟信号中提取基波信号，获取与基波信号对应的基波反馈信号后，将基波反馈信号输入预失真模型系数训练模块404；
预失真模型系数训练模块404，用于接收基波反馈信号和自预失真处理模块输出的预失真信号，根据自预失真处理模块405输出的预失真信号、自射频信号反馈模块输出的基波反馈信号，对预失真模型系数训练模块404的预失真系数进行训练，并将训练得到的预失真系数传递给预失真处理模块405；
预失真处理模块405，用于根据第一辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
可选的，在上述图4对应的实施例的基础上，非线性失真信号可以包括谐波信号，
辅助反馈模块401具体用于从放大后的模拟信号中提取谐波信号，对放大后的谐波信号进行采样处理，得到与谐波信号对应的谐波反馈信号，并将谐波反馈信号输入辅助模型系数训练模块402；
本发明实施例中，预失真处理模块405包括第五辅助滤波子模块4051和第三预失真处理子模块4052，预失真模型系数训练模块404包括第六辅助滤 波子模块4041和第三预失真系数处理子模块4042，
辅助模型系数训练模块402具体用于利用输入辅助模型系数训练模块402的预失真信号、谐波反馈信号进行建模，对辅助模型系数训练模块402的辅助系数进行训练，收敛后得到第四辅助系数，并将第四辅助系数传递给第五辅助滤波子模块4051和第六辅助滤波子模块4041；
第五辅助滤波子模块4051还用于根据输入预失真处理模块404的中频信号构造与中频信号对应的谐波参考信号，并利用第四辅助系数对谐波参考信号进行滤波处理，并将得到的第五输出信号输入预失真处理子模块4052；
第三预失真处理子模块4052还用于根据预设预失真模型，利用第五辅助滤波子模块4051输出的第五输出信号和输入预失真处理模块405的削波处理后的基带信号生成预失真向量，将预失真向量和预失真系数相乘，得到预失真信号并输出。
第六辅助滤波子模块4041，用于对自射频反馈信号模块403输入的基波反馈信号，利用第四辅助系数进行滤波处理，并将得到的第六输出信号输入预失真模型处理子模块4032；
第三预失真系数处理子模块4042，用于根据第六输出信号、输入第三预失真系数处理子模块4042的基波反馈信号、预失真信号进行非线性建模，并对预失真模型系数训练模块404的预失真系数进行训练，得到预失真系数。
可选的，装置还包括：
预处理模块406，用于将输入的基带信号进行数字上变频，并将同一频段的所有载波合并后进行削波处理，并将削波处理后的基带信号输入预失真处理模块405，基带信号包括多个频段的载波。
信号合并和传输模块407，用于接收自预失真处理模块输出的预失真信号，并将自预失真处理模块输出的预失真信号转化为模拟信号后输出至功率放大模块408；
功率放大模块408，用于对输入功率放大模块406的模拟信号进行放大后，输出至辅助反馈模块401及射频信号反馈模块407。
本发明实施例中，通过辅助反馈模块401从模拟信号中提取谐波信号，利用谐波信号得到谐波反馈信号，实现实时反馈谐波信号，辅助模型系数训练模块402利用包络反馈信号及预失真信号对辅助系数进行训练，并将得到的第四 辅助系数传递给第五辅助滤波子模块4051和第六辅助滤波子模块4041，预失真处理模块405利用第四辅助系数和第三预失真系数处理子模块4042训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，对中频信号进行预失真处理，得到预失真信号，使得输入预失真处理装置的中频信号经过预失真处理后，所产生的非线性量与功率放大器产生的非线性量相互抵消，提升宽带情况下数字预失真技术的性能。
为便于理解，下面以频带1、频带2为预处理电路的输出信号即中频信号(信号X1、信号X2；X1对应频带1，X2对应频带2)，频带1和频带2均为包括多个频段的载波的基带信号，以及辅助反馈模块同时反馈包络信号及谐波信号为例对本发明实施例中一种预失真处理的装置内部运作流程进行详细的描述。
需要说明的是，包络信号的频率范围为f2-f1-40MHz≤fe≤f2-f1+40MHz，频带1和频带2对应的二次谐波的反馈信号频率范围为2f1-40MHz≤fh1≤2f1+40MHz和2f2-40MHz≤fh2≤2f2+40MHz，请参阅图4和图5，本发明实施例中一种预失真处理的装置一具体应用场景如下：
预处理模块406将输入的频带1、频带2进行数字上变频，并将同一频段的所有载波合并后进行削波处理，得到信号X1、信号X2，并将信号X1、信号X2输入预失真处理模块405。
信号合并和传输模块407将预失真处理模块405输出的预失真信号y1、y2转化为模拟信号后输出至功率放大模块408；
其中，信号合并和传输模块407将y1、y2转化为模拟信号具体实现方式如下：
将预失真信号y1、y2中每个频带的信号分别进行数字上变频、合并和转化为模拟信号。
功率放大模块408对输入功率放大模块403的模拟信号进行放大后，输出至辅助反馈模块401及射频信号反馈模块403；
射频信号反馈模块403接收功率放大模块408输入的放大后的模拟信号，并从放大后的模拟信号中提取基波信号Z1、Z2，并对基波信号Z进行采样处 理，得到与基波信号Z1对应的基波反馈信号df1、与基波信号Z2对应的基波反馈信号df2后输入预失真模型系数训练模块407。
其中，利用基波信号Z1、Z2得到基波反馈信号df1、df2具体实现方式如下，分别对两个基波信号Z1、Z2与预失真信号y1、y2进行时延相位对齐后，得到对应的基波反馈信号df1、df2。
辅助反馈模块401从放大后的模拟信号中提取包络信号E和谐波信号F1、F2，对E、F1、F2分别进行采样处理，得到与包络信号E对应的包络反馈信号de，与F1对应的包络反馈信号df1、与F2对应的包络反馈信号dh2，并将de、dh1及dh2输入辅助模型系数训练模块407；
其中，利用E和F得到de、dh1及dh2具体实现方式如下：
分别将E、F1、F2这三个信号的数字中频反馈信号，其对应的参考信号即分别为
Figure PCTCN2014091668-appb-000001
进行时延估计、相位对齐和功率调整，分别得到三个调整后的包络反馈信号和谐波反馈信号，分别记为de、dh1及dh2。
辅助模型系数训练模块402根据de、dh1、dh1及自预失真处理模块405输出的y1、y2对辅助模型系数训练模块402的辅助系数进行训练，收敛后得到目标辅助系数，并将目标辅助系数传递给预失真处理模块405和预失真模型系数训练模块404；
其中，利用de、dh1、dh2、y1及y2对辅助模型系数训练模块402的辅助系数进行训练具体实现方式可以为：
分别对de、dh1、dh2进行建模，分别拟合三个参考信号
Figure PCTCN2014091668-appb-000002
分别记为
Figure PCTCN2014091668-appb-000003
和
Figure PCTCN2014091668-appb-000004
其中，Fi(i),i＝1,2,3为建模方式，该建模方式可以为线性建模也可以为非线性建模，取决于该模型拟合的精确度和复杂度，当该模型为线性模型时，具体的表达式如下具体表达式如下：
Figure PCTCN2014091668-appb-000005
Figure PCTCN2014091668-appb-000006
Figure PCTCN2014091668-appb-000007
其中，we、wh1、wh2分别为包络信号、频带1和频带2的二次谐波的滤波系数，即目标辅助系数，通过利用连续的N个采样点构造三个线性方程组，并采用最小二乘法或者LMS，RLS等类似的自适应算法，计算出目标辅助系数we、wh1、wh2，并将目标辅助系数传递给预失真处理模块405中的第五辅助滤波子模块4051和预失真模型系数训练模块404中的第六辅助滤波子模块4041，目标辅助系数用于辅助滤波。
需要说明的是，本实施例中的模型也可采用非线性建模，如记忆多项式模型，具体建模方式本文中均不作限定。
预失真模型系数训练模块404根据y1、y2、自射频信号反馈模块输出的基波反馈信号df1、df2，对预失真模型系数训练模块404的预失真系数进行训练，收敛后得到目标预失真系数，并将训练得到的预失真系数传递给预失真处理模块405；
其中，利用df1、df2对预失真系数进行训练具体实现方式如下：
利用基波反馈信号df1、df2构造包络信号的包络参考信号以及频带1、2对应的二次谐波信号的谐波参考信号，分别为
Figure PCTCN2014091668-appb-000008
并将这三个信号输入第六辅助滤波子模块4041，将得到的三个输出信号记为pe、pf1、pf2，且分别与
Figure PCTCN2014091668-appb-000009
对应，本文中出现第六滤波子模块之处均不再赘述。
其中，第六辅助滤波子模块4041完成功能可由下述公式描述：
Figure PCTCN2014091668-appb-000010
Figure PCTCN2014091668-appb-000011
Figure PCTCN2014091668-appb-000012
然后，预失真处理模块405中的第三预失真处理子模块4052利用第六辅助滤波子模块4041的输出信号pe、pf1、pf2和基波反馈信号df1、df2进行非线性建模，拟合预失真电路输出信号y1、y2。为了便于描述，仅考虑三阶非线性分量的情况下，具体的非线性模型如下所示：
Figure PCTCN2014091668-appb-000013
Figure PCTCN2014091668-appb-000014
其中
Figure PCTCN2014091668-appb-000015
表示第i个频带第j部分记忆深度为m的预失真系数，此实施例中，i＝1，2，3，4，5，故频带1频带2的预失真系数w1、w2表达式为：
Figure PCTCN2014091668-appb-000016
Figure PCTCN2014091668-appb-000017
利用连续的N个采样点，构造两个线性方程组，利用最小二乘法或者LMS，RLS等其他的自适应算法，求解频带1频带2的预失真系数w1、w2，并将预失真系数传递给预失真处理模块405，用于预失真处理。
预失真处理模块405根据目标辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
其中，利用目标辅助系数we、wh1、wh2和预失真系数进行非线性建模具体实现方式如下：
预失真处理子模块4052利用频带1和频带2的数字中频信号x1、x2构造对应的包络信号的参考信号和二次谐波信号的参考信号，分别如下
Figure PCTCN2014091668-appb-000018
并将这三个信号输入第五辅助滤波子模块4051，分别得到三个输出信号记为qe、qf1、qf2，第五辅助滤波子模块4051完成功能可由下述公式描述：
Figure PCTCN2014091668-appb-000019
Figure PCTCN2014091668-appb-000020
Figure PCTCN2014091668-appb-000021
第三预失真处理子模块4052利用第五辅助滤波子模块4051输出的qe、qf1、qf2和频带1频带2信号x1、x2进行非线性建模，并与预失真系数w1、w2对应相乘，并求和得到预失真处理模块405的输出信号y1、y2，其中，预失真处理模块405实现功能可由下述两式表示：
Figure PCTCN2014091668-appb-000022
Figure PCTCN2014091668-appb-000023
将预失真信号y1、y2输入信号合并和传输模块407，以使信号合并和传输模块407将y1、y2转化为模拟信号后输出至功率放大模块408。
收敛判断：
可以通过检测基波反馈信号df1、df2的ACLR(Adjacent Channel Leakage Ratio，相邻频道泄漏比)性能；
也可以通过检测基波反馈信号df1、df2与频带1频带2信号x1、x2的NMSE(Normalized Mean Squared Error，归一化均方误差，用于表征计算的收敛速度和计算精度，其表达式为：
Figure PCTCN2014091668-appb-000024
判断DPD是否收敛，如果收敛，则结束流程；如果不收敛，重复辅助模型 系数训练模块404对辅助系数进行训练、预失真模型系数训练模块404对预失真系数进行训练及预失真处理模块405根据建模的步骤，直到收敛。
需要说明的是，本文中的预失真处理模块具体可以是预失真电路，第五辅助滤波子模块和第六辅助滤波子模块具体可以是辅助滤波电路，功率放大模块可以是功率放大器，辅助反馈模块具体可以是辅助反馈电路，预失真模型系数训练模块可以是预失真模型系数学习电路，辅助模型系数训练模块可以是辅助模型系数学习电路，信号合并和传输模块可以是信号合并和传输电路，射频信号反馈模块可以是射频信号反馈电路，具体在实际应用时，根据需要选择相关电路来实现本发明实施例方案的目的，具体可以参考图5所示的信号传输流程图。
本发明实施例中，通过辅助反馈模块401从模拟信号中提取包络信号和谐波信号，对包络信号和谐波信号进行采样处理，得到包络反馈信号和谐波反馈信号，实现实时反馈包络信号和谐波信号，辅助模型系数训练模块402利用包络反馈信号、谐波反馈信号及预失真信号对辅助系数进行训练，并将得到的目标辅助系数传递给预失真处理模块405和预失真模型系数训练模块404，预失真处理模块405利用目标辅助系数和预失真模型系数训练模块403训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，对中频信号进行预失真处理，得到预失真信号，使得输入预失真处理装置的中频信号经过预失真处理后，所产生的非线性量与功率放大模块408产生的非线性量相互抵消，提升宽带情况下数字预失真技术的性能。
上面对本发明实施例中一种预失真处理的装置进行了详细的描述，下面对本发明实施例中一种预失真处理的方法进行说明，请参阅图7，本发明实施例中一种预失真处理的方法一实施例包括：
501、辅助反馈模块接收放大后的模拟信号，并从模拟信号中提取非线性失真信号，获取与非线性失真信号对应的反馈信号，并将反馈信号输入辅助模 型系数训练模块；
其中，从模拟信号中提取非线性失真信号后，对该非线性失真信号进行采样处理可得到与非线性失真信号对应的反馈信号，采样处理具体实现方式可以是：
首先将非线性失真信号转化为数字信号，然后可以采用ADC(Digital to Analog conversion，模数转换)采样方法对该数字信号进行数字下变频和时延相位对齐，具体可以使用下变频器对该数字信号进行下变频，使用移相器对该数字信号进行时延相位对齐，从而可以得到至少一个频带对应的反馈信号，具体采样处理的方式本文中均不作限定，该非线性失真信号可以是单频带、双频带，也可以包含不同频带等，具体成分本文中均不作限定。
502、辅助模型系数训练模块接收反馈信号和自预失真处理模块输出的预失真信号，根据反馈信号、自预失真处理模块输入的预失真信号对辅助模型系数训练模块的辅助系数进行训练，并将训练得到的第一辅助系数传递给预失真处理模块；
503、射频信号反馈模块接收放大后的模拟信号，并从放大后的模拟信号中提取基波信号，获取与基波信号对应的基波反馈信号后，将基波反馈信号输入预失真模型系数训练模块；
504、预失真模型系数训练模块接收基波反馈信号和自预失真处理模块输出的预失真信号，根据基波反馈信号、自射频信号反馈模块输入的基波反馈信号，对预失真模型系数训练模块的预失真系数进行训练，并将训练得到的预失真系数传递给预失真处理模块；
505、预失真处理模块根据第一辅助系数和预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
需要说明的是，本文中，辅助模型系数训练模块将辅助系数传递给预失真处理模块和预失真模型系数训练模块、以及预失真模型系数训练模块将预失真系数传递给预失真处理模块的传递方式可以这样实现：
在FPGA(Field－Programmable Gate Array，现场可编程门阵列)中实现时，将在电路A对应存储系数的寄存器中的数据，复制到电路B对应存储系数的寄存器，完成的仅仅是数据的传递，当然还可以通过改变电路B的读取寄存器的地址进行改变，同样也是完成系数的复制，具体实现方式本文中均不 作限定。
本发明实施例中，通过辅助反馈模块从模拟信号中提取非线性失真信号，对非线性失真信号进行采样处理，得到与非线性失真信号对应的反馈信号，实现实时反馈非线性失真信号，辅助模型系数训练模块利用辅助反馈模块输出的反馈信号及预失真处理模块输出的预失真信号对辅助系数进行训练，并将得到的第一辅助系数传递给预失真处理模块和预失真模型系数训练模块，预失真处理模块利用第一辅助系数和预失真模型系数训练模块训练得到的预失真系数进行非线性建模，有效提高非线性建模的准确度，对中频信号进行预失真处理，得到预失真信号，使得输入预失真处理装置的中频信号经过预失真处理后，所产生的非线性量与功率放大器产生的非线性量相互抵消，线性放大，提升宽带情况下数字预失真技术的性能。
可选地，在上述图5对应的实施例的基础上，本发明实施例提供的预失真处理的方法的第一个可选实施例中，非线性失真信号包括包络信号，预失真处理模块包括第一辅助滤波子模块和第一预失真处理子模块，预失真模型系数训练模块包括第二辅助滤波子模块和第一预失真系数处理子模块；
辅助反馈模块从模拟信号中提取包络信号，并对包络信号进行采样处理，得到与包络信号对应的反馈信号，并将包络反馈信号输入辅助模型系数训练模块；
辅助模型系数训练模块利用输入辅助模型系数训练模块的预失真信号、包络反馈信号进行建模，对辅助系数进行训练，收敛后得到第二辅助系数，并将第二辅助系数传递给第一辅助滤波子模块和第二辅助滤波子模块；
其中，第一辅助滤波子模块根据输入预失真处理模块的中频信号构造与中频信号对应的的包络参考信号，利用第二辅助系数对包络参考信号进行滤波处理，并将得到的第一输出信号输入第一预失真处理子模块；
第一预失真处理子模块根据预设预失真模型，利用第一辅助滤波子模块输出的第一输出信号和输入第一预失真处理子模块的削波处理后的基带信号生成预失真向量，将预失真向量和预失真系数相乘，得到预失真信号并输出；
第二辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用第二辅助系数进行滤波处理，并将得到的第二输出信号输入预失真模型处理子模块；
第一预失真系数处理子模块根据第二输出信号、输入第一预失真系数处理子模块的基波反馈信号、预失真信号进行非线性建模，并对预失真模型系数训练模块的预失真系数进行训练，得到预失真系数。
可选地，在上述图5对应的实施例的基础上，本发明实施例提供的预失真处理的方法的第二个可选实施例中，
非线性失真信号包括包络信号和谐波信号，预失真处理模块包括第三辅助滤波子模块和第二预失真处理子模块，预失真模型系数训练模块包括第四辅助滤波子模块和第二预失真系数处理子模块；
辅助反馈模块从模拟信号中提取包络信号和谐波信号，并对包络信号和谐波信号进行采样处理，得到与包络信号对应的包络反馈信号，以及与谐波信号对应的谐波反馈信号，并将包络反馈信号和谐波反馈信号输入辅助模型系数训练模块；
辅助模型系数训练模块利用输入辅助模型系数训练模块的预失真信号、包络反馈信号和谐波反馈信号对辅助系数进行训练，收敛后得到第三辅助系数，并将第三辅助系数传递给第三辅助滤波子模块和第四辅助滤波子模块；
其中，第三辅助滤波子模块根据输入预失真处理模块的中频信号构造与中频信号对应的包络参考信号和谐波参考信号，利用第二辅助系数对包络参考信号和谐波参考信号进行滤波处理，并将得到的第三输出信号输入第二预失真处理子模块；
第二预失真处理子模块根据预设预失真模型，利用第三辅助滤波子模块输出的第三输出信号和输入第二预失真处理子模块的削波处理后的基带信号生成预失真向量，将预失真向量和预失真系数相乘，得到预失真信号并输出；
第四辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用第二辅助系数进行滤波处理，并将得到的第四输出信号输入第二预失真系数处理子模块；
第二预失真系数处理子模块根据第四输出信号、输入第二预失真系数处理子模块的基波反馈信号、预失真信号进行非线性建模，并对预失真模型系数训练模块的预失真系数进行训练，得到预失真系数。
可选地，在上述图5对应的实施例的基础上，本发明实施例提供的预失真处理的方法的第三个可选实施例中，
非线性失真信号为谐波信号，预失真处理模块包括第五辅助滤波子模块和第三预失真处理子模块，预失真模型系数训练模块包括第六辅助滤波子模块和第三预失真系数处理子模块；
辅助反馈模块从模拟信号中提取谐波信号，并对谐波信号进行采样处理，得到与谐波信号对应的谐波反馈信号，并将谐波反馈信号输入辅助模型系数训练模块；
辅助模型系数训练模块利用输入辅助模型系数训练模块的预失真信号和谐波反馈信号对辅助系数进行训练，收敛后得到第四辅助系数，并将第四辅助系数传递给第五辅助滤波子模块和第六辅助滤波子模块；
其中，第五辅助滤波子模块根据输入预失真处理模块的中频信号构造与中频信号对应的谐波参考信号，利用第四辅助系数对谐波参考信号进行滤波处理，并将得到的第五输出信号输入第三预失真处理子模块；
第三预失真处理子模块根据预设预失真模型，利用第五辅助滤波子模块输出的第五输出信号和输入第三预失真处理子模块的削波处理后的基带信号生成预失真向量，将预失真向量和预失真系数相乘，得到预失真信号并输出；
第六辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用第二辅助系数进行滤波处理，并将得到的第六输出信号输入预失真模型处理子模块；
第三预失真系数处理子模块根据第六输出信号、输入第三预失真系数处理子模块的基波反馈信号、预失真信号进行非线性建模，并对预失真模型系数训练模块的预失真系数进行训练，得到预失真系数。
可选地，在上述预失真处理的方法图5对应的实施例或任一可选实施例的基础上，本发明实施例提供的预失真处理的方法的第四个可选实施例中，
预处理模块将输入预处理模块的基带信号进行数字上变频，将同一频段的所有载波合并后进行削波处理，并将削波处理后的基带信号输入预失真处理模块，基带信号包括多个频段的载波；
信号合并和传输模块接收自预失真处理模块输出的预失真信号，并将自预失真处理模块输出的预失真信号转化为模拟信号后输出至功率放大模块；
功率放大模块对输入功率放大模块的模拟信号进行放大，并将放大后的模拟信号输出至辅助反馈模块及射频信号反馈模块；
本发明实施例中，可以通过辅助反馈模块反馈包络信号或谐波信号中的至少一个，并将包络信号转换为包络反馈信号或将谐波信号转换为谐波反馈信号，辅助模型系数训练模块利用包络反馈信号或谐波反馈信号进行非线性建模，将训练得到的辅助系数传递给预失真模型系数训练模块和预失真处理模块，以使预失真模型系数训练模块利用辅助系数和基波反馈信号建模并将训练的预失真系数传递给预失真处理模块，提高预失真建模的准确度，使得预失真处理模块利用预失真系数对将中频信号进行预失真处理，使中频信号所产生的非线性量与功率放大器产生的非线性量相互抵消，从而提升宽带情况下数字预失真技术的性能。
需要说明的是，本文中的预失真模型系数训练模块所训练出来的预失真系数只要在第一预设阈值范围内，预失真处理模块就可以一直使用该预失真系数，可以不用实时的更新预失真系数，并且辅助模型系数训练模块对辅助系数可以进行定期的训练，也可以不定期的进行训练，该辅助系数只要在第二预设阈值范围内，就可以不用更新，也可以更新，具体预失真模型系数训练模块训练预失真系数、以及辅助模型系数训练模块训练辅助系数的周期根据电路设计以及客户需求来定，具体实现方式本文中均不作限定。
本文中应用了具体实施例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上，本说明书内容不应理解为对本发明的限制。在本发明实施例中使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本发明。在本发明实施例和所附权利要求书中所使用的单数形式的“一种”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。还应当理解，本文中使用的术语“a或b中至少一个”是指并包含只有a、只有b、或者既有a又有b中任意一种可能的情况。
本文中应用了具体实施例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上，本说明书内容不应理解为对本发明的限制。在本发明实施例中使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本发明。在本发 明实施例和所附权利要求书中所使用的单数形式的“一种”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。还应当理解，本文中使用的术语“a或b中至少一个”是指并包含只有a、只有b、或者既有a又有b中任意一种可能的情况。
以上，以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (22)

  1. 一种预失真处理的装置，其特征在于，所述装置包括：
    辅助反馈模块，用于接收放大后的模拟信号，并从所述放大后的模拟信号中提取非线性失真信号，获取与所述非线性失真信号对应的反馈信号，并将所述反馈信号输入辅助模型系数训练模块；
    所述辅助模型系数训练模块，用于接收所述反馈信号和自预失真处理模块输出的预失真信号，根据所述反馈信号、自预失真处理模块输出的预失真信号对所述辅助模型系数训练模块的辅助系数进行训练，将训练得到的第一辅助系数传递给所述预失真处理模块；
    射频信号反馈模块，用于接收放大后的模拟信号，并从所述放大后的模拟信号中提取基波信号，获取与所述基波信号对应的基波反馈信号后，将所述基波反馈信号输入所述预失真模型系数训练模块；
    所述预失真模型系数训练模块，用于接收基波反馈信号和自所述预失真处理模块输出的预失真信号，根据所述基波反馈信号、自所述预失真处理模块输出的预失真信号，对所述预失真模型系数训练模块的预失真系数进行训练，并将训练得到的预失真系数传递给所述预失真处理模块；
    所述预失真处理模块，用于接收中频信号，并根据所述第一辅助系数和所述预失真系数进行非线性建模，对所述中频信号进行预失真处理，并将得到的预失真信号输出。
  2. 根据权利要求1所述的装置，其特征在于，所述非线性失真信号包括包络信号，所述辅助反馈模块具体用于接收放大后的模拟信号，从放大后的模拟信号中提取包络信号，对所述包络信号进行采样处理，得到与所述包络信号对应的包络反馈信号，并将所述包络反馈信号输入辅助模型系数训练模块。
  3. 根据权利要求1或2所述的装置，其特征在于，所述预失真处理模块包括第一辅助滤波子模块和第一预失真处理子模块，所述预失真模型系数训练模块包括第二辅助滤波子模块和第一预失真系数处理子模块，
    所述辅助模型系数训练模块具体用于利用输入所述辅助模型系数训练模块的预失真信号、包络反馈信号进行建模，对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第二辅助系数，并将所述第二辅助系数传递给所述第一辅助滤波子模块和所述第二辅助滤波子模块。
  4. 根据权利要求3所述的装置，其特征在于，所述第一辅助滤波子模块用于根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号，并利用所述第二辅助系数对所述包络参考信号进行滤波处理，并将得到的第一输出信号输入所述第一预失真处理子模块；
    所述第一预失真处理子模块用于根据预设预失真模型，利用所述第一辅助滤波子模块输出的第一输出信号和输入所述预失真处理模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
    所述第二辅助滤波子模块用于对自射频反馈信号模块输入的基波反馈信号，利用所述第二辅助系数进行滤波处理，并将得到的第二输出信号输入所述第一预失真系数处理子模块；
    所述第一预失真系数处理子模块用于根据所述第二输出信号、输入所述第一预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
  5. 根据权利要求1所述的装置，其特征在于，所述非线性失真信号包括包络信号和谐波信号，所述辅助反馈模块具体用于接收放大后的模拟信号，从放大后的模拟信号中提取包络信号和谐波信号，对所述包络信号和所述谐波信号进行采样处理，得到与所述包络信号对应的包络反馈信号，及与所述谐波信号对应的谐波反馈信号，并将所述包络反馈信号、所述谐波反馈信号输入所述辅助模型系数训练模块。
  6. 根据权利要求1或5所述的装置，其特征在于，所述预失真处理模块包括第三辅助滤波子模块和第二预失真处理子模块，所述预失真模型系数训练模块包括第四辅助滤波子模块和第二预失真系数处理子模块，
    所述辅助模型系数训练模块具体用于利用输入所述辅助模型系数训练模块的预失真信号、包络反馈信号和谐波反馈信号进行建模，对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第三辅助系数，并将所述第三辅助系数传递给所述第三辅助滤波子模块和所述第四辅助滤波子模块。
  7. 根据权利要求6所述的装置，其特征在于，所述第三辅助滤波子模块用于根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包 络参考信号和谐波参考信号，并利用所述第三辅助系数对所述包络参考信号和所述谐波参考信号进行滤波处理，并将得到的第三输出信号输入所述第二预失真处理子模块；
    所述第二预失真处理子模块用于根据预设预失真模型，利用所述第一辅助滤波子模块输出的第三输出信号和输入所述预失真处理模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
    所述第四辅助滤波子模块用于对自射频反馈信号模块输入的基波反馈信号，利用所述第三辅助系数进行滤波处理，并将得到的第四输出信号输入所述第二预失真系数处理子模块；
    所述第二预失真系数处理子模块用于根据所述第四输出信号、输入所述第二预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
  8. 根据权利要求1所述的装置，其特征在于，所述非线性失真信号包括谐波信号，所述辅助反馈模块具体用于从放大后的模拟信号中提取谐波信号，并对所述谐波信号进行采样处理，得到与所述谐波信号对应的谐波反馈信号，并将所述谐波反馈信号输入辅助模型系数训练模块。
  9. 根据权利要求1或8所述的装置，其特征在于，所述预失真处理模块包括第五辅助滤波子模块和第三预失真处理子模块，所述预失真模型系数训练模块包括第六辅助滤波子模块和第三预失真系数处理子模块，
    所述辅助模型系数训练模块具体用于利用输入所述辅助模型系数训练模块的预失真信号、谐波反馈信号进行建模，对所述辅助模型系数训练模块的辅助系数进行训练，收敛后得到第四辅助系数，并将所述第四辅助系数传递给所述第五辅助滤波子模块和所述第六辅助滤波子模块。
  10. 根据权利要求9所述的装置，其特征在于，第五辅助滤波子模块用于根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的谐波参考信号，并对所述谐波参考信号进行滤波处理，并将利用所述第四辅助系数得到的第五输出信号输入所述第三预失真处理子模块；
    所述第三预失真处理子模块用于根据预设预失真模型，利用所述第五辅助 滤波子模块输出的第五输出信号和输入所述预失真处理模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
    所述第六辅助滤波子模块用于对自射频反馈信号模块输入的基波反馈信号，利用所述第四辅助系数进行滤波处理，并将得到的第六输出信号输入预失真模型处理子模块；
    所述第三预失真系数处理子模块用于根据所述第六输出信号、输入所述第三预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
  11. 根据权利要求1至10任一所述的装置，其特征在于，所述装置还包括：
    预处理模块，用于将输入所述预处理模块的基带信号进行数字上变频，并将同一频段的所有载波合并后进行削波处理，并将削波处理后的基带信号输入所述预失真处理模块，所述基带信号包括多个频段的载波；
    信号合并和传输模块，用于接收自所述预失真处理模块输出的预失真信号，并将自所述预失真处理模块输出的预失真信号转化为模拟信号后输出至功率放大模块；
    所述功率放大模块，用于对输入所述功率放大模块的模拟信号进行放大，并将放大后的模拟信号输出至所述辅助反馈模块及所述射频信号反馈模块。
  12. 一种预失真处理的方法，其特征在于，所述方法包括：
    辅助反馈模块接收放大后的模拟信号，并从放大后的模拟信号中提取非线性失真信号，获取与所述非线性失真信号对应的反馈信号，并将所述反馈信号输入辅助模型系数训练模块；
    所述辅助模型系数训练模块接收所述反馈信号和自预失真处理模块输出的预失真信号，根据所述反馈信号、自预失真处理模块输入的预失真信号对所述辅助模型系数训练模块的辅助系数进行训练，并将训练得到的第一辅助系数传递给所述预失真处理模块；
    射频信号反馈模块接收放大后的模拟信号，并从所述放大后的模拟信号中提取基波信号，获取与所述基波信号对应的基波反馈信号后，将所述基波反馈 信号输入所述预失真模型系数训练模块；
    所述预失真模型系数训练模块接收基波反馈信号和自所述预失真处理模块输出的预失真信号，根据所述基波反馈信号、自所述预失真处理模块输出的预失真信号，对所述预失真模型系数训练模块的预失真系数进行训练，并将训练得到的预失真系数传递给所述预失真处理模块；
    所述预失真处理模块根据所述第一辅助系数和所述预失真系数进行非线性建模，并对输入的中频信号进行预失真处理，并将预失真处理得到的预失真信号输出。
  13. 根据权利要求12所述的方法，其特征在于，所述非线性失真信号包括包络信号，所述辅助反馈模块接收放大后的模拟信号，从所述模拟信号中提取非线性失真信号，并对所述非线性失真信号进行采样处理，得到与所述非线性失真信号对应的反馈信号，并将所述反馈信号输入辅助模型系数训练模块具体包括：
    所述辅助反馈模块接收放大后的模拟信号，从所述模拟信号中提取所述包络信号，并对所述包络信号进行采样处理，得到与所述包络信号对应的反馈信号，并将所述包络反馈信号输入辅助模型系数训练模块。
  14. 根据权利要求13所述的方法，其特征在于，所述预失真处理模块包括第一辅助滤波子模块和第一预失真处理子模块，所述预失真模型系数训练模块包括第二辅助滤波子模块和第一预失真系数处理子模块，所述方法还包括：
    所述辅助模型系数训练模块利用输入所述辅助模型系数训练模块的预失真信号、包络反馈信号进行建模，对所述辅助系数进行训练，收敛后得到第二辅助系数，并将所述第二辅助系数传递给所述第一辅助滤波子模块和所述第二辅助滤波子模块。
  15. 根据权利要求14所述的方法，其特征在于，所述方法还包括：
    所述第一辅助滤波子模块根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号，对所述包络参考信号进行滤波处理，并将利用所述第二辅助系数得到的第一输出信号输入所述第一预失真处理子模块；
    所述第一预失真处理子模块根据预设预失真模型，利用所述第一辅助滤波子模块输出的第一输出信号和输入所述第一预失真处理子模块的削波处理后 的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
    所述第二辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用所述第二辅助系数进行滤波处理，并将得到的第二输出信号输入预失真模型处理子模块；
    所述第一预失真系数处理子模块根据所述第二输出信号、输入所述第一预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
  16. 根据权利要求12所述的方法，其特征在于，所述非线性失真信号包括包络信号和谐波信号，所述方法还包括：
    所述辅助反馈模块从所述模拟信号中提取包络信号和谐波信号，并对所述包络信号和谐波信号进行采样处理，得到与所述包络信号对应的包络反馈信号，以及与所述谐波信号对应的谐波反馈信号，并将所述包络反馈信号和所述谐波反馈信号输入所述辅助模型系数训练模块。
  17. 根据权利要求12或16所述的方法，其特征在于，所述预失真处理模块包括第三辅助滤波子模块和第二预失真处理子模块，所述预失真模型系数训练模块包括第四辅助滤波子模块和第二预失真系数处理子模块，所述方法还包括：
    所述辅助模型系数训练模块利用输入所述辅助模型系数训练模块的预失真信号、包络反馈信号和谐波反馈信号对所述辅助系数进行训练，收敛后得到第三辅助系数，并将所述第三辅助系数传递给所述第三辅助滤波子模块和所述第四辅助滤波子模块。
  18. 根据权利要求17所述的方法，其特征在于，所述方法还包括：所述第三辅助滤波子模块根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的包络参考信号和谐波参考信号，对所述包络参考信号和所述谐波参考信号进行滤波处理，并将利用所述第三辅助系数得到的第三输出信号输入所述第二预失真处理子模块；
    所述第二预失真处理子模块根据预设预失真模型，利用所述第三辅助滤波子模块输出的第三输出信号和输入所述第二预失真处理子模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到 所述预失真信号并输出；
    所述第四辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利用所述第三辅助系数进行滤波处理，并将得到的第四输出信号输入所述第二预失真系数处理子模块；
    所述第二预失真系数处理子模块根据所述第四输出信号、输入所述第二预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
  19. 根据权利要求12所述的方法，其特征在于，所述非线性失真信号包括谐波信号，所述方法还包括：
    所述辅助反馈模块从所述模拟信号中提取谐波信号，并对谐波信号进行采样处理，得到与所述谐波信号对应的谐波反馈信号，并将所述谐波反馈信号输入所述辅助模型系数训练模块。
  20. 根据权利要求12或19所述的方法，其特征在于，所述预失真处理模块包括第五辅助滤波子模块和第三预失真处理子模块，所述预失真模型系数训练模块包括第六辅助滤波子模块和第三预失真系数处理子模块，所述方法还包括：
    所述辅助模型系数训练模块利用输入所述辅助模型系数训练模块的预失真信号和谐波反馈信号对所述辅助系数进行训练，收敛后得到第四辅助系数，并将所述第四辅助系数传递给所述第五辅助滤波子模块和所述第六辅助滤波子模块。
  21. 根据权利要求20所述的方法，其特征在于，所述方法还包括：
    所述第一辅助滤波子模块根据输入所述预失真处理模块的中频信号构造与所述中频信号对应的谐波参考信号，对所述谐波参考信号进行滤波处理，并将利用所述第四辅助系数得到的第五输出信号输入所述第三预失真处理子模块；
    所述第三预失真处理子模块根据预设预失真模型，利用所述第一辅助滤波子模块输出的第一输出信号和输入所述第三预失真处理子模块的削波处理后的基带信号生成预失真向量，将所述预失真向量和所述预失真系数相乘，得到所述预失真信号并输出；
    所述第六辅助滤波子模块对自射频反馈信号模块输入的基波反馈信号，利 用所述第四辅助系数进行滤波处理，并将得到的第六输出信号输入预失真模型处理子模块；
    所述第三预失真系数处理子模块根据所述第六输出信号、输入所述第三预失真系数处理子模块的基波反馈信号、所述预失真信号进行非线性建模，并对所述预失真模型系数训练模块的预失真系数进行训练，得到所述预失真系数。
  22. 根据权利要求12至21任一所述的方法，其特征在于，所述方法还包括：
    预处理模块将输入所述预处理模块的基带信号进行数字上变频，将同一频段的所有载波合并后进行削波处理，并将削波处理后的基带信号输入所述预失真处理模块，所述基带信号包括多个频段的载波；
    所述信号合并和传输模块接收自所述预失真处理模块输出的预失真信号，并将自所述预失真处理模块输出的预失真信号转化为模拟信号后输出至功率放大模块；
    所述功率放大模块对输入所述功率放大模块的模拟信号进行放大，并将放大后的模拟信号输出至所述辅助反馈模块及所述射频信号反馈模块。
PCT/CN2014/091668 2014-11-19 2014-11-19 一种预失真处理的装置及方法 WO2016078038A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP14906288.7A EP3208938B1 (en) 2014-11-19 2014-11-19 Pre-distortion processing device and method
CN201480082743.6A CN107078702B (zh) 2014-11-19 2014-11-19 一种预失真处理的装置及方法
PCT/CN2014/091668 WO2016078038A1 (zh) 2014-11-19 2014-11-19 一种预失真处理的装置及方法
US15/599,956 US10075324B2 (en) 2014-11-19 2017-05-19 Predistortion processing apparatus and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2014/091668 WO2016078038A1 (zh) 2014-11-19 2014-11-19 一种预失真处理的装置及方法

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/599,956 Continuation US10075324B2 (en) 2014-11-19 2017-05-19 Predistortion processing apparatus and method

Publications (1)

Publication Number Publication Date
WO2016078038A1 true WO2016078038A1 (zh) 2016-05-26

Family

ID=56013072

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/091668 WO2016078038A1 (zh) 2014-11-19 2014-11-19 一种预失真处理的装置及方法

Country Status (4)

Country Link
US (1) US10075324B2 (zh)
EP (1) EP3208938B1 (zh)
CN (1) CN107078702B (zh)
WO (1) WO2016078038A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108365862A (zh) * 2018-01-22 2018-08-03 中国科学院微电子研究所 一种消除射频电路谐波的方法及射频电路

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10447211B2 (en) * 2015-06-17 2019-10-15 Telefonaktiebolaget Lm Ericsson (Publ) Least mean squares adaptation of a concurrent multi-band pre-distorter using overlapping spines
US10581469B1 (en) * 2017-04-17 2020-03-03 DeepSig Inc. Machine learning-based nonlinear pre-distortion system
CN108988794A (zh) * 2018-09-28 2018-12-11 京信通信***（中国）有限公司 一种模拟预失真电路及模拟预失真线性化对消方法
CN113852576A (zh) * 2020-06-28 2021-12-28 中兴通讯股份有限公司 信号失真预校正方法、装置及非易失性存储介质
CN112859611B (zh) * 2021-01-19 2023-05-16 重庆邮电大学 一种自适应预失真***及方法
CN115529049A (zh) * 2021-06-25 2022-12-27 广州海格通信集团股份有限公司 基于闭环的短波通信方法、装置、设备及可读存储介质
CN116938281A (zh) * 2022-04-02 2023-10-24 华为技术有限公司 通信方法以及相关装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1689295A (zh) * 2002-10-31 2005-10-26 中兴通讯股份有限公司 一种宽带预失真线性化的方法与***
CN101156159A (zh) * 2004-03-03 2008-04-02 电力波技术公司 用于高效率发射机的数字预失真***和方法
JP2011103540A (ja) * 2009-11-10 2011-05-26 Hitachi Kokusai Electric Inc 電力増幅器
CN102413083A (zh) * 2010-09-26 2012-04-11 电信科学技术研究院 一种信号处理方法及装置
CN102487367A (zh) * 2010-12-02 2012-06-06 中国科学院微电子研究所 一种自适应的功放数字基带预失真方法
CN103581082A (zh) * 2012-07-31 2014-02-12 富士通株式会社 基带预失真的系数更新装置及方法、预失真设备及发射机

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5892397A (en) * 1996-03-29 1999-04-06 Spectrian Adaptive compensation of RF amplifier distortion by injecting predistortion signal derived from respectively different functions of input signal amplitude
US5923712A (en) * 1997-05-05 1999-07-13 Glenayre Electronics, Inc. Method and apparatus for linear transmission by direct inverse modeling
US20020181611A1 (en) * 2001-06-01 2002-12-05 Lg Electronics Inc. Analog quadrature modulator (AQM) error compensating apparatus and method
JP4015455B2 (ja) * 2002-03-29 2007-11-28 富士通株式会社 歪補償装置
US8811917B2 (en) * 2002-05-01 2014-08-19 Dali Systems Co. Ltd. Digital hybrid mode power amplifier system
KR100472070B1 (ko) * 2002-10-16 2005-03-10 한국전자통신연구원 선형화가 가능한 적응 배열 안테나 시스템 및 그 선형화방법
US7848717B2 (en) * 2004-12-21 2010-12-07 Zte Corporation Method and system for out of band predistortion linearization
CN102143107B (zh) * 2011-02-25 2013-10-09 华为技术有限公司 一种实现数字基带预失真的方法及装置
EP3236589B1 (en) * 2012-03-09 2019-05-08 Huawei Technologies Co., Ltd. Method, apparatus, device, and system for cancelling multi-carrier transmission interference

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1689295A (zh) * 2002-10-31 2005-10-26 中兴通讯股份有限公司 一种宽带预失真线性化的方法与***
CN101156159A (zh) * 2004-03-03 2008-04-02 电力波技术公司 用于高效率发射机的数字预失真***和方法
JP2011103540A (ja) * 2009-11-10 2011-05-26 Hitachi Kokusai Electric Inc 電力増幅器
CN102413083A (zh) * 2010-09-26 2012-04-11 电信科学技术研究院 一种信号处理方法及装置
CN102487367A (zh) * 2010-12-02 2012-06-06 中国科学院微电子研究所 一种自适应的功放数字基带预失真方法
CN103581082A (zh) * 2012-07-31 2014-02-12 富士通株式会社 基带预失真的系数更新装置及方法、预失真设备及发射机

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108365862A (zh) * 2018-01-22 2018-08-03 中国科学院微电子研究所 一种消除射频电路谐波的方法及射频电路
CN108365862B (zh) * 2018-01-22 2020-10-16 中国科学院微电子研究所 一种消除射频电路谐波的方法及射频电路

Also Published As

Publication number Publication date
US10075324B2 (en) 2018-09-11
US20170257251A1 (en) 2017-09-07
EP3208938A4 (en) 2017-10-18
EP3208938A1 (en) 2017-08-23
CN107078702A (zh) 2017-08-18
EP3208938B1 (en) 2019-08-28
CN107078702B (zh) 2019-11-29

Similar Documents

Publication Publication Date Title
WO2016078038A1 (zh) 一种预失真处理的装置及方法
US8605814B2 (en) Distortion-compensated RF transmitter and method therefor
US7848451B2 (en) Digital pre-distortion technique using nonlinear filters
US8787494B2 (en) Modeling digital predistorter
US10797750B2 (en) System architecture for supporting digital pre-distortion and full duplex in cable network environments
WO2015096735A1 (zh) 一种数字预失真参数的求取方法及预失真***
US20110080216A1 (en) Systems and Methods of Power Amplifier Digital Pre-Distortion
CN102143108A (zh) 一种改进的自适应预失真技术
EP2837093B1 (en) Digital predistorter (dpd) structure based on dynamic deviation reduction (ddr)-based volterra series
EP2641324A1 (en) Joint process estimator with variable tap delay line for use in power amplifier digital predistortion
CN111064439A (zh) 一种改善短波数字预失真性能的***及方法
US8295790B2 (en) Apparatus and method for pre-distorting and amplifying a signal
Babaroglu et al. Cellular digital post-distortion: Signal processing methods and rf measurements
KR102097521B1 (ko) 고주파 증폭 장치 및 왜곡보상 방법
KR101071317B1 (ko) 전치 왜곡을 이용한 전력 증폭 장치
WO2015100523A1 (zh) 干扰抑制方法以及装置
KR101451912B1 (ko) 전력증폭기 모델을 포함하는 광대역 신호 송신장치
KR101204964B1 (ko) 로컬 오실레이터 커플링 효과와 비선형 전력 증폭기의 선형화를 위한 전치왜곡 장치 및 그 방법
Ding et al. Time-domain adaptive predistortion for nonlinear amplifiers
Long et al. Indirect learning hybrid memory predistorter based on polynomial and look-up-table
Bulusu et al. Power amplifier effects and peak-to-average power mitigation
KR20140118130A (ko) 복수의 비선형 증폭기에 대하여 단일 피드백 회로를 사용하는 전치보상 장치 및 방법
EP3288182B1 (en) System architecture for supporting digital pre-distortion and full duplex in cable network environments
CN103066927B (zh) 用于数字预矫正放大器的失真的方法及装置
Long et al. The design and implementation of baseband predistorter based on FPGA and ARM

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: 14906288

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2014906288

Country of ref document: EP