Antenna signal decoding method and device

Abstract

The embodiment of the invention provides an antenna signal decoding method and device, and the method comprises the steps: carrying out the channel estimation of a received signal, obtaining a channel estimation matrix, and enabling the received signal to be an uplink signal, received by a plurality of receiving antennas, of a plurality of transmitting antennas; determining an adjustment parameter which enables lattice Gaussian sampling probability search to converge; according to the adjustment parameter and the channel estimation matrix, carrying out lattice Gaussian sampling probability-based search decoding on the received signal to obtain a candidate vector set containing a plurality of candidate vectors; and determining a decoding result of the receiving antenna according to a plurality of candidate vectors in the candidate vector set. The problem that antenna signal decoding cannot meet high performance and low calculation complexity at the same time in the prior art can be solved at least.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communication technology, and more specifically, to a method and device for decoding antenna signals. Background Art

[0002] Signal detection in multiple-input, multiple-output (MIMO) communication systems is a key technology for 5G. While the Maximum Likelihood (ML) algorithm can ensure optimal MIMO signal detection performance, its computational complexity increases exponentially with the number of antennas, making it difficult to apply in practice. Based on the mathematical expression of ML, the MIMO signal detection problem can be transformed into the Closest Vector Problem (CVP) in the field of lattice decoding. Typical algorithms for solving this problem include lattice reduction-assisted decoding, sphere decoding (SD), fixed-complexity SD, and K-best decoding. However, as the system dimensionality increases, detection algorithms within traditional detection frameworks, such as lattice reduction-assisted decoding, suffer significant performance losses. Approximate ML algorithms, such as SD and algorithms designed to reduce SD complexity (such as fixed-complexity SD and K-best decoding), also face the problem of rapidly increasing complexity or performance degradation.

[0003] Sampling decoding converts the traditional decoding problem into a sampling problem, and its optimal solution has the shortest Euclidean distance, which corresponds to the maximum sampling probability. However, unlike the continuous Gaussian distribution, it is not easy to sample the low-dimensional discrete Gaussian distribution; therefore, the key to the success of sampling decoding lies in how to sample the discrete Gaussian distribution. In the past, the main idea to solve this problem was based on the Monte Carlo-Markov chain (MCMC). However, due to the inherent randomness in the discrete Gaussian distribution, the possibility of missing the optimal solution always exists, resulting in unavoidable performance loss. In addition, independent and identically distributed sampling also requires repeated calculations, which also increases unnecessary computational complexity.

[0004] There is no solution to the problem in related technologies that antenna signal decoding cannot simultaneously meet high performance and low computational complexity. Summary of the Invention

[0005] The embodiments of the present application provide a method and apparatus for decoding an antenna signal, so as to at least solve the problem in the related art that antenna signal decoding cannot simultaneously meet high performance and low computational complexity.

[0006] According to one embodiment of the present application, a method for decoding an antenna signal is provided, the method comprising:

[0007] Performing channel estimation on a received signal to obtain a channel estimation matrix, wherein the received signal is an uplink signal of a plurality of transmitting antennas received by a plurality of receiving antennas;

[0008] Determining adjustment parameters that make the Gaussian sampling probability search converge;

[0009] performing Gaussian sampling probability search decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including a plurality of candidate vectors;

[0010] Determine a decoding result of the receiving antenna according to multiple candidate vectors in the candidate vector set.

[0011] According to another embodiment of the present application, a decoding device for an antenna signal is provided, the device comprising:

[0012] a channel estimation module, configured to perform channel estimation on a received signal to obtain a channel estimation matrix, wherein the received signal is an uplink signal of multiple transmitting antennas received by multiple receiving antennas;

[0013] A first determination module is used to determine the adjustment parameters that make the Gaussian sampling probability search converge;

[0014] a decoding module, configured to perform Gaussian sampling probability search decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including a plurality of candidate vectors;

[0015] The second determining module is configured to determine a decoding result of the receiving antenna according to a plurality of candidate vectors in the candidate vector set.

[0016] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is configured to execute the steps of any one of the above method embodiments when running.

[0017] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any one of the above method embodiments.

[0018] According to another embodiment of the present application, a computer program product is provided, including computer program instructions, wherein the computer program instructions enable a computer to implement the steps in any one of the above method embodiments.

[0019] In an embodiment of the present application, channel estimation is performed on a received signal to obtain a channel estimation matrix. The received signal is an uplink signal of multiple transmitting antennas received by multiple receiving antennas; an adjustment parameter that makes the Gaussian sampling probability search converge is determined, and the received signal is decoded based on the Gaussian sampling probability search according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set containing multiple candidate vectors; the decoding result of the receiving antenna is determined based on the multiple candidate vectors in the candidate vector set, which can solve the problem in the related art that antenna signal decoding cannot simultaneously meet high performance and low computational complexity. While ensuring high performance, it has lower complexity and can be flexibly adapted to various application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a hardware structure block diagram of a computer device for the antenna signal decoding method according to an embodiment of the present application;

[0021] Figure 2 is a flowchart of a method for decoding antenna signals according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a massive MIMO detection algorithm based on Gaussian sampling probability search of this embodiment;

[0023] Figure 4 2 is a schematic diagram of a two-dimensional Gaussian distribution of a massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment;

[0024] Figure 5 3 is a schematic diagram comparing the probability distribution of the massive MIMO detection algorithm based on Gaussian sampling probability search under different adjustment parameters of this embodiment;

[0025] Figure 6 4 is a flowchart of the implementation of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment;

[0026] Figure 7 2 is a schematic diagram of the entire search and decoding process of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment;

[0027] Figure 8 Statistical diagram of the number of candidate vectors of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment under different initial search range conditions;

[0028] Figure 9 is a block diagram of a decoding device for antenna signals according to an embodiment of the present application;

[0029] Figure 10 4 is a system structure diagram of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment. DETAILED DESCRIPTION

[0030] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0031] It should be noted that the terms "first", "second", etc. in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0032] The method embodiments provided in the embodiments of the present application can be executed in a computer device or similar computing device. Taking running on a computer device as an example, Figure 1 This is a hardware structure block diagram of a computer device for the antenna signal decoding method according to an embodiment of the present application. Figure 1 As shown, the computer device may include one or more ( Figure 1 Only one is shown) a processor 102 (the processor 102 may include but is not limited to a microprocessor MCU or a processing device such as a programmable logic device) and a memory 104 for storing data. The computer device may also include a transmission device 106 and an input / output device 108 for communication functions. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer device. For example, the computer device may also include Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0033] Memory 104 can be used to store computer programs, such as software programs and modules of application software, such as the computer program corresponding to the antenna signal decoding method in the embodiment of the present application. Processor 102 executes the computer program stored in memory 104 to execute various functional applications and single-board matching, thereby implementing the above-mentioned method. Memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 may further include memory remotely located relative to processor 102, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0034] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by a communications provider of a computer device. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] In this embodiment, a method for decoding antenna signals running on the above-mentioned computer device is provided. Figure 2 is a flowchart of a method for decoding antenna signals according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0036] Step S202, performing channel estimation on the received signal to obtain a channel estimation matrix, wherein the received signal is an uplink signal of multiple transmitting antennas received by multiple receiving antennas;

[0037] Step S204, determining adjustment parameters that make the Gaussian sampling probability search converge;

[0038] Step S206 , performing probabilistic searching on lattice Gaussian distribution decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including a plurality of candidate vectors;

[0039] Step S208: Determine a decoding result of the receiving antenna according to multiple candidate vectors in the candidate vector set.

[0040] In this embodiment, the above step S208 may specifically include: respectively determining the Euclidean distances of multiple candidate vectors in the candidate vector set; and determining a candidate vector with the smallest Euclidean distance among the multiple candidate vectors as the target decoding result.

[0041] Through the above steps S202 to S208, the problem in related technologies that antenna signal decoding cannot simultaneously meet high performance and low computational complexity can be solved. While ensuring high performance, it has lower complexity and can be flexibly adapted to various application scenarios.

[0042] In this embodiment, the above step S206 may specifically include: performing the following steps S1-S4 from the last stream to the first stream of the received signal:

[0043] S1, determine the candidate constellation points of the current stream based on the candidate vector set before the current stream;

[0044] Furthermore, the above step S1 may specifically include: selecting candidate vectors of the current stream in sequence from the candidate vector set before the current stream; calculating the Babai point of the candidate vector of the current stream, and determining the candidate constellation point of the current stream according to the Babai point. Furthermore, the Babai point of the candidate vector may be calculated by the following method: in, represents the Babai point of the candidate vector of the current stream i, Represents the candidate constellation points of the k-th stream, i+1≤k≤n; determine multiple constellation points closest to the Babai point of the candidate vector as the candidate constellation points of the current stream.

[0045] S2, determining the lattice point Gaussian distribution probability of the candidate constellation points of the current stream according to the adjustment parameters, and determining the search range of the candidate constellation points of the current stream according to the lattice point Gaussian distribution probability of the candidate constellation points of the current stream;

[0046] S3, performing pruning, recursive processing, or serial interference cancellation (SIC) detection on the candidate constellation points of the current stream according to the search range of the candidate constellation points of the current stream to obtain a target candidate vector;

[0047] S4, updating the candidate vector set of the current stream according to the target candidate vector.

[0048] Furthermore, the above step S2 may specifically include:

[0049] S21, respectively determining the difference between each candidate constellation point of the current stream and the Babai point of the candidate vector of the current stream;

[0050] S22, determining a grid point Gaussian distribution probability of candidate constellation points of the current stream according to the adjustment parameter and the difference;

[0051] Furthermore, it can be achieved by a normalized Gaussian-like function, that is, a Gaussian-like function is introduced, and the adjustment parameter and the difference are used as the exponential term of the Gaussian-like function; the Gaussian-like function is normalized to obtain the grid point Gaussian distribution probability of the candidate constellation point of the current stream. Specifically, Among them, the numerator is the exponential term of the Gaussian function, and the denominator is the normalization process, that is, the normalized Gaussian-like function. represents the candidate constellation point j of the current stream i, is the grid point Gaussian distribution probability of the candidate constellation point j of the current stream i, is the above difference, σ is the adjustment parameter;

[0052] The search range of candidate constellation points of the current stream is determined according to the above product in the following manner: in, is the search range of candidate constellation point j of current stream i, is the search range of the candidate constellation point j of the i+1th stream, is the lattice Gaussian distribution probability of the candidate constellation point j of the previous stream i+1 of the current stream i, K is the initial value of the search range, which can be determined through simulation. The initial value of the search range is used to achieve a trade-off between performance and complexity. The higher the performance, the larger K is, and the lower the complexity, the smaller K is.

[0053] S23, determining the product of the search range of the lattice point Gaussian distribution probability of the candidate constellation points of the previous stream of the current stream and the lattice point Gaussian distribution probability of the candidate constellation points of the current stream;

[0054] S24, determining that the product is a search range for candidate constellation points of the current stream.

[0055] Furthermore, the above-mentioned step S3 may specifically include: for the candidate constellation points of the current stream, when the search range of the candidate constellation points of the current stream is less than 1, performing a pruning operation, that is, pruning the candidate constellation points of the current stream; when the search range of the candidate constellation points of the current stream is greater than or equal to 1 and less than 2, performing SIC detection, that is, decoding the remaining streams to obtain a candidate vector, and adding the candidate vector to the candidate vector set, and selecting a point closest to the Babai point as the candidate constellation point during the decoding process of the remaining streams; when the search range of the candidate constellation points of the current stream is greater than or equal to 2, performing a recursive operation, that is, decoding the next stream of the current stream, until the candidate vector is output for the first stream and added to the candidate vector set.

[0056] In this embodiment, the above step S204 may specifically include: performing real number processing on the channel estimation matrix to obtain a processed channel estimation matrix; further, performing real number processing in the following manner:

[0057] Where H is the channel estimation matrix with dimension m′×n′, m′ is the number of receive antennas, n′ is the number of transmit antennas, and B is the processed channel estimation matrix with dimension m×n, where m=2m′ and n=2n′.

[0058] Performing QR decomposition on the processed channel estimation matrix to obtain a first matrix and a second matrix. Further, decomposing the processed channel estimation matrix into an m×n dimensional orthogonal matrix to obtain a first matrix Q, and decomposing the processed channel estimation matrix into an n×n dimensional upper triangular matrix to obtain a second matrix R.

[0059] The adjustment parameters in this embodiment are used to ensure algorithm convergence and relatively reasonable performance and complexity. The adjustment parameters can be determined based on performance requirements and computational complexity requirements. Further, the adjustment parameters can be determined based on the second matrix. Specifically, the adjustment parameters can be determined based on the second matrix in the following manner: Among them, c is the received signal after processing, σ is the adjustment parameter, r ii is the element in the i-th row and i-th column of the second matrix R. It is sufficient if one of the adjustment parameters can meet the performance requirements and computational complexity. The above determination method is only an example, and other methods are also possible and are not specifically limited here.

[0060] In an optional embodiment, the method further includes: performing real number processing on the received signal to obtain a processed received signal; further, the real number processing can be performed in the following manner: Where c is the processed received signal, c′ is the received signal;

[0061] The processed received signal is transformed according to the first matrix in the following manner to obtain a transformed received signal: y = Q T c,y∈R n , where y is the transformed received signal, c is the processed received signal, R is the second matrix, n=2n′, n′ is the number of transmitting antennas, and the transformed received signal is then subjected to probability search decoding based on Gaussian sampling, which can further reduce the computational complexity.

[0062] The massive MIMO detection based on Gaussian sampling probability search proposed in this embodiment introduces additional degrees of freedom (parameters) to solve the detection problem by constructing a probabilistic twin model. On this basis, a deterministic sampling scheme is designed to effectively overcome the problem of random sampling decoding falling into local suboptimal solutions. Both theoretical analysis and simulations show that the method proposed in this embodiment improves performance while reducing computational complexity. Specifically, it includes the following steps:

[0063] Step 1, data preprocessing, mainly realizes the acquisition of the real system channel estimation matrix and the received signal, decoupling the received signal and initializing the algorithm parameters. The specific steps include:

[0064] For a MIMO system with n′ transmitting antennas and m′ receiving antennas, channel estimation is performed using pilot signals to obtain a channel estimation matrix H with dimensions of m′×n′.

[0065] To facilitate the calculation of Gaussian probability, the complex system is converted into an equivalent real system. This is divided into two parts: the realization of the channel estimation matrix and the realization of the received signal.

[0066] The complex channel estimation result is converted to real number, and the channel estimation result of the real system is obtained as follows:

[0067]

[0068] The matrix dimension is m×n, where m=2m′ and n=2n′. The specific real number conversion process is to divide the channel estimation matrix of the real system into blocks, where the matrix block of the first m′ rows and n′ columns is the real part of H, the matrix block of the first m′ rows and n′ columns is the imaginary part of -H, the matrix block of the last m′ rows and n′ columns is the imaginary part of H, and the matrix block of the last m′ rows and n′ columns is the real part of H.

[0069] Similarly, the received signal is realized. The received system of the real system is an m×1-dimensional vector, where the first m′ elements are the real part of the received complex signal c′, and the last m′ elements are the imaginary part of the received complex signal c′, that is,

[0070]

[0071] To achieve the decoupling of the received signal, perform narrow QR decomposition on the B matrix to obtain matrices Q and R, where Q is an m×n dimensional orthogonal matrix and R is an n×n dimensional upper triangular matrix. Apply Q to c to obtain the transformed received signal y∈R n , y=Q T c.

[0072] Set the initial search range K and set adjustable parameters Among them, r ii Initialize the candidate vector set L for the i-th row and i-th column element of the R matrix, that is, L = φ.

[0073] Step 2: Start decoding from the last stream (nth stream), then decode the n-1th stream, until the decoding of the 1st stream is completed. The following takes the i-th stream as an example to illustrate the decoding process. Candidate vectors are selected in the candidate vector sets of the nth stream to the i+1th stream in turn, and the one-dimensional grid point Gaussian distribution probability of the Babai point and the candidate constellation point (all constellation points mentioned in this embodiment refer to the vector point set in the constellation diagram of the real number system after modulation) is calculated. and search scope Assume that the candidate vectors selected from the nth stream to the i+1th stream are The specific steps include:

[0074] Calculate Babai Point In order to obtain candidate constellation points and calculate the one-dimensional grid point Gaussian distribution probability, when 1≤i<n, in, represents the candidate constellation point of the kth stream, i+1≤k≤n; when i=n,

[0075] Calculate the one-dimensional grid point Gaussian distribution probability of candidate constellation points For subsequent calculations Search scope One-dimensional grid point Gaussian distribution probability The expression is

[0076]

[0077] The superscript j represents the candidate constellation point ranked at the jth position after all constellation points to the Babai point are sorted from smallest to largest. J represents the number of valid candidate constellation points. Theoretically, it can be proved that J can meet the performance requirements by taking 3. Note that the one-dimensional grid point Gaussian distribution probability mentioned in this embodiment is different from the common one-dimensional grid point Gaussian distribution probability. The common one-dimensional grid point Gaussian distribution probability is expressed as:

[0078]

[0079] The denominator summation term is performed on an integer set or a constellation point set. In this embodiment, only valid constellation points are selected.

[0080] calculate Search scope As the basis for subsequent candidate vector selection pruning, recursion or protection operations, when 1≤i<n, here for The parent node (the corresponding node when searching at layer i+1); when i=n,

[0081] Step 3: According to The size of the candidate vector is pruned, recursively or protected, that is, for each candidate node judge Which of the following situations applies:

[0082] like Then Pruning is performed, that is, the constellation point is considered not to belong to the candidate constellation points;

[0083] like Then perform SIC detection on the remaining stream, that is, the remaining stream only retains the point closest to the Babai point as the candidate constellation point, and finally obtains a candidate vector And join the set L;

[0084] like Then enter the next layer (i-1 layer) detection and repeat the above steps until the candidate vector is output when i=1 And join the set L.

[0085] Step 4: Calculate the Euclidean distance of all candidate sets and obtain the final decoding result. Set L contains all candidate vectors, and calculate the Euclidean distance of each candidate vector. Select the candidate vector corresponding to the minimum distance As the decoding result,

[0086] The present application is further described below with reference to specific embodiments.

[0087] Figure 3 Schematic diagram of the massive MIMO detection algorithm based on Gaussian sampling probability search of this embodiment, as shown in FIG. Figure 3 As shown, the MIMO complex system includes n′ transmitting antennas and m′ receiving antennas. The relationship between the signal c′ received by the receiving antenna and the transmitted signal x′ is as follows:

[0088] c′=Hx′+w′;

[0089] Where H is the complex channel matrix of dimension m′×n′, w′ is the received m′×1-dimensional complex Gaussian white noise with mean zero and standard deviation σ w′ Gaussian distribution, x′ is an n′×1 vector, representing the transmitted complex signal, each element in the vector comes from the set X of constellation points, c′ is an m′×1 vector, representing the received signal.

[0090] The optimal MIMO detection is the ML algorithm, which is expressed as follows:

[0091]

[0092] The above formula essentially seeks to minimize the Euclidean distance. However, this embodiment constructs a probabilistic twin model from a probability perspective, converting the minimization of the Euclidean distance into the maximization of the Gaussian distribution probability. The expression of the Gaussian distribution probability is as follows:

[0093]

[0094] Figure 4 Schematic diagram of the two-dimensional Gaussian distribution of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment, as shown in FIG. Figure 4 As shown, when the prior probability satisfies the uniform distribution, the maximum a posteriori probability expression is similar in form to the Gaussian probability, but the physical meaning of the parameters is completely different. The σ in the maximum a posteriori expression represents the noise power, while the adjustment parameter σ of the probabilistic twin model constructed in this embodiment is an adjustable parameter, which introduces an additional degree of freedom. By changing the size of σ, the discrimination of the Gaussian distribution can be made greater. Figure 5: is a schematic diagram of the probability distribution comparison of the massive MIMO detection algorithm based on Gaussian sampling probability search under different adjustment parameters of this embodiment, such as Figure 5 As shown in the figure, the calculation and solution of Gaussian distribution are difficult. Although the classical sampling search algorithm can solve the problem of Gaussian distribution to a certain extent, the randomness of sampling will cause the optimization problem to fall into a local suboptimal solution. To overcome this problem, this embodiment uses the fact that the Klein sampling probability feature is similar to the probability feature of Gaussian distribution, and uses Klein sampling probability instead of Gaussian distribution for solution, that is,

[0095]

[0096] Where, the Klein sampling probability is

[0097] in, represents the Babai point of the n-i+1th stream. On this basis, a deterministic sampling strategy is designed based on the Klein sampling probability to achieve the effect that the sampling probability in the candidate vector set is higher than the deterministic threshold. The candidate vector set can be represented by the following expression:

[0098]

[0099] Probability threshold P threshold is set to Obviously, the larger K is, the greater the probability threshold P is. threshold The smaller it is, the more candidate vectors the candidate vector set L contains, and the larger the corresponding probability search range.

[0100] The following is an example of a MIMO system with 2 transmitting antennas n′ and 4 receiving antennas m′. Figure 6 is a flow chart of the implementation of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment, as shown in FIG. Figure 6 As shown, including:

[0101] Step S61, data preprocessing, mainly realizes the acquisition of the real system channel estimation matrix and the received signal, decoupling the received signal and initializing the algorithm parameters. Specifically includes:

[0102] Step S611 performs channel estimation to obtain a channel estimation matrix H. For a MIMO system including n′=2 transmitting antennas and m′=4 receiving antennas, channel estimation is performed using pilot signals to obtain a channel estimation matrix H with dimensions of m′×n′, which can be expressed as

[0103]

[0104] Where h ijis the element in the i-th row and j-th column of the matrix H, representing the channel response from the j-th transmitting antenna to the i-th receiving antenna.

[0105] Step S612: convert the complex system into an equivalent real system, obtain the channel estimation matrix B and the received signal c in the real system, and define B = [b ij ] m×n , m=2m′=8,n=2n′=4, then B and H satisfy:

[0106]

[0107] here Indicates h ij The real part of Indicates h ij The imaginary part of the real system. The received signal c=[c ij ] n×1 .

[0108] Step S613: Decouple the received signal and perform narrow QR decomposition on the matrix B to obtain a first matrix Q and a second matrix R. The matrix Q is applied to the received signal c, where Q = [q ij ] n×m , where n = 8, m = 4, R is an upper triangular matrix of m × m dimensions, which can be expressed as

[0109]

[0110] Apply this matrix to c to obtain the transformed received signal y, y = Q T c, where y=[y1,y2,…,y4] T .

[0111] Step S614: Set the initial search range K. Assume K = 9, then the number of candidate points in the nth stream is 9. Set the adjustable parameter Among them, r ii is the element in row i and column i of the R matrix, It refers to the minimum value of the diagonal element modulus, which initializes the candidate vector set L, that is, L = φ.

[0112] Step S62: Decoding is started from the bottom up, starting from the last stream (stream 4), then decoding the third stream and the second stream in sequence, until the decoding of the first stream is completed. Figure 7 FIG. 1 is a schematic diagram of the entire search and decoding process of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment. Figure 7 As shown. The following takes the second stream as an example to illustrate the decoding process. Assuming that the decoding results of the third and fourth streams have two candidate vectors, respectively denoted as and Select candidate vectors in turn to decode the second stream. Suppose the candidate vector is selected The decoding process is as follows:

[0113] Step S621, calculate the Babai point

[0114] Step S622: Calculate the one-dimensional grid point Gaussian distribution probability of the candidate constellation points Select the three candidate points closest to the BaBai point and record them as j is the candidate point number, j = 1, 2, 3, first calculate the normalization factor of the second stream

[0115] j is taken from 1 to 3, and calculated in sequence

[0116] calculate Search scope here for The parent node of It has been calculated in the third stream and does not need to be calculated again.

[0117] Step S63, according to calculation The size of the candidate vector is pruned, recursively or protected, that is, for each candidate node judge If the size satisfies any of the following (a), (b), or (c), perform the corresponding steps:

[0118] Step S631, if Then Pruning is performed, that is, the constellation point is considered not to belong to the candidate constellation points;

[0119] Step S632: If Then perform SIC detection on the remaining stream, that is, the remaining stream only retains the point closest to the Babai point as the candidate constellation point, and finally obtains a candidate vector And add to the set L, such as Figure 4 shown.

[0120] Step S633, if Then enter the next stream (i-1) detection and repeat the above steps until the candidate vector is output when i=1 And join the set L.

[0121] Step S64, calculate the Euclidean distance of all candidate sets and obtain the final decoding result. Set L contains all candidate vectors, calculate the Euclidean distance of each candidate vector, Select the candidate vector corresponding to the minimum distance As the decoding result,

[0122] Given The total number of access nodes |S| and the number of candidate vectors |L| of the algorithm of this embodiment satisfy:

[0123] |L|<K;

[0124] |S|<n|L|<nK;

[0125] In a typical communication scenario, Figure 8 : is a statistical diagram of the number of candidate vectors of the massive MIMO detection algorithm based on Gaussian sampling probability search in this embodiment under different initial search range conditions. The number of candidate vectors under different initial search ranges is as follows: Figure 8 As shown, we can see that the number of candidate vectors |L| is significantly smaller than K.

[0126] In terms of performance, decoding performance can be measured by decoding. The decoding radius of this embodiment is:

[0127]

[0128] Obviously, the larger K is, the larger the decoding radius of this embodiment is, the higher the probability of searching for the ML solution is, and thus the better detection performance is achieved. However, increasing K means raising the upper limit of the number of access nodes, increasing the number of potential access nodes, and thus increasing the complexity of the search. Therefore, there is a potential trade-off between performance and complexity in the algorithm of this embodiment.

[0129] This embodiment has low complexity while ensuring performance.

[0130] The embodiment of the present application also provides a decoding device for antenna signals, Figure 9 is a block diagram of a decoding device for antenna signals according to an embodiment of the present application, such as Figure 9 As shown, the device includes:

[0131] a channel estimation module 92 for performing channel estimation on a received signal to obtain a channel estimation matrix, wherein the received signal is an uplink signal of multiple transmitting antennas received by multiple receiving antennas;

[0132] A first determining module 94 is used to determine an adjustment parameter that makes the Gaussian sampling probability search converge;

[0133] A decoding module 96 is configured to perform Gaussian sampling probability search decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including a plurality of candidate vectors;

[0134] The second determining module 98 is configured to determine a decoding result of the receiving antenna according to a plurality of candidate vectors in the candidate vector set.

[0135] In one embodiment, the decoding module 96 is further configured to perform the following steps from the last stream to the first stream of the transformed received signal:

[0136] Determine the candidate constellation points of the current stream based on the candidate vector set before the current stream;

[0137] Determining a lattice point Gaussian distribution probability of candidate constellation points of the current stream according to the adjustment parameter, and determining a search range of candidate constellation points of the current stream according to the lattice point Gaussian distribution probability of candidate constellation points of the current stream;

[0138] Performing pruning, recursive processing, or serial interference cancellation (SIC) detection on the candidate constellation points of the current stream according to the search range of the candidate constellation points of the current stream to obtain a target candidate vector;

[0139] The candidate vector set of the current flow is updated according to the target candidate vector.

[0140] In one embodiment, the decoding module 96 is further configured to respectively determine the difference between each candidate constellation point of the current stream and the Babai point of the candidate vector of the current stream;

[0141] Determine the lattice point Gaussian distribution probability of the candidate constellation points of the current stream according to the adjustment parameter and the difference;

[0142] Determine the product of a search range of the lattice point Gaussian distribution probability of the candidate constellation points of the previous stream of the current stream and the lattice point Gaussian distribution probability of the candidate constellation points of the current stream;

[0143] The product is determined to be a search range of candidate constellation points of the current stream.

[0144] In one embodiment, the decoding module 96 is further configured to introduce a Gaussian-like function, use the adjustment parameter and the difference as an exponential term of the Gaussian-like function, and perform normalization processing on the Gaussian-like function to obtain a grid point Gaussian distribution probability of the candidate constellation points of the current stream.

[0145] In one embodiment, the decoding module 96 is further configured to prune the candidate constellation points of the current stream if the search range of the candidate constellation points of the current stream is less than 1;

[0146] When the search range of the candidate constellation point of the current stream is greater than or equal to 1 and less than 2, the remaining stream is decoded to obtain a candidate vector, and the candidate vector is added to the candidate vector set. In the decoding process of the remaining stream, the point closest to the Babai point is selected as the candidate constellation point;

[0147] When the search range of the candidate constellation points of the current stream is greater than or equal to 2, decoding of the next stream of the current stream is performed until the candidate vector is output for the first stream and added to the candidate vector set.

[0148] In one embodiment, the decoding module 96 is further configured to sequentially select candidate vectors of the current stream from a set of candidate vectors preceding the current stream;

[0149] Calculate the Babai point of the candidate vector of the current stream and determine the candidate constellation point of the current stream based on the Babai point.

[0150] In one embodiment, the decoding module 96 is further configured to determine a plurality of constellation points that are closest to the Babai point of the candidate vector as candidate constellation points of the current stream.

[0151] In one embodiment, the first determining module 94 is further configured to perform real number processing on the channel estimation matrix to obtain a processed channel estimation matrix; perform QR decomposition on the processed channel estimation matrix to obtain a first matrix and a second matrix;

[0152] An adjustment parameter is determined according to the second matrix.

[0153] In one embodiment, the first determining module 94 is further configured to determine adjustment parameters according to performance requirements and computational complexity requirements.

[0154] In one embodiment, the first determination module 94 is further used to perform real number processing on the received signal to obtain a processed received signal; and convert the processed received signal according to the first matrix to obtain a transformed received signal, wherein the transformed received signal is used to perform Gaussian sampling probability search decoding.

[0155] In one embodiment, the second determining module 98 is further configured to respectively determine the Euclidean distances of multiple candidate vectors in the candidate vector set; and determine a candidate vector with the smallest Euclidean distance among the multiple candidate vectors as the target decoding result.

[0156] Figure 10 : is a system structure diagram of the massive MIMO detection algorithm based on Gaussian sampling probability search of this embodiment, such as Figure 10As shown, it mainly includes three parts: receiving antenna, RF processing module and baseband processing module. Among them, the baseband processing module is used to realize channel estimation and signal detection. The uplink signals received by multiple receiving antennas are processed by the RF processing module and sent to the baseband processing module. The baseband processing module is used to realize some or all of the functions of the above modules. In the baseband processing module, the channel H is estimated using the pilot signal, and H is sent to the signal detection. Using the channel estimation matrix H and the received data signal c′, the large-scale MIMO detection algorithm based on Gaussian sampling probability search provided in the embodiment of the present application is adopted to perform signal detection.

[0157] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above method embodiments when run.

[0158] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0159] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0160] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0161] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0162] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing device, they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices, they can be implemented using program code executable by the computing device, and thus, they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be performed in a different order than herein, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0163] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for decoding antenna signals, characterized in that: The method comprises: Performing channel estimation on a received signal to obtain a channel estimation matrix, wherein the received signal is an uplink signal of a plurality of transmitting antennas received by a plurality of receiving antennas; Determining adjustment parameters that make the Gaussian sampling probability search converge; performing Gaussian sampling probability search decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including a plurality of candidate vectors; Determine a decoding result of the receiving antenna according to multiple candidate vectors in the candidate vector set.

2. The method according to claim 1, characterized in that Performing Gaussian sampling probability search decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including multiple candidate vectors includes: From the last stream to the first stream of the received signal, perform the following steps: Determine candidate constellation points of the current stream according to a candidate vector set before the current stream; determining a lattice point Gaussian distribution probability of candidate constellation points of the current stream according to the adjustment parameter, and determining a search range of candidate constellation points of the current stream according to the lattice point Gaussian distribution probability of candidate constellation points of the current stream; Performing a pruning, recursive, or serial interference cancellation (SIC) operation on the candidate constellation points of the current stream according to a search range of the candidate constellation points of the current stream to obtain a target candidate vector; The candidate vector set of the current stream is updated according to the target candidate vector.

3. The method according to claim 2, characterized in that Determining the lattice point Gaussian distribution probability of the candidate constellation points of the current stream according to the adjustment parameter, and determining the search range of the candidate constellation points of the current stream according to the lattice point Gaussian distribution probability of the candidate constellation points of the current stream includes: respectively determining a difference between each candidate constellation point of the current stream and a Babai point of a candidate vector of the current stream; Determining a lattice point Gaussian distribution probability of candidate constellation points of the current stream according to the adjustment parameter and the difference; Determine the product of a search range of lattice point Gaussian distribution probabilities of candidate constellation points of a previous stream of the current stream and the lattice point Gaussian distribution probabilities of candidate constellation points of the current stream; The product is determined to be a search range of candidate constellation points of the current stream.

4. The method according to claim 3, characterized in that Determining the lattice point Gaussian distribution probability of the candidate constellation points of the current stream according to the adjustment parameter and the difference includes: A Gaussian-like function is introduced, and the adjustment parameter and the difference are used as exponential terms of the Gaussian-like function; Normalizing the Gaussian-like function to obtain a grid point Gaussian distribution probability of the candidate constellation points of the current stream.

5. The method according to claim 2, characterized in that Performing pruning, recursion, or protection operations on the candidate constellation points of the current stream according to the search range of the candidate constellation points of the current stream to obtain a target candidate vector includes: For the candidate constellation points of the current stream, if the search range of the candidate constellation points of the current stream is less than 1, pruning the candidate constellation points of the current stream; When the search range of the candidate constellation point of the current stream is greater than or equal to 1 and less than 2, decoding the remaining stream to obtain a candidate vector, and adding the candidate vector to the candidate vector set, wherein, during the decoding process of the remaining stream, a point closest to the Babai point is selected as the candidate constellation point; When the search range of the candidate constellation points of the current stream is greater than or equal to 2, decoding of the next stream of the current stream is performed until a candidate vector is output for the first stream and added to the candidate vector set.

6. The method according to claim 2, characterized in that Determining the candidate constellation points of the current stream based on the candidate vector set before the current stream includes: Selecting candidate vectors of the current stream in sequence from a set of candidate vectors preceding the current stream; Calculate the Babai points of the candidate vectors of the current stream, and determine the candidate constellation points of the current stream according to the Babai points.

7. The method according to claim 6, characterized in that Determining the candidate constellation points of the current stream according to the Babai point includes: Determine a plurality of constellation points that are closest to the Babai point of the candidate vector as candidate constellation points of the current stream.

8. The method according to claim 1, characterized in that The tuning parameters that determine the convergence of the Gaussian sampling probability search include: Performing real number processing on the channel estimation matrix to obtain a processed channel estimation matrix; Performing QR decomposition on the processed channel estimation matrix to obtain a first matrix and a second matrix; The adjustment parameter is determined according to the second matrix.

9. The method according to claim 8, characterized in that The method further comprises: The adjustment parameters are determined according to performance requirements and computational complexity requirements.

10. The method according to claim 8, characterized in that The method further comprises: Performing real number processing on the received signal to obtain a processed received signal; The processed received signal is converted according to the first matrix to obtain the transformed received signal, wherein the transformed received signal is used for performing Gaussian sampling-based probability search decoding.

11. The method according to any one of claims 1 to 10, characterized in that Determining a decoding result of the receiving antenna according to multiple candidate vectors in the candidate vector set includes: respectively determining the Euclidean distances of a plurality of candidate vectors in the candidate vector set; A candidate vector having the smallest Euclidean distance among the multiple candidate vectors is determined as the target decoding result.

12. A decoding device for antenna signals, characterized in that: The device comprises: a channel estimation module, configured to perform channel estimation on a received signal to obtain a channel estimation matrix, wherein the received signal is an uplink signal of multiple transmitting antennas received by multiple receiving antennas; A first determination module is used to determine the adjustment parameters that make the Gaussian sampling probability search converge; a decoding module, configured to perform Gaussian sampling probability search decoding on the received signal according to the adjustment parameter and the channel estimation matrix to obtain a candidate vector set including a plurality of candidate vectors; The second determining module is configured to determine a decoding result of the receiving antenna according to a plurality of candidate vectors in the candidate vector set.

13. A computer-readable storage medium storing a computer program, wherein: The computer program is configured to execute the method according to any one of claims 1 to 11 when executed.

14. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 11.

Images