A System Complexity Optimization Method for Low-Altitude Large-Scale MIMO Fusion Networks

Abstract

The present invention discloses a method for optimizing system complexity for a low-altitude large-scale MIMO fusion network. In a large-scale multiple-input multiple-output system based on the angular domain of the downlink, there is a base station with M antennas and randomly distributed single-antenna users. All users are divided into K groups after user scheduling, and each group has users. It is assumed that the base station knows the channel set and the corresponding channel response information in advance. An approximate message passing algorithm is adopted to obtain a precoding design for reducing high PAPR. In addition, a new optimization iteration scheme based on the principle of minimizing the optimal mean square error is proposed to reduce the PAPR in the angular domain considering a finite set. It has important application value for reducing the peak-to-average power ratio of large-scale multiple-input multiple-output communication systems.

Description

Technical Field

[0001] The present invention belongs to the technical field of communication networks, and particularly relates to a method for optimizing system complexity for a low-altitude large-scale MIMO fusion network. Background Art

[0002] In large-scale MIMO communication systems, the transmitted signals of orthogonal frequency division multiplexing (OFDM) systems often face the problem of high peak-to-average power ratio (PAPR). High PAPR poses a great challenge to digital-to-analog converters (DACs) and power amplifiers, reducing the power efficiency of the system. Although various methods have been tried to solve the PAPR problem in OFDM systems, in complex scenarios such as large-scale MIMO angle division multiple access (ADMA) systems, the PAPR problem has not been effectively solved. At the same time, the system also needs to take into account multiple performance requirements such as low symbol error rate (SER), low inter-user interference (IUI), and computational complexity. Therefore, the research and development of efficient signal processing technologies is crucial for improving the performance of large-scale MIMO communication systems. Summary of the Invention

[0003] Object of the Invention: To address the problem of reducing the peak-to-average power ratio (PAPR) in large-scale multiple-input multiple-output (MIMO) communication systems. To solve this problem, an approximate message passing (AMP) algorithm based on Bayesian theory is developed. In addition, based on the characteristics of user sparsity, a signal recovery technology based on precoding in angle division multiple access (ADMA) large-scale multiple-input multiple-output systems is studied.

[0004] Technical Solution: A method for optimizing system complexity for a low-altitude large-scale MIMO fusion network, comprising the following steps:

[0005] Step S1: Construct a system model. Under the two assumptions that in a large-scale multiple-input multiple-output system based on the angle domain in the downlink, the base station knows the channel set and the corresponding channel response information in advance, set the system architecture and the signal transmission model;

[0006] Step S2: Construct a precoding scheme based on generalized approximate message passing. Use the generalized approximate message passing precoding scheme to estimate the posterior distribution parameters , based on the posterior likelihood probability , estimate the posterior message to obtain a posterior estimator , which is an intermediate variable related to signal processing; use the expectation-maximization process to calculate the noise variance estimator and update it to minimize the Euclidean distance ; continuously repeat the iteration process until the iteration termination condition is satisfied;

[0007] Step S3: Construct a precoding technical solution based on the optimized minimum mean square error to obtain the linear minimum mean square error estimation result; introduce a penalty factor to calculate relevant quantities, including calculating and iteratively updating the signal estimation value; calculate the precoding factor to simplify the optimal MMSE.

[0008] Preferably, the specific steps of Step S1 are as follows:

[0009] Step 101: In a large-scale multiple-input multiple-output system based on the downlink angular domain, there is a base station with M antennas and randomly distributed single-antenna users. After user scheduling, all users are divided into K groups, and each group has users. Assume that the base station knows the channel set and the corresponding channel response information in advance. Define the angular domain channel set as , and the channel set of the th group is . In addition, use to represent the channel state information in the downlink, and the corresponding angular domain CSI is represented as , and the angle rotation factor is .

[0010] Then the received signal of the kth group is

[0011] ;

[0012] where is the downlink channel coefficient of each group, is the downlink channel coefficient of the kth group. Assume that the channel responses in the downlink follow an independent and identical distribution, with an expected value of zero and a unit variance. In addition, the received signal is , the transmitted signal at the base station is represented by , and the noise vector is represented as ;

[0013] Step 102: Since multiple access is considered in the angle division, the relationship between and is

[0014] ;

[0015] where and are rotation matrices with the rotation factor , is the discrete Fourier transform matrix;

[0016] Then, the symbol vector sent to the th group of users is represented as

[0017] ;

[0018] wherein , is the transmission symbol of the k-th group, and a PAPR reduction mapping function based on precoding is given

[0019] ;

[0020] is the downlink CSI in the angular domain, can be written as

[0021] ;

[0022] wherein, and are the th column of the unit DFT matrix, the rotation factor is the th column of the inverse DFT matrix, and can be expressed as

[0023] ;

[0024] gives the precoding constraint for eliminating multi-user interference. Therefore, the received signal of the th group can be rewritten as

[0025] ;

[0026] wherein, .

[0027] Preferably, the specific steps according to step S2 are: input the received signal y, the downlink channel coefficient H, the transmission symbol vector r, and the number of iterations T; initialize, set the value of the number of iterations t to 0, and initialize the variable , and the initialization method is to set to 0 to meet the subsequent calculation requirements;

[0028] Step 201: To reduce PAPR, use the following formula:

[0029] ;

[0030] In the formula is the Frobenius norm of ;

[0031] To reduce the high PAPR in the large-scale MIMO system based on the angular domain, the posterior likelihood probability is obtained by using the GAMP framework to obtain the posterior estimator, that is , and applying the procedure and Steps are taken to update the parameters and construct a precoding scheme based on Generalized Approximate Message Passing (GAMP), demonstrating the iterative steps for iteratively reducing the PAPR based on precoding in the angular domain. The variable is initialized as , satisfying . Therefore can be minimized, where is the signal estimate value. According to the derivation of the precoding scheme based on Generalized Approximate Message Passing (GAMP), is updated as

[0032] ;

[0033] where is the number of iterations, and the value of the step size can be calculated as

[0034] ;

[0035] where .

[0036] Preferably, the iterative termination condition in step S2 is to reach a preset maximum number of iterations or the difference between two adjacent iteration results is less than a set threshold. After satisfying the termination condition, the final result is output, completing the calculation process of the GAMP-based precoding scheme to reduce PAPR.

[0037] Preferably, the specific steps according to step S3 are as follows: Define relevant parameters including precoding parameters and noise variance, and construct a channel model, where n is the noise vector, and the precoding parameters are limited within a finite set; Calculate the correlation matrix and vector;

[0038] Step 301: Since , according to the following optimization scheme, another approximate estimate can be obtained;

[0039] ;

[0040] where , is the precoding parameter, is the noise variance, and limited in a finite set is defined as ;

[0041] First, define the channel symbol vector model as

[0042] ;

[0043] where ;

[0044] Step 302: The linear MMSE estimate is

[0045] ;

[0046] Wherein,

[0047] and ;

[0048] Then there is

[0049] and

[0050] ;

[0051] Wherein, E is the expected value in probability calculation;

[0052] Step 303: Let be the penalty factor, and thus

[0053] ;

[0054] Wherein is the iteration index;

[0055] Therefore, the optimal linear MMSE iterative estimate can be written as

[0056] ;

[0057] The precoding factor is

[0058] ;

[0059] And the optimal estimate can be simplified to

[0060] ;

[0061] In the precoding technical solution based on optimized minimum mean square error (MMSE), the mapping function is represented by and the damping coefficient is represented by It can be seen from the precoding technical solution based on optimized minimum mean square error that the computational complexity of the calculation is and the overall computational complexity is

[0062] ;

[0063] Wherein, M is the total number of base station antennas; M1 is the number of user groups, that is, the antennas are divided into M1 groups; T1 is the number of iteration times, that is, the total number of iterations in the optimization process; T2 is the number of sub-iteration times, that is, the number of sub-iterations in each iteration.

[0064] Compared with the prior art, the method for analyzing the channel transmission characteristics of RIS-assisted macrocell wireless communication in the Internet of Things provided by the present invention has the following advantages:

[0065] (1) PAPR reduction

[0066] Both the AMP algorithm based on Bayesian theory and the signal recovery technology based on precoding can effectively reduce the PAPR in large-scale MIMO systems. The AMP algorithm optimizes the precoding scheme through iteration, and the precoding-based technology reduces the pressure on DACs and power amplifiers and improves the system power efficiency by reasonably designing precoding factors and other operations.

[0067] (2) Performance improvement

[0068] While reducing the PAPR, the two technologies help to achieve low SER, improve the signal transmission accuracy, reduce the interference between users, and enhance the overall performance of the system, enabling multiple users to communicate efficiently in the same system.

[0069] (3) Reduction of computational complexity

[0070] Compared with some traditional methods, the two technologies of the present invention have lower computational complexity on the premise of ensuring the system performance. The AMP algorithm utilizes Bayesian theory and the approximate message passing mechanism, and the precoding-based technology reduces the system computational burden and resource consumption by optimizing the iterative scheme, and is more suitable for the complex computational requirements of large-scale MIMO systems. Description of the drawings

[0071] Figure 1 : SER of different PAPR reduction schemes and different signal-to-noise ratios;

[0072] Figure 2 : PAPR performance of different precoding-based PAPR reduction schemes;

[0073] Figure 3 : IUI performance of Algorithm 1 and Algorithm 2, where (a) is Algorithm 1 and (b) is Algorithm 2;

[0074] Figure 4 : SER performance under different signal-to-noise ratios and different conditions;

[0075] Figure 5 : When the modulation mode is 16QAM and M = 128, the peak-to-average power ratio (PAPR) performance under different (a) is and (b) is ;

[0076] Figure 6: When the modulation mode is 16QAM, different The inter-user interference (IUI) performance under , (a) is . Detailed implementation manners

[0077] The present invention will be further illustrated below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. After reading the present invention, various equivalent modifications made by those skilled in the art to the present invention all fall within the scope defined by the appended claims of this application.

[0078] In the solution of the present invention, a method for optimizing the system complexity of a low-altitude large-scale MIMO fusion network.

[0079] Specifically, it includes the following 3 steps:

[0080] The first step is to construct a system model. Under the two assumptions of a large-scale multiple-input multiple-output system based on the angle domain in the downlink and the base station knowing the channel set and the corresponding channel response information in advance, the system architecture is set and the signal transmission model is established.

[0081] The acquisition method of the channel coefficient is determined based on the pre-known channel information, and certain statistical characteristics (such as independent and identically distributed with zero expectation and unit variance, etc.) are always followed in the calculation for subsequent calculations.

[0082] Step 101: Assume that in a large-scale multiple-input multiple-output (MIMO) system based on the downlink angle domain, there is a base station with M antennas and randomly distributed single-antenna users. All users are divided into K groups after user scheduling, and each group has users. Assume that the base station knows the channel set and the corresponding channel response information in advance. Define the angle domain channel set as , and the channel set of the th group is . In addition, use to represent the channel state information (CSI) in the downlink, and the corresponding angle domain CSI is represented as , and the angle rotation factor is .

[0083] The received signal of the kth group is

[0084] ;

[0085] where is the downlink channel coefficient of each group. Assume that the channel responses in the downlink follow an independent and identical distribution, with an expected value of zero and a unit variance. In addition, the received signal is , the transmitted signal at the base station is represented by . The noise vector is expressed as .

[0086] Step 102: Since multiple access is considered in the angle division, the relationship between and is

[0087] ;

[0088] where and are rotation matrices with rotation factor , is the discrete Fourier transform (DFT) matrix.

[0089] Then, the symbol vector sent to the k-th group of users is represented as

[0090] ;

[0091] where , given a PAPR reduction mapping function based on precoding

[0092] ;

[0093] is the downlink CSI in the angular domain, can be written as

[0094] ;

[0095] where and are the m-th columns of the unit DFT matrix. The rotation factor is the n-th column of the inverse DFT (IDFT) matrix, which can be expressed as

[0096] ;

[0097] Next, the precoding constraint for eliminating multi-user interference is given. Therefore, the received signal of the k-th group can be rewritten as

[0098] ;

[0099] where .

[0100] Second step, construct a precoding scheme based on generalized AMP (GAMP), input the received signal y, downlink channel coefficient H, transmitted symbol vector r, number of iterations T; initialize, set the value of the iteration number t to 0, and initialize the variable in the following way Set it to 0 to meet the subsequent calculation requirements; use the Generalized Approximate Message Passing (GAMP) scheme to estimate the posterior distribution , and based on the posterior likelihood probability , estimate the posterior message to obtain the posterior estimator , which is an intermediate variable related to signal processing; use the Expectation-Maximization (EM) process to calculate the noise variance estimator and update it to minimize ; continuously repeat the above iterative process until the iterative termination condition is met; the iterative termination condition is reaching the preset maximum number of iterations or the difference between the results of two adjacent iterations is less than the set threshold; after the termination condition is met, output the final result to complete the calculation process of the GAMP-based precoding scheme to reduce PAPR.

[0101] Specifically, the application of the Approximate Message Passing (AMP) algorithm based on Bayesian theory in large-scale MIMO communication systems can effectively reduce the Peak-to-Average Power Ratio (PAPR), while achieving low Symbol Error Rate (SER) and low Inter-User Interference (IUI), and has lower computational complexity compared to traditional methods, which is suitable for the signal processing requirements in large-scale MIMO systems.

[0102] Step 201: To reduce PAPR, we have:

[0103] ;

[0104] where is 's Frobenius norm. The convex optimization problem in the above equation can be solved by designing .

[0105] The Generalized AMP (GAMP) algorithm is a Bayesian iterative algorithm based on the factor graph principle for obtaining likelihood estimates. The Expectation-Maximization (EM) scheme can be used to estimate certain variables and parameters.

[0106] Here, the EM-TGM-GAMP technique is extended to reduce the high PAPR in angle-domain based large-scale MIMO systems. The posterior likelihood probability is obtained by using the GAMP framework to obtain the posterior estimator, that is . In addition, the procedure and steps in the embedded EM algorithm are also applied to update the parameters. Algorithm 1 demonstrates the iterative steps for the iterative reduction of PAPR based on precoding in the angle domain. Algorithm 1 is to construct a precoding scheme based on the Generalized Approximate Message Passing AMP (GAMP). It should be noted that the variable is initialized as to satisfy , so can be minimized, where is the signal estimation value. According to the derivation of Algorithm 1, is updated to

[0107] ;

[0108] where is the number of iterations, and the value of the step size can be calculated as

[0109] ;

[0110] where

[0111] .

[0112] In the third step, a precoding technical solution based on the optimized minimum mean square error (MMSE) is constructed. Define relevant parameters, including precoding parameters and noise variance, and construct a channel model, where n is the noise vector, and the precoding parameters are limited within a finite set; calculate the correlation matrix and vector to obtain the linear minimum mean square error (MMSE) estimation result; introduce a penalty factor to calculate relevant quantities, including calculating the iteratively updated signal estimation value; calculate the precoding factor, and simplify the optimal MMSE under specific conditions (such as when the value is large, it is a parameter related to the system).

[0113] Specifically, the optimized iterative scheme based on the optimal MMSE principle further optimizes the signal recovery effect and improves the overall system performance by reasonably selecting and updating relevant parameters (such as precoding parameters, penalty factors, etc.) during the calculation process. Then, calculate the correlation matrix and vector to obtain the linear minimum mean square error (MMSE) estimation result; next, introduce a penalty factor to calculate relevant quantities, including calculating the iteratively updated signal estimation value.

[0114] Step 301: Since , according to the following optimization scheme, another approximate estimate can be obtained.

[0115] ;

[0116] where , is the precoding parameter, is the noise variance, and defined within a finite set is .

[0117] First, we define the channel model as

[0118] ;

[0119] where ;

[0120] Step 302: The linear MMSE estimate is

[0121] ;

[0122] where

[0123] and ;

[0124] Then there is

[0125] and

[0126] ;

[0127] Step 303: Let be the penalty factor, from which we can get

[0128] ;

[0129] is the iteration index.

[0130] Therefore, the optimal linear MMSE iterative estimate can be written as

[0131] ;

[0132] The precoding factor is

[0133] ;

[0134] In addition, the optimal estimate of the optimal linear MMSE iteration can be simplified to

[0135] ;

[0136] At being larger:

[0137] The simplified MMSE-based optimal precoding PAPR reduction algorithm is shown in Algorithm 2. Algorithm 2 represents the technical solution of constructing precoding based on optimized minimum mean square error (MMSE). In Algorithm 2, the mapping function is represented by and the damping coefficient is represented by ; It can be seen from Algorithm 2 that the computational complexity of the calculation is and the overall computational complexity is

[0138] .

[0139] Based on the above derivation, Figure 1The results show that all SER curves decrease as the SNR increases. In the case of using QPSK or 16-QAM and equipping the BS with antennas, when the SNR varies from -10 dB to 30 dB, the SER of Algorithm 1 changes slowly. Specifically, when Algorithm 1 uses 16-QAM at high SNR values, the SER value drops from approximately 0.1 dB to approximately dB. It can also be seen from Figure 1 that the SER value of Algorithm 2 is relatively small at low SNR. For example, when SNR = 10 dB and QPSK, the SER value using Algorithm 2 is less than dB; when the SNR value is greater than 20 dB, when the BS has or antennas, the SER curve of 16-QAM using Algorithm 2 remains almost unchanged; in addition, the clipping ratios of the MF and ZF methods are equal to 1.5. It is observed that when the SNR is less than 30 dB, the SER performance of MF and ZF is slightly better than that of Algorithm 1. In addition, the SER performance of the traditional MMSE scheme is close to that of Algorithm 2. However, it is obtained at the cost of higher PAPR suppression performance.

[0140] Figure 2 In Figure 3 shows the IUI performance of Algorithm 1 and Algorithm 2 when the modulation mode is 16-QAM and is equal to 256; it can be seen that the IUI value of Algorithm 1 is approximately between -55 dB and -25 dB. On the contrary, in the same simulation environment, the IUI value of Algorithm 2 is between -16 dB and -7 dB. It should be noted that Figures 1-3 is implemented based on constant envelope precoding with finite symbols.

[0141] Figure 4 It can be seen from

[0142] Figure 5 and Figure 6 show the PAPR performance and IUI performance in the same simulation environment as Figure 4 . It can be seen from Figure 5 's (a) and (b) that when the radius difference value ranges from 0.05 to 0.1, the PAPR value expands from (0 dB, 0.045 dB) to (0 dB, 0.06 dB). Similarly, Figure 6 's IUI result in (a) is approximately from -22 dB to -9 dB, while Figure 6 's in (b) is approximately from -24 dB to -9 dB.

[0143] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for optimizing the system complexity of a low-altitude large-scale MIMO fusion network, characterized in that, It includes the following steps: Step S1: Construct a system model. Under the two assumptions that in a downlink angle-domain massive multiple-input multiple-output system, the base station knows the channel set and the corresponding channel response information in advance, set the system architecture and signal transmission model; Step S2: Construct a precoding scheme based on generalized approximate message passing, and use the generalized approximate message passing precoding scheme to estimate the posterior distribution parameters , based on the posterior likelihood probability , estimate the posterior message to obtain a posterior estimator , is an intermediate variable related to signal processing; use the expectation maximization process to calculate the noise variance estimator and update it to minimize the Euclidean distance ; Continuously repeat the iterative process until the iterative termination condition is met; Step S3: Construct a precoding technical solution based on optimizing the minimum mean square error to obtain the linear minimum mean square error estimation result; Introduce a penalty factor to calculate relevant quantities, including calculating and iteratively updating the signal estimation value; Calculate the precoding factor and simplify the optimal MMSE.

2. The system complexity optimization method for a low-altitude large-scale MIMO fusion network according to claim 1, characterized in that The specific steps of step S1 are as follows: Step 101: In a large-scale multiple-input multiple-output system based on the downlink angular domain, there is a base station with M antennas and randomly distributed single-antenna users. After user scheduling, all users are divided into K groups, and each group has users. Assuming that the base station knows the channel set and the corresponding channel response information in advance, the angular domain channel set is defined as , and the channel set of the th group is . In addition, use to represent the channel state information in the downlink, and the corresponding angular domain CSI is expressed as , and the angle rotation factor is . Then the received signal of the kth group is ； wherein is the downlink channel coefficient for each group, is the downlink channel coefficient for the k-th group. It is assumed that the channel responses in the downlink follow an independent and identical distribution with an expected value of zero and a unit variance. In addition, the received signal is , and the transmitted signal at the base station is denoted by , and the noise vector is denoted as ; Step 102: Since multiple access is considered in the angle division, therefore and The relationship between them is ； Among them, and are rotation matrices with rotation factors , is a discrete Fourier transform matrix; Then, the symbol vector representation sent to the group of users is ； Among them , is the transmission symbol of the k-th group, and a PAPR reduction mapping function based on precoding is given ； is the downlink CSI in the angular domain, can be written as ； Among them, and are the th columns of the unit DFT matrix, and the rotation factor is the th column of the inverse DFT matrix, which can be expressed as ； The precoding constraints for eliminating multi-user interference are given. Therefore, the received signal of the th group can be rewritten as ； Among them, 。 3. The system complexity optimization method for a low-altitude large-scale MIMO fusion network according to claim 1, characterized in that The specific steps according to step S2 are as follows: input the received signal y, the downlink channel coefficient H, the transmitted symbol vector r, and the number of iterations T; initialize, set the value of the number of iterations t to 0, and initialize the variable , and the initialization method is to set to 0 to meet the requirements of subsequent calculations; Step 201: To reduce PAPR, use the following formula: ； In the formula is the Frobenius norm of; Reducing High PAPR in Angular Domain-based Massive MIMO Systems, Posterior Likelihood Probability is to obtain a posterior estimator by using the GAMP framework, that is , applying the procedure and steps in the embedded EM algorithm to update the parameters, construct a precoding scheme based on Generalized Approximate Message Passing (AMP), demonstrate the iterative steps of the precoding-based PAPR reduction iteratively in the angular domain, and the variable is initialized to , satisfying , so is minimized, where is the signal estimate value. According to the derivation of the precoding scheme based on Generalized Approximate Message Passing (AMP), is updated to ； Among them is the number of iterations, and the value of the step size can be calculated as ； Among them 。 4. The method for optimizing the system complexity of a low-altitude large-scale MIMO fusion network according to claim 1, wherein The iteration termination condition in step S2 is to reach the preset maximum number of iterations or the difference between the results of two adjacent iterations is less than the set threshold. After meeting the termination condition, output the final result to complete the calculation process of the GAMP-based precoding scheme to reduce PAPR.

5. The system complexity optimization method for a low-altitude large-scale MIMO fusion network according to claim 1, characterized in that The specific steps according to step S3 are as follows: Define relevant parameters including precoding parameters and noise variance, and construct a channel model, where n is a noise vector, and the precoding parameters are limited within a finite set; calculate the correlation matrix and vector; Step 301: Since , according to the following optimization scheme, obtain another approximate estimate; ； Among them , is a precoding parameter, is the noise variance, which is restricted in a finite set and is defined as ; First, define the channel symbol vector model as ； Among them ; Step 302: The linear MMSE estimate is ； where, and ; then there is and ; where E is the expected value in probability calculation; Step 303: Set as the penalty factor, and thus we can obtain ； Among them is the iteration exponent; Therefore, the optimal linear MMSE iterative estimation can be written as ； Precoding factor is ； and the optimal estimation can be simplified to ； In the pre - coding technical solution based on optimized minimum mean - square error (MMSE), the mapping function is represented by , and the damping coefficient is represented by , which is obtained from the pre - coding technical solution based on optimized minimum mean - square error. The computational complexity of the calculation is , and the overall computational complexity is ； where M is the total number of base station antennas; M1 is the number of user groups, that is, the antennas are divided into M1 groups; T1 is the number of iterations, that is, the total number of iterations in the optimization process; T2 is the number of sub-iterations, that is, the number of sub-iterations in each iteration.

Images