Multi-source error calibration system and method for broadband vector modulation system

Abstract

The invention provides a broadband vector modulation system multi-source error calibration system and method, and relates to the technical field of electronic testing. The calibration system comprises a vector input end, an error / noise parameter adaptive acquisition-identification-updating channel, a broadband vector modulation system output channel and a vector output end, and the error / noise parameter adaptive acquisition-identification-updating channel comprises an error / noise parameter identification-updating module and a characteristic parameter acquisition module. An output channel of the broadband vector modulation system comprises a system error / noise parameter calibration transfer function calculation module connected with the error / noise parameter identification-updating module, a digital pre-distortion broadband calibration module, a vector modulator connected with the characteristic parameter acquisition module and a digital-to-analog converter. According to the multi-source error calibration method for the broadband vector modulation system provided by the invention, accurate calibration of multiple types of error parameters in the broadband vector modulation system is realized, and the calibration speed of the broadband vector modulation system is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic testing, in particular to a wideband vector modulation system multi-source error calibration system and method. BACKGROUND

[0002] The wideband vector modulation system is a core component of the vector signal generator, and in actual application, the EVM index thereof is affected by multiple error parameters such as modulator amplitude / phase imbalance, DAC spur noise, carrier leakage, and power amplifier nonlinearity. The wideband vector modulation system error calibration method optimizes the EVM index of the system through estimation and calibration of the error parameters in the system.

[0003] Patent CN112600522A discloses a method for digital pre-distortion with power-specific capture selection, which makes the model more representative of the entire signal by specifically targeting the lower power area and combining it with the high power capture, thereby improving the signal quality. Patent CN113196653A discloses a multi-band digital compensator for a nonlinear system, which accurately compensates for the nonlinearity of the radio frequency transmission chain and applies a pre-distortion device, uses different sets of real signals derived from individual band signals that make up the input signal, the output of the nonlinear transformation is used as a gain term for the complex signal set, the pre-distortion signal is calculated as an input function superposition, thereby selecting a match for a specific system using the complex signal and the derived real signal, thereby improving the calculation efficiency and reducing the complexity of adapting the pre-distortion by adapting the nonlinear transformation.

[0004] However, the conventional wideband vector modulation system error calibration algorithm can usually only correct a single type of error parameter, such as mixer amplitude / phase imbalance calibration or power amplifier pre-distortion calibration. Therefore, there is an urgent need to propose a wideband vector modulation system multi-source error calibration system and method suitable for estimating and calibrating multiple error parameters such as amplitude / phase imbalance, DAC spur noise, and power amplifier nonlinearity in the wideband vector modulation system. SUMMARY

[0005] In order to solve the problem that the wideband vector modulation system error calibration method in the prior art can only correct a single type of error parameter and meet the accurate calibration of multiple error parameters in the wideband vector modulation system, the present application proposes a wideband vector modulation system multi-source error calibration system and method, which realizes accurate calibration of multiple error parameters such as mixer amplitude / phase imbalance, DAC spur noise, and power amplifier nonlinearity in the wideband vector modulation system.

[0006] To achieve the above purpose, the present application adopts the following technical solutions:

[0007] A wideband vector modulation system multi-source error calibration system, comprising a vector input end, an error / noise parameter adaptive acquisition-identification-update path, a wideband vector modulation system output path and a vector output end;

[0008] The vector input end is connected with the error / noise parameter adaptive acquisition-identification-update path and the wideband vector modulation system output path respectively, and the wideband vector modulation system output path is connected with the vector output end.

[0009] The error / noise parameter adaptive acquisition-identification-update path comprises an error / noise parameter identification-update module and a characteristic parameter acquisition module, and the error / noise parameter identification-update module is connected with the vector input end and the characteristic parameter acquisition module respectively.

[0010] The wideband vector modulation system output path comprises a system error / noise parameter calibration transfer function calculation module, a digital pre-distortion wideband calibration module, a vector modulator and a digital-to-analog converter, and the system error / noise parameter calibration transfer function calculation module, the digital pre-distortion wideband calibration module, the vector modulator and the digital-to-analog converter are connected in sequence, the system error / noise parameter calibration transfer function calculation module is connected with the vector input end and the error / noise parameter identification-update module respectively, and the vector modulator and the digital-to-analog converter are connected with the characteristic parameter acquisition module and the vector output end respectively.

[0011] Preferably, an analog-to-digital converter and a characteristic parameter extraction matrix S for extracting a sub-target function are arranged in the characteristic parameter acquisition module.

[0012] A wideband vector modulation system multi-source error calibration method, which adopts the wideband vector modulation system multi-source error calibration system as described above, and the calibration process is as follows:

[0013] The input signal x(n) is input to the error / noise parameter identification-update module of the error / noise parameter adaptive acquisition-identification-update path and the system error / noise parameter calibration transfer function calculation module of the wideband vector modulation system output path through the vector input end respectively.

[0014] The input signal x(n) is processed in the output path of the wideband vector modulation system in turn through a system error / noise parameter calibration transfer function calculation module, a digital pre-distortion wideband calibration module, a vector modulator and a digital-to-analog converter to obtain a vector output y. The vector output y is input into a characteristic parameter acquisition module of an error / noise parameter adaptive acquisition-identification-update path. A plurality of sub-objective functions and an error / noise parameter matrix b are extracted by the characteristic parameter acquisition module and input into an error / noise parameter identification-update module. The error / noise parameter identification-update module constructs a system comprehensive objective function by using the input signal x(n) and the plurality of sub-objective functions, and obtains an optimal solution b of the error / noise parameter matrix b by solving the system comprehensive objective function * and input into the system error / noise parameter calibration transfer function calculation module. The system transfer function H is solved by the system error / noise parameter calibration transfer function calculation module and input into the digital pre-distortion wideband calibration module. The signal after pre-distortion correction by the digital pre-distortion wideband calibration module is input into the vector modulator and the digital-to-analog converter. The calibrated vector signal after processing by the vector modulator and the digital-to-analog converter is input into the vector output end. The output signal after calibration by the wideband vector modulation system multi-source error calibration system is output by the vector output end.

[0015] Preferably, when the characteristic parameter acquisition module processes the vector output y processed by the wideband vector modulation system output path, the analog-to-digital converter in the characteristic parameter acquisition module is used to sample the vector output y to obtain a sampling sequence y(n). The sampling sequence y(n) is subjected to fast Fourier transform. The sampling sequence Y after fast Fourier transform is input into a characteristic parameter extraction matrix S. The sub-objective function is extracted by the characteristic parameter extraction matrix S, and the following is obtained:

[0016]

[0017] In the formula, f P (b) is the Pth sub-objective function; b is the error / noise parameter matrix, and the dimension is Mx1.

[0018] Preferably, the error / noise parameter identification-update module constructs a system comprehensive objective function by using the input signal x(n) and the plurality of sub-objective functions, and obtains the following:

[0019]

[0020] In the formula, min is the minimum value function; F(b) is the system comprehensive objective function; i is the serial number, P is the total number of sub-objective functions; P1 is a self-defined number, and the value is less than the total number P of sub-objective functions; w i is the weight coefficient of the ith sub-objective function; f i (b) is the ith sub-objective function.

[0021] Preferably, the whale algorithm is used to solve the optimization problem in the system synthesis objective function, and the maximum number of iterations M is set I , a random number initialization matrix X(t) with a dimension of MxPN is generated, PN column vectors of the random number initialization matrix X(t) are sequentially substituted into the system synthesis objective function for optimization calculation, the minimum value of the system synthesis objective function and the optimal solution vector X corresponding thereto are solved * , the optimal solution matrix X is obtained based on the optimal solution vector X corresponding to each column vector * X * (t) = [X * …X * ]; *

[0022] In each optimization calculation process, random numbers r1, r2 and p between 0 and 1 are randomly generated to obtain a first intermediate variable and a second intermediate variable C, as shown in formula (3):

[0023]

[0024] In the formula, a is a self-defined number, and the value a decreases linearly from 2 to 0;

[0025] When the random number p < 0.5 and , the following is obtained:

[0026]

[0027] In the formula, X(t+1) is the (t+1)th optimization calculation result;

[0028] When the random number p < 0.5 and , the following is obtained:

[0029]

[0030] In the formula, X rand is a random matrix with a dimension of MxPN;

[0031] When p > 0.5, the following is obtained:

[0032] X(t+1) = |X * (t) - X(t)|x e l cos(2πl) + X * (t) (6)

[0033] In the formula, l is a random number between -1 and 1;

[0034] ​After each optimization calculation, the optimal solution vector obtained by the last optimization calculation is replaced by the result of this optimization calculation until the number of calculations reaches the preset maximum iteration number, and the optimal solution b of the error / noise parameter matrix b is obtained * .

[0035] Preferably, based on the optimal solution b of the error / noise parameter matrix b * , the system error / noise parameter calibration transfer function calculation module is used to solve the system transfer function, and the following is obtained:

[0036] min:||AH-b * || (7)

[0037] In the formula, A is a system matrix; H is a system transfer function;

[0038] In order to reduce the ill-conditioned nature of the system transfer function H solving process, a preconditioning matrix M is constructed to solve the system transfer function, as shown in formula (8):

[0039]

[0040] In the formula, A T is the transpose matrix of the system matrix A; I is the unit matrix; ω, γ, μ are all adjustment parameters; is the diagonal matrix of A T A+μI; is the lower triangular matrix of A T A+μI; M2 are intermediate variable matrices; M is a preconditioning matrix.

[0041] Preferably, the digital pre-distortion wideband calibration module performs pre-distortion correction according to the system transfer function, as shown in formula (9):

[0042]

[0043] In the formula, n is the signal number, is the signal after pre-distortion correction; h Q-1 is the Q-1th element in the system transfer function.

[0044] The beneficial technical effects brought by the present application are:

[0045] (1) The present application proposes a wideband vector modulation system multi-source error calibration system and method, which realizes accurate calibration of multi-source error parameters such as mixer amplitude and phase imbalance, DAC spurious noise, power amplifier nonlinearity in the wideband vector modulation system by fusing multi-objective function construction, preconditioned function solving, error / noise parameter adaptive acquisition-identification-update feedback architecture and FIR filter architecture, solves the deficiency that the existing wideband vector modulation system error calibration method can only correct single type error parameters, and meets the calibration requirements of multi-source error parameters in the wideband vector modulation system.

[0046] (2) The present application proposes a wideband vector modulation system multi-source error calibration system and method, which uses the whale algorithm to solve the optimization problem constructed by the multi-objective function, has faster convergence speed, and effectively improves the correction speed of the wideband vector modulation system multi-source error calibration.

[0047] (3) The present application proposes a wideband vector modulation system multi-source error calibration system and method, which solves the system transfer function by constructing a preconditioned matrix in the system transfer function solving process, effectively avoids the ill-conditioned problem in the system transfer function solving process, and guarantees the accuracy of the system transfer function solving. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 FIG. 1 is a schematic diagram of a wideband vector modulation system multi-source error calibration system.

[0049] Figure 2 FIG. 4 is the calibration effect of the method of the present application on the DAC spurious noise in the wideband vector modulation system; in the figure, (a) is the DAC output spectrum without using the method of the present application for spurious noise calibration, and (b) is the DAC output spectrum after using the method of the present application for spurious noise calibration.

[0050] Figure 3 FIG. 5 is the calibration effect of the method of the present application on the mixer amplitude and phase imbalance calibration in the wideband vector modulation system; in the figure, (a) is the mixer amplitude characteristic without using the method of the present application for calibration, (b) is the mixer phase characteristic without using the method of the present application for calibration, (c) is the mixer amplitude characteristic after using the method of the present application for calibration, and (d) is the mixer phase characteristic after using the method of the present application for calibration.

[0051] Figure 4 FIG. 6 is an effect diagram of the method of the present application on the power amplifier predistortion calibration in the wideband vector modulation system. DETAILED DESCRIPTION

[0052] The present application will be further described in detail below in combination with the drawings and embodiments.

[0053] This embodiment proposes a broadband vector modulation system multi-source error calibration system, such as Figure 1 As shown, it includes a vector input terminal, an error / noise parameter adaptive acquisition-identification-update path, a broadband vector modulation system output path and a vector output terminal.

[0054] The vector input end is respectively connected to the error / noise parameter adaptive acquisition-identification-update path and the broadband vector modulation system output path, and the broadband vector modulation system output path is connected to the vector output end.

[0055] The error / noise parameter adaptive acquisition-identification-update path includes an error / noise parameter identification-update module and a characteristic parameter acquisition module. The error / noise parameter identification-update module is connected to the vector input end and the characteristic parameter acquisition module respectively. The characteristic parameter acquisition module is provided with an analog-to-digital converter ADC and is preset with a characteristic parameter extraction matrix S for extracting the sub-objective function.

[0056] The output path of the broadband vector modulation system includes a system error / noise parameter calibration transfer function calculation module, a digital predistortion broadband calibration module, a vector modulator and a digital-to-analog converter DAC. The system error / noise parameter calibration transfer function calculation module, the digital predistortion broadband calibration module, the vector modulator and the digital-to-analog converter are connected in sequence. The system error / noise parameter calibration transfer function calculation module is respectively connected to the vector input end and the error / noise parameter identification-update module. The vector modulator and the digital-to-analog converter are respectively connected to the characteristic parameter acquisition module and the vector output end.

[0057] This embodiment also proposes a broadband vector modulation system multi-source error calibration method, using the broadband vector modulation system multi-source error calibration system described above. The specific calibration process is as follows:

[0058] The input signal x(n) is inputted into the error / noise parameter identification-update module of the error / noise parameter adaptive acquisition-identification-update path and the system error / noise parameter calibration transfer function calculation module of the wideband vector modulation system output path respectively via the vector input terminal;

[0059] The input signal x(n) is processed in the output path of the wideband vector modulation system in turn by a system error / noise parameter calibration transfer function calculation module, a digital pre-distortion wideband calibration module, a vector modulator and a digital-to-analog converter to obtain a vector output y, the vector output y is input into a characteristic parameter acquisition module of an error / noise parameter adaptive acquisition-identification-update path, a plurality of sub-objective functions and an error / noise parameter matrix b are extracted by the characteristic parameter acquisition module and input into an error / noise parameter identification-update module, the error / noise parameter identification-update module constructs a system comprehensive objective function by using the input signal x(n) and the plurality of sub-objective functions, and an optimal solution b of the error / noise parameter matrix b is obtained by solving the system comprehensive objective function * and input into the system error / noise parameter calibration transfer function calculation module, a system transfer function H is solved by the system error / noise parameter calibration transfer function calculation module and input into the digital pre-distortion wideband calibration module, a signal after pre-distortion correction by the digital pre-distortion wideband calibration module is input into the vector modulator and the digital-to-analog converter, a calibrated vector signal after processing by the vector modulator and the digital-to-analog converter is input into a vector output end, and an output signal after calibration by the wideband vector modulation system multi-source error calibration system is output by the vector output end.

[0060] When the wideband vector modulation system is calibrated by the multi-source error calibration, the vector output y processed by the output path of the wideband vector modulation system is processed by the characteristic parameter acquisition module, the vector output y is sampled by an analog-to-digital converter in the characteristic parameter acquisition module to obtain a sampling sequence y(n), the sampling sequence y(n) is subjected to fast Fourier transform, the sampling sequence Y after fast Fourier transform is input into a characteristic parameter extraction matrix S, a sub-objective function is extracted by the characteristic parameter extraction matrix S, and the following is obtained:

[0061]

[0062] In the formula, f P (b) is the Pth sub-objective function; b is an error / noise parameter matrix with a dimension of Mx1.

[0063] The error / noise parameter identification-update module constructs a system comprehensive objective function by using the input signal x(n) and the plurality of sub-objective functions, and the following is obtained:

[0064]

[0065] In the formula, min is a minimum value function; F(b) is a system comprehensive objective function; i is a serial number, P is the total number of sub-objective functions; P1 is a self-defined number with a value less than the total number P of sub-objective functions; w i is a weight coefficient of the ith sub-objective function; f i(b) is the i-th sub-objective function.

[0066] Furthermore, the whale algorithm is used to solve the optimization problem in the system comprehensive objective function, and the maximum number of iterations M is set. I , generate a random number initialization matrix X(t) of dimension M×PN, substitute the PN column vectors of the random number initialization matrix X(t) into the system comprehensive objective function in turn for optimization calculation, and solve the minimum value of the system comprehensive objective function and its corresponding optimal solution vector X * , based on the optimal solution vector X corresponding to each column vector * Get the optimal solution matrix X * (t), X * (t) = [X * …X * ].

[0067] During each optimization calculation, random numbers r1, r2, and p between [0, 1] are randomly generated to obtain the first intermediate variable And the second intermediate variable C, as shown in formula (3):

[0068]

[0069] Where a is a custom number, and the value of a decreases linearly from 2 to 0.

[0070] When the random number p is less than 0.5 and When , we get:

[0071]

[0072] Where X(t+1) is the result of the t+1th optimization calculation.

[0073] When the random number p is less than 0.5 and When , we get:

[0074]

[0075] Where, X rand is a random matrix with dimension M×PN.

[0076] When p>0.5, we get:

[0077] X(t+1)=|X * (t)-X(t)|×e l cos(2πl)+X * (t) (6)

[0078] Where l is a random number between [-1,1].

[0079] After each optimization calculation, the optimal solution vector obtained by the previous optimization calculation is replaced by the result of this optimization calculation until the number of calculations reaches the preset maximum iteration number, and the optimal solution b of the error / noise parameter matrix b is obtained * .

[0080] Based on the optimal solution b of the error / noise parameter matrix b * , the system transfer function is solved by using the system error / noise parameter calibration transfer function calculation module, and the following is obtained:

[0081] min:||AH-b * || (7)

[0082] In the formula, A is the system matrix; H is the system transfer function.

[0083] In order to reduce the ill-conditioned nature of the system transfer function H solving process, a preconditioning matrix M is constructed to solve the system transfer function, as shown in formula (8):

[0084]

[0085] In the formula, A T is the transpose matrix of the system matrix A; I is the unit matrix; ω, γ, μ are all adjustment parameters; is the diagonal matrix of A T A+μI; is the lower triangular matrix of A T A+μI; M2 are all intermediate variable matrices; M is the preconditioning matrix.

[0086] The digital pre-distortion wideband calibration module is used to perform pre-distortion correction according to the system transfer function, as shown in formula (9):

[0087]

[0088] In the formula, n is the signal number, is the signal after pre-distortion correction; h Q-1 is the Q-1th element in the system transfer function.

[0089] Finally, the signal after pre-distortion correction of the digital pre-distortion wideband calibration module is input into the vector modulator and the digital-to-analog converter, and the calibrated vector signal is obtained after processing by the vector modulator and the digital-to-analog converter, and the output signal calibrated by the wideband vector modulation system multi-source error calibration system is output by the vector output end.

[0090] Further, in order to verify the correction effect of the multi-source error calibration system and method of the wideband vector modulation system proposed in the embodiment, the DAC spur noise, the mixer amplitude and phase imbalance and the nonlinearity of the power amplifier in the wideband vector modulation system are calibrated by using the multi-source error calibration method of the wideband vector modulation system proposed in the embodiment, Figure 2 The calibration effect of the method on the DAC spur noise in the wideband vector modulation system, Figure 3 The calibration effect of the method on the mixer amplitude and phase imbalance in the wideband vector modulation system, Figure 4 The effect diagram of the method on the pre-distortion calibration of the power amplifier in the wideband vector modulation system, and comparison can find that the DAC spur noise, the mixer amplitude and phase imbalance and the nonlinearity of the power amplifier in the wideband vector modulation system can be accurately corrected by using the method, and it is further verified that the multi-source error parameters in the wideband vector modulation system can be accurately corrected by using the method, and the calibration demand of the multi-error parameters of the wideband vector modulation system is met.

[0091] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "arrangement", "installation", "connection", "connection", "fixing" should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected, it can be mechanically connected, or it can be electrically connected, it can be directly connected, or it can be indirectly connected through an intermediate medium, it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0092] Of course, the above description is not a limitation of the present application, and the present application is not limited to the above examples, and the changes, modifications, additions or replacements made by those skilled in the art within the essential scope of the present application should also belong to the protection scope of the present application.

Claims

1. A broadband vector modulation system multi-source error calibration system, characterized in that: It includes a vector input terminal, an error / noise parameter adaptive acquisition-identification-update path, a broadband vector modulation system output path and a vector output terminal; The vector input end is respectively connected to the error / noise parameter adaptive acquisition-identification-update path and the broadband vector modulation system output path, and the broadband vector modulation system output path is connected to the vector output end; The error / noise parameter adaptive acquisition-identification-update path includes an error / noise parameter identification-update module and a characteristic parameter acquisition module, and the error / noise parameter identification-update module is connected to the vector input terminal and the characteristic parameter acquisition module respectively; The output path of the broadband vector modulation system includes a system error / noise parameter calibration transfer function calculation module, a digital pre-distortion broadband calibration module, a vector modulator and a digital-to-analog converter. The system error / noise parameter calibration transfer function calculation module, the digital pre-distortion broadband calibration module, the vector modulator and the digital-to-analog converter are connected in sequence. The system error / noise parameter calibration transfer function calculation module is respectively connected to the vector input end and the error / noise parameter identification-update module. The vector modulator and the digital-to-analog converter are respectively connected to the characteristic parameter acquisition module and the vector output end.

2. The broadband vector modulation system multi-source error calibration system according to claim 1, characterized in that: The characteristic parameter acquisition module is provided with an analog-to-digital converter and a characteristic parameter extraction matrix S for extracting the sub-objective function.

3. A method for calibrating multi-source errors in a broadband vector modulation system, characterized in that: Using the broadband vector modulation system multi-source error calibration system as claimed in claim 2, the calibration process is as follows: The input signal x(n) is inputted into the error / noise parameter identification-update module of the error / noise parameter adaptive acquisition-identification-update path and the system error / noise parameter calibration transfer function calculation module of the wideband vector modulation system output path respectively via the vector input terminal; The input signal x(n) is processed in sequence by a system error / noise parameter calibration transfer function calculation module, a digital predistortion broadband calibration module, a vector modulator, and a digital-to-analog converter in the output path of the broadband vector modulation system to obtain a vector output y. The vector output y is input into a characteristic parameter acquisition module of an error / noise parameter adaptive acquisition-identification-update path. The characteristic parameter acquisition module is used to extract a plurality of sub-objective functions and an error / noise parameter matrix b and input them into the error / noise parameter identification-update module. The error / noise parameter identification-update module constructs a system comprehensive objective function using the input signal x(n) and the plurality of sub-objective functions, and solves the system comprehensive objective function to obtain an optimal solution b of the error / noise parameter matrix b. * And input it into the system error / noise parameter calibration transfer function calculation module, use the system error / noise parameter calibration transfer function calculation module to solve the system transfer function H and input it into the digital pre-distortion broadband calibration module, use the digital pre-distortion broadband calibration module to perform pre-distortion correction on the signal and input it into the vector modulator and the digital-to-analog converter, after being processed by the vector modulator and the digital-to-analog converter, the calibrated vector signal is input into the vector output end, and the vector output end is used to output the output signal calibrated by the broadband vector modulation system multi-source error calibration system.

4. The method for calibrating multi-source errors in a broadband vector modulation system according to claim 3, wherein: When the characteristic parameter acquisition module processes the vector output y obtained by processing the output path of the broadband vector modulation system, the analog-to-digital converter in the characteristic parameter acquisition module is used to sample the vector output y to obtain a sampling sequence y(n), and then the sampling sequence y(n) is fast Fourier transformed. The sampling sequence Y after the fast Fourier transform is input into the characteristic parameter extraction matrix S, and the characteristic parameter extraction matrix S is used to extract the sub-objective function to obtain: Where, f P (b) is the Pth sub-objective function; b is the error / noise parameter matrix with dimension M×1.

5. The method for calibrating multi-source errors in a broadband vector modulation system according to claim 4, wherein: The error / noise parameter identification-update module constructs the system comprehensive objective function using the input signal x(n) and multiple sub-objective functions to obtain: Where min is the minimum value function; F(b) is the system comprehensive objective function; i is the sequence number, and P is the total number of sub-objective functions; P1 is a custom number, and its value is less than the total number of sub-objective functions P; w i is the weight coefficient of the i-th sub-objective function; f i (b) is the i-th sub-objective function.

6. The method for calibrating multi-source errors in a broadband vector modulation system according to claim 5, wherein: The whale algorithm is used to solve the optimization problem in the system's comprehensive objective function, and the maximum number of iterations M is set. I , generate a random number initialization matrix X(t) of dimension M×PN, substitute the PN column vectors of the random number initialization matrix X(t) into the system comprehensive objective function in turn for optimization calculation, and solve the minimum value of the system comprehensive objective function and its corresponding optimal solution vector X * , based on the optimal solution vector X corresponding to each column vector * Get the optimal solution matrix X * (t), X * (t) = [X * …X * ]; During each optimization calculation, random numbers r1, r2, and p between [0, 1] are randomly generated to obtain the first intermediate variable And the second intermediate variable C, as shown in formula (3): Where a is a custom number, and the value of a decreases linearly from 2 to 0; When the random number p is less than 0.5 and When , we get: Where, X(t+1) is the result of the t+1th optimization calculation; When the random number p is less than 0.5 and When , we get: Where, X rand is a random matrix with dimension M×PN; When p>0.5, we get: X(t+1)=|X * (t)-X(t)|×e l cos(2πl)+X * (t) (6) Where l is a random number between [-1,1]; After each optimization calculation is completed, the optimal solution vector obtained by the previous optimization calculation is replaced by the result of this optimization calculation until the number of calculations reaches the preset maximum number of iterations, and the optimal solution b of the error / noise parameter matrix b is obtained. * .

7. The method for calibrating multi-source errors in a broadband vector modulation system according to claim 6, wherein: The optimal solution b based on the error / noise parameter matrix b * , use the system error / noise parameter calibration transfer function calculation module to solve the system transfer function and obtain: my:||AH-b * || (7) Where A is the system matrix; H is the system transfer function; In order to reduce the pathological nature of the solution process of the system transfer function H, a preconditioning matrix M is constructed to solve the system transfer function, as shown in formula (8): Where A T is the transposed matrix of the system matrix A; I is the identity matrix; ω, γ, and μ are all adjustment parameters; A T A+μI diagonal matrix; A T The lower triangular matrix of A+μI; M2 is the intermediate variable matrix; M is the preconditioning matrix.

8. The method for calibrating multi-source errors in a broadband vector modulation system according to claim 7, wherein: The digital predistortion broadband calibration module performs predistortion correction according to the system transfer function, as shown in formula (9): Where n is the signal number, is the signal after predistortion correction; h Q-1 is the Q-1th element in the system transfer function.

Images