WO2015096735A1 - 一种数字预失真参数的求取方法及预失真系统 - Google Patents

一种数字预失真参数的求取方法及预失真系统 Download PDF

Info

Publication number
WO2015096735A1
WO2015096735A1 PCT/CN2014/094808 CN2014094808W WO2015096735A1 WO 2015096735 A1 WO2015096735 A1 WO 2015096735A1 CN 2014094808 W CN2014094808 W CN 2014094808W WO 2015096735 A1 WO2015096735 A1 WO 2015096735A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
predistortion
feedback signal
conjugate
feedback
Prior art date
Application number
PCT/CN2014/094808
Other languages
English (en)
French (fr)
Inventor
熊军
王新民
段滔
王静怡
Original Assignee
大唐移动通信设备有限公司
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 大唐移动通信设备有限公司 filed Critical 大唐移动通信设备有限公司
Priority to US15/108,108 priority Critical patent/US9647717B2/en
Priority to KR1020167020510A priority patent/KR101680207B1/ko
Priority to JP2016542951A priority patent/JP6159890B2/ja
Priority to EP14873505.3A priority patent/EP3089414B1/en
Publication of WO2015096735A1 publication Critical patent/WO2015096735A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/62—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission for providing a predistortion of the signal in the transmitter and corresponding correction in the receiver, e.g. for improving the signal/noise ratio
    • 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
    • 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
    • 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
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02—Transmitters
    • H04B1/04—Circuits
    • H04B1/0475—Circuits with means for limiting noise, interference or distortion
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/38—Synchronous or start-stop systems, e.g. for Baudot code
    • H04L25/40—Transmitting circuits; Receiving circuits
    • H04L25/49—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems
    • 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
    • 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
    • H03F2200/00—Indexing scheme relating to amplifiers
    • H03F2200/198—A hybrid coupler being used as coupling circuit between stages of an amplifier 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
    • H03F2201/3209—Indexing scheme relating to modifications of amplifiers to reduce non-linear distortion the amplifier comprising means for compensating memory effects
    • 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
    • H03F2201/3224—Predistortion being done for compensating memory effects
    • 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
    • H03F2201/3227—Adaptive predistortion based on amplitude, envelope or power level feedback from the output of the main amplifier
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02—Transmitters
    • H04B1/04—Circuits
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02—Transmitters
    • H04B1/04—Circuits
    • H04B2001/0408—Circuits with power amplifiers
    • H04B2001/0416—Circuits with power amplifiers having gain or transmission power control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02—Transmitters
    • H04B1/04—Circuits
    • H04B2001/0408—Circuits with power amplifiers
    • H04B2001/0425—Circuits with power amplifiers with linearisation using predistortion
    • 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
    • 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

Definitions

  • the invention relates to the field of digital predistortion processing, and in particular to a method for obtaining digital predistortion parameters and a predistortion system.
  • the modified volterra series can more clearly describe the physical meaning of the nonlinear system, but its number of model parameters is exponential with the increase of system nonlinearity and memory length. The increase is only applicable to the study of weakly nonlinear systems, otherwise it will cause computational convergence problems.
  • the memory effect produced by the power amplifier of the ultra-wideband signal is very serious.
  • the memory effect of the power amplifier is caused by the inconsistent signal response of the power amplifier to each frequency point.
  • the form of the power amplifier output signal is not only related to the current point signal, but also to the front of the power amplifier. At the moment, obviously, as the signal bandwidth increases, the memory depth of the amplifier is also significantly deepened.
  • the amplifier as an analog device is itself a nonlinear system with amplitude-amplitude (AM-AM) and amplitude-phase (AM-PM) nonlinear distortion.
  • AM-AM distortion refers to the distortion of the amplitude of the output signal and the input signal.
  • AM-PM distortion refers to the change in the amplitude of the nonlinear power amplifier input signal, resulting in a change in the phase difference between the output and the input signal.
  • the narrowband signal is input, the influence of the memory effect is relatively small, and the AM-AM and AM-PM distortion of the power amplifier can be corrected to achieve a better effect.
  • the bandwidth of the signal increases, especially for ultra-wideband signals such as 100M in the next generation of mobile communications, the memory effect of the power amplifier is very serious, making the power amplifier become a very complex system of linear and nonlinear distortion, for such a system.
  • the embodiment of the invention provides a method for obtaining predistortion parameters and a predistortion system for simplifying the computational complexity of the model while achieving good signal processing effects.
  • the predistortion processed predistortion signal and the process are obtained.
  • a first feedback signal processed by the power amplifier, the predistortion signal being obtained according to the following predistortion model:
  • z(n) represents the predistorted signal output at time n
  • x(n) represents the original signal input at time n
  • n represents the input time of the original signal
  • m represents the memory moment of the original signal
  • w represents the Distortion parameter
  • M represents the memory depth
  • Q represents the nonlinear order
  • L represents the maximum cross-sampling point
  • q represents the nonlinear order index
  • * represents the conjugate of the signal
  • l represents the cross-sampling point
  • x (nm) represents the original signal
  • x*(nm) represents a conjugate signal of the original signal
  • the predistortion parameter index table is updated according to the determined predistortion parameters.
  • the original signal is used to replace the conjugate signal of the original signal in the predistortion model according to a signal vector relationship between the original signal and the conjugate signal of the original signal, and the replaced predistortion model is:
  • exp(-j2 ⁇ m1 +j2 ⁇ m2 ) represents the vector relationship between the original signal and the conjugate signal of the original signal
  • ⁇ represents the complex angle of the original signal
  • the above technical solution in the embodiment of the present invention reduces the computational complexity of the overall predistortion model by simplifying the time conjugate interleaving model in the predistortion model, thereby saving multiplier resources.
  • the existing algorithm can be used when acquiring the signal amplitude (ie, the first feedback signal), and the complex angle of the signal can be obtained while acquiring the signal amplitude. Therefore, the simplified process of the model in the embodiment of the present invention is based on the existing The resource is realized without additional resources, and the implementation is simple and convenient.
  • the replaced predistortion model is further changed according to a correspondence between a predistortion parameter in the predistortion parameter index table and an original signal amplitude:
  • the LUT represents a predistortion parameter index table
  • ) represents the signal amplitude
  • the predistortion model has a corresponding relationship between the predistortion parameter and the signal amplitude of the original signal, and the predistortion model is further performed by extracting a common factor. Simplification reduces the overall complexity of the predistortion model.
  • performing the canceling the rated linear gain on the first feedback signal to obtain the second feedback signal specifically includes:
  • the second feedback signal is formed by a first feedback signal that cancels the nominal linear gain and a conjugate signal of the first feedback signal.
  • the conjugate signal of the first feedback signal needs to be cancelled by the rated linear gain.
  • the pre-distortion parameter is determined according to the matrix formed by the second feedback signal and the matrix formed according to the pre-distortion signal, and specifically includes:
  • a predistortion parameter is determined based on a least squares solution of the predistortion parameter.
  • the solution of the linear equation is determined by using the least squares principle.
  • the matrix decomposition method or the fast Cholesky decomposition method may be used to solve the matrix coefficients.
  • the embodiment of the invention further provides a digital predistortion processing system, the system comprising:
  • a predistorter for performing predistortion processing on the input original signal after the periodic filtering process starts, outputting a predistortion signal to the power amplifier; and updating a predistortion parameter index table according to the predistortion parameter sent by the operator, the pre The distortion signal is obtained according to the following predistortion model:
  • z(n) represents the predistorted signal output at time n
  • x(n) represents the original signal input at time n
  • n represents the input time of the original signal
  • m represents the memory moment of the original signal
  • w represents the Distortion parameter
  • M represents memory depth
  • Q represents nonlinear order
  • L represents maximum crossover Sample point
  • q represents a nonlinear order index
  • * denotes the conjugate of the signal
  • l denotes a cross-sample point
  • x(n-m) denotes the original signal
  • x*(n-m) denotes a conjugate signal of the original signal
  • a power amplifier configured to perform power amplifier on the predistortion signal output by the predistorter, and output a first feedback signal to the operator;
  • a novel digital predistortion processing model is proposed.
  • the predistorter uses the model proposed by the embodiment of the present invention to process the original signal, thereby achieving the overall signal processing performance of the system. It also simplifies the complexity of the operation.
  • the embodiment of the invention provides a novel predistortion model, which is a croppable PVS model, compares the MP model, adds a cross-project model, and adopts an architecture near the MP model, thereby achieving a greatly reduced model operation on the one hand.
  • the complexity on the other hand, effectively reflects the main nonlinearity of the power amplifier.
  • FIG. 1 is a schematic flowchart of a method for obtaining a digital predistortion parameter according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of a phase relationship between an original signal and a conjugate signal thereof according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart diagram of a detailed implementation example of a method for obtaining a digital predistortion parameter according to an embodiment of the present disclosure
  • FIG. 4 is a schematic structural diagram of a system of a digital predistortion processing system according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a signal flow of a digital predistortion processing system according to an embodiment of the present invention.
  • the embodiment of the present invention provides a method for obtaining pre-distortion parameters and a pre-distortion system for Simplify the computational complexity of the model while achieving good signal processing.
  • an embodiment of the present invention provides a method for determining a digital predistortion parameter. As shown in FIG. 1, the method includes:
  • z(n) represents the predistorted signal output at time n
  • x(n) represents the original signal input at time n
  • n represents the input time of the original signal
  • m represents the memory moment of the original signal
  • w represents the Distortion parameter
  • M represents the memory depth
  • Q represents the nonlinear order
  • L represents the maximum cross-sampling point
  • q represents the nonlinear order index
  • * represents the conjugate of the signal
  • l represents the cross-sampling point
  • x (nm) represents the original signal
  • x*(nm) represents a conjugate signal of the original signal
  • the predistortion model proposed by the embodiment of the present invention is actually a PVS model, and the PVS model is A PVS model between the MP model and the Volterra series model.
  • This model more fully reflects the nonlinear characteristics of the power amplifier by using the sequence of intersecting items at the adjacent time.
  • This model adopts the sequence of intersecting items at the adjacent time. More comprehensive reflection of the nonlinear characteristics of the power amplifier.
  • the PVS model can also be said to be a model cut out from the Volterra model. In use, only the nonlinear order Q, the memory depth M, and the cross-sampling point L need to be configured to fully characterize the model. Looking closely at the predistortion model of the embodiment of the present invention, the total large model includes three small models that are cropped from the Volterra model, specifically:
  • the embodiment of the present invention adopts these three small models to form a total predistortion model of the embodiment of the present invention.
  • the computational complexity is significantly reduced, and the applicant's test uses the pre-preparation of the embodiment of the present invention.
  • the four-carrier (80MHz) performance test of the Long Term Evolution (LTE) system is as follows.
  • the Adjacent Channel Power Ratio (ACPR) can be up to 5-6dBc.
  • the predistortion model provided by the embodiment of the present invention requires a large number of multipliers due to the square multiplication of the conjugate item and the signal.
  • the original signal is used to replace the conjugate signal of the original signal in the predistortion model, and the replaced predistortion model is:
  • exp(-j2 ⁇ m1 +j2 ⁇ m2 ) represents the vector relationship between the original signal and the conjugate signal of the original signal
  • ⁇ represents the complex angle of the original signal
  • the relationship between the conjugate signal and the original signal is clearly shown in the figure.
  • the amplitude of the conjugate signal in the time-interleaved conjugate memory polynomial can be the original signal.
  • the amplitude is replaced by the specific process:
  • ⁇ 1 actan(imag(x(nm))/real(x(nm)));
  • ⁇ 2 atan(imag(x(nml))/real(x(nml)));
  • the embodiment of the present invention reduces the computational complexity of the overall predistortion model by simplifying the time conjugate interleaving model in the predistortion model, thereby saving multiplier resources.
  • an existing algorithm can be used when acquiring the signal amplitude (ie, the original signal), and the complex angle of the signal can be obtained while acquiring the amplitude of the signal.
  • the CORDIC algorithm uses the CORDIC algorithm to synchronization of the amplitude of the acquired signal.
  • the complex angle of the signal is also obtained. Therefore, the simplified process of the model in the embodiment of the present invention is implemented according to existing resources, and no additional resources are needed, and the implementation is simple and convenient.
  • the simplified pre-loss in the embodiment of the present invention is carefully observed.
  • the true model shows that it can be further simplified by extracting the common factor, so the predistortion model can be further simplified as:
  • the LUT represents an index table of predistortion parameters.
  • the LUT table generation process is given:
  • the amplitude interval in the LUT table is generated as follows:
  • the storage space of one LUT is the length of A*(4L+1)*M.
  • ) is a predistortion parameter corresponding to the input original signal amplitude
  • the reason why the predistortion model can be simplified into the formula (4) in the above embodiment is achieved by simplifying the time-interleaved conjugate memory polynomial in the predistortion model.
  • the embodiment of the present invention can be implemented by simplifying the time interleaved memory polynomial model part based on the formula (3).
  • the time interleaving item in the time-interleaved memory polynomial has less and less influence as the memory depth increases, that is, the interlaced part of the signal leaves with the increase of the memory depth.
  • the predistortion model provided by the embodiment of the present invention may finally be:
  • the above embodiments of the present invention provide a memory in the polynomial model by omitting some time interleaving memory according to the existing rules of time interleaving in the time-interleaved memory polynomial as the memory depth increases, which is less and less affected.
  • a deeper time interleaving project achieves the goal of reducing the overall computational complexity of the overall predistortion model.
  • the applicant proves that the object of the invention can be achieved by using the formula (1), (4) or (6), that is, improving the signal performance. At the same time, reduce the computational complexity of the predistortion model.
  • the second feedback signal is obtained by canceling the rated linear gain of the first feedback signal, and the method includes:
  • the second feedback signal is formed by a first feedback signal that cancels the nominal linear gain and a conjugate signal of the first feedback signal.
  • y represents the first feedback signal
  • u represents the first feedback signal that cancels the rated linear gain of the power amplifier
  • G represents the rated linear gain
  • n represents the input moment of the first feedback signal
  • m represents the memory moment of the first feedback signal
  • M represents Memory depth
  • Q represents a nonlinear order
  • L represents the maximum cross-sampling point
  • q represents a nonlinear order index
  • l represents a cross-sampling point
  • y* represents the conjugate signal of the first feedback signal
  • u* represents the conjugate signal of the first feedback signal after canceling the rated linear gain of the power amplifier
  • G represents the rated linear gain
  • n represents the input timing of the first feedback signal
  • m Representing the memory moment of the first feedback signal
  • M represents the memory depth
  • Q represents the nonlinear order
  • L represents the maximum cross-sample point
  • q represents the nonlinear order index
  • l represents the cross-sample point.
  • the pre-distortion model includes the conjugate signal of the first feedback signal and the first feedback signal, when performing the rated linear gain cancellation, the two formulas are separately calculated. And finally all with Together form a matrix U of the second feedback signal.
  • the pre-distortion parameter is determined according to the matrix formed by the second feedback signal and the matrix formed according to the pre-distortion signal, and specifically includes:
  • a predistortion parameter is determined based on a least squares solution of the predistortion parameter.
  • the predistortion model is infinitely approximated to the power amplifier model, it can be considered that the predistortion signal z and the first feedback signal y have the following relationship:
  • the foregoing embodiment of the embodiment of the present invention calculates the predistortion parameter by using the above relationship between the first feedback signal and the predistortion signal, and obtains the matrix and the predistortion signal of the second feedback signal.
  • the above matrix relation can be changed to calculate the least squares solution of the predistortion parameter. Since the inverse of the matrix is involved in the calculation of the least squares solution of the predistortion parameter, the operation is more complicated.
  • the matrix decomposition method can avoid the inversion of the matrix.
  • the common matrix decomposition method has QR (Orthogonal-triangular Decomposition). Decomposition, SVD, Singular Value Decomposition decomposition, etc.
  • matrix inversion can be avoided by using the adaptive LS algorithm, such as Recursive Least Square (RLS).
  • the model of the cross-project is added, and the architecture near the MP model is adopted, which can greatly save the complexity of the model on the one hand, and can effectively reflect the main non-power of the power amplifier on the other hand. Linear.
  • the PVS model proposed by the embodiment of the present invention is based on a croppable model, and the parameter configuration (non-linear order, memory depth, and cross-sampling point) corresponding to the power amplifier can be adaptively adjusted according to the actual situation of the power amplifier to meet different power amplifiers. Claim.
  • formula (3) is used as a predistortion model:
  • x(n) represents the original signal input at time n
  • z(n) represents the pre-distorted signal output at time n
  • y(n) represents the first feedback signal
  • n represents the input moment of the signal
  • m represents the signal.
  • Memory moment w represents the predistortion parameter
  • M represents the memory depth
  • Q represents the nonlinear order
  • L represents the maximum cross-sampling point
  • q represents the nonlinear order index
  • * represents the conjugate of the signal
  • l represents the cross-sampling point
  • x (nm) represents the signal amplitude of the original signal
  • x*(nm) represents the signal amplitude of the conjugate signal of the original signal
  • y(nm) represents the signal amplitude of the first feedback signal
  • y*(nm) represents the first The signal amplitude of the conjugate signal of the feedback signal
  • G represents the rated linear gain
  • u represents the first feedback signal after canceling the rated linear gain of the power
  • the predistorter processes the original signal according to the previously updated LUT table and the predistortion model, and outputs a predistortion signal.
  • the predistorter extracts corresponding predistortion parameters in the last updated LUT table according to different amplitudes of the input original signals
  • the predistortion signal is changed from a digital signal to an analog signal by a digital-to-analog conversion module;
  • S205 Receive a first feedback signal that passes through a power amplifier process through a radio frequency receiver.
  • the first feedback signal converted into the digital signal is cancelled by the rated linear gain to obtain a second feedback signal.
  • formula (7) can be transformed into The matrix formed by the predistortion signal and the matrix formed by the second feedback signal are brought into the above formula to determine the least squares solution of the predistortion parameter w (since this formula is a set of overdetermined equations.
  • This embodiment may adopt the principle of least squares Determining the solution of the linear equation, using the QR decomposition method of the matrix or the fast Cholesky decomposition method to solve the matrix coefficient in the actual process);
  • the predistorter performs digital predistortion processing on the next original signal according to the updated LUT table.
  • the embodiment of the present invention further provides a digital pre-distortion processing system.
  • the system includes:
  • the predistorter 1 is configured to perform predistortion processing on the input original signal after the periodic filtering process starts, output a predistortion signal to the power amplifier, and update the predistortion parameter index table according to the predistortion parameter sent by the operator;
  • the power amplifier 2 is configured to perform power amplifier on the predistortion signal output by the predistorter, and output a first feedback signal to the operator, where the predistortion signal is obtained according to the following predistortion model:
  • z(n) represents the predistorted signal output at time n
  • x(n) represents the original signal input at time n
  • n represents the input time of the original signal
  • m represents the memory moment of the original signal
  • w represents the Distortion parameter
  • M represents the memory depth
  • Q represents the nonlinear order
  • L represents the maximum cross-sampling point
  • q represents the nonlinear order index
  • * represents the conjugate of the signal
  • l represents the cross-sampling point
  • x (nm) represents the original signal
  • x*(nm) represents a conjugate signal of the original signal
  • the operator 3 is configured to acquire the predistortion signal and the first feedback signal, and perform a cancellation of a nominal linear gain on the first feedback signal to obtain a second feedback signal; a matrix formed according to the second feedback signal and And determining a predistortion parameter according to the matrix formed by the predistortion signal, and transmitting the determined predistortion parameter to a predistorter.
  • a novel digital predistortion processing model is proposed.
  • the predistorter uses the model proposed by the embodiment of the present invention to process the signal, thereby achieving the signal processing performance of the system as a whole. It also simplifies the complexity of the operation.
  • the original signal is used to replace the conjugate signal of the original signal in the predistortion model according to the signal vector relationship between the original signal and the conjugate signal of the original signal.
  • the replaced predistortion model is:
  • exp(-j2 ⁇ m1 +j2 ⁇ m2 ) represents the vector relationship between the original signal and the conjugate signal of the original signal
  • ⁇ represents the complex angle of the original signal
  • the pre-distortion of the replacement is performed according to a correspondence between a predistortion parameter in the predistortion parameter index table and a signal amplitude of an original signal.
  • the model is further changed to:
  • the LUT represents a predistortion parameter index table
  • ) represents the signal amplitude
  • the operator performs a second linear feedback signal on the first feedback signal to obtain a second feedback signal, which specifically includes:
  • the second feedback signal is formed by a first feedback signal that cancels the nominal linear gain and a conjugate signal of the first feedback signal.
  • the operator passes the following Eliminating the nominal linear gain of the first feedback signal:
  • y represents the first feedback signal
  • u represents the first feedback signal that cancels the rated linear gain of the power amplifier
  • G represents the rated linear gain
  • n represents the input moment of the first feedback signal
  • m represents the memory moment of the first feedback signal
  • M represents Memory depth
  • Q represents a nonlinear order
  • L represents the maximum cross-sampling point
  • q represents a nonlinear order index
  • l represents a cross-sampling point
  • the operator eliminates the nominal linear gain of the conjugate signal of the first feedback signal by the following formula:
  • y* represents the conjugate signal of the first feedback signal
  • u* represents the conjugate signal of the first feedback signal after canceling the rated linear gain of the power amplifier
  • G represents the rated linear gain
  • n represents the input timing of the first feedback signal
  • m Representing the memory moment of the first feedback signal
  • M represents the memory depth
  • Q represents the nonlinear order
  • L represents the maximum cross-sample point
  • q represents the nonlinear order index
  • l represents the cross-sample point.
  • the operator determines a predistortion parameter according to the matrix formed by the second feedback signal and the matrix formed according to the predistortion signal, and specifically includes:
  • the predistortion parameter is determined based on the least squares solution of the predistortion parameter.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the present invention may employ computer-usable storage media (including but not limited to disks) in one or more of the computer-usable program code embodied therein. A form of computer program product embodied on a memory and optical storage, etc.).
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Nonlinear Science (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Amplifiers (AREA)

Abstract

本发明实施例公开了一种预失真参数的求取方法及预失真系统，涉及数字预失真处理领域，用以在达到良好信号处理效果的同时，简化模型的计算复杂度。该方法包括：在周期性滤波处理开始后，获取经过预失真处理后的预失真信号和经过功放处理后的第一反馈信号（S101）；对所述第一反馈信号进行消除额定线性增益得到第二反馈信号（S102）；根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数（S103）；根据确定的预失真参数更新预失真参数索引表（S104）。

Description

一种数字预失真参数的求取方法及预失真系统
本申请要求在2013年12月26日提交中国专利局、申请号为201310741067.6、发明名称为“一种数字预失真参数的求取方法及预失真系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及数字预失真处理领域，具体涉及一种数字预失真参数的求取方法及预失真系统。
背景技术
随着现代通信技术的发展，功率放大器的各种非线性特性越来越受到关注，行为模型也成为微波电路领域研究的热点。相比较传统的晶体管级的电路模型，行为模型极大地简化了模型的分析和计算，并保持了足够的非线性电路分析的精度，使其特别适用于宽带数字信号系统的性能分析；因此，在大规模集成电路和预失真技术研究中具有很好的应用前景。对于宽带通信系统而言，由于必须考虑功率放大器记忆效应，因此传统的非线性模型不再适用。目前通常采用带记忆的多项式或者人工神经网络等模型来描述功率放大器的动态特性。相比较基于人工神经网络的行为模型，Volterra级数模型(modified volterra series)可以更清晰地描述非线性系统的物理意义，但它的模型参数数目随着系统非线性及记忆长度的增加呈指数形式增加，只适用于弱非线性系统的研究，否则，将会引起计算收敛性问题。
超宽带信号的功放产生的记忆效应非常的严重，功放记忆效应产生的原因是功放对各个频率点的信号响应不一致，其表现的形式为功放输出信号不但与当前点信号有关，而且与功放前面的时刻点有关，显然，随着信号带宽的增加，功放的记忆深度也显著加深。做为模拟器件的功放本身是一个非线性系统，存在幅度-幅度(AM-AM)和幅度-相位(AM-PM)的非线性失真， AM-AM失真是指输出信号和输入信号幅度上的失真，比如当输入信号摆幅进入阈值电压之下或者饱和电压之上时，输出电压信号就会发生截断或削顶，即为AM-AM失真。AM-PM失真是指，非线性功放输入信号幅度上的变化，导致了输出和输入信号之间的相位差的变化。当窄带信号输入时，记忆效应的影响相对较小，通过校正功放的AM-AM和AM-PM失真就可以达到较好的效果。随着信号的带宽增加，尤其是下一代移动通信中100M这样的超宽带信号，功放的记忆效应非常严重，使得功放变成一个非常复杂的线性与非线性失真相互糅合的系统，对于这样一个系统其理论上的完备表达式是一个Volterra级数模型。显然Volterra级数模型目前来看是不可实现的，需要对其进行简化和优化处理，如何提取功放的主要失真模型，并建立一个有效的、可实现的、开销小的功放预失真模型是一项非常具有挑战意义的工作。
为解决该问题，近年来通常采用一些简化的模型建立放大器的行为模型，其中最常用的是Wiener维纳模型和Hammerstein哈默斯坦模型，这两种模型极大地降低了模型的复杂度，并且能够应用于强非线性系统，因此，在功率放大器非线性行为模型研究中得到广泛的应用。但该两种模型并不能完全地描述功率放大器的非线性特性，尤其是很难精确地表示功率放大器的包络记忆效应；此外，Wiener模型和Hammerstein模型对于模型参数求解来说，均不是线性方程，对模型的参数提取困难。内存多项式(Memory Polynomial，MP)是另一种常用的行为模型，它可以看作是一种扩展的Hammerstein模型，但有时也不能得到符合要求的模型精度。因此，建立精度更高的参数线性的行为模型受到关注。
发明内容
本发明实施例提供了一种预失真参数的求取方法及预失真系统，用以在达到良好信号处理效果的同时，简化模型的计算复杂度。
本发明实施例提供的一种数字预失真参数的求取方法，该方法包括：
在周期性滤波处理开始后，获取经过预失真处理后的预失真信号和经过 功放处理后的第一反馈信号，所述预失真信号根据如下预失真模型获得：
Figure PCTCN2014094808-appb-000001
其中，z(n)表示n时刻输出的经过预失真处理后的信号，x(n)表示n时刻输入的原始信号，n表示原始信号的输入时刻，m表示原始信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号，x*(n-m)表示原始信号的共轭信号；
对所述第一反馈信号进行消除额定线性增益得到第二反馈信号；
根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数；
根据确定的预失真参数更新预失真参数索引表。
在本发明实施例上述技术方案中，提出了一种新型的数字预失真处理模型，以在保证信号处理性能的前提下，简化运算复杂程度。
优选地，根据原始信号与原始信号的共轭信号二者之间信号向量关系，采用原始信号替代所述预失真模型中的原始信号的共轭信号，替代后的所述预失真模型为：
Figure PCTCN2014094808-appb-000002
其中，exp(-j2θm1+j2θm2)表示原始信号与原始信号的共轭信号之间的向量关系，θ表示原始信号的复角。
通过上述公式推导可知，本发明实施例上述技术方案通过简化预失真模型中的时间共轭交错模型，而使整体预失真模型的计算复杂度下降，节省了乘法器资源。而且，在获取信号幅值(即第一反馈信号)时可以采用现有算法，均可以在获取信号幅值的同时获取信号的复角，所以本发明实施例对模型的简化过程是根据现有资源实现的，无需额外增加资源，实现简单方便。
更佳地，根据所述预失真参数索引表中预失真参数与原始的信号幅值之间的对应关系，将所述替代后的所述预失真模型进一步变化为：
Figure PCTCN2014094808-appb-000003
其中，LUT表示预失真参数索引表，LUTm(|x(n-m)|)表示原始信号的信号幅值|x(n-m)|在LUT表中所对应的预失真参数。
在本发明实施例上述技术方案中，通过所述预失真参数索引表中预失真参数与原始信号的信号幅值之间存在对应关系，将所述预失真模型通过提取公因式的方式进一步进行了简化，使得预失真模型整体的复杂度下降。
一般地，对所述第一反馈信号进行消除额定线性增益得到第二反馈信号，具体包括：
消除的所述第一反馈信号的额定线性增益；
消除所述第一反馈信号的共轭信号的额定线性增益；
经过消除额定线性增益的第一反馈信号和第一反馈信号的共轭信号组成所述第二反馈信号。
由于本发明实施例技术方案所提出的预失真模型设计共轭信号，所以在对第一反馈信号进行消除额定线性增益时，还需要对第一反馈信号的共轭信号进行额定线性增益的消除。
优选地，根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，具体包括：
将所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵带入到公式
Figure PCTCN2014094808-appb-000004
确定预失真参数的最小二乘解，其中，
Figure PCTCN2014094808-appb-000005
表示预失真参数的最小二乘解，z表示预失真信号形成的矩阵，U表示第二反馈信号形成的矩阵，UH表示矩阵的U的共轭矩阵；
根据所述预失真参数的最小二乘解确定预失真参数。
本发明实施例上述技术方案应用最小二乘原理确定线性方程的解，实际过程中可以采用矩阵的QR分解方法或者快速Cholesky分解方法求解矩阵系数。
本发明实施例还提供了一种数字预失真处理系统，该系统包括：
预失真器，用于在周期性滤波处理开始后，对输入的原始信号进行预失真处理，向功放器输出预失真信号；根据运算器发送的预失真参数更新预失真参数索引表，所述预失真信号根据如下预失真模型获得：
Figure PCTCN2014094808-appb-000006
其中，z(n)表示n时刻输出的经过预失真处理后的信号，x(n)表示n时刻输入的原始信号，n表示原始信号的输入时刻，m表示原始信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉 采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号，x*(n-m)表示原始信号的共轭信号；
功放器，用于对预失真器输出的预失真信号进行功放，并向运算器输出第一反馈信号；
运算器，用于获取所述预失真信号和所述第一反馈信号，并对所述第一反馈信号进行消除额定线性增益得到第二反馈信号；根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，将所述确定的预失真参数发送到预失真器。
在本发明实施例上述技术方案中，提出了一种新型的数字预失真处理模型，预失真器使用本发明实施例所提出的模型对原始信号进行处理，达到了既保证系统整体的信号处理性能，又简化了运算复杂程度的效果。
本发明实施例提供了一种新型的预失真模型，是一种可裁剪的PVS模型，对比MP模型，增加了交叉项目的模型，采用MP模型附近的架构，从而实现了一方面大大降低模型运算的复杂度，另外一方面有效的反映出功放的主要非线性。
附图说明
图1为本发明实施例提供的一种数字预失真参数的求取方法的方法流程示意图；
图2为本发明实施例提供的原始信号与其共轭信号之间的相位关系示意图；
图3为本发明实施例提供的一种数字预失真参数的求取方法的详细实施例流程示意图；
图4为本发明实施例提供的一种数字预失真处理系统的系统结构示意图；
图5为本发明实施例提供的一种数字预失真处理系统的信号流程示意图。
具体实施方式
由于现有技术常用的数字预失真的处理模型对信号的处理性能和计算的复杂度上不能兼顾，所以本发明实施例提供了一种预失真参数的求取方法及预失真系统，用以在达到良好信号处理效果的同时，简化模型的计算复杂度。
首先，本发明实施例提供了一种数字预失真参数的求取方法，如图1所示，该方法包括：
S101，在周期性滤波处理开始后，获取经过预失真处理后的预失真信号和经过功放处理后的第一反馈信号，所述预失真信号根据如下预失真模型获得：
Figure PCTCN2014094808-appb-000007
其中，z(n)表示n时刻输出的经过预失真处理后的信号，x(n)表示n时刻输入的原始信号，n表示原始信号的输入时刻，m表示原始信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号，x*(n-m)表示原始信号的共轭信号；
S102，对所述第一反馈信号进行消除额定线性增益得到第二反馈信号；
S103，根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数；
S104，根据确定的预失真参数更新预失真参数索引表。
在本发明实施例上述实施例中，提出了一种新型的数字预失真处理模型，以在保证信号处理性能的前提下，简化运算复杂程度。
本发明实施例所提出来的预失真模型其实是一种PVS模型，PVS模型是 一种介于MP模型和Volterra级数模型的PVS模型，这种模型通过采用相邻时刻交叉项目序列，更全面的反映出功放的非线性特性，这种模型通过采用相邻时刻交叉项目序列，更全面的反映出功放的非线性特性。PVS模型也可以说是一种从Volterra模型中裁剪出来的模型，在使用时仅需要配置好非线性阶数Q，记忆深度M，和交叉采样点L就可以完全表征模型。仔细查看本发明实施例的预失真模型可知，总的大模型中包含了三个从Volterra模型中裁剪出来的小模型，具体为：
1)记忆多项式MP模型zmp(n)；
2)时间交错记忆多项式zmp-cl(n)zmp+cl(n)；
3)时间交错共轭记忆多项式zmp-tl(n)和zmp+tl(n)，这些模型的具体描述如下：
Figure PCTCN2014094808-appb-000008
本发明实施例采用这三种小模型组成了本发明实施例的总的预失真模型，相比于原始的Volterra模型，计算复杂度明显下降，经过申请人的测试，采用本发明实施例的预失真模型测试长期演进(Long Term Evolution，LTE)系统的4个载波(80MHz)性能测试如下，相对传统的MP模型邻信道功率比(Adjacent Channel Power Ratio，ACPR)能有5～6dBc的抬升。
可以理解的是，本发明实施例所提供的预失真模型由于涉及共轭项目和信号的平方相乘，需要耗费大量的乘法器，为了进一步达到节省运算的复杂度的目的，在上述实施例的基础上，根据原始信号与原始信号的共轭信号二 者之间信号向量关系，采用原始信号替代所述预失真模型中的原始信号的共轭信号，替代后的所述预失真模型为：
Figure PCTCN2014094808-appb-000009
其中，exp(-j2θm1+j2θm2)表示原始信号与原始信号的共轭信号之间的向量关系，θ表示原始信号的复角。
如图2所示，图中清楚的展示了共轭信号与原始信号的关系，根据二者之间的复角关系，时间交错共轭记忆多项式中的共轭信号的幅值可以采用原始信号的幅值来代替，具体的过程为：
如图所示，可知：
θ1＝actan(imag(x(n-m))/real(x(n-m)))；
θ2＝atan(imag(x(n-m-l))/real(x(n-m-l)))；
所以共轭信号的幅值x*(n-m)＝x(n-m)·exp(-j2θ1)；
而原始信号幅值的平方为：
x2(n-m-l)＝x(n-m-l)x*(n-m-l)·exp(j2θ2)＝|x(n-m-l)|2·exp(j2θ2)。
通过上述公式推导可知，本发明实施例通过简化预失真模型中的时间共轭交错模型，而使整体预失真模型的计算复杂度下降，节省了乘法器资源。而且，在获取信号幅值(即原始信号)时可以采用现有算法，均可以在获取信号幅值的同时获取信号的复角，例如，采用CORDIC算法，即能实现在获取信号幅值的同步也获取信号的复角，所以本发明实施例对模型的简化过程是根据现有资源实现的，无需额外增加资源，实现简单方便。
进一步地，因为所述预失真参数索引表中预失真参数与原始信号的信号幅值之间存在对应关系，所以通过仔细观察本发明实施例中经过简化的预失 真模型可知，可以通过提取公因式的方式进一步进行简化，所以预失真模型可以进一步简化为：
首先，我们知道LUT表示预失真参数的索引表，为了便于理解，对LUT表的生成过程进行简要说明：
假设输入信号最大值为mv＝max(|y(n)|)，LUT的最大尺寸A，那么LUT表中的幅度间隔是
Figure PCTCN2014094808-appb-000010
在上述假设的基础上LUT的生成方式如下：
Figure PCTCN2014094808-appb-000011
根据上式可知，一个LUT的存储空间是A*(4L+1)*M的长度。
LUTm(|x(n-m)|)是按照输入的原始信号幅度|x(n-m)|为索引对应的预失真参数，而又因为：
Figure PCTCN2014094808-appb-000012
Figure PCTCN2014094808-appb-000013
Figure PCTCN2014094808-appb-000014
Figure PCTCN2014094808-appb-000015
所以，上述公式(2)可简化为：
Figure PCTCN2014094808-appb-000017
在公式(3)的基础上，进一步地，预失真模型在提取公因式后可简化为：
Figure PCTCN2014094808-appb-000018
由于公式(4)中的预失真模型实现了公因式的提取工作，极大的降低了运算的难度。
总体来说，在上述实施例中预失真模型之所以能够简化成公式(4)，是通过对预失真模型中的时间交错共轭记忆多项式进行简化得以实现的。
如果根据实际情况，需要预失真模型进一步降低计算难度，本发明实施例在公式(3)基础上可以通过简化时间交错记忆多项式模型部分来实现。
首先，在进行简化前，需要进行简化根据的阐述，时间交错记忆多项式中的时间交错项目随着记忆深度的增加，其影响越来越小，也就是信号的交错部分随着记忆深度的增加离开对角线越远，对系统的影响就越小，所以出于简化计算复杂度的目的，可以将交错记忆多项式中的记忆深度过大的时间交错项目省略掉，以达到降低运算复杂度的目的。就是时间交错项目的数量由最大记忆深度，变为最大记忆深度减去最大交叉采样点，所以本发明实施例预失真模型中的交错记忆多项式模型部分zmpcl(n)可简化为：
Figure PCTCN2014094808-appb-000019
总体来说根据公式(3)与公式(5)，本发明实施例提供的预失真模型最终可为：
Figure PCTCN2014094808-appb-000020
本发明实施例上述实施例根据时间交错记忆多项式中的时间交错项目随着记忆深度的增加，其影响越来越小的现有规则，提供了一种通过省略一些时间交错记忆多项式模型中的记忆深度较大的时间交错项目来实现降低总体预失真模型整体运算复杂度的目的。
综上所述，根据上述各实施例的论证和公式的推导，申请人证明了不论使用公式(1)、(4)或(6)均能够达到本发明实施例的发明目的，即提升信号性能的同时，降低预失真模型的运算复杂度。
上述各实施例证明了本发明实施例提供预失真模型的性能优良性，下面通过一些具体的实施例来对本发明实施例的预失真参数求取过程进行详细解释。
一般地，在进行预失真参数的求取时为了求取出的预失真参数准确，对所述第一反馈信号进行消除额定线性增益得到第二反馈信号，具体包括：
消除所述第一反馈信号的额定线性增益；
消除所述第一反馈信号的共轭信号的额定线性增益；
经过消除额定线性增益的第一反馈信号和第一反馈信号的共轭信号组成所述第二反馈信号。
一般地，通过下列公式消除所述第一反馈信号的额定线性增益：
Figure PCTCN2014094808-appb-000021
其中，y表示第一反馈信号，u表示消除功放额定线性增益的第一反馈信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表 示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点；
通过下列公式消除所述第一反馈信号的共轭信号的额定线性增益：
Figure PCTCN2014094808-appb-000022
其中，y*表示第一反馈信号的共轭信号，u*表示消除功放额定线性增益后的第一反馈信号的共轭信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点。
在本发明实施例中，可以理解的是，由于预失真模型包含了第一反馈信号和第一反馈信号的共轭信号，所以在进行额定线性增益消除时，需要分别通过上述两个公式进行计算，最后所有的
Figure PCTCN2014094808-appb-000023
和
Figure PCTCN2014094808-appb-000024
一起组成了第二反馈信号的矩阵U。
优选地，根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，具体包括：
将所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵带入到公式
Figure PCTCN2014094808-appb-000025
确定预失真参数的最小二乘解，其中，
Figure PCTCN2014094808-appb-000026
表示预失真参数的最小二乘解，z表示预失真信号形成的矩阵，U表示第二反馈信号形成的矩阵，UH表示矩阵的U的共轭矩阵；
根据所述预失真参数的最小二乘解确定预失真参数。
可以理解的是，因为预失真模型无限逼近于功放模型，所以可以认为预失真信号z与第一反馈信号y存在如下关系式：
Figure PCTCN2014094808-appb-000027
正因为存在如上关系，所以本发明实施例上述实施例在计算预失真参数时是通过第一反馈信号与预失真信号的上述关系式求取的，在获得第二反馈信号的矩阵和预失真信号矩阵后，上述关系式可变化成矩阵关系式z＝Uw，根据LS算法，上述矩阵关系式可变化为计算预失真参数的最小二乘解
Figure PCTCN2014094808-appb-000028
由于在计算预失真参数的最小二乘解时涉及到矩阵的求逆，运算比较复杂，采用矩阵分解的方法可以避免对矩阵的求逆，比如常见的矩阵分解方法有QR(Orthogonal-triangular Decomposition)分解、奇异值分解(SVD，Singular Value Decomposition)分解等。另外，采用自适应的LS算法也可以避免矩阵求逆，比如递归最小二乘(RLS，Recursive Least Square)。
综上所述，通过申请的上述说明，证明了本发明实施例所提出预失真模型具有以下优点：
1、本发明实施例提出的PVS模型对比MP模型，增加了交叉项目的模型，采用MP模型附近的架构，一方面能够大大节省模型的复杂度，另外一方面可以有效的反映出功放的主要非线性。
2、本发明实施例提出的PVS模型是一种基于可裁剪的模型，根据功放实际情况可以自适应调整功放对应的参数配置(非线性阶数、记忆深度和交叉采样点)，满足不同功放的要求。
3、通过本发明实施例提出的PVS模型中各小模型的共同点，简化整体模型算法的复杂度，提高信号的处理性能。
为了更好的理解本发明实施例提供的一种预失真参数求取方法，如图3、 图5所示，提供一个详细的实施过程，在本实施例中采用公式(3)为预失真模型：
首先，对下述实施例中出现的字母或符合统一进行解释：
x(n)表示n时刻输入的原始信号，z(n)表示n时刻输出的经过预失真处理后的信号，y(n)表示第一反馈信号，n表示信号的输入时刻，m表示信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号的信号幅值，x*(n-m)表示原始信号的共轭信号的信号幅值，y(n-m)表示第一反馈信号的信号幅值，y*(n-m)表示第一反馈信号的共轭信号的信号幅值，G表示额定线性增益，u表示消除功放额定线性增益后的第一反馈信号，y*表示第一反馈信号的共轭信号，u*表示消除功放额定线性增益后的第一反馈信号的共轭信号，U表示第二反馈信号的矩阵，
Figure PCTCN2014094808-appb-000029
表示预失真参数的最小二乘解，UH表示矩阵的U的共轭矩阵；
S201，在周期性滤波处理开始后，原始信号输入预失真器；
S202，预失真器根据前一次更新后的LUT表及预失真模型对原始信号进行处理，输出预失真信号；
(1)，预失真器根据输入的原始信号的不同幅值在上一次更新的LUT表中提取出对应的预失真参数；
(2)，根据下列公式对原始信号进行信号处理获得预失真信号；
Figure PCTCN2014094808-appb-000030
S203，将预失真信号通过数模转换模块由数字信号变为模拟信号；
S204，将模拟信号的预失真信号通过射频发射器发射到功放模块进行功放过程；
S205，通过射频接收器接收经过功放过程的第一反馈信号；
S206，通过模数转换器将第一反馈信号由模拟信号转换为数字信号；
S207，将转换为数字信号的第一反馈信号消除额定线性增益获得第二反馈信号；
(1)，通过下列公式消除所述第一反馈信号的额定线性增益：
Figure PCTCN2014094808-appb-000031
(2)，通过下列公式消除所述第一反馈信号的共轭信号的额定线性增益：
Figure PCTCN2014094808-appb-000032
(3)，第二反馈信号的矩阵表示为
Figure PCTCN2014094808-appb-000033
(4)，至此，第二反馈信号与预失真信号的关系式可简化为z(n)＝Uw(7)。
S208，将根据所述第二反馈信号形成的矩阵U及根据预失真信号形成的矩阵Z确定预失真参数w；
(1)，公式(7)可变换为
Figure PCTCN2014094808-appb-000034
将预失真信号形成的矩阵及第二反馈信号形成的矩阵带入到上述公式，确定预失真参数w的最小二乘解(因为此公式为超定方程组。本实施例可以采用最小二乘原理确定线性方程的解，实际过程中采用矩阵的QR分解方法或者快速Cholesky分解方法求解矩阵系数)；
S209，将确定的预失真参数发送到预失真器以对LUT表进行更新。
S210，预失真器根据更新后的LUT表对下一次原始信号进行数字预失真处理。
对应本发明实施例上述方法，本发明实施例还提供了一种数字预失真处理系统，如图4所示，该系统包括：
预失真器1，用于在周期性滤波处理开始后，对输入的原始信号进行预失真处理，向功放器输出预失真信号；根据运算器发送的预失真参数更新预失真参数索引表；
功放器2，用于对预失真器输出的预失真信号进行功放，并向运算器输出第一反馈信号，所述预失真信号根据如下预失真模型获得：
Figure PCTCN2014094808-appb-000035
其中，z(n)表示n时刻输出的经过预失真处理后的信号，x(n)表示n时刻输入的原始信号，n表示原始信号的输入时刻，m表示原始信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号，x*(n-m)表示原始信号的共轭信号；
运算器3，用于获取所述预失真信号和所述第一反馈信号，并对所述第一反馈信号进行消除额定线性增益得到第二反馈信号；根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，将所述确定的预失真参数发送到预失真器。
在本发明实施例上述实施例中，提出了一种新型的数字预失真处理模型，预失真器使用本发明实施例所提出的模型对信号进行处理，达到了既保证系统整体的信号处理性能，又简化了运算复杂程度的效果。
优选地，在本发明实施例上述实施例的基础上，根据原始信号与原始信号的共轭信号二者之间信号向量关系，采用原始信号替代所述预失真模型中的原始信号的共轭信号，替代后的所述预失真模型为：
Figure PCTCN2014094808-appb-000036
其中，exp(-j2θm1+j2θm2)表示原始信号与原始信号的共轭信号之间的向量关系，θ表示原始信号的复角。
优选地，在本发明实施例上述实施例的基础上，根据所述预失真参数索引表中预失真参数与原始信号的信号幅值之间的对应关系，将所述替代后的所述预失真模型进一步变化为：
Figure PCTCN2014094808-appb-000037
其中，LUT表示预失真参数索引表，LUTm(|x(n-m)|)表示原始信号的信号幅值|x(n-m)|在LUT表中所对应的预失真参数。
优选地，在本发明实施例上述实施例的基础上，所述运算器对所述第一反馈信号进行消除额定线性增益得到第二反馈信号，具体包括：
消除所述第一反馈信号的额定线性增益；
消除所述第一反馈信号的共轭信号的额定线性增益；
经过消除额定线性增益的第一反馈信号和第一反馈信号的共轭信号组成所述第二反馈信号。
优选地，在本发明实施例上述实施例的基础上，所述运算器通过下列公 式消除所述第一反馈信号的额定线性增益：
Figure PCTCN2014094808-appb-000038
其中，y表示第一反馈信号，u表示消除功放额定线性增益的第一反馈信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点；
所述运算器通过下列公式消除所述第一反馈信号的共轭信号的额定线性增益：
Figure PCTCN2014094808-appb-000039
其中，y*表示第一反馈信号的共轭信号，u*表示消除功放额定线性增益后的第一反馈信号的共轭信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点。
优选地，在本发明实施例上述实施例的基础上，所述运算器根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，具体包括：
将所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵带入到公式
Figure PCTCN2014094808-appb-000040
确定预失真参数的最小二乘解，其中，
Figure PCTCN2014094808-appb-000041
表示预失真参数的最小二乘解，z表示预失真信号形成的矩阵，U表示第二反馈信号形成的矩阵，UH表示矩阵的U的共轭矩阵；
根据预失真参数的最小二乘解确定预失真参数。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘 存储器和光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。

Claims (12)

  1. 一种数字预失真参数的求取方法，其特征在于，该方法包括：
    在周期性滤波处理开始后，获取经过预失真处理后的预失真信号和经过功放处理后的第一反馈信号，所述预失真信号根据如下预失真模型获得：
    Figure PCTCN2014094808-appb-100001
    其中，z(n)表示n时刻输出的经过预失真处理后的信号，x(n)表示n时刻输入的原始信号，n表示原始信号的输入时刻，m表示原始信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号，x*(n-m)表示原始信号的共轭信号；
    对所述第一反馈信号进行消除额定线性增益得到第二反馈信号；
    根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数；
    根据确定的预失真参数更新预失真参数索引表。
  2. 如权利要求1所述的方法，其特征在于，根据原始信号与原始信号的共轭信号二者之间信号向量关系，采用原始信号替代所述预失真模型中的原始信号的共轭信号，替代后的所述预失真模型为：
    Figure PCTCN2014094808-appb-100002
    其中，exp(-j2θm1+j2θm2)表示原始信号与原始信号的共轭信号之间的向量关系，θ表示原始信号的复角。
  3. 如权利要求2所述的方法，其特征在于，根据所述预失真参数索引表中预失真参数与原始的信号幅值之间的对应关系，将所述替代后的所述预失真模型进一步变化为：
    Figure PCTCN2014094808-appb-100003
    其中，LUT表示预失真参数索引表，LUTm(|x(n-m)|)表示原始信号的信号幅值|x(n-m)|在LUT表中所对应的预失真参数。
  4. 如权利要求3所述的方法，其特征在于，对所述第一反馈信号进行消除额定线性增益得到第二反馈信号，具体包括：
    消除所述第一反馈信号的额定线性增益；
    消除所述第一反馈信号的共轭信号的额定线性增益；
    经过消除额定线性增益的第一反馈信号和第一反馈信号的共轭信号组成所述第二反馈信号。
  5. 如权利要求4所述的方法，其特征在于，通过下列公式消除所述第一 反馈信号的额定线性增益：
    Figure PCTCN2014094808-appb-100004
    (0≤l≤L，m＝1...M，q＝1...Q)其中，y表示第一反馈信号，u表示消除功放额定线性增益的第一反馈信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点；
    通过下列公式消除所述第一反馈信号的共轭信号的额定线性增益：
    Figure PCTCN2014094808-appb-100005
    (0≤l≤L，m＝1...M，q＝3...Q)
    其中，y*表示第一反馈信号的共轭信号，u*表示消除功放额定线性增益后的第一反馈信号的共轭信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点。
  6. 如权利要求1所述的方法，其特征在于，根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，具体包括：
    将所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵带入到公式
    Figure PCTCN2014094808-appb-100006
    确定预失真参数的最小二乘解，其中，
    Figure PCTCN2014094808-appb-100007
    表示预失真参数的最小二乘解，z表示预失真信号形成的矩阵，U表示第二反馈信号形成的矩阵，UH表示矩阵的U的共轭矩阵；
    根据所述预失真参数的最小二乘解确定预失真参数。
  7. 一种数字预失真处理系统，其特征在于，该系统包括：
    预失真器，用于在周期性滤波处理开始后，对输入的原始信号进行预失真处理，向功放器输出预失真信号；根据运算器发送的预失真参数更新预失真参数索引表，所述预失真信号根据如下预失真模型获得：
    Figure PCTCN2014094808-appb-100008
    其中，z(n)表示n时刻输出的经过预失真处理后的信号，x(n)表示n时刻输入的原始信号，n表示原始信号的输入时刻，m表示原始信号的记忆时刻，w表示预失真参数，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，*表示信号的共轭，l表示交叉采样点，x(n-m)表示原始信号，x*(n-m)表示原始信号的共轭信号；
    功放器，用于对预失真器输出的预失真信号进行功放，并向运算器输出第一反馈信号；
    运算器，用于获取所述预失真信号和所述第一反馈信号，并对所述第一反馈信号进行消除额定线性增益得到第二反馈信号；根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，将所述确定的预失真参数发送到预失真器。
  8. 如权利要求7所述的系统，其特征在于，根据原始信号与原始信号的共轭信号二者之间信号向量关系，采用原始信号替代所述预失真模型中的原始信号的共轭信号，替代后的所述预失真模型为：
    Figure PCTCN2014094808-appb-100009
    其中，exp(-j2θm1+j2θm2)表示原始信号与原始信号的共轭信号之间的向量 关系，θ表示原始信号的复角。
  9. 如权利要求8所述的系统，其特征在于，根据所述预失真参数索引表中预失真参数与原始信号的信号幅值之间的对应关系，将所述替代后的所述预失真模型进一步变化为：
    Figure PCTCN2014094808-appb-100010
    其中，LUT表示预失真参数索引表，LUTm(|x(n-m)|)表示原始信号的信号幅值|x(n-m)|在LUT表中所对应的预失真参数。
  10. 如权利要求9所述的系统，其特征在于，所述运算器对所述第一反馈信号进行消除额定线性增益得到第二反馈信号，具体包括：
    消除所述第一反馈信号的额定线性增益；
    消除所述第一反馈信号的共轭信号的额定线性增益；
    经过消除额定线性增益的第一反馈信号和第一反馈信号的共轭信号组成所述第二反馈信号。
  11. 如权利要求10所述的系统，其特征在于，所述运算器通过下列公式消除所述第一反馈信号的额定线性增益：
    Figure PCTCN2014094808-appb-100011
    (0≤l≤L，m＝1...M，q＝1...Q)其中，y表示第一反馈信号，u表示消除功放额定线性增益的第一反馈信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点；
    所述运算器通过下列公式消除所述第一反馈信号的共轭信号的额定线性增益：
    Figure PCTCN2014094808-appb-100012
    (0≤l≤L，m＝1...M，q＝3...Q)其中，y*表示第一反馈信号的共轭信号，u*表示消除功放额定线性增益后的第一反馈信号的共轭信号，G表示额定线性增益，n表示第一反馈信号的输入时刻，m表示第一反馈信号的记忆时刻，M表示记忆深度，Q表示非线性阶数，L表示最大交叉采样点，q表示非线性阶数索引，l表示交叉采样点。
  12. 如权利要求7所述的系统，其特征在于，所述运算器根据所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵确定预失真参数，具体包括：
    将所述第二反馈信号形成的矩阵及根据所述预失真信号形成的矩阵带入到公式
    Figure PCTCN2014094808-appb-100013
    确定预失真参数的最小二乘解，其中，
    Figure PCTCN2014094808-appb-100014
    表示预失真参数的最小二乘解，z表示预失真信号形成的矩阵，U表示第二反馈信号形成的矩阵，UH表示矩阵的U的共轭矩阵；
    根据预失真参数的最小二乘解确定预失真参数。
PCT/CN2014/094808 2013-12-26 2014-12-24 一种数字预失真参数的求取方法及预失真系统 WO2015096735A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US15/108,108 US9647717B2 (en) 2013-12-26 2014-12-24 Digital pre-distortion parameter obtaining method and pre-distortion system
KR1020167020510A KR101680207B1 (ko) 2013-12-26 2014-12-24 디지털 전치 왜곡 파라미터를 산출하는 방법 및 전치 왜곡 시스템
JP2016542951A JP6159890B2 (ja) 2013-12-26 2014-12-24 デジタルプリディストーションパラメーターを求める方法及びプリディストーションシステム
EP14873505.3A EP3089414B1 (en) 2013-12-26 2014-12-24 Digital pre-distortion parameter obtaining method and pre-distortion system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310741067.6 2013-12-26
CN201310741067.6A CN103685111B (zh) 2013-12-26 2013-12-26 一种数字预失真参数的求取方法及预失真系统

Publications (1)

Publication Number Publication Date
WO2015096735A1 true WO2015096735A1 (zh) 2015-07-02

Family

ID=50321456

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/094808 WO2015096735A1 (zh) 2013-12-26 2014-12-24 一种数字预失真参数的求取方法及预失真系统

Country Status (6)

Country Link
US (1) US9647717B2 (zh)
EP (1) EP3089414B1 (zh)
JP (1) JP6159890B2 (zh)
KR (1) KR101680207B1 (zh)
CN (1) CN103685111B (zh)
WO (1) WO2015096735A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017054153A1 (zh) * 2015-09-30 2017-04-06 华为技术有限公司 一种非线性干扰的抵消方法和装置
CN107276546A (zh) * 2016-04-08 2017-10-20 大唐移动通信设备有限公司 一种数字预失真处理方法及装置
JP2018523940A (ja) * 2015-07-27 2018-08-23 ノースロップ グラマン システムズ コーポレーション 送信機による非線形の事前符号化

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103685111B (zh) 2013-12-26 2017-01-11 大唐移动通信设备有限公司 一种数字预失真参数的求取方法及预失真系统
KR102296415B1 (ko) * 2015-02-13 2021-09-02 한국전자통신연구원 넓은 범위의 이득 조절을 제공하는 대역통과 필터
CN104869091A (zh) * 2015-04-29 2015-08-26 大唐移动通信设备有限公司 一种数字预失真系数训练方法及系统
CN105024960B (zh) * 2015-06-23 2018-11-09 大唐移动通信设备有限公司 一种dpd系统
CN106453173A (zh) * 2016-11-24 2017-02-22 希诺麦田技术（深圳）有限公司 预失真参数估计系统及无线发射系统
US10075201B1 (en) * 2017-07-12 2018-09-11 Intel IP Corporation Adaptive nonlinear system control using robust and low-complexity coefficient estimation
US10469109B2 (en) * 2017-09-19 2019-11-05 Qualcomm Incorporated Predistortion for transmitter with array
US10985951B2 (en) 2019-03-15 2021-04-20 The Research Foundation for the State University Integrating Volterra series model and deep neural networks to equalize nonlinear power amplifiers
CN111082756B (zh) * 2019-12-11 2023-06-06 电子科技大学 一种用于mimo发射机的数模混合预失真结构
CN111092602B (zh) * 2019-12-27 2023-10-20 京信网络系统股份有限公司 功率放大器的建模方法、装置、计算机设备和存储介质
CN113659937B (zh) * 2021-05-10 2022-06-07 中兴通讯股份有限公司 预失真处理方法及其装置、通信设备、存储介质
CN115529211A (zh) * 2021-06-24 2022-12-27 华为技术有限公司 一种更新预处理参数的方法及装置
CN114900244B (zh) * 2022-05-13 2023-07-04 中国电子科技集团公司第三十研究所 一种基于开环数字预失真的信号失真控制系统
CN114629756B (zh) * 2022-05-16 2022-08-16 成都凯腾四方数字广播电视设备有限公司 一种多模5g广播发射机自适应预失真方法及系统
CN115987729B (zh) * 2022-06-24 2023-09-29 上海星思半导体有限责任公司 相位对齐方法、相位对齐装置与计算机可读存储介质
CN115278746B (zh) * 2022-07-28 2023-03-28 北京邮电大学 面向5g宽带功放的自适应快速误差修正数字预失真方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN102511153A (zh) * 2011-11-16 2012-06-20 华为技术有限公司 一种微波预失真信号生成方法和装置
CN103685111A (zh) * 2013-12-26 2014-03-26 大唐移动通信设备有限公司 一种数字预失真参数的求取方法及预失真系统

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6246286B1 (en) * 1999-10-26 2001-06-12 Telefonaktiebolaget Lm Ericsson Adaptive linearization of power amplifiers
US6903604B2 (en) * 2001-06-07 2005-06-07 Lucent Technologies Inc. Method and apparatus for modeling and estimating the characteristics of a power amplifier
US7289773B2 (en) * 2003-01-23 2007-10-30 Powerwave Technologies, Inc. Digital transmitter system employing self-generating predistortion parameter lists and adaptive controller
US7873172B2 (en) * 2005-06-06 2011-01-18 Ntt Docomo, Inc. Modified volterra-wiener-hammerstein (MVWH) method for loudspeaker modeling and equalization
US7479828B2 (en) * 2005-11-15 2009-01-20 Alcatel-Lucent Usa Inc. Selecting samples for amplifier digital predistortion estimation
US7929927B2 (en) * 2007-10-29 2011-04-19 Freescale Semiconductor, Inc. Adaptive pre-distortion with interference detection and mitigation
CN100594669C (zh) * 2008-07-18 2010-03-17 东南大学 基于模糊神经网络的哈默斯坦模型的功放预失真方法
JP5299958B2 (ja) * 2008-12-01 2013-09-25 日本無線株式会社 プリディストータ
CN101771639B (zh) * 2008-12-31 2012-10-10 大唐移动通信设备有限公司 一种预失真参数的处理方法和装置
CA2815541A1 (en) * 2010-11-16 2012-05-24 Telefonaktiebolaget Lm Ericsson (Publ) Joint process estimator with variable tap delay line for use in power amplifier digital predistortion
US9190963B2 (en) * 2013-08-22 2015-11-17 Blackberry Limited System and method for fast polynomial pre-distortion

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN102511153A (zh) * 2011-11-16 2012-06-20 华为技术有限公司 一种微波预失真信号生成方法和装置
CN103685111A (zh) * 2013-12-26 2014-03-26 大唐移动通信设备有限公司 一种数字预失真参数的求取方法及预失真系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
DENNIS, R.M. ET AL.: "A Generalized Memory Polynomial Model for Digital Predistortion of RF Power Amplifiers", IEEE TRANSACTION ON SIGNAL PROCESSING, vol. 10, no. 54, 31 October 2006 (2006-10-31), pages 3852 - 3860, XP055187301 *
NAN, JINGCHANG ET AL.: "New LMEC Model for Behavior Modeling and Predistortion for RF Power Amplifiers", APPLICATION RESEARCH OF COMPUTERS, vol. 30, no. 8, 31 August 2013 (2013-08-31), pages 2448 - 2449, XP008181358 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018523940A (ja) * 2015-07-27 2018-08-23 ノースロップ グラマン システムズ コーポレーション 送信機による非線形の事前符号化
JP2021036683A (ja) * 2015-07-27 2021-03-04 ノースロップ グラマン システムズ コーポレーション 送信機による非線形の事前符号化
JP7198251B2 (ja) 2015-07-27 2022-12-28 ノースロップ グラマン システムズ コーポレーション 送信機による非線形の事前符号化
WO2017054153A1 (zh) * 2015-09-30 2017-04-06 华为技术有限公司 一种非线性干扰的抵消方法和装置
CN108141243A (zh) * 2015-09-30 2018-06-08 华为技术有限公司 一种非线性干扰的抵消方法和装置
CN108141243B (zh) * 2015-09-30 2019-11-29 华为技术有限公司 一种非线性干扰的抵消方法和装置
CN107276546A (zh) * 2016-04-08 2017-10-20 大唐移动通信设备有限公司 一种数字预失真处理方法及装置

Also Published As

Publication number Publication date
CN103685111A (zh) 2014-03-26
KR101680207B1 (ko) 2016-11-28
JP2017509179A (ja) 2017-03-30
EP3089414B1 (en) 2017-12-20
KR20160096208A (ko) 2016-08-12
EP3089414A1 (en) 2016-11-02
US20160329927A1 (en) 2016-11-10
JP6159890B2 (ja) 2017-07-05
US9647717B2 (en) 2017-05-09
EP3089414A4 (en) 2016-11-02
CN103685111B (zh) 2017-01-11

Similar Documents

Publication Publication Date Title
WO2015096735A1 (zh) 一种数字预失真参数的求取方法及预失真系统
CN101459636B (zh) 自适应预失真方法
Raich et al. Orthogonal polynomials for power amplifier modeling and predistorter design
JP5753272B2 (ja) タップ出力の正規化を伴う非線形モデル
JP5236661B2 (ja) 多チャンネル広帯域通信システムにおけるベースバンドプリディストーション線形化の方法及びシステム
US8787494B2 (en) Modeling digital predistorter
JP5120178B2 (ja) 非線形システム逆特性同定装置及びその方法、電力増幅装置及び電力増幅器のプリディストータ
JP2014533017A (ja) デジタル・プリディストーション（ｄｐｄ）および他の非線形アプリケーションのためのユーザ定義の非線形関数を含む命令セットを有するプロセッサ
EP2641324A1 (en) Joint process estimator with variable tap delay line for use in power amplifier digital predistortion
CN102075469B (zh) 用于数字预失真系统的信号延迟时间的估计方法
Pan et al. A predistortion algorithm based on accurately solving the reverse function of memory polynomial model
Liu et al. On the robustness of look-up table digital predistortion in the presence of loop delay error
CN103179074A (zh) 基于正交多项式的自适应预失真系统及方法
CN103929212A (zh) 一种宽带接收机非线性盲辨识及补偿方法
Dvorak et al. Adaptive-order polynomial methods for power amplifier model estimation
CN115529212A (zh) 基于开环的短波通信方法、装置、设备及可读存储介质
Khawam et al. Accurate and low complexity polynomial and neural network models for PA digital pre-distortion
Zeng et al. A crest factor reduction method in digital predistortion for improvement of power efficiency
Ma et al. An enhanced digital predistortion algorithm based on polynomial model identification
KR20110048086A (ko) 전치 왜곡을 이용한 전력 증폭 장치
Gao et al. Approaches for Improving Efficiency: Adaptive Predistortion of Dual-Level Voltage Power Amplifiers
Chang An efficient compensation of TWTA's nonlinear distortion in wideband OFDM systems
Xia et al. Polynomial Model with Feedback for Digital Predistortion of Power Amplifiers
Hormozi et al. Multi-Level 2D LUT as digital pre-distorter for linearizing memory affected RF power amplifiers
CN117424568A (zh) 基于极坐标多项式模型的数字信号预失真处理方法及系统

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: 14873505

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2016542951

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 15108108

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2014873505

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2014873505

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20167020510

Country of ref document: KR

Kind code of ref document: A