Peak-to-average ratio suppression method, device, system and storage medium based on symbol-level precoding scheme

Abstract

The present invention discloses a peak-to-average ratio suppression method, device, system and storage medium based on a symbol-level precoding scheme, which belongs to the field of wireless communication technology, and includes: obtaining a downlink model of a MIMO-OFDM system, and establishing a frequency domain signal model received by a user based on the downlink model; obtaining known channel state information of the MIMO-OFDM system and establishing a symbol-level precoding scheme based on the downlink model; formalizing and equivalently converting the peak-to-average ratio suppression problem based on the symbol-level precoding scheme to obtain an objective function; solving the objective function based on a Dinkelbach algorithm, combining a preset PDD algorithm in the solving process, and combining a preset mAPG algorithm in the PDD algorithm to finally obtain an optimal solution of the objective function; calculating a minimum peak-to-average ratio value according to the optimal solution, optimizing the MIMO-OFDM system according to the minimum peak-to-average ratio value, and realizing peak-to-average ratio suppression; the present invention can suppress the peak-to-average ratio to a minimum.

Description

Technical Field

[0001] The invention relates to a peak-to-average ratio suppression method, device, system and storage medium based on a symbol-level precoding scheme, and belongs to the technical field of wireless communications. Background Art

[0002] Orthogonal frequency division multiplexing (OFDM) is an effective method to deal with frequency selective fading in broadband wireless communication systems, and is also one of the key technologies of 4G and 5G. However, since the phases of each subcarrier of OFDM are independent of each other, it is susceptible to high peak-to-average power ratio (PAPR). Too high PAPR may cause linear RF components (such as power amplifiers) to operate in a nonlinear amplification range. This will not only cause signal distortion and out-of-band radiation, but also lead to excessive system energy consumption, which has become an important issue restricting the development of the communications industry. In 4G communication systems, the intermediate frequency peak clipping algorithm is a major solution for controlling PAPR. However, the intermediate frequency peak clipping algorithm usually sacrifices the error vector magnitude (EVM) index of the signal in exchange for a reduction in the signal output PAPR. Especially in high-order modulation modes, the limitations of intermediate frequency peak clipping are becoming more and more obvious. In addition, with the rapid popularization and development of 5G systems, the definition of communication system bandwidth by low frequency and high frequency is getting wider and wider, and the number of transmission channels is also increasing. The application of intermediate frequency peak clipping algorithms has encountered more and more bottlenecks. In order not to affect the overall efficiency of system equipment, we hope to seek more ways and methods to reduce the peak-to-average ratio of the signal. Focusing on OFDM system, this paper studies how to use symbol-level precoding to improve the PAPR control problem in Multi-User Multiple Input Multiple Output (MU-MIMO) system.

[0003] As one of the key technologies of the physical layer of 5G wireless systems, massive antennas have gradually moved from academia to industry and entered the stage of large-scale deployment. The increase in antenna size leads to an increase in the number of RF channels. From the perspective of cost control, it is expected that each channel will use cheaper RF devices, including cheaper high-power amplifiers (HPA), which may result in a smaller linear operating range. From the perspective of reducing system operating power consumption, each amplifier is expected to work in the linear operating area. The above contradiction will continue to worsen with the increase in antenna size, thus bringing extremely severe challenges to the design of high-power amplifiers. Therefore, in 5G systems, especially large-scale MIMO systems, reducing the transmitter PAPR becomes the only feasible solution.

[0004] At present, researchers have developed many PAPR control technologies for single-input single-output orthogonal frequency division multiplexing (SISO-OFDM) wireless systems. Among them, the more famous technologies include amplitude clipping, clipping filtering, subcarrier retention (TR), subcarrier injection (TI), active constellation extension (ACE), and multiple signal representation technologies (such as partial transmit sequence (PTS), selective mapping (SLM) and interleaving, etc.). These technologies achieve PAPR reduction at the cost of increased transmission signal power, increased bit error rate (BER), data rate loss, and increased computational complexity. Although the above PAPR control technology can be easily extended to point-to-point MIMO systems. However, it is difficult to directly extend it to 5G MU-MIMO systems. The main reason is that in MU-MIMO systems, the transmitter needs to use precoding technology to eliminate inter-user interference (MUI). The mutual coupling of two different types of targets, PAPR control and MUI elimination, greatly increases the difficulty of transmitter design. Therefore, in actual MIMO-OFDM systems, there is a problem of high peak-to-average ratio and difficulty in suppression. Summary of the invention

[0005] The purpose of the present invention is to provide a method, device, system and storage medium for peak-to-average ratio suppression based on a symbol-level precoding scheme, so as to solve the problem that it is difficult to suppress the peak-to-average ratio in a MIMO-OFDM system in the prior art.

[0006] To achieve the above objectives, the present invention is implemented by adopting the following technical solutions:

[0007] In a first aspect, the present invention provides a peak-to-average ratio suppression method based on a symbol-level precoding scheme, comprising:

[0008] Acquire a downlink model of the MIMO-OFDM system, and establish a frequency domain signal model received by the user based on the downlink model;

[0009] Acquire known channel state information of the MIMO-OFDM system, and establish a symbol-level precoding scheme according to the known channel state information;

[0010] Based on the symbol-level precoding scheme, the peak-to-average ratio suppression problem is formalized and equivalently transformed to obtain an objective function;

[0011] Solving the objective function based on the Dinkelbach algorithm, combining the preset PDD algorithm in the solving process, combining the preset mAPG algorithm in the PDD algorithm, and finally obtaining the optimal solution of the objective function;

[0012] Calculating a peak-to-average ratio minimum value according to the optimal solution, optimizing the MIMO-OFDM system according to the peak-to-average ratio minimum value to achieve peak-to-average ratio suppression;

[0013] The acquiring known channel state information of the MIMO-OFDM system and establishing a symbol-level precoding scheme according to the known channel state information includes:

[0014] When the data symbol sent to user k Extracted from the M-PSK constellation diagram, we get:

[0015]

[0016] in, represents the set of all data symbols of the M-PSK constellation diagram, e is a natural constant, j represents the complex phase part identifier, m represents the mth M-PSK constellation diagram symbol, M represents the PSK constellation diagram order, It means equivalent to;

[0017] Define the first intermediate variable Note SINR k is the SINR constraint of the received signal of user k and is recorded as in, represents the complex domain, W represents the total number of OFDM subcarriers, N represents the number of base station antennas, t k represents the measure of SINR received by user k, σ represents the square root of noise variance;

[0018] Using basic linear algebra The symbol-level precoding scheme is expressed as:

[0019]

[0020] in, Indicates sending a signal, [.] T represents the transpose operation of the matrix, and Represents the real and imaginary parts of a complex number, (.) * represents the conjugate operation, (.) H represents the conjugate transpose operation, |.| represents the modulus operation,

[0021] is the channel matrix The kth row of for The wth column of is the W-dimensional discrete Fourier transform matrix, w is the index of the OFDM subcarrier, and cot represents the cotangent function;

[0022] The peak-to-average ratio suppression problem is formalized and equivalently transformed based on the symbol-level precoding scheme to obtain an objective function, including:

[0023] Formalizing the peak-to-average ratio suppression problem, we get the first problem:

[0024]

[0025]

[0026] t≥t 0 (6)

[0027]

[0028] Among them, (P 1 ) represents the first problem, min represents the minimization operation, st represents the constraint condition, ‖.‖ ∞ represents the infinite norm, represents the first intermediate variable, t represents the SINR metric, and Represents the real and imaginary parts of a complex number, (.) * represents the conjugate operation, (.) H represents the conjugate transpose operation, |.| represents the modulus operation, is the channel matrix The kth row of for The wth column of is the W-dimensional discrete Fourier transform matrix, w is the index of the OFDM subcarrier, cot represents the cotangent function, represents the data symbol sent to user k, M represents the order of the PSK constellation diagram, t 0 ≥0,t 0 is a measure of the preset SINR, ‖.‖ 2 represents the two-norm, P T is the upper limit of the transmission power;

[0029] The first problem is converted into an equivalent form to obtain the objective function, including:

[0030] definition:

[0031]

[0032]

[0033] Among them, vec(.) means to vectorize the matrix. for The result of vectorization, a means The vector consisting of the real and imaginary parts of represents the channel symbol information vector, represents the Kronecker product operation, I KW Represents a KW-dimensional unit matrix, where all vectors Composition Matrix Re-realize the matrix Finally get represents the second intermediate variable, represents the third intermediate variable, represents the fourth intermediate variable;

[0034] The symbol-level precoding scheme represented by formula (5) is equivalently converted into the following linear matrix inequality:

[0035]

[0036] Among them, ι 2KW×1 Represents a unit vector of 2KW×1 dimension;

[0037] The first problem is converted into the second problem:

[0038]

[0039]

[0040] t≥t 0 (10)

[0042]

[0043] Among them, P 2 It is the second question;

[0044] Use the log-sum-exp function to approximate the infinite norm in the second problem:

[0045]

[0046] Where η is the smoothing parameter, η>0, a i Represents the fifth intermediate parameter, a i =[a i ,a i+NW ] T , a i for a i The i-th element of ;

[0047] The objective function is obtained by using an approximation for the second problem:

[0048]

[0049] t≥t 0 (15)

[0050]

[0051] Among them, P 3is the objective function, W represents the total number of OFDM subcarriers, and N represents the number of base station antennas;

[0052] The solving of the objective function based on the Dinkelbach algorithm includes:

[0053] Create a constrained fractional optimization problem as follows:

[0054]

[0055] Where A(θ) is a concave function with θ as a variable, B(θ) is a convex function with θ as a variable, and θ is the feasible domain corresponding to the constraints in the objective function;

[0056] Mapping formula (17) to the objective function, the vector θ consists of a and t, and the optimization problem is:

[0057]

[0058] Among them, max represents the maximization operation, and λ is the auxiliary variable of the Dinkelbach algorithm;

[0059] Mapping formula (18) to the objective function, we get the fourth problem:

[0060]

[0061] t≥t 0 (twenty one)

[0062]

[0063] Among them, min represents the minimization operation, η is the smoothing parameter, η>0;

[0064] The optimal solution is obtained by solving the fourth problem through the Dinkelbach algorithm;

[0065] The method combines the preset PDD algorithm in the solution process, and combines the preset mAPG algorithm in the PDD algorithm to finally obtain the optimal solution of the objective function, including:

[0066] During the iteration of the Dinkelbach algorithm, the preset PDD algorithm is used each time formula (18) is calculated;

[0067] In the preset PDD algorithm, the preset mAPG algorithm is used for each calculation of the sixth question during the iteration process;

[0068] Combining the preset PDD algorithm and the preset mAPG algorithm, the objective function is finally solved to obtain the optimal solution;

[0069] The sixth problem is obtained by introducing the PDD algorithm auxiliary variable u into the fourth problem, and the expression is:

[0070]

[0071] stu≤0

[0072] t≥t 0

[0073]

[0074] Among them, ρ is the penalty factor, and the variable γ is the dual variable corresponding to the equality constraint. is the objective function.

[0075] Furthermore, the frequency domain signal model is:

[0076]

[0077] in, is the received signal of K users on the wth OFDM subcarrier, to are the received signals of users 1 to K on the wth OFDM subcarrier, Indicates sending a signal. N represents the number of base station antennas, represents the channel matrix of the MIMO-OFDM system, represents Gaussian white noise and follows the distribution σ 2 represents the noise variance, 0 represents the zero vector, I represents the unit vector, w is the index of the OFDM subcarrier, W represents the total number of OFDM subcarriers, represents a complex domain, [.] T Represents the transpose operation on a matrix.

[0078] In a second aspect, the present invention provides a peak-to-average ratio suppression device based on a symbol-level precoding scheme based on the method described in any one of the first aspects, characterized in that it includes:

[0079] A frequency domain signal model building module is used to obtain a downlink model of the MIMO-OFDM system and to build a frequency domain signal model received by the user based on the downlink model;

[0080] A symbol-level precoding scheme establishing module: used to obtain known channel state information of the MIMO-OFDM system and establish a symbol-level precoding scheme according to the known channel state information;

[0081] An objective function building module is used to formalize and equivalently transform the peak-to-average ratio suppression problem based on the symbol-level precoding scheme to obtain an objective function;

[0082] Objective function solving module: used to solve the objective function based on the Dinkelbach algorithm, combine the preset PDD algorithm in the solving process, and combine the preset mAPG algorithm in the PDD algorithm to finally obtain the optimal solution of the objective function;

[0083] Peak-to-average ratio suppression module: used to calculate the minimum peak-to-average ratio according to the optimal solution, optimize the MIMO-OFDM system according to the minimum peak-to-average ratio, and achieve peak-to-average ratio suppression.

[0084] In a third aspect, the present invention provides a peak-to-average ratio suppression system based on a symbol-level precoding scheme, comprising:

[0085] A memory for storing instructions;

[0086] A processor is used to execute the instructions to implement the operation of the peak-to-average ratio suppression method based on the symbol-level precoding scheme as described in any one of the first aspects.

[0087] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a peak-to-average ratio suppression method based on a symbol-level precoding scheme as described in any one of the first aspects.

[0088] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0089] The present invention provides a peak-to-average ratio suppression method, device, system and storage medium based on a symbol-level precoding scheme, and proposes a joint symbol-level precoding scheme and peak-to-average ratio suppression. Through the design of the symbol-level precoding scheme, under the constraint of known channel state information, the peak-to-average ratio is minimized while the signal-to-interference-noise ratio of the system is maximized, so as to promote the correlation development of low-cost and low-power hardware implementation and provide good communication performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 It is a flow chart of a peak-to-average ratio suppression method based on a symbol-level precoding scheme provided by an embodiment of the present invention;

[0091] Figure 2 It is a downlink model of a MIMO-OFDM system provided by an embodiment of the present invention;

[0092] Figure 3 is a comparison chart of the convergence performance of three algorithms provided by the embodiments of the present invention;

[0093] Figure 4is a PAPR complementary cumulative distribution function diagram provided by an embodiment of the present invention;

[0094] Figure 5 is a graph showing changes in transmit power / SINR / PAPR with Dinkelbach iterations provided by an embodiment of the present invention;

[0095] Figure 6 is a graph showing changes in transmit power / SINR / PAPR versus transmit power threshold provided by an embodiment of the present invention;

[0096] Figure 7 It is a schematic diagram of a user receiving signal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0097] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the protection scope of the present invention.

[0098] Example 1

[0099] In recent years, symbol-level precoding (SLP) has been widely studied. Unlike the traditional linear block-level precoding (BLP) technology that aims to suppress or eliminate MUI, SLP technology regards MUI as a source of useful signals that can improve system performance, and significantly increases the symbol error rate (SER) at the receiving end. In this context, the present invention uses SLP to reduce PAPR, which has great potential. Aiming at the actual MIMO-ODFM system, the symbol-level information of the transmitted data is deeply mined, and the transmission theory and efficient design mechanism based on symbol-level information in wireless communication systems are studied. We consider the joint design of SLP and PAPR, hoping to maximize the system signal-to-noise ratio (SINR), so that the received signal has a higher SINR and the probability of being correctly detected, and provides good communication performance under low complexity. At the same time, the system PAPR is reduced to promote the correlation to develop low-cost and low-power hardware implementation. In addition, in order to save base station energy consumption, the transmission power is limited. According to the research content, an efficient PAPR suppression algorithm based on optimization theory is explored, and the performance of the proposed system and the effectiveness of the algorithm are verified by simulation results.

[0100] like Figure 1 As shown, an embodiment of the present invention provides a peak-to-average ratio suppression method based on a symbol-level precoding scheme, including:

[0101] S1. Obtain a downlink model of the MIMO-OFDM system, and establish a frequency domain signal model received by a user based on the downlink model.

[0102] The downlink model of the MIMO-OFDM system is as follows: Figure 2As shown in the figure, a base station equipped with N antennas simultaneously serves K single-antenna users (K≤N) through space division multiplexing. The frequency domain signal received by the user can be expressed as:

[0103]

[0104] in, is the received signal of K users on the wth OFDM subcarrier, to are the received signals of users 1 to K on the wth OFDM subcarrier, Indicates sending a signal. N represents the number of base station antennas, represents the channel matrix of the MIMO-OFDM system, represents Gaussian white noise and follows the distribution σ 2 represents the noise variance, 0 represents the zero vector, I represents the unit vector, w is the index of the OFDM subcarrier, W represents the total number of OFDM subcarriers, represents a complex domain, [.] T Represents the transpose operation on a matrix.

[0105] S2. Acquire known channel state information of the MIMO-OFDM system, and establish a symbol-level precoding scheme according to the known channel state information.

[0106] When the data symbol sent to user k Extracted from the M-PSK constellation diagram, we get:

[0107]

[0108] in, represents the set of all data symbols of the M-PSK constellation diagram, e is a natural constant, j represents the complex phase part identifier, m represents the mth M-PSK constellation diagram symbol, M represents the PSK constellation diagram order, It means equivalent to;

[0109] Define the first intermediate variable Note SINR k is the SINR constraint of the received signal of user k and is recorded as in, represents the complex domain, W represents the total number of OFDM subcarriers, N represents the number of base station antennas, t k represents the measure of SINR received by user k, σ represents the square root of noise variance;

[0110] Using basic linear algebra The symbol-level precoding scheme is expressed as:

[0111]

[0112] in, Indicates sending a signal, [.] T represents the transpose operation of the matrix, and Represents the real and imaginary parts of a complex number, (.) * represents the conjugate operation, (.) H represents the conjugate transpose operation, |.| represents the modulus operation, is the channel matrix The kth row of for The wth column of is the W-dimensional discrete Fourier transform matrix, w is the index of the OFDM subcarrier, and cot represents the cotangent function.

[0113] S3. Based on the symbol-level precoding scheme, the peak-to-average ratio suppression problem is formalized and equivalently transformed to obtain an objective function.

[0114] Based on the symbol-level precoding scheme obtained in step S2, the joint design of SLP and PAPR is considered, and the peak-to-average ratio suppression problem is formalized to obtain the first problem:

[0115]

[0116] t≥t 0 (6)

[0117]

[0118] Among them, (P 1 ) represents the first problem, min represents the minimization operation, st represents the constraint condition, ‖.‖ ∞ represents the infinite norm, represents the first intermediate variable, t represents the SINR metric, and Represents the real and imaginary parts of a complex number, (.) * represents the conjugate operation, (.) H represents the conjugate transpose operation, |.| represents the modulus operation, is the channel matrix The kth row of for The wth column of is the W-dimensional discrete Fourier transform matrix, w is the index of the OFDM subcarrier, cot represents the cotangent function, represents the data symbol sent to user k, M represents the order of the PSK constellation diagram, t0 ≥0,t 0 is a measure of the preset SINR, ‖.‖ 2 represents the two-norm, P T is the upper limit of the transmission power;

[0119] The first problem is converted into an equivalent form to obtain the objective function, including:

[0120] definition:

[0121]

[0122] Among them, vec(.) means to vectorize the matrix. for The result of vectorization, a means The vector consisting of the real and imaginary parts of represents the channel symbol information vector, represents the Kronecker product operation, I KW Represents a KW-dimensional unit matrix, where all vectors Composition Matrix Re-realize the matrix Finally get represents the second intermediate variable, represents the third intermediate variable, represents the fourth intermediate variable;

[0123] The symbol-level precoding scheme represented by formula (5) is equivalently converted into the following linear matrix inequality:

[0124]

[0125] Among them, ι 2kw×1 represents a unit vector of 2kw×1 dimension;

[0126] The first problem is converted into the second problem:

[0127]

[0128] T≥t 0 (10)

[0129]

[0130] Among them, P 2 It is the second question;

[0131] Use the log-sum-exp function to approximate the infinite norm in the second problem:

[0132]

[0133] Where η is the smoothing parameter, η>0, a i Represents the fifth intermediate parameter, a i =[a i ,a i+nw ] t , a i for a i The i-th element of ;

[0134] The objective function is obtained by using an approximation for the second problem:

[0135]

[0136] t≥t 0 (15)

[0137]

[0138] Among them, P 3 is the objective function, W represents the total number of OFDM subcarriers, and N represents the number of base station antennas.

[0139] S4. Solve the objective function based on the Dinkelbach algorithm, combine the preset PDD algorithm in the solving process, and combine the preset mAPG algorithm in the PDD algorithm to finally obtain the optimal solution of the objective function.

[0140] Problem (P 3 ) is a fraction, so we use the Dinkelbach algorithm to solve it. For a constrained fractional optimization problem:

[0141]

[0142] Among them, A(θ) is a concave function with θ as a variable, B(θ) is a convex function with θ as a variable, θ is the feasible domain corresponding to the constraints in the objective function, and the vector θ consists of a and t.

[0143] When A(θ) is a concave function with θ as a variable and B(θ) is a convex function with θ as a variable, iterative algorithm 1 can converge to the global optimal solution of the problem, where λ is the auxiliary variable of the Dinkelbach algorithm.

[0144]

[0145] The key step 4 in Algorithm 1 is to solve the optimization problem:

[0146]

[0147] Mapping formula (18) to the objective function, we get the fourth problem:

[0148]

[0149] t≥t 0 (twenty one)

[0150]

[0151] Among them, min represents the minimization operation, η is the smoothing parameter, and η>0.

[0152] Then, update a and t: The Dinkelbach algorithm solves the problem (P 4 ) optimal solution to update a and t.

[0153] Then, update λ 1 :By solving the problem (P 4 ) to get the optimal solution a * and t * To update λ 1 ,get:

[0154]

[0155] In order to solve the problem efficiently (P 4 ), we will give a customized PDD algorithm. Consider a general constrained optimization problem:

[0156]

[0157] sth(θ)=0 (24)

[0158] g(θ)≤0 (25)

[0159] The algorithm is as follows:

[0160]

[0161] Among them, P ρ,v For the problem (P 5 ) is the augmented Lagrange function optimization problem (AL), ρ is the penalty factor, and v is the dual variable corresponding to the equality constraint formula (24).

[0162] We first introduce the PDD algorithm auxiliary variable u, (P 4 ) is expressed as:

[0163]

[0164] stu≤0

[0165] t≥t 0

[0166]

[0167] Among them, ρ is the penalty factor, and the variable γ is the dual variable corresponding to the equality constraint. is the objective function.

[0168] The PDD algorithm solves the problem (P 6 ) The optimal solution is used to update a, t and u, see step 3 in Algorithm 2;

[0169] The PDD algorithm solves the problem (P 6 ) Optimal solution new a, t and u, threshold judgment Thus the penalty factor ρ and variable γ are updated.

[0170] In the PDD algorithm, the subproblem (P 6 ) We use the mAPG algorithm to solve efficiently. For general constrained optimization problems:

[0171]

[0172] The mAPG algorithm is given by Algorithm 3 below, where the step size β in step 4 is r is selected via a one-dimensional linear search.

[0173] Mapping formula (26) to the problem (P 6 ), the vector θ consists of variables a, t and u, and Θ is the feasible domain corresponding to the constraints in the objective function.

[0174] The partial derivative operations of a, t and u in Algorithm 3 are as follows:

[0175] (1) Find the partial derivative of variable a:

[0176] Given t,u,ρ andγ, let b=u-t1 2KW×1 +ργ, then the objective function The gradient is as follows

[0177]

[0178] (2) Find the partial derivative of variable t:

[0179] Given u, a, ρ and γ, let Then the objective function The gradient is as follows:

[0180]

[0181] (3) Find the partial derivative of variable u

[0182] Given a, t, ρ and γ, let Then the objective function The gradient is as follows:

[0183]

[0184] The projection operations of a, t and u in Algorithm 3 are as follows:

[0185]

[0186] When η→0, the LSE function becomes larger, making it more difficult to solve. So we make the corresponding transformation of the LSE function as follows:

[0187]

[0188] in,

[0189] Finally, we get the derivative as follows:

[0190]

[0191]

[0192] S5. Calculate a minimum peak-to-average ratio value according to the optimal solution, and optimize the MIMO-OFDM system according to the minimum peak-to-average ratio value to achieve peak-to-average ratio suppression.

[0193] The inverse discrete Fourier transform (IDFT) at the transmitter will cause the OFDM signal to have a large dynamic range. Such signals are easily affected by the nonlinear distortion of the RF link in the actual system. Therefore, it is necessary to reduce the PAPR of the transmitted signal, that is:

[0194]

[0195] The MIMO-OFDM system is optimized according to the calculated minimum peak-to-average ratio value to achieve peak-to-average ratio suppression.

[0196] In order to verify the performance of the above solution, we used MATLAB to conduct simulation experiments on corresponding scenarios.

[0197] The simulation design is as follows: We consider a MISO-OFDM system, with W = 16 OFDM subcarriers, and the BS is equipped with N = 16 antennas to send QPSK signals to K = 12 single-antenna users. The total transmission power is set to P T =5dB, the channel parameters follow an independent and identically distributed (iid) Gaussian distribution Noise power σ 2 =1, received SINR measurement value threshold t 0 =-1dB.

[0198] First, we Figure 3 Demonstrating the convergence of the proposed algorithm, the left figure shows the problem (P 3 ) in the Dinkelbach framework. The middle figure shows the equality constraint function, namely ‖h(θ k )‖ ∞ The internal iteration in the PDD algorithm, the right figure shows the PDD algorithm subproblem (P 6 ) in the iterative change of the target value in the mAPG algorithm solution. We see that the problem (P 3 ) The target value converges monotonically after a small number of iterations in the Dinkelbach framework, and the equality constraint function will drop to the termination condition after a finite number of iterations, that is, the equality constraint condition is satisfied. In the figure on the right, we tested the convergence of the mAPG algorithm, and the simulation values ​​show that the acceleration effect of mAPG is obvious.

[0199] CCDF is the abbreviation of Complementary Cumulative Distribution Function. It is a concept introduced to represent the statistical characteristics of the peak-to-average PAPR in OFDM systems. It is defined as the probability that the peak-to-average value exceeds a certain threshold in a multi-carrier transmission system. Based on the above parameter settings, we obtain the statistical characteristic curve of CCDF for improving PAPR in a multi-user multiple input multiple output (MU-MIMO) system using symbol-level precoding, as shown in the figure: Figure 4 .

[0200] Next Figure 5 The SLP-PAPR algorithm based on the Dinkelbach framework shows some important system indicators: transmit power, SINR and PAPR iterative changes. This patent considers the joint design of SLP and PAPR, hoping to maximize the system signal-to-noise ratio (SINR), so that the received signal has a higher SINR, while reducing the system PAPR. It can be seen from the figure that with the iterative changes of Dinkelbach, SINR increases, while PAPR decreases, and the transmit power is limited within the power threshold, thus saving base station energy consumption.

[0201] Next Steps Figure 6 The figure shows the changes of relevant system performance indicators under different total transmission power thresholds. As the total transmission power threshold increases, PAPR decreases, while the transmission power and SINR increase. The performance of PAPR can be improved and the system SINR can be increased by properly adjusting the total transmission power threshold.

[0202] The present invention uses constructive interference based scheme to design precoders using available knowledge of channel information and expected symbols, so that the user received signal is located in the corresponding expected construction area. The construction area is defined in the same constellation domain as the expected information symbol, thereby reducing the bit error rate to ensure the reliability of communication. Figure 7 As shown, the user receives the signal in the corresponding structural area, which improves the communication quality.

[0203] Example 2

[0204] The embodiment of the present invention further provides a peak-to-average ratio suppression device based on a symbol-level precoding scheme, comprising:

[0205] A frequency domain signal model building module is used to obtain a downlink model of the MIMO-OFDM system and to build a frequency domain signal model received by the user based on the downlink model;

[0206] Symbol-level precoding scheme establishment module: used to obtain known channel state information of the MIMO-OFDM system and establish a symbol-level precoding scheme based on the known channel state information;

[0207] Objective function building module: used to formalize and equivalently transform the peak-to-average ratio suppression problem based on the symbol-level precoding scheme to obtain the objective function;

[0208] Objective function solving module: used to solve the objective function based on the Dinkelbach algorithm, combining the preset PDD algorithm in the solving process, and combining the preset mAPG algorithm in the PDD algorithm, and finally obtaining the optimal solution of the objective function;

[0209] Peak-to-average ratio suppression module: used to calculate the minimum peak-to-average ratio according to the optimal solution, optimize the MIMO-OFDM system according to the minimum peak-to-average ratio, and achieve peak-to-average ratio suppression.

[0210] Example 3

[0211] The embodiment of the present invention provides a peak-to-average ratio suppression system based on a symbol-level precoding scheme, including:

[0212] A memory for storing instructions;

[0213] A processor is used to execute instructions so that the device performs operations to implement the following peak-to-average ratio suppression method based on a symbol-level precoding scheme:

[0214] Obtain a downlink model of the MIMO-OFDM system, and establish a frequency domain signal model received by the user based on the downlink model;

[0215] Acquire known channel state information of the MIMO-OFDM system, and establish a symbol-level precoding scheme according to the known channel state information;

[0216] Based on the symbol-level precoding scheme, the peak-to-average ratio suppression problem is formalized and equivalently transformed to obtain the objective function;

[0217] The objective function is solved based on the Dinkelbach algorithm, and the preset PDD algorithm is combined in the solving process, and the preset mAPG algorithm is combined in the PDD algorithm to finally obtain the optimal solution of the objective function;

[0218] The minimum peak-to-average ratio is calculated according to the optimal solution, and the MIMO-OFDM system is optimized according to the minimum peak-to-average ratio to achieve peak-to-average ratio suppression.

[0219] Example 4

[0220] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the following peak-to-average ratio suppression method based on a symbol-level precoding scheme is implemented:

[0221] Obtain a downlink model of the MIMO-OFDM system, and establish a frequency domain signal model received by the user based on the downlink model;

[0222] Acquire known channel state information of the MIMO-OFDM system, and establish a symbol-level precoding scheme according to the known channel state information;

[0223] Based on the symbol-level precoding scheme, the peak-to-average ratio suppression problem is formalized and equivalently transformed to obtain the objective function;

[0224] The objective function is solved based on the Dinkelbach algorithm, and the preset PDD algorithm is combined in the solving process, and the preset mAPG algorithm is combined in the PDD algorithm to finally obtain the optimal solution of the objective function;

[0225] The minimum peak-to-average ratio is calculated according to the optimal solution, and the MIMO-OFDM system is optimized according to the minimum peak-to-average ratio to achieve peak-to-average ratio suppression.

[0226] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0227] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0228] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0229] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0230] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A peak-to-average ratio suppression method based on a symbol-level precoding scheme, characterized in that: include: Acquire a downlink model of the MIMO-OFDM system, and establish a frequency domain signal model received by the user based on the downlink model; Acquire known channel state information of the MIMO-OFDM system, and establish a symbol-level precoding scheme according to the known channel state information; Based on the symbol-level precoding scheme, the peak-to-average ratio suppression problem is formalized and equivalently transformed to obtain an objective function; Solving the objective function based on the Dinkelbach algorithm, combining the preset PDD algorithm in the solving process, combining the preset mAPG algorithm in the PDD algorithm, and finally obtaining the optimal solution of the objective function; Calculating a peak-to-average ratio minimum value according to the optimal solution, optimizing the MIMO-OFDM system according to the peak-to-average ratio minimum value to achieve peak-to-average ratio suppression; The acquiring known channel state information of the MIMO-OFDM system and establishing a symbol-level precoding scheme according to the known channel state information includes: When the data symbol sent to user k Extracted from the M-PSK constellation diagram, we get: in, represents the set of all data symbols of the M-PSK constellation diagram, e is a natural constant, j represents the complex phase part identifier, m represents the mth M-PSK constellation diagram symbol, M represents the PSK constellation diagram order, It means equivalent to; Define the first intermediate variable Note SINR k is the SINR constraint of the received signal of user k and is recorded as in, represents the complex domain, W represents the total number of OFDM subcarriers, N represents the number of base station antennas, t k represents the measure of SINR received by user k, σ represents the square root of noise variance; Using basic linear algebra The symbol-level precoding scheme is expressed as: in, Indicates sending a signal, [.] T represents the transpose operation of the matrix, and Represents the real and imaginary parts of a complex number, (.) * represents the conjugate operation, (.) H represents the conjugate transpose operation, |.| represents the modulus operation, is the channel matrix The kth row of for The wth column of is the W-dimensional discrete Fourier transform matrix, w is the index of the OFDM subcarrier, and cot represents the cotangent function; The peak-to-average ratio suppression problem is formalized and equivalently transformed based on the symbol-level precoding scheme to obtain an objective function, including: Formalizing the peak-to-average ratio suppression problem, we get the first problem: t≥t0 (6) Among them, (P1) represents the first problem, min represents the minimization operation, st represents the constraint condition, ||.|| ∞ represents the infinite norm, represents the first intermediate variable, t represents the SINR metric, and Represents the real and imaginary parts of a complex number, (.) * represents the conjugate operation, (.) H represents the conjugate transpose operation, |.| represents the modulus operation, is the channel matrix The kth row of for The wth column of is the W-dimensional discrete Fourier transform matrix, w is the index of the OFDM subcarrier, cot represents the cotangent function, represents the data symbol sent to user k, M represents the order of the PSK constellation diagram, t0≥0, t0 is the measure of the preset SINR, ||.||2 represents the two-norm, P T is the upper limit of the transmission power; The first problem is converted into an equivalent form to obtain the objective function, including: definition: Among them, vec(.) means to vectorize the matrix. for The result of vectorization, a means The vector consisting of the real and imaginary parts of represents the channel symbol information vector, represents the Kronecker product operation, I KW Represents a KW-dimensional unit matrix, where all vectors Composition Matrix Re-realize the matrix Finally get represents the second intermediate variable, represents the third intermediate variable, represents the fourth intermediate variable; The symbol-level precoding scheme represented by formula (5) is equivalently converted into the following linear matrix inequality: Among them, ι 2KW×1 Represents a unit vector of 2KW×1 dimension; The first problem is converted into the second problem: t≥t0(10) Among them, P2 is the second problem; Use the log-sum-exp function to approximate the infinite norm in the second problem: Among them, η is the smoothing parameter, η>0, a i Represents the fifth intermediate parameter, a i =[a i , a i+NW ] T , a i for a i The i-th element of ; The objective function is obtained by using an approximation for the second problem: t≥t0(15) Among them, P3 is the objective function, W represents the total number of OFDM subcarriers, and N represents the number of base station antennas; The solving of the objective function based on the Dinkelbach algorithm includes: Create a constrained fractional optimization problem as follows: Where A(θ) is a concave function with θ as a variable, B(θ) is a convex function with θ as a variable, and θ is the feasible domain corresponding to the constraints in the objective function; Mapping formula (17) to the objective function, the vector θ consists of a and t, and the optimization problem is: Among them, max represents the maximization operation, and λ is the auxiliary variable of the Dinkelbach algorithm; Mapping formula (18) to the objective function, we get the fourth problem: t≥t0 (21) Among them, min represents the minimization operation, η is the smoothing parameter, η>0; The optimal solution is obtained by solving the fourth problem through the Dinkelbach algorithm; The method combines the preset PDD algorithm in the solution process, and combines the preset mAPG algorithm in the PDD algorithm to finally obtain the optimal solution of the objective function, including: During the iteration of the Dinkelbach algorithm, the preset PDD algorithm is used each time formula (18) is calculated; In the preset PDD algorithm, the preset mAPG algorithm is used for each calculation of the sixth question during the iteration process; Combining the preset PDD algorithm and the preset mAPG algorithm, the objective function is finally solved to obtain the optimal solution; The sixth problem is obtained by introducing the PDD algorithm auxiliary variable u into the fourth problem, and the expression is: stu≤0 t≥t0 Among them, ρ is the penalty factor, and the variable Υ is the dual variable corresponding to the equality constraint. is the objective function.

2. The peak-to-average ratio suppression method based on the symbol-level precoding scheme according to claim 1, characterized in that: The frequency domain signal model is: in, is the received signal of K users on the wth OFDM subcarrier, to are the received signals of users 1 to K on the wth OFDM subcarrier, Indicates sending a signal. N represents the number of base station antennas, represents the channel matrix of the MIMO-OFDM system, represents Gaussian white noise and follows the distribution σ 2 represents the noise variance, 0 represents the zero vector, I represents the unit vector, w is the index of the OFDM subcarrier, W represents the total number of OFDM subcarriers, represents a complex domain, [.] T Represents the transpose operation on a matrix.

3. A peak-to-average ratio suppression device based on a symbol-level precoding scheme based on the method according to any one of claims 1 to 2, characterized in that: include: A frequency domain signal model building module is used to obtain a downlink model of the MIMO-OFDM system and to build a frequency domain signal model received by the user based on the downlink model; A symbol-level precoding scheme establishing module: used to obtain known channel state information of the MIMO-OFDM system and establish a symbol-level precoding scheme according to the known channel state information; An objective function building module is used to formalize and equivalently transform the peak-to-average ratio suppression problem based on the symbol-level precoding scheme to obtain an objective function; Objective function solving module: used to solve the objective function based on the Dinkelbach algorithm, combine the preset PDD algorithm in the solving process, and combine the preset mAPG algorithm in the PDD algorithm to finally obtain the optimal solution of the objective function; Peak-to-average ratio suppression module: used to calculate the minimum peak-to-average ratio according to the optimal solution, optimize the MIMO-OFDM system according to the minimum peak-to-average ratio, and achieve peak-to-average ratio suppression.

4. A peak-to-average ratio suppression system based on a symbol-level precoding scheme, characterized in that: include: A memory for storing instructions; A processor is used to execute the instructions to implement the operation of the peak-to-average ratio suppression method based on the symbol-level precoding scheme as described in any one of claims 1 to 2.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the peak-to-average ratio suppression method based on a symbol-level precoding scheme as claimed in any one of claims 1 to 2 is implemented.

Images