A signal distortion pre-correction method, device, system and composite system

Abstract

The present application discloses a signal distortion pre-correction device and method, system and composite system. The signal distortion pre-correction method comprises: performing time-sharing acquisition of the output signals of multiple channels of a power amplifier module to obtain multiple analog feedback signals; converting the multiple analog feedback signals into multiple digital feedback signals; determining correction parameters according to the multiple digital feedback signals and the input forward signal; and correcting the forward signal based on the correction parameters to obtain a pre-correction signal. The solution provided in this embodiment estimates the correction parameters based on multiple feedback signals, which reduces the requirements for the consistency of the characteristics of each power amplifier in the system, and can well solve the signal distortion pre-correction problem of nonlinear systems in analog beamforming application scenarios.

Description

Technical Field

[0001] The embodiments of the present invention relate to, but are not limited to, a signal distortion pre-correction method and device, a system and a composite system. Background Art

[0002] In modern mobile communication systems, multi-carrier transmission technology and high-order digital modulation methods will lead to a higher peak-to-average ratio and a larger signal transmission bandwidth. When the power amplifier (PA) works in a near-saturation region, it will cause serious nonlinear distortion and memory effects in the PA. An effective way to avoid PA nonlinear distortion is to back off the power of the input signal, but this will greatly reduce the efficiency of the PA and increase its power consumption. At present, due to its advantages of low cost and obvious distortion improvement effect, digital pre-distortion technology has become the preferred method for distortion pre-correction of nonlinear systems.

[0003] In 5G (5 th In the fifth generation (5th generation) millimeter wave communication, in order to obtain larger system capacity, higher spectrum utilization and higher beamforming gain, massive multiple input multiple output (Massive Multiple InMultiple Out, Massive MIMO) technology and analog beamforming technology are used. The use of analog beamforming technology will cause a digital channel to be connected to multiple analog RF channels. In this case, if digital pre-distortion technology is used, it is necessary to design a pre-distorter to correct the nonlinear distortion of multiple PAs at the same time. Moreover, in this system, since there is no circulator between the PA and the antenna array, the output load mismatch and inter-antenna coupling interference will have a great impact on the power amplifier characteristics, which also poses a greater challenge to the design of the pre-distorter. Traditional pre-distortion algorithms are more suitable for scenarios where one digital channel corresponds to one analog RF channel. In a multiple input multiple output (Multiple In Multiple Out, MIMO) analog beamforming system, if the traditional pre-distortion method is used, the structural complexity and volume of the transmitter will be greatly increased, and its power consumption will also increase. Summary of the invention

[0004] At least one embodiment of the present invention provides a signal distortion pre-correction method and device, system and composite system to achieve signal distortion pre-correction.

[0005] An embodiment of the present invention provides a signal distortion pre-correction device, comprising: a signal distortion correction network module, a correction parameter trainer, a data acquisition module and a first conversion module, wherein the data acquisition module is connected to an external power amplifier module, wherein:

[0006] The data acquisition module is used to perform time-sharing acquisition on the output signals of multiple channels of the power amplifier module, and output analog feedback signals to the first conversion module;

[0007] The first conversion module is used to convert the multiple analog feedback signals into multiple digital feedback signals and send them to the correction parameter trainer;

[0008] The correction parameter trainer is used to determine the correction parameters according to the multiple digital feedback signals and the input forward signal, and output the correction parameters to the signal distortion correction network module;

[0009] The signal distortion correction network module is used to output a pre-corrected signal after correcting the forward signal based on the correction parameter.

[0010] Compared with the related art, an embodiment of the present invention includes time-sharing acquisition of output signals of multiple channels of a power amplifier module to obtain multiple analog feedback signals; converting the multiple analog feedback signals into multiple digital feedback signals; determining correction parameters according to the multiple digital feedback signals and the input forward signal; and correcting the forward signal based on the correction parameters to obtain a pre-correction signal. The solution provided in this embodiment estimates the correction parameters based on multiple feedback signals, which reduces the requirement for characteristic consistency of each power amplifier in the system.

[0011] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0013] Figure 1 A conventional signal distortion prevention device architecture diagram applicable to MIMO millimeter wave communication systems;

[0014] Figure 2 A schematic diagram of a signal distortion pre-correction device according to an embodiment of the present invention;

[0015] Figure 3 A schematic diagram of a signal distortion pre-correction device provided by another embodiment of the present invention;

[0016] Figure 4 A basic structural diagram of a correction parameter trainer provided by an embodiment of the present invention;

[0017] Figure 5 A basic structural diagram of a signal distortion correction network module provided by an embodiment of the present invention;

[0018] Figure 6 A basic structural diagram of a signal distortion correction network module provided by another embodiment of the present invention;

[0019] Figure 7 A flow chart of a signal distortion pre-correction method provided by an embodiment of the present invention;

[0020] Figure 8 A flow chart of a signal distortion pre-correction method provided by another embodiment of the present invention;

[0021] Fig. 9 A block diagram of a signal distortion pre-correction composite system provided by an embodiment of the present invention;

[0022] Fig.10 A block diagram of a signal distortion pre-correction composite system provided for a specific embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present invention more clear, the embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other arbitrarily without conflict.

[0024] The steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions. Also, although a logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a sequence different from that shown here.

[0025] The industry has also proposed some solutions to the distortion pre-correction problem of MIMO analog beamforming systems. These solutions can pre-correct the nonlinear signal distortion in multi-antenna analog beamforming systems to a certain extent, but they still have great shortcomings.

[0026] In one solution, only one feedback link channel is used to collect and feedback the output signal of one power amplifier, perform data preprocessing and model parameter extraction, and to correct the overall nonlinear distortion of all power amplifiers. This places very high demands on the consistency of the characteristics of all power amplifiers. Figure 1As shown, a conventional signal distortion pre-device architecture diagram applicable to MIMO millimeter wave communication system is provided, and the device includes: baseband signal 010, general DPD (Digital Pre-distortion) module 011, DAC (Digital to Analog Converter) module 012, up-conversion module 013, analog beamforming module 014, power amplifier group 015, down-conversion module 016, ADC (Analog to Digital Converter) module 017 and data calibration processing module 018. The device only performs feedback collection and data calibration processing on the output of one power amplifier, extracts DPD parameters based on the general DPD architecture, and is used to correct and compensate for the nonlinear distortion of all power amplifiers.

[0027] In another scheme, the feedback acquisition signals of multiple power amplifiers need to be de-beamformed and combined in the analog domain, which places very high demands on the consistency of the power amplifier characteristics and the consistency of the transmission link from each power amplifier output port to the combiner; the output signals of all power amplifiers need to be classified in the analog domain. If the classification standard is not selected appropriately, it will greatly affect the performance of the signal distortion budget method; and a power amplifier control module is required to adjust the supply voltage or current of each power amplifier in real time, which is not conducive to the stability of the power amplifier characteristics and will increase the complexity of the system.

[0028] In addition, the above two solutions do not consider the impact of factors such as output load mismatch and antenna coupling interference on the signal distortion pre-correction performance.

[0029] At least one embodiment of the present invention proposes a signal distortion pre-correction method and device suitable for a MIMO millimeter wave communication system, which performs time-sharing acquisition on the output signals of each power amplifier, and then performs multi-channel data parallel linear processing, establishes a correction model, constructs a signal matrix and weighted processing, and extracts correction parameters. Finally, the correction parameters in the distortion correction network are updated in real time, thereby achieving the purpose of pre-correcting the distortion of the nonlinear system.

[0030] An embodiment of the present invention provides a signal distortion pre-correction device, such as Figure 2As shown, it includes: a signal distortion correction network module 100, a first conversion module 170, a correction parameter trainer 180 and a data acquisition module 190. The device is connected to an external second conversion module 110, an analog beamforming processor 120, a power amplifier module 130 and an antenna array module 150. The antenna array module 150 includes a plurality of antenna elements. The data acquisition module 190 is connected to the power amplifier module 130. The signal distortion correction network module 100 is connected to the second conversion module 110. The second conversion module 110 is connected to the analog beamforming processor 120. The power amplifier module 130 includes multiple channels (analog channels). For example, one channel includes one power amplifier.

[0031] The data acquisition module 190 is used to perform time-sharing acquisition on the output signals of multiple channels of the power amplifier module 130 and output analog feedback signals to the first conversion module 170;

[0032] The first conversion module 170 is used to convert the analog feedback signal into a digital feedback signal and send it to the correction parameter trainer 180; wherein the first conversion module 170 is, for example, an analog-to-digital conversion module; in addition, the first conversion module 170 may also have a frequency conversion function, such as an ADC with an integrated frequency conversion function.

[0033] The correction parameter trainer 180 is used to determine the correction parameters according to the digital feedback signal and the input forward signal, and output the correction parameters to the signal distortion correction network module 100; specifically, it includes: performing data linear processing on the forward signal and the digital feedback signal, and solving and iteratively training the signal distortion pre-correction model parameters. The signal distortion pre-compensation model can be multiple groups of low-order filters, Volterra series, memory crossover polynomials, neural networks (such as BP networks, ART networks, etc.), wavelet networks, support vector bases, etc.

[0034] The signal distortion correction network module 100 is used to correct the input forward signal based on the correction parameter (or distortion pre-compensation processing), and then output the pre-corrected signal to the second conversion module 110 .

[0035] The solution provided in this embodiment collects signals of multiple channels in a time-sharing manner and performs correction based on feedback signals of the multiple channels, and has low requirements on the characteristic consistency of the power amplifiers of each channel in the system.

[0036] In one embodiment, if Figure 3 As shown, the data acquisition module 190 includes: a signal coupling network module 140 and a signal sampling controller 160, wherein the signal coupling network module 140 is connected to an external power amplifier module 130, wherein:

[0037] The signal sampling controller 160 is used to control the signal coupling network module 140 to realize the time-sharing acquisition of the output signals of the multiple channels of the power amplifier module 130. Specifically, the signal coupling network module 140 is controlled so that the output of a certain channel of the power amplifier module 130 is connected to the feedback link to realize the acquisition of the output signal of the channel. It should be noted that the multiple channels for time-sharing acquisition may include all channels of the power amplifier module 130 or some of the channels.

[0038] The signal coupling network module 140 is used to time-couple the output signals of the channels of the power amplifier module 130 according to the control of the signal sampling controller 160, and output analog feedback signals to the first conversion module 170; the output signals of each channel are, for example, the output signals of each power amplifier of the power amplifier module.

[0039] In one embodiment, the correction parameter trainer 180 includes a data linear processing module 181 and a correction parameter calculation module 182, wherein:

[0040] The data linear processing module 181 is used to perform linear processing on the forward signal and the digital feedback signal to obtain a linearly processed forward signal and a linearly processed digital feedback signal; wherein the linear processing includes but is not limited to at least one of the following: delay alignment, mirror calibration, frequency alignment, feedback non-flatness compensation, and complex gain compensation. In one embodiment, the linear processing is parallel linear processing.

[0041] The correction parameter calculation module 182 is used to construct multiple signal matrices based on a preset signal distortion pre-correction model, a linearly processed forward signal, a linearly processed digital feedback signal, and the pre-correction signal, wherein the signal matrix corresponds to the collected channel, and the multiple signal matrices are weighted using weighting factors, and a combination matrix is ​​generated using the weighted signal matrix, and nonlinear distortion pre-compensation parameters are obtained based on the combination matrix, and link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed digital feedback signal, and the weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix. In one embodiment, the weighting factor is the adjacent channel power ratio (ACPR) of the channel corresponding to the signal matrix. It should be noted that the weighting factor is not limited to ACPR, and can also be obtained by other means.

[0042] The solution provided in this embodiment performs weighted processing on the constructed signal matrix during the correction parameter extraction process, so that the performance index in the beamforming direction can be optimized; the feedback signals of each power amplifier are processed linearly in parallel, so as to eliminate the influence of the delay, phase and amplitude difference of the coupled feedback link on the performance of the distortion correction algorithm; because the feedback signals of each power amplifier can be processed linearly, the beam angle can be any angle when extracting the correction parameters, and is not limited to the 0° beam, and the distortion pre-correction performance in the far-field beam direction can be guaranteed.

[0043] In one embodiment, the correction parameter calculation module 182 is used to determine the mismatch filter coefficient and the mutual coupling filter coefficient according to the power amplifier model of the power amplifier module, the forward signal after linear processing, and the digital feedback signal after linear processing, and construct multiple signal matrices based on the preset signal distortion pre-correction model, the mismatch filter coefficient, the mutual coupling filter coefficient, the forward signal after linear processing, the digital feedback signal after linear processing, and the pre-correction signal, wherein there is a corresponding relationship between the signal matrix and the collected channels, and the multiple signal matrices are weighted using weighting factors respectively. Then, a combination matrix is ​​generated using the weighted signal matrix, and nonlinear distortion pre-compensation parameters, mismatch compensation parameters, and mutual coupling compensation parameters are obtained based on the combination matrix, and link imbalance calibration parameters are determined based on the forward signal after linear processing and the digital feedback signal after linear processing, and the weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix. The mismatch filter coefficient, mutual coupling filter coefficient, mismatch compensation parameter and mutual coupling compensation parameter can be used to correct the distortion caused by load mismatch and the distortion caused by mutual coupling interference between antennas, and can reduce the impact of output load mismatch and mutual coupling interference between antennas on system performance.

[0044] It should be noted that, in another embodiment, the mismatch filter coefficient, the mutual coupling filter coefficient, the mismatch compensation parameter and the mutual coupling compensation parameter may not be calculated.

[0045] In one embodiment, the signal matrix corresponding to the j-th channel is:

[0046]

[0047]

[0048] in, x j is the forward signal after linear processing, z j is the pre-correction signal, y′ j is the digital feedback signal after linear processing, h a is the mismatch filter coefficient, h bis the mutual coupling filter coefficient, L and Q are the memory depth and nonlinear order of the signal distortion pre-correction model respectively; N+1 is the data length used to construct the signal matrix.

[0049] In one embodiment, after weighting the plurality of signal matrices using weighting factors respectively, generating a combination matrix using the weighted signal matrices includes:

[0050] W=[α1W1,α2W2,…,α K W K ]

[0051] V=[α1V1 T ,α2V2 T ,…,α K V K T ] T

[0052] Among them, α1, α2, ..., α K is the weighting factor.

[0053] Figure 4 The figure shows a basic structure diagram of a correction parameter trainer 180 provided by an embodiment of the present invention, which includes: a data linear processing module 181 and a correction parameter calculation module 182. The data linear processing module 181 includes a plurality of parallel data linear processing submodules 181(1), 181(2), ... 181(K). The correction parameter calculation module 182 includes a mismatch and mutual coupling coefficient estimation module 1821, a matrix generation module 1822, a matrix weighting processing module 1823 and a parameter estimation module 1824. Among them,

[0054] The data linear processing submodules 181(1), 181(2), ... 181(K) are based on the forward signal x j (j=1,2,...,K), and at the same time, the feedback signal y output by each power amplifier j Linear processing operations such as delay alignment, image calibration, complex gain compensation, frequency alignment, and feedback link uneven equalization are performed in parallel to obtain the feedback signal after linear processing. The mathematical expression of each power amplifier output signal is as follows:

[0055]

[0056] Among them, y i (t) and x i (t) are the analog input signal and output signal of the ith power amplifier respectively; G[x i (t)] and Φ[x i (t)] are the amplitude distortion and phase distortion of the ith power amplifier respectively; ω i and τi are the frequency offset and transmission delay of the ith analog channel respectively. Since the transmission delay, amplitude value and phase value of each feedback signal after transmission through the link are different, and the difference may be large, linear processing of each feedback signal can eliminate the difference between transmission links, which is conducive to the subsequent extraction and training of correction model parameters.

[0057] The mismatch and mutual coupling coefficient estimation module 1821 is used to estimate the mismatch filter coefficient and the mutual coupling filter coefficient. The mismatch filter coefficient is the coefficient used by the mismatch filter module for filtering; the mutual coupling filter coefficient is the coefficient used by the mutual coupling filter module for filtering.

[0058] When considering the influence of power amplifier output load mismatch and antenna coupling interference, the forward signal x j and feedback signal y′ j The relationship expression between them is as follows:

[0059] y′ j =f(x j ,x′ j ,x″ j )

[0060] x′ j =h a *x j ,x″ j =h b *x j

[0061] Where, j = 1, 2, ..., K, f(·) is the function describing the characteristics of the power amplifier; * is the linear convolution operator; x′ j is the standing wave signal fitted based on the forward signal; x″ j is the mutual coupling interference signal between antennas fitted based on the forward signal; h a is the mismatch filter coefficient; h b is the mutual coupling filter coefficient.

[0062] According to the specific power amplifier model (the power amplifier model can be selected according to the power amplifier characteristics in the power amplifier module, such as using a memory cross polynomial model. Of course, this is only an example, and other power amplifier models can be selected as needed), select the forward signal and feedback signal (x j and y′ j ), the model coefficients, mismatch filter coefficients and mutual coupling filter coefficients are cross-estimated using the least squares algorithm (LS) until the coefficients h that meet the accuracy requirements are obtained. a and h bThen, multiple groups of linearly processed data are selected to perform the same operation as above to obtain multiple groups of mismatch filter coefficients and mutual coupling filter coefficients, and they are averaged to obtain the final mismatch filter coefficients and mutual coupling filter coefficients.

[0063] The matrix generation module 1822 is used to use the forward signal x j (j=1, 2, ..., K), feedback signal y j '(j=1,2,...,K), pre-correction signal z j (i=1,2,...,K), mismatch filter coefficient h a and mutual coupling filter coefficient h b , and construct the signal matrix W1,W2,...,W in parallel K and V1,V2,...,V K Specifically, the matrix generation module 1822 includes a plurality of matrix generation submodules 1822(1), 1822(2), ... 1822(K). The matrix generation submodule 1822(1) constructs signal matrices W1 and V1, the matrix generation submodule 1822(2) constructs signal matrices W2 and V2, and the matrix generation submodule 1822(K) constructs signal matrices W K and V K , where the matrix W j and V j The specific form of depends on the signal distortion precompensation model used. If the memory polynomial model is used, the matrix W j and V j for:

[0064]

[0065]

[0066] In the formula, L and Q are the memory depth and nonlinear order of the signal distortion pre-compensation model respectively; N+1 is the data length used to construct the signal matrix.

[0067] It should be noted that the above W j and V j The matrix construction is only an example and can be constructed in other ways as needed.

[0068] In another embodiment, when the nonlinear distortion caused by load imbalance and mutual coupling is not compensated,

[0069] Matrix weighting processing module 1823, used to use weighting factors α1, α2, ..., α K For matrices W1, W2, ..., WK and V1,V2,...,V K After weighted processing, the weighted matrices are used to generate combined matrices W and V, namely:

[0070] W=[α1W1,α2W2,…,α K W K ]

[0071] V=[α1V1 T ,α2V2 T ,…,α K V K T ] T

[0072] The purpose of weighting the matrix is ​​to ensure that the MIMO beamforming system can obtain better performance indicators in the far field direction; the weighting factors α1, α2, ...., α K It is positively correlated with the performance of the corresponding channel, that is, the better the performance of the channel, the greater the weighting factor of the corresponding signal matrix.

[0073] One method for determining the weighting factor is as follows: the weighting factor α of the signal matrix corresponding to the jth channel is j for:

[0074]

[0075]

[0076] Among them, K is the number of channels (that is, the number of channels for time-sharing acquisition), ACPR j is the adjacent channel power ratio of the jth channel, which is based on the jth feedback signal y j The calculated value is the ratio of the signal power of the main channel to the signal power of the adjacent channel. It should be noted that the method for calculating the weighting factor is only an example, and other methods can be used for calculation as needed.

[0077] The parameter estimation module 1824 establishes an error objective function based on matrices W and V, V=W*c, uses a parameter solving method to extract correction parameters, and performs multiple iterations to obtain the optimal correction parameter value c. The correction parameters include nonlinear distortion pre-compensation parameters, mismatch compensation parameters, and mutual coupling compensation parameters. The link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed feedback signal. The link imbalance calibration parameters are used to eliminate IQ imbalance, local oscillator leakage, and unevenness of the link. The method for determining the link imbalance calibration parameters can refer to the relevant technology.

[0078] The parameter solving iterative algorithm used includes but is not limited to the following methods:

[0079]

[0080] c(n)=μ(n)·x(n)·e H (n)

[0081]

[0082] In the above formula, μ and λ are the adjustment factors of the iterative algorithm; c(n) is the parameter estimation value of the nth iteration; x(n) is the input signal of the distortion pre-correction model; (·) H is the conjugate transpose operation of the complex vector; e(n) is the fitting error value at the current moment obtained by the pre-correction model; h(n) is the weighting coefficient of the error signal e(n).

[0083] In one embodiment, if Figure 5 As shown, the signal distortion correction network module 100 includes: a first filter group 101, a nonlinear distortion correction module 102, a link imbalance correction module 108 and a second filter group 109, wherein:

[0084] The first filter group 101 is used to perform rate conversion processing on the forward signal and output the converted forward signal to the nonlinear distortion correction module 102; the first filter group 101, for example, includes multiple FIR (Finite Impulse Response) filters, which perform filtering processing on the forward signal x in parallel and convert the output signals s1, s2, ..., s M Arrange in time order to get the signal x a , thereby completing the rate conversion processing of the forward signal x.

[0085] The nonlinear distortion correction module 102 is used to correct the transformed forward signal according to the nonlinear distortion precompensation parameter in the correction parameter to generate a first correction signal and output the first correction signal to the link imbalance correction module 108; specifically, the output signal of the first filter group 101 is subjected to nonlinear basis function construction and multiplication operation of each basis function with the nonlinear distortion precompensation parameter, so as to correct the signal distortion introduced by the nonlinear characteristics of the power amplifier. One implementation method is: using the output signal x of the first filter group 101 a To construct nonlinear basis functions T1, T2, T3, ..., T N , N is the number of nonlinear basis functions; then, these nonlinear basis functions are multiplied by the corresponding model coefficients (i.e., nonlinear distortion precompensation parameters) to obtain the first correction signal x d, used to offset the nonlinear signal distortion generated by the forward signal excitation power amplifier. The specific presentation form of the nonlinear basis function depends on the adopted signal distortion pre-compensation model, which can be multiple sets of low-order filters, Volterra series, memory crossover polynomials, neural networks (such as BP networks, ART networks, etc.), wavelet networks and support vector bases, etc. The signal distortion pre-compensation model of the present invention is not limited to the above models.

[0086] The link imbalance correction module 108 is used to correct the first correction signal x according to the link imbalance calibration parameter in the correction parameter. d Processing is performed to output the fifth correction signal x h To the second filter group 109; to achieve correction processing for I / Q imbalance, local oscillator leakage, and link unevenness.

[0087] The second filter group 109 is used to filter the fifth correction signal x h The second filter group 109 includes, for example, a plurality of FIR filters, which are used to process the signal x. h Parallel filtering is performed to complete the rate conversion of the transmission signal to meet the rate requirement of the subsequent second conversion module 110.

[0088] In another embodiment, if Figure 6 As shown, the signal distortion correction network module 100 includes: a first filter group 101, a nonlinear distortion correction module 102, a merging module 107, a link imbalance correction module 108 and a second filter group 109, and also includes at least one of a mismatch correction unit composed of a mismatch filter module 103 and a mismatch error compensation module 105, and a mutual coupling correction unit composed of a mutual coupling filter module 104 and a mutual coupling error compensation module 106, that is, it can only include the mismatch filter module 103 and the mismatch error compensation module 105, and can also only include the mutual coupling filter module 104 and the mutual coupling error compensation module 106. , or, it includes a mismatch filter module 103 and a mismatch error compensation module 105, a mutual coupling filter module 104 and a mutual coupling error compensation module 106 at the same time, that is, only mismatch correction can be performed, only mutual coupling correction can be performed, or both mismatch correction and mutual coupling correction can be performed. The functions of the first filter group 101, the nonlinear distortion correction module 102 and the second filter group 109 refer to the previous embodiment and are not repeated here. It should be noted that, in this embodiment, the first correction signal output by the nonlinear distortion correction module 102 is not input into the link imbalance correction module 108, but is input into the merging module 107. Wherein:

[0089] The mismatch filtering module 103 is used to filter the transformed forward signal (causal filtering or non-causal filtering) according to the mismatch filtering coefficient in the correction parameter to generate a first intermediate signal x b , where x b Used to simulate the standing wave signal generated by the load mismatch at the output end of the power amplifier module, and output the first intermediate signal x b To the mismatch error compensation module 105; that is, the mismatch filtering module 103 is used to simulate the standing wave signal generated by the load mismatch at the output end of the power amplifier module;

[0090] The mismatch error compensation module 105 is used to correct the first intermediate signal x according to the mismatch compensation parameter in the correction parameter. b Correction is performed to obtain the second correction signal x e , output the second correction signal x e To the merging module 107; specifically, the output signals of the mismatch filtering module and the first filter group are subjected to nonlinear basis function construction and multiplication operation of each basis function with the mismatch filtering coefficient, so as to correct the signal distortion introduced by the load mismatch. First, the output signal x of the first filter group 101 is used a and the output signal x of the mismatch filter module 103 b , to construct nonlinear basis functions S1,S2,S3,...,S K , K is the number of nonlinear basis functions; then, these nonlinear basis functions are multiplied by the corresponding coefficients to obtain the error compensation signal x e ; Used to offset the nonlinear signal distortion caused by the standing wave signal exciting the power amplifier caused by the output load mismatch. Among them, the specific presentation form of the nonlinear basis function depends on the mismatch error pre-compensation model adopted, and the mismatch error pre-compensation model can be multiple groups of low-order filters, Volterra series, memory cross polynomials, neural networks (such as BP networks, ART networks, etc.), wavelet networks and support vector bases, etc. The mismatch error pre-compensation model of the present invention is not limited to the above models. In this embodiment, the distortion caused by the load mismatch is corrected to reduce the impact of the output load mismatch on the system performance.

[0091] The mutual coupling filtering module 104 is used to filter (causal filtering or non-causal filtering) the transformed forward signal according to the mutual coupling filtering coefficient in the correction parameter to generate a second intermediate signal x c , where x c Used to simulate the mutual coupling interference signal between the output antennas of the power amplifier module, and output the second intermediate signal x c to the mutual coupling error compensation module 106; the mutual coupling filter module 104 is used to simulate the mutual coupling interference signal between the output antennas of the power amplifier module;

[0092] The mutual coupling error compensation module 106 is used to compensate the second intermediate signal x according to the mutual coupling compensation parameter in the correction parameter. c Correction is performed to obtain a third correction signal x f , output the third correction signal x f To the merging module 107; specifically, the output signals of the mutual coupling filter module and the first filter group are subjected to the construction of nonlinear basis functions and the multiplication operation of each basis function with the mutual coupling filter coefficient, so as to correct the signal distortion introduced by the mutual coupling between the antennas. Specifically, firstly, the output signal x of the first filter group 101 is used a and the output signal x of the mutual coupling filter module 104 c , to construct nonlinear basis functions R1, R2, R3, ..., R M , M is the number of nonlinear basis functions; then, these nonlinear basis functions are multiplied by the corresponding coefficient values ​​to obtain the mutual coupling error compensation signal x f ; Used to offset the nonlinear distortion caused by the mutual coupling interference signal between antennas exciting the power amplifier. Among them, the specific presentation form of the nonlinear basis function depends on the mutual coupling error pre-compensation model adopted, and the mutual coupling error pre-compensation model can be multiple groups of low-order filters, Volterra series, memory cross polynomials, neural networks (such as BP networks, ART networks, etc.), wavelet networks and support vector bases, etc. The mutual coupling error pre-compensation model of the present invention is not limited to the above models. In this embodiment, the distortion caused by the mutual coupling interference between antennas is corrected, which can reduce the impact of the mutual coupling interference between antennas on the system performance.

[0093] The merging module 107 is used to merge the input signals to obtain a fourth correction signal x g , output the fourth correction signal x g To the link imbalance correction module 108, wherein the input signal may be the first correction signal and the second correction signal, or the first correction signal and the third correction signal, or the first correction signal, the second correction signal and the third correction signal. The merging method is to add the input signals.

[0094] The link imbalance correction module 108 is used to correct the fourth correction signal x according to the link imbalance calibration parameter in the correction parameter. g Processing is performed to output the fifth correction signal x h To the second filter group 109; specifically, filtering processing (linear processing) is performed to correct I / Q imbalance, in-band unevenness, local oscillator leakage, etc. of the system transmission link.

[0095] The signal distortion correction network module provided in this embodiment can correct the signal distortion introduced by nonlinear analog devices, output load mismatch and coupling interference between antennas.

[0096] For a MIMO millimeter wave communication system using analog beamforming technology, an embodiment of the present invention proposes a nonlinear system signal distortion pre-correction device. In the correction device, a signal sampling controller is used to control the signal coupling network, collect and feed back the output signals of each power amplifier, and after multi-channel data linear processing operations, calculation of mismatch factors and mutual coupling factors, establishment of a signal distortion pre-correction model, and extraction of model parameters, the signal distortion of the MIMO system is effectively pre-compensated, and the mutual coupling interference between antennas can be offset, which can further ensure the performance advantages of the MIMO beamforming system. Compared with traditional pre-distortion technology, this device can better solve the problem of pre-compensation of nonlinear system signal distortion in application scenarios where one pre-distorter corresponds to multiple power amplifiers.

[0097] The signal distortion pre-correction device provided by the embodiment of the present invention has the following advantages:

[0098] 1) The signal sampling controller controls the coupling network to collect and feedback the output signals of each power amplifier in a time-sharing manner, which has low requirements on the consistency of the characteristics of each power amplifier in the system;

[0099] 2) During the correction parameter extraction process, weighted processing is performed on the constructed signal matrix to optimize the performance index in the beamforming direction;

[0100] 3) Parallel linear processing of the feedback signals of each power amplifier can eliminate the impact of the delay, phase and amplitude differences of the coupled feedback link on the performance of the distortion correction algorithm;

[0101] 4) Since the feedback signals of each power amplifier can be processed linearly, the beam angle can be any angle when extracting the correction parameters, not limited to the 0° beam, and the distortion pre-correction performance in the far-field beam direction can be guaranteed.

[0102] 5) It can reduce the impact of output load mismatch and mutual coupling interference between antennas on system performance.

[0103] Compared with the distortion correction device in the related art, the correction device provided in this embodiment proposes a new signal distortion pre-correction algorithm, which can realize the correction of the overall nonlinear distortion of multiple parallel power amplifiers, and can also reduce the influence of factors such as output load mismatch and antenna coupling interference; it uses a feedback link channel to complete the time-sharing acquisition of the output signals of each channel. The solution provided in this embodiment can be applied to wireless communication systems using analog beamforming technology, such as millimeter wave communication systems, and can also be applied to communication systems where the power amplifier characteristics are greatly affected by standing waves.

[0104] In the correction device provided by the embodiment of the present invention, the input data signal (i.e., the forward signal) can be both a service signal and a training sequence, and the consistency requirement for the characteristics of each power amplifier is relatively low. Moreover, when the consistency of the characteristics of each power amplifier is good, the feedback link selection switch in the signal coupling network module can be fixed to a certain channel, and the feedback signal can be collected and subsequent operation processing can be performed.

[0105] An embodiment of the present invention provides a signal distortion pre-correction method, such as Figure 7 As shown, including:

[0106] Step 701, performing time-sharing acquisition on output signals of multiple channels of a power amplifier module to obtain multiple analog feedback signals;

[0107] Step 702, converting the multiple analog feedback signals into multiple digital feedback signals;

[0108] Step 703, determining a correction parameter according to the multiple digital feedback signals and the input forward signal;

[0109] Step 704, correcting the forward signal based on the correction parameter to obtain a pre-corrected signal;

[0110] In another embodiment, it also includes:

[0111] Step 705: convert the pre-correction signal into an analog signal, and output the analog signal to an analog beamforming processor.

[0112] In one embodiment, determining the correction parameter according to the plurality of digital feedback signals and the input forward signal comprises:

[0113] The forward signal and the digital feedback signal are linearly processed to obtain the linearly processed forward signal and the linearly processed digital feedback signal. Based on the signal distortion pre-correction model, multiple signal matrices are constructed based on the linearly processed forward signal, the linearly processed digital feedback signal and the pre-correction signal, wherein the signal matrix corresponds to the collected channel, the multiple signal matrices are weighted using weighting factors respectively, a combination matrix is ​​generated using the weighted signal matrix, nonlinear distortion pre-compensation parameters, mismatch compensation parameters, and mutual coupling compensation parameters are obtained based on the combination matrix, and link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed digital feedback signal, wherein the weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix.

[0114] In one embodiment, the weighting factor α of the signal matrix corresponding to the jth channel is j for:

[0115]

[0116]

[0117] Where K is the number of channels, ACPR j is the ACPR of the jth channel, that is, the ratio of the signal power of the adjacent channel to the main channel of the jth analog channel.

[0118] In one embodiment, determining the correction parameter according to the plurality of digital feedback signals and the input forward signal comprises:

[0119] Performing linear processing on the input forward signal and the digital feedback signal to obtain a linearly processed forward signal and a linearly processed digital feedback signal;

[0120] Determine the mismatch filter coefficient and the mutual coupling filter coefficient according to the power amplifier model of the power amplifier module, the forward signal after linear processing, and the digital feedback signal after linear processing;

[0121] Constructing a plurality of signal matrices based on a signal distortion pre-correction model, a mismatch filter coefficient, a mutual coupling filter coefficient, a linearly processed forward signal, a linearly processed digital feedback signal, and the pre-correction signal, wherein the signal matrix corresponds to the collected channel;

[0122] The multiple signal matrices are weighted using weighting factors respectively, and a combination matrix is ​​generated using the weighted signal matrix. Nonlinear distortion pre-compensation parameters, mismatch compensation parameters, and mutual coupling compensation parameters are obtained based on the combination matrix. Link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed digital feedback signal. The weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix.

[0123] In one embodiment, the signal matrix corresponding to the j-th channel is:

[0124]

[0125]

[0126] in, Among them, x j is the forward signal after linear processing, z j is the pre-correction signal, y′ j is the digital feedback signal after linear processing, h a is the mismatch filter coefficient, h b is the mutual coupling filter coefficient, L and Q are the memory depth and nonlinear order of the signal distortion pre-correction model respectively; N+1 is the data length used to construct the signal matrix.

[0127] In one embodiment, the correcting the input forward signal based on the correction parameter to obtain the pre-corrected signal comprises:

[0128] Performing rate conversion processing on the forward signal to obtain a transformed forward signal;

[0129] Correcting the transformed forward signal according to the nonlinear distortion pre-compensation parameter in the correction parameter to generate a first correction signal;

[0130] Processing the first correction signal according to the link imbalance calibration parameter in the correction parameter to obtain a fifth correction signal;

[0131] The fifth corrected signal is subjected to rate conversion processing to obtain the pre-corrected signal, and the pre-corrected signal is output.

[0132] In one embodiment, the correcting the input forward signal based on the correction parameter to obtain the pre-corrected signal comprises:

[0133] Performing rate conversion processing on the forward signal to obtain a transformed forward signal;

[0134] Correcting the transformed forward signal according to the nonlinear distortion pre-compensation parameter in the correction parameter to generate a first correction signal;

[0135] Filtering the transformed forward signal according to the mismatch filter coefficient in the correction parameter to generate a first intermediate signal, and correcting the first intermediate signal according to the mismatch compensation parameter in the correction parameter to obtain a second correction signal;

[0136] Filtering the transformed forward signal according to the mutual coupling filter coefficient in the correction parameter to generate a second intermediate signal, and correcting the second intermediate signal according to the mutual coupling compensation parameter in the correction parameter to obtain a third correction signal;

[0137] Combining the first correction signal and the second correction signal, or the first correction signal and the third correction signal, or the first correction signal, the second correction signal and the third correction signal to obtain a fourth correction signal;

[0138] Processing the fourth correction signal according to the link imbalance calibration parameter in the correction parameter to generate a fifth correction signal;

[0139] The fifth corrected signal is subjected to rate conversion processing to obtain the pre-corrected signal, and the pre-corrected signal is output.

[0140] An embodiment of the present invention provides a signal distortion pre-correction system, referring to Figure 1, comprising the signal distortion pre-correction device in any embodiment, further comprising a second conversion module 110, an analog beamforming processor 120 and a power amplifier module 130, wherein:

[0141] The signal distortion pre-correction device is further used to output the pre-correction signal to the second conversion module 110;

[0142] The second conversion module 110 is used to convert the pre-correction signal into an analog signal and output it to the analog beamforming processor 120;

[0143] The analog beamforming processor 120 is used to perform amplitude and phase weighting processing on the input analog signal, generate a weighted analog signal and output it to the power amplifier module; the weighting processing is to facilitate the subsequent output signal of the antenna array to form a beam in a preset direction;

[0144] The power amplifier module 130 is used to amplify the weighted analog signal and then output it.

[0145] The device of the present invention is described in detail below in conjunction with the accompanying drawings and embodiments applied to a 5G millimeter wave communication system.

[0146] like Figure 8 As shown, an embodiment of the present invention provides a signal distortion pre-correction method, which is applied to the signal distortion pre-correction device provided in an embodiment of the present invention, including:

[0147] Step 801: The signal sampling controller 160 realizes time-sharing collection and feedback of output signals of each power amplifier by controlling the signal coupling network module 140, and outputs an analog feedback signal to the first conversion module 170;

[0148] Step 802: The first conversion module 170 converts the analog feedback signal into a digital feedback signal and sends it to the correction parameter trainer 180;

[0149] Step 803: the correction parameter trainer 180 obtains correction parameters according to the digital feedback signal and the input forward signal; the correction parameters include: nonlinear distortion pre-compensation parameters, mismatch compensation parameters, mutual coupling compensation parameters and link imbalance calibration parameters.

[0150] Specifically, first based on the forward signal x=[x1,x2,...,x K ], the output signal y of the first conversion module 170 is y=[y1, y2, ..., y K ] Perform multi-channel data linear processing operations to obtain pre-processed signals;

[0151] Then, the obtained pre-processed signal is used to solve the mismatch filter coefficient and the mutual coupling filter coefficient;

[0152] Finally, the objective function of the signal distortion correction model is constructed, and the final correction parameter c is obtained by using the parameter solving iterative algorithm, where K is the number of channels for time-sharing acquisition, and when one channel includes one power amplifier, it is the number of parallel power amplifiers. It should be noted that K is less than or equal to the number of channels of the power amplifier module 130.

[0153] In addition, a link imbalance calibration parameter is obtained according to the digital feedback signal and the input forward signal.

[0154] Step 804: According to the correction parameter c provided by the correction parameter trainer 180, the correction parameter in the signal distortion correction network module 100 is updated in real time; the signal distortion correction network module 100 performs nonlinear distortion and linear distortion pre-correction processing on the signal x under the condition of high sampling rate according to the forward signal x and its amplitude information, and sends the obtained pre-correction signal z to the second conversion module 110; the second conversion module 110 converts the pre-correction signal (which is a digital signal) into an analog signal and sends it to the analog beamforming processor 120;

[0155] Step 805: The analog beamforming processor 120 performs weighted processing on the phase and amplitude of the input signal and then outputs the signal. The output signal of the analog beamforming processor 120 is amplified by the power amplifier of the power amplifier module 130 and then directly transmitted to the antenna array module 150 and radiated into space. The analog beamforming processor 120 can perform weighted processing on the phase and amplitude of the input signal using analog devices such as an adjustable attenuator and a phase shifter.

[0156] Fig. 9 FIG. 1 is a schematic diagram of a signal distortion pre-correction composite system provided by an embodiment of the present invention. Fig. 9 As shown, the signal distortion pre-correction composite system includes multiple signal distortion pre-correction systems, and the multiple signal distortion pre-correction systems reuse the same correction parameter trainer 180, and the correction parameter trainer 180 is connected to the relevant modules in each signal distortion pre-correction system through a selection switch. Specifically, it is connected to the input end of each signal distortion pre-correction system through a first selection switch 801, connected to the signal distortion correction network module through a second selection switch 802, and connected to the first conversion module 170 through a third selection switch 803.

[0157] Fig.10 The figure shows a specific embodiment of a signal distortion pre-correction device suitable for a MIMO millimeter wave communication system according to the present invention.

[0158] In this embodiment, in the MIMO downlink transmission system using analog shaping technology, the number of digital channels and analog RF channels are set to 4 and 32 respectively. Each digital channel corresponds to 8 analog RF channels, and the output of one analog RF channel is connected to two antenna arrays. At the same time, it is assumed that the transmission signals of the four digital channels are all 400MHz baseband signals containing 10 subcarriers, that is, x a 、x b 、x c With x d .

[0159] Taking the signal transmission of digital channel 1 as an example, the correction process of the signal distortion of the nonlinear system is described in detail. The signal transmission of the remaining digital channels is similar and will not be repeated here. The related signals of different digital channels are distinguished by the subscripts a, b, c, and d.

[0160] Step 1001: The signal sampling controller 160a controls the signal coupling network module 140a to collect the output signals of each power amplifier in a time-sharing manner and obtain the feedback signal y i (t), (i=1,2,...,8), and at the same time, the feedback signal y is fed back by the feedback VGA200a. i The gain of (t) is adjusted to ensure that the amplitude of the feedback signal is not too small to affect the extraction and training of distortion correction parameters;

[0161] Step 1002: Module 210a generates a local oscillation signal with a frequency of fr = 27 GHz, and transmits the RF signal y i (t) Move to zero intermediate frequency and output digital feedback signal y after passing through ADC module 170a a =[y1,y2,...,y8];

[0162] Step 1003: The calibration parameter trainer 180 selects open loops X and Y to be connected to the forward link and feedback link of the digital channel 1, and first respectively performs calibration on the forward signal x i and feedback acquisition signal y i , (i=1,2,...,8) perform data linear processing operations such as delay alignment, mirror calibration, and complex gain compensation to obtain the preprocessed signal y i ', (i = 1, 2, ..., 8); then, based on the power amplifier model, using the forward signal x i For signal y i 'Perform approximate fitting and use LS method to calculate the mismatch filter coefficient h a and the mutual coupling filter coefficient h b; secondly, construct signal matrices W1, W2, ..., W8 and V1, V2, ..., V8 according to the adopted signal distortion pre-correction model, determine the size of weighting factor values ​​α1, α2, ..., α8, and perform weighted processing on the constructed matrices to obtain composite matrices W and V; finally, use an iterative method to perform multiple iterative calculations on the distortion pre-correction parameters to obtain the optimal value c of the correction parameter;

[0163] Step 1004: copy the correction parameter value c obtained by the correction parameter trainer 180 to the signal distortion correction network module 100a through the selection switch Z, and update the correction parameter in the signal distortion correction network module 100a accordingly;

[0164] Step 1005: According to the 400 MHz baseband signal x sent on digital channel 1 a =[x1,x2,...,x K ] and its amplitude, the signal distortion correction network module 100a performs mismatch and mutual coupling filtering under high sampling rate conditions, constructs nonlinear basis functions, and multiplies them with corresponding correction parameters, thereby completing the forward signal x a The distortion pre-compensation processing operation is performed to obtain the pre-corrected signal z a , the signal is converted into an analog signal after being processed by module 110a;

[0165] Step 1006: Module 210a provides a frequency of f r =27GHz local oscillation signal, the output signal of DAC module 110a is moved to the RF end; after the RF transmission signal is processed by power divider group 190a, it is transmitted in parallel on 8 analog RF channels, and the signal power becomes one eighth of the original; the analog beamforming processor 120a uses the steering vector [β0,β1e -j2πdsinθ ,...,β7e -j7*2πdsinθ ] The amplitude and phase of the RF transmission signal in the 8 analog channels are weighted to form a beam in a specific direction; and the power amplifier module 130a uses 8 parallel power amplifiers to amplify the output signal of the analog beamforming processor 120a, directly transmit it to the antenna array module 150a and radiate it into space. Note that β0, β1, ..., β7 are the signal amplitude weighted values ​​on each analog RF channel, d is the distance between antenna elements, and θ is the beam direction angle.

[0166] In a specific embodiment of the correction device, the antenna array modules 150a, 150b, 150c and 150d all use 16 antenna elements, and the layout is a 4*4 antenna array; the four digital channels share a correction parameter trainer, which is connected to the forward transmission link, signal distortion correction network and coupling feedback link of each digital channel through a selection switch, which can not only reduce the hardware resource requirements of the transmitter, but also greatly reduce the power consumption of the system.

[0167] Here, the present invention has been described in more detail in conjunction with specific embodiments. The purpose of providing the detailed description of the above embodiments is to enable those skilled in the art to make or use the present invention. Various modifications to the above embodiments are readily understood by those skilled in the art.

[0168] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

Claims

1. A signal distortion pre-correction device, comprising: A signal distortion correction network module, a correction parameter trainer, a data acquisition module and a first conversion module, wherein the data acquisition module is connected to an external power amplifier module, wherein: The data acquisition module is used to perform time-sharing acquisition on the output signals of multiple channels of the power amplifier module, and output analog feedback signals to the first conversion module; The first conversion module is used to convert the multiple analog feedback signals into multiple digital feedback signals and send them to the correction parameter trainer; The correction parameter trainer is used to determine the correction parameters according to the multiple digital feedback signals and the input forward signal, and output the correction parameters to the signal distortion correction network module; The signal distortion correction network module is used to output a pre-corrected signal after correcting the forward signal based on the correction parameter; The correction parameter trainer comprises: A data linear processing module, used for performing linear processing on the forward signal and the digital feedback signal to obtain a linearly processed forward signal and a linearly processed digital feedback signal; A correction parameter calculation module is used to construct multiple signal matrices based on a preset signal distortion pre-correction model, a linearly processed forward signal, a linearly processed digital feedback signal, and the pre-correction signal. The signal matrix corresponds to the collected channel. The multiple signal matrices are weighted using weighting factors respectively, and a combination matrix is ​​generated using the weighted signal matrix. Nonlinear distortion pre-compensation parameters are obtained based on the combination matrix, and link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed digital feedback signal. The weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix.

2. The signal distortion pre-correction device according to claim 1, characterized in that: No. The weighting factor of the signal matrix corresponding to the channels for ， ， ； ， in is the number of channels, It is Adjacent channel power ratio of each channel.

3. The signal distortion pre-correction device according to claim 1, characterized in that: The correction parameter trainer includes a data linear processing module and a correction parameter calculation module, wherein: The data linear processing module is used to perform linear processing on the forward signal and the digital feedback signal to obtain a linearly processed forward signal and a linearly processed digital feedback signal; The correction parameter calculation module is used to determine the mismatch filter coefficient and the mutual coupling filter coefficient according to the power amplifier model of the power amplifier module, the forward signal after linear processing and the digital feedback signal after linear processing; construct multiple signal matrices based on the preset signal distortion pre-correction model, the mismatch filter coefficient, the mutual coupling filter coefficient, the forward signal after linear processing, the digital feedback signal after linear processing and the pre-correction signal, wherein there is a corresponding relationship between the signal matrix and the collected channels, the multiple signal matrices are weighted using weighting factors respectively, and a combination matrix is ​​generated using the signal matrix after weighted processing, nonlinear distortion pre-compensation parameters, mismatch compensation parameters and mutual coupling compensation parameters are obtained based on the combination matrix, and link imbalance calibration parameters are determined based on the forward signal after linear processing and the digital feedback signal after linear processing, and the weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix.

4. The signal distortion pre-correction device according to claim 3, characterized in that: No. The signal matrix corresponding to each channel is: in, , is the forward signal after linear processing, is the pre-correction signal, is the digital feedback signal after linear processing, is the mismatch filter coefficient, is the mutual coupling filter coefficient, , are the memory depth and nonlinear order of the signal distortion pre-correction model, respectively; The data length used to construct the signal matrix.

5. The signal distortion pre-correction device according to claim 1, characterized in that: The signal distortion correction network module comprises: a first filter group, a nonlinear distortion correction module, a link imbalance correction module and a second filter group, wherein: The first filter group is used to perform rate conversion processing on the forward signal and output the converted forward signal to the nonlinear distortion correction module; The nonlinear distortion correction module is used to correct the transformed forward signal according to the nonlinear distortion pre-compensation parameter in the correction parameter to generate a first correction signal and output the first correction signal to the link imbalance correction module; The link imbalance correction module is used to process the first correction signal according to the link imbalance calibration parameter in the correction parameter, and output a fifth correction signal to the second filter group; The second filter group is used to perform rate conversion processing on the fifth correction signal and output the pre-correction signal.

6. The signal distortion pre-correction device according to claim 3, characterized in that: The signal distortion correction network module includes: a first filter group, a nonlinear distortion correction module, a merging module, a link imbalance correction module and a second filter group, and also includes at least one of a mismatch correction unit composed of a mismatch filter module and a mismatch error compensation module, and a mutual coupling correction unit composed of a mutual coupling filter module and a mutual coupling error compensation module, wherein: The first filter group is used to perform rate conversion processing on the forward signal, and output the converted forward signal to the nonlinear distortion correction module, the mismatch filtering module and the mutual coupling filtering module; The nonlinear distortion correction module is used to correct the transformed forward signal according to the nonlinear distortion pre-compensation parameter in the correction parameter to generate a first correction signal and output the first correction signal to the merging module; The mismatch filtering module is used to filter the transformed forward signal according to the mismatch filtering coefficient in the correction parameter to generate a first intermediate signal, and output the first intermediate signal to the mismatch error compensation module; The mismatch error compensation module is used to correct the first intermediate signal according to the mismatch compensation parameter in the correction parameter to obtain a second correction signal, and output the second correction signal to the merging module; The mutual coupling filtering module is used to filter the transformed forward signal to a second intermediate signal according to the mutual coupling filtering coefficient in the correction parameter, and output the second intermediate signal to the mutual coupling error compensation module; The mutual coupling error compensation module is used to correct the second intermediate signal according to the mutual coupling compensation parameter in the correction parameter to obtain a third correction signal, and output the third correction signal to the merging module; The merging module is used to merge the input signals to obtain a fourth correction signal, and output the fourth correction signal to the link imbalance correction module; The link imbalance correction module is used to process the fourth correction signal according to the link imbalance calibration parameter in the correction parameter, and output a fifth correction signal to the second filter group; The second filter group is used to perform rate conversion processing on the fifth correction signal and output the pre-correction signal.

7. The signal distortion pre-correction device according to any one of claims 1 to 6, characterized in that: The data acquisition module includes: a signal coupling network module and a signal sampling controller, wherein the signal coupling network module is connected to an external power amplifier module, wherein: The signal sampling controller is used to control the signal coupling network module to achieve time-sharing acquisition of output signals of multiple channels of the power amplifier module; The signal coupling network module is used to perform time-sharing coupling on the output signals of multiple channels of the power amplifier module according to the control of the signal sampling controller, and output analog feedback signals to the first conversion module.

8. A signal distortion precorrection method, comprising: The output signals of multiple channels of the power amplifier module are collected in time division to obtain multiple analog feedback signals; converting the plurality of analog feedback signals into a plurality of digital feedback signals; determining a correction parameter according to the plurality of digital feedback signals and an input forward signal; Correcting the forward signal based on the correction parameter to obtain a pre-corrected signal; Wherein, determining the correction parameter according to the multiple digital feedback signals and the input forward signal includes: Performing linear processing on the input forward signal and the digital feedback signal to obtain a linearly processed forward signal and a linearly processed digital feedback signal; Constructing a plurality of signal matrices based on a signal distortion pre-correction model, a linearly processed forward signal, a linearly processed digital feedback signal, and the pre-correction signal, wherein the signal matrices correspond to the collected channels; The multiple signal matrices are weighted using weighting factors respectively, a combination matrix is ​​generated using the weighted signal matrices, nonlinear distortion pre-compensation parameters are obtained based on the combination matrix, link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed digital feedback signal, and the weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix.

9. The signal distortion pre-correction method according to claim 8, characterized in that: No. The weighting factor of the signal matrix corresponding to the channels for ， ， ； ， in is the number of channels, It is Adjacent channel power ratio of each channel.

10. The signal distortion pre-correction method according to claim 8, characterized in that: Determining the correction parameter according to the plurality of digital feedback signals and the input forward signal comprises: Performing linear processing on the input forward signal and the digital feedback signal to obtain a linearly processed forward signal and a linearly processed digital feedback signal; Determine the mismatch filter coefficient and the mutual coupling filter coefficient according to the power amplifier model of the power amplifier module, the forward signal after linear processing, and the digital feedback signal after linear processing; Constructing a plurality of signal matrices based on a signal distortion pre-correction model, a mismatch filter coefficient, a mutual coupling filter coefficient, a linearly processed forward signal, a linearly processed digital feedback signal, and the pre-correction signal, wherein the signal matrix corresponds to the collected channel; The multiple signal matrices are weighted using weighting factors respectively, and a combination matrix is ​​generated using the weighted signal matrices. Nonlinear distortion pre-compensation parameters, mismatch compensation parameters and mutual coupling compensation parameters are obtained based on the combination matrix. In addition, link imbalance calibration parameters are determined based on the linearly processed forward signal and the linearly processed digital feedback signal. The weighting factor of the signal matrix is ​​positively correlated with the performance of the channel corresponding to the signal matrix.

11. The signal distortion pre-correction method according to claim 10, characterized in that: No. The signal matrix corresponding to each channel is: in, , is the forward signal after linear processing, is the pre-correction signal, is the digital feedback signal after linear processing, is the mismatch filter coefficient, is the mutual coupling filter coefficient, , are the memory depth and nonlinear order of the signal distortion pre-correction model, respectively; The data length used to construct the signal matrix.

12. The signal distortion pre-correction method according to claim 8, characterized in that: The correcting the input forward signal based on the correction parameter to obtain the pre-corrected signal comprises: Performing rate conversion processing on the forward signal to obtain a transformed forward signal; Correcting the transformed forward signal according to the nonlinear distortion pre-compensation parameter in the correction parameter to generate a first correction signal; Processing the first correction signal according to the link imbalance calibration parameter in the correction parameter to obtain a fifth correction signal; The fifth corrected signal is subjected to rate conversion processing to obtain the pre-corrected signal, and the pre-corrected signal is output.

13. The signal distortion pre-correction method according to claim 10, characterized in that: The correcting the input forward signal based on the correction parameter to obtain the pre-corrected signal comprises: Performing rate conversion processing on the forward signal to obtain a transformed forward signal; Correcting the transformed forward signal according to the nonlinear distortion pre-compensation parameter in the correction parameter to generate a first correction signal; Filtering the transformed forward signal according to the mismatch filter coefficient in the correction parameter to generate a first intermediate signal, and correcting the first intermediate signal according to the mismatch compensation parameter in the correction parameter to obtain a second correction signal; Filtering the transformed forward signal according to the mutual coupling filter coefficient in the correction parameter to generate a second intermediate signal, and correcting the second intermediate signal according to the mutual coupling compensation parameter in the correction parameter to obtain a third correction signal; Combining the first correction signal and the second correction signal, or the first correction signal and the third correction signal, or the first correction signal, the second correction signal and the third correction signal to obtain a fourth correction signal; Processing the fourth correction signal according to the link imbalance calibration parameter in the correction parameter to generate a fifth correction signal; The fifth corrected signal is subjected to rate conversion processing to obtain the pre-corrected signal, and the pre-corrected signal is output.

14. A signal distortion pre-correction system, characterized in that: The device comprises the signal distortion pre-correction device according to any one of claims 1 to 7, and further comprises a second conversion module, an analog beamforming processor and a power amplifier module, wherein: The signal distortion pre-correction device is further used to output the pre-correction signal to the second conversion module; The second conversion module is used to convert the pre-correction signal into an analog signal, and output the analog signal to the analog beamforming processor; The analog beamforming processor is used to perform amplitude and phase weighting processing on the input analog signal, generate a weighted analog signal and output it to the power amplifier module; The power amplifier module is used to amplify the weighted analog signal and then output it.

15. A signal distortion pre-correction composite system, characterized in that: It comprises a plurality of signal distortion pre-correction systems as claimed in claim 14, and the plurality of signal distortion pre-correction systems reuse the same correction parameter trainer, and the correction parameter trainer is connected to the relevant modules in each signal distortion pre-correction system through a selection switch.

Images