Signal processing method and device, equipment, storage medium and program product

Abstract

The invention relates to a signal processing method, device and equipment, a readable storage medium and a program product. The method comprises the following steps: acquiring a baseband common mode algorithm model of wireless communication and a local performance function corresponding to each baseband module in a receiving end; according to the local performance function corresponding to each baseband module, converting the baseband common-mode algorithm model into a propagation function of each baseband module; processing a to-be-processed signal received by a receiving end by using the propagation function corresponding to each baseband module to obtain bit soft information of the to-be-processed signal; and according to the bit soft information, determining an original sending symbol corresponding to the to-be-processed signal. By adopting the method, the bit error rate performance of the system can be improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a signal processing method, device, equipment, storage medium and program product. Background Art

[0002] At present, 5G and 6G mobile communications have technical characteristics such as large bandwidth, massive connections, high reliability and low latency, which can be deeply integrated with vertical industries and fully penetrated. As vertical industries have rich application scenarios and widely varying needs, 5G and 6G mobile communications require customization of base stations and terminals based on application scenarios.

[0003] In the related art, due to the customization of base stations and terminals, the baseband signal processing schemes of different base stations and terminals vary greatly in the algorithms of each baseband module, resulting in the need to optimize the algorithms of each baseband module separately. Since the algorithms of each baseband module are based on different message dimensions (for example, logarithmic domain messages and real domain messages), the interfaces between baseband modules are not unified, resulting in a large amount of useful information being lost in the interfaces between baseband modules, which limits the overall bit error rate performance of the system. Summary of the invention

[0004] Based on this, it is necessary to provide a signal processing method, device, equipment, readable storage medium and program product that can improve the system bit error rate performance in response to the above technical problems.

[0005] In a first aspect, the present application provides a signal processing method, comprising:

[0006] Obtaining a baseband common mode algorithm model for wireless communication and local function functions corresponding to each baseband module in a receiving end;

[0007] According to the local functional functions corresponding to the baseband modules, the baseband common mode algorithm model is converted into the propagation functions of the baseband modules respectively;

[0008] Using the propagation functions corresponding to the baseband modules, the signal to be processed received by the receiving end is processed to obtain the bit soft information of the signal to be processed;

[0009] An original transmitted symbol corresponding to the signal to be processed is determined according to the bit soft information.

[0010] In one of the embodiments, before obtaining the baseband common mode algorithm model of wireless communication and the local functional functions corresponding to each baseband module in the receiving end, the method further includes:

[0011] Obtain a baseband signal receiving model;

[0012] Decomposing the baseband signal receiving model according to the Bayesian probability formula to obtain a system-level baseband algorithm model, wherein the system-level baseband algorithm model is used to solve the estimated value of the original transmitted symbol according to the signal to be processed;

[0013] Splitting the signal to be processed into local signals to be processed corresponding to each baseband module, and converting the system-level baseband algorithm model into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module;

[0014] Splitting the local signal to be processed into input signals of corresponding functional nodes in the factor graph, and converting the module-level baseband algorithm model into a node-level baseband algorithm model according to the input signals of corresponding functional nodes in the factor graph;

[0015] Converting the node-level baseband algorithm model to the baseband common-mode algorithm model according to probability information transmitted along the edge between the function node in the factor graph and the variable node in the factor graph;

[0016] Among them, the variable node is used to characterize the local signal to be processed of the corresponding baseband module, and the functional node is used to characterize the constraint conditions of the corresponding baseband module on the local signal to be processed; the system-level baseband algorithm model includes at least one of the following: maximum a posteriori probability MAP algorithm model, message passing algorithm model, and linear transformation algorithm model.

[0017] In one embodiment, the baseband common mode algorithm model includes a function to be evolved; and converting the baseband common mode algorithm model into propagation functions of each baseband module according to the local function functions corresponding to each baseband module includes:

[0018] The functions to be evolved in the baseband common-mode algorithm model are replaced with the local functional functions corresponding to the respective baseband modules, so as to convert the baseband common-mode algorithm model into the propagation functions of the respective baseband modules.

[0019] In one embodiment, the using the propagation functions of the respective baseband modules to process the signal to be processed received by the receiving end to obtain the target bit soft information of the signal to be processed includes:

[0020] The soft information generated by the propagation functions of the baseband modules is subjected to multiple loop iterations between the propagation functions of the baseband modules until an early stopping condition is met or the number of loops reaches a maximum number of iterations, so as to obtain the bit soft information of the signal to be processed;

[0021] Among them, the signal to be processed is the input of the first propagation function in the first loop iteration process, and each loop iteration process includes forward message passing and backward message passing. The soft information generated by the propagation function of the baseband module corresponding to each transmission node in the forward message passing and the backward message passing is used as the input of the propagation function of the baseband module corresponding to the next transmission node.

[0022] In one of the embodiments, the propagation function is at least one of the following: Gaussian approximate message propagation GAMP algorithm function, belief propagation BP detection algorithm function, soft modulation and demodulation algorithm, BP decoding algorithm function.

[0023] In one embodiment, the forward messaging includes:

[0024] Use the GAMP algorithm function to process the pilot reception signal or the signal feedback information of the previous cycle to generate the channel information of the current cycle;

[0025] Using the BP detection algorithm function, processing the channel information of the current cycle and the signal to be processed to generate symbol soft information of the current cycle;

[0026] Using a soft modulation and demodulation algorithm, demodulating the symbol soft information of the current cycle to generate demodulation information of the current cycle;

[0027] The demodulation information of the current cycle is interleaved to generate bit soft information of the current cycle.

[0028] In one embodiment, the backward message delivery includes:

[0029] Use the BP decoding algorithm function to process the bit information of the current cycle to generate the bit prior information of the current cycle;

[0030] Performing deinterleaving processing on the bit prior information of the current cycle to generate deinterleaved bit prior information of the current cycle;

[0031] Using a soft modulation and demodulation algorithm, modulating the deinterleaved bit prior information of the current cycle to generate symbol prior soft information of the current cycle;

[0032] The BP detection algorithm function is used to process the channel information of the current cycle and the signal to be processed to generate signal feedback information of the current cycle.

[0033] In a second aspect, the present application provides a signal processing device, comprising:

[0034] An acquisition module is used to acquire a baseband common mode algorithm model of wireless communication and local function functions corresponding to each baseband module in a receiving end;

[0035] A conversion module, used for converting the baseband common mode algorithm model into the propagation function of each baseband module according to the local functional function corresponding to each baseband module;

[0036] A processing module is used to process the signal to be processed received by the receiving end using the propagation functions corresponding to the various baseband modules to obtain the bit soft information of the signal to be processed; and determine the original transmission symbol corresponding to the signal to be processed based on the bit soft information.

[0037] In one embodiment, the signal processing device further includes:

[0038] A model generation module is used to obtain a baseband signal receiving model; according to a Bayesian probability formula, the baseband signal receiving model is decomposed to obtain a system-level baseband algorithm model, and the system-level baseband algorithm model is used to solve the estimated value of the original transmitted symbol according to the signal to be processed; the signal to be processed is split into local signals to be processed corresponding to each baseband module, and the system-level baseband algorithm model is converted into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module; the local signals to be processed are split into input signals of corresponding functional nodes in a factor graph, and the module-level baseband algorithm model is converted into a node-level baseband algorithm model according to the input signals of corresponding functional nodes in the factor graph; according to the probability information transmitted along the edge between the functional nodes in the factor graph and the variable nodes in the factor graph, the node-level baseband algorithm model is converted into the baseband common-mode algorithm model;

[0039] Among them, the variable node is used to characterize the local signal to be processed of the corresponding baseband module, and the functional node is used to characterize the constraint conditions of the corresponding baseband module on the local signal to be processed; the system-level baseband algorithm model includes at least one of the following: maximum a posteriori probability MAP algorithm model, message passing algorithm model, and linear transformation algorithm model.

[0040] In one of the embodiments, the baseband common-mode algorithm model includes functions to be evolved; the conversion module is also used to replace the functions to be evolved in the baseband common-mode algorithm model with local functional functions corresponding to each baseband module, so as to convert the baseband common-mode algorithm model into propagation functions of each baseband module.

[0041] In one embodiment, the processing module is also used to perform multiple loop iterative processing on the soft information generated by the propagation functions of the baseband modules between the propagation functions of the baseband modules until an early stopping condition is met or the number of loops reaches a maximum number of iterations, so as to obtain the bit soft information of the signal to be processed.

[0042] Among them, the signal to be processed is the input of the first propagation function in the first loop iteration process, and each loop iteration process includes forward message passing and backward message passing. The soft information generated by the propagation function of the baseband module corresponding to each transmission node in the forward message passing and the backward message passing is used as the input of the propagation function of the baseband module corresponding to the next transmission node.

[0043] In one of the embodiments, the propagation function includes at least one of the following: Gaussian approximate message propagation GAMP algorithm function, belief propagation BP detection algorithm function, soft modulation and demodulation algorithm, BP decoding algorithm function.

[0044] In one embodiment, the forward messaging includes:

[0045] Use the GAMP algorithm function to process the pilot reception signal or the signal feedback information of the previous cycle to generate the channel information of the current cycle; use the BP detection algorithm function to process the channel information of the current cycle and the signal to be processed to generate the symbol soft information of the current cycle; use the soft modulation and demodulation algorithm to demodulate the symbol soft information of the current cycle to generate the demodulation information of the current cycle; interleave the demodulation information of the current cycle to generate the bit soft information of the current cycle.

[0046] In one embodiment, the backward message delivery includes:

[0047] The bit information of the current cycle is processed using a BP decoding algorithm function to generate a priori bit information of the current cycle; the bit prior information of the current cycle is deinterleaved to generate a priori deinterleaved bit information of the current cycle; the deinterleaved bit prior information of the current cycle is modulated using a soft modulation and demodulation algorithm to generate a priori symbol soft information of the current cycle; the channel information of the current cycle and the signal to be processed are processed using a BP detection algorithm function to generate signal feedback information of the current cycle.

[0048] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the signal processing method of the first aspect when executing the computer program.

[0049] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the signal processing method of the first aspect described above.

[0050] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the signal processing method of the first aspect when executed by a processor.

[0051] The above-mentioned signal processing method, device, equipment, readable storage medium and program product first obtain the baseband common-mode algorithm model of wireless communication and the local function corresponding to each baseband module in the receiving end. Secondly, according to the local function corresponding to each baseband module, the baseband common-mode algorithm model is converted into the propagation function of each baseband module respectively. Thirdly, the propagation function corresponding to each baseband module is used to process the signal to be processed received by the receiving end to obtain the bit soft information of the signal to be processed. Finally, according to the bit soft information, the original transmitted symbol corresponding to the signal to be processed is determined. Since the propagation function corresponding to each baseband module is evolved and generated through the baseband common-mode algorithm model, the loss of useful information caused by the different message dimensions based on the algorithms corresponding to each baseband module can be avoided, thereby improving the system bit error rate performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 An application environment diagram of a signal processing method provided in an embodiment of the present application;

[0054] Figure 2 A flowchart of a signal processing method provided in an embodiment of the present application;

[0055] Figure 3 A principle derivation diagram of a baseband common mode algorithm model provided in an embodiment of the present application;

[0056] Figure 4 A schematic diagram of a receiver based on a baseband common mode algorithm model provided in an embodiment of the present application;

[0057] Figure 5 A schematic diagram of a factor graph provided in an embodiment of the present application;

[0058] Figure 6 A flowchart of another signal processing method provided in an embodiment of the present application;

[0059] Figure 7 A structural block diagram of a signal processing device provided in an embodiment of the present application;

[0060] Figure 8 An internal structure diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0062] The signal processing method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. The transmitting end 101 communicates with the receiving end 102. When the transmitting end 101 wants to send a signal to be processed to the receiving end 102, the receiving end 102 can first obtain the baseband common mode algorithm model of the wireless communication and the local function function corresponding to each baseband module in the receiving end 102. Secondly, the receiving end 102 can convert the baseband common mode algorithm model into the propagation function of each baseband module according to the local function function corresponding to each baseband module. Thirdly, the receiving end 102 can use the propagation function corresponding to each baseband module to process the signal to be processed received by the receiving end to obtain the bit soft information of the signal to be processed. Finally, the receiving end 102 can determine the original transmitted symbol corresponding to the signal to be processed according to the bit soft information.

[0063] The transmitting end 101 may be a terminal or a network device, and the receiving end 102 may also be a terminal or a network device. The terminal may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The network devices may include base stations.

[0064] In an exemplary embodiment, Figure 2 As shown, a signal processing method is provided, which is applied to Figure 1 The receiving end in FIG. 1 is taken as an example to explain, including S201-S204:

[0065] S201, obtaining a baseband common mode algorithm model of wireless communication and local functional functions corresponding to each baseband module in a receiving end.

[0066] In the present application, when the receiving end receives a signal to be processed, the baseband common mode algorithm model of the wireless communication and the local functional functions corresponding to each baseband module in the receiving end can be obtained.

[0067] The baseband common mode algorithm model can be a general mathematical model of the baseband module algorithm derived by deeply mining the algorithm commonalities of each baseband module. The baseband common mode algorithm model can be used to evolve the propagation function of each baseband module.

[0068] The following describes how to generate a baseband common-mode algorithm model.

[0069] Figure 3 A principle derivation diagram of a baseband common mode algorithm model provided in an embodiment of the present application, such as Figure 3 As shown, when generating a baseband common-mode algorithm model, the baseband signal receiving model can be first obtained, and the baseband signal receiving model can be decomposed according to the Bayesian probability formula to obtain a system-level baseband algorithm model, which is used to solve the estimated value of the original transmitted symbol according to the signal to be processed. Secondly, the signal to be processed is split into local signals to be processed corresponding to each baseband module, and the system-level baseband algorithm model is converted into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module. Thirdly, the local signals to be processed are split into input signals of corresponding functional nodes in the factor graph, and the module-level baseband algorithm model is converted into a node-level baseband algorithm model according to the input signals of corresponding functional nodes in the factor graph. Finally, the node-level baseband algorithm model is converted into a baseband common-mode algorithm model according to the probability information transmitted along the edge between the functional nodes in the factor graph and the variable nodes in the factor graph.

[0070] It should be understood that the embodiments of the present application do not limit the system-level baseband algorithm model. In some embodiments, the system-level baseband algorithm model includes at least one of the following: a maximum a posteriori (MAP) algorithm model, a message passing algorithm model, and a linear transformation algorithm model.

[0071] Accordingly, based on different types of system-level baseband algorithm models, different methods can be used to split the signal to be processed and the local signal to be processed. For example, if the system-level baseband algorithm model is a MAP algorithm model, the signal to be processed can be split into the local signal to be processed by MAP, and the local signal to be processed can be split into the input signals of the corresponding functional nodes in the factor graph by MAP.

[0072] The baseband signal receiving model may be a general mathematical model corresponding to the signal to be processed received by the receiving end after the bit signal to be transmitted is processed and transmitted by each baseband module of the transmitting end device.

[0073] Exemplarily, the baseband signal receiving model may be as shown in formula (1):

[0074] y=H·u·∏k G k +n (1)

[0075] Where y is the signal to be processed received by the receiving device, H is the MIMO channel matrix, G k is the mathematical matrix model abstracted from the kth baseband module, j∈{1,2,…,M}, M is the total number of baseband modules in the transmitting device, u is the bit signal to be sent, and n is the system noise vector.

[0076] Exemplarily, the process in which the bit signal to be transmitted is processed and sent by each baseband module of the transmitting device includes the process in which the bit signal to be transmitted undergoes signal processing processes such as channel coding, interleaving, and symbol modulation, and the process in which the processed bit signal to be transmitted is sent to the air interface through a multiple-in multiple-out (MIMO) antenna and is received by the receiving device.

[0077] In some embodiments, if the system-level baseband algorithm model is a MAP algorithm model, based on the process in which the bit signal to be transmitted is processed and transmitted by each baseband module of the transmitting end device, after obtaining the baseband signal reception model, the Bayesian probability formula can be used to decompose the baseband signal reception model in combination with the maximum a posteriori (MAP) probability to be solved for the bit signal to be transmitted, and obtain the system-level baseband algorithm model. The system-level baseband algorithm model is used to solve the estimated value of the original transmitted symbol according to the signal to be processed.

[0078] Among them, the MAP probability that needs to be solved for the bit signal to be transmitted can be shown as formula (2), and the system-level baseband algorithm model can be shown as formula (3). As shown in formulas (2) and (3), the direct solution of the bit signal to be transmitted in formula (2) can be decomposed into the solution of the output signal of the general baseband module, the solution of the output signal of the MIMO detection module and the solution of the prior probability in formula (3).

[0079]

[0080] Where u is the bit signal to be sent, H is the MIMO channel matrix, G k The mathematical matrix model abstracted from the kth baseband module is: is the estimated value of the bit signal u to be sent, x k is the output signal of the kth baseband module, k∈{1,2,...,M}, M is the total number of baseband modules in the transmitting device; p is the a posteriori function, that is, given y, H, G1, G M etc., and solve the maximum a posteriori probability of the bit signal to be sent through p; Used to find the output signal of the universal baseband module, p(y|x M ,H) is used to solve the output signal of the MIMO detection module, p(x M ) is used to solve for the prior probability.

[0081] For example, starting from the received signal to be processed, the estimated value of the bit signal u to be sent can be derived from each baseband module in reverse from the transmitting end. The signal to be processed is first solved by the MIMO detection module to obtain the MAP solution of the signal to be processed. Subsequently, according to Find the MAP solution for the M-1th module By analogy, according to the solution formula of the output signal of the general baseband module, the MAP solution is obtained module by module until the estimated value of the bit signal u to be sent is obtained.

[0082] In some embodiments, the signal to be processed may be split into local signals to be processed corresponding to each baseband module, and the system-level baseband algorithm model may be converted into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module.

[0083] The local signal to be processed may correspond to an input signal of a baseband module.

[0084] It should be understood that the processing of the signal to be processed received by the receiving end can be decomposed into the processing process of each baseband module in a unified form. Correspondingly, the signal to be processed can also be decomposed into local signals to be processed corresponding to each baseband module.

[0085] It should be understood that the embodiments of the present application do not limit how to decompose the signal to be processed into local signals to be processed corresponding to each baseband module. In some embodiments, this can be done based on the type of system-level baseband algorithm model.

[0086] For example, if the system-level baseband algorithm model is a MAP algorithm model, then in formula (3), for each baseband module, the output signal x k and G k , find the local signal to be processed (i.e., the input signal x k-1 Based on this, the signal estimation value of the signal to be processed can be split into the maximum a posteriori probability MAP of the local signal to be processed corresponding to each baseband module, thereby converting the system-level baseband algorithm model into a module-level baseband algorithm model.

[0087] Exemplarily, the module-level baseband algorithm model can be shown as formula (4):

[0088]

[0089] Among them, x k is the output signal of the kth baseband module, k∈{1,2,...,M}, B is the local signal to be processed (input signal) x k-1 The value space of G k The mathematical matrix model abstracted from the kth baseband module can be used as a constraint matrix, p(x k ∣x k-1 ,G k ) is the likelihood probability, p(x k-1 ) is the a priori probability of the input signal of the general baseband module.

[0090] It should be noted that, among the multiple baseband modules in the receiving end, the output signal of the previous baseband module can be used as the local signal to be processed (input signal) of the next baseband module, that is, the output signal x of the k-1th baseband module k-1 , can be used as the local signal to be processed (input signal) x of the kth baseband module k-1 .

[0091] It should be understood that the multiplication of the likelihood probability term and the prior probability term in the above formula evaluates the probability that the output signal of the kth baseband module is x k When the kth baseband module has different local signals to be processed (input signals) x k-1 In addition, due to the difference in the mathematical matrix expression of the MIMO detection module and the general baseband module, the signal solution process of the MIMO detection module in the system-level baseband algorithm model is separated out, but the signal solution of the MIMO detection module can still be performed in the form of the module-level baseband algorithm model.

[0092] In some embodiments, during the processing of the signal to be processed, a message passing algorithm may be used to solve the problem. Each baseband module may be generally expressed by a baseband common mode algorithm model, and provide a local signal to be processed x of the relevant baseband module. k-1 That is, using the factor graph, the message passing algorithm is used to solve the local signal x to be processed. k-1 The marginal distribution probability is the marginal distribution probability.

[0093] The factor graph includes variable nodes and function nodes. The variable nodes are used to represent the local signals to be processed of the corresponding baseband module, and the function nodes are used to represent the constraints of the corresponding baseband module on the local signals to be processed. The variable nodes and function nodes are connected by edges.

[0094] It should be understood that the embodiments of the present application do not limit how to split the local signal to be processed into input signals of corresponding functional nodes in the factor graph. In some embodiments, this can be done based on the type of system-level baseband algorithm model.

[0095] For example, if the system-level baseband algorithm model is a MAP algorithm model, in the factor graph, the function node corresponds to the constraint matrix G k The row vector of the variable node corresponds to the constraint matrix G k The function node also provides the likelihood probability of the corresponding constraint condition for the variable node, and the variable node also provides the prior probability of the variable to be solved for the function node. For example, if the constraint matrix G k The row vector composed of the elements in the i-th row is g i , corresponding to the i-th functional node. Based on the constraint matrix G k The maximum a posteriori probability solution is split into g-based i The maximum a posteriori probability formula of the row vector. Based on this, the module-level baseband algorithm model can be converted into a node-level baseband algorithm model. The node-level baseband algorithm model can be as shown in formula (5):

[0096]

[0097] Where Θ is the input signal x of the function node j The value space of y i is the i-th element of the output signal of baseband module k, and x is the input signal of the function node to be solved. The message passing algorithm iteratively updates the likelihood probability p(y i |x,g i ) and the prior probability P(x) are used to solve the maximum a posteriori probability of the bit signal u to be sent.

[0098] In some embodiments, the probability information transmitted along the edge between the functional node and the variable node and the probability information transmitted along the edge between the variable node and the functional node can be determined respectively. Subsequently, based on the probability information transmitted along the edge between the functional node and the variable node and the probability information transmitted along the edge between the variable node and the functional node, the node-level baseband algorithm model can be converted into a baseband common-mode algorithm model.

[0099] For example, the probability information transmitted from the i-th function node to the j-th variable node along the edge in the factor graph can be defined as r ij (μ k ), μ k is the input signal of the function node to be solved. According to the calculation rules of the sum-product algorithm (SPA) of the factor graph, r ij (μ k) can be determined by formula (6):

[0100]

[0101] Among them, i is a function node, j is a variable node, and F i (x) is the local function of the function node, q li is the prior probability of the corresponding signal to be processed.

[0102] For example, the probability information transmitted from the jth variable node to the ith function node along the edge in the factor graph can be defined as q ji (μ k ), μ k is the input signal of the function node to be solved. According to the calculation rules of the sum-product algorithm (SPA) of the factor graph, q ji (μ k ) can be determined by formula (7):

[0103] q ji (μ k )=ξ ji ×∏ l≠i r lj (μ k ) (7)

[0104] Among them, i is a function node, j is a variable node, ξ ji is a scaling factor to speed up the convergence of the algorithm.

[0105] For example, in order to facilitate hardware implementation, the log-likelihood ratio (LLR) can be used to represent each probability information. ij (μ k ) is the probability information r that the i-th function node is transmitted to the j-th variable node along the edge ij (μ k ) in LLR form, Q ji (μ k ) is the probability message q sent by the h-th variable node to the i-th function node ji (μ k ) in LLR form. ij (μ k ) and Q ji (μ k ) can be expressed as formula (8):

[0106]

[0107] Among them, μ kand μ0 are both signals x to be processed j It should be understood that by x j Take μ k The probability of dividing by x j Taking μ0 and finding the logarithm, we can determine the log-likelihood ratio.

[0108] Exemplarily, based on formula (8), the node-level baseband algorithm model can be converted into a baseband common-mode algorithm model, and the baseband common-mode algorithm model can be as follows (9):

[0109]

[0110] Among them, R ij (μ k ) is the probability information r that the i-th function node is transmitted to the j-th variable node along the edge ij (μ k ) in LLR form, Q ji (μ k ) is the probability message q sent by the j-th variable node to the i-th function node ji (μ k ) in LLR form, x k is the output signal of the kth baseband module, k∈{1,2,...,M}, μ k is the input signal of the function node to be solved.

[0111] In this application, the baseband common-mode algorithm model is implemented by combining a general mathematical model and a flexible and configurable local functional function. The baseband common-mode algorithm has a unified underlying hardware architecture and can achieve multi-module hardware architecture reuse through hardware folding and configuration, thereby improving hardware design efficiency.

[0112] It should be understood that the embodiments of the present application do not limit the local functional functions corresponding to the baseband module. In some embodiments, the baseband module may include a channel estimation module, a MIMO detection module, a modulation / demodulation module, an interleaving code module, and a channel decoding module.

[0113] The channel decoding module may include a low-density parity check (LDPC) decoding module and a polarization decoding module. The interleaving code module may be an interleave division multiple access (IDMA) detection module.

[0114] In some embodiments, different baseband modules may have different local functional functions, and the local functional functions may be pre-configured based on the different functions of each baseband module.

[0115] S202 . According to the local functional functions corresponding to the baseband modules, the baseband common mode algorithm model is converted into the propagation functions of the baseband modules respectively.

[0116] In this step, after obtaining the baseband common-mode algorithm model of wireless communication and the local functional functions corresponding to each baseband module in the receiving end, the baseband common-mode algorithm model can be converted into the propagation functions of each baseband module according to the local functional functions corresponding to each baseband module.

[0117] The propagation function includes at least one of the following: a Gaussian approximate message propagation (Generalized Approximate Message Passing, GAMP) algorithm function, a belief propagation (Belief Propagation, BP) detection algorithm function, a soft modulation and demodulation algorithm, and a BP decoding algorithm function.

[0118] In some embodiments, the baseband common mode algorithm model includes a function to be evolved. Accordingly, the function to be evolved in the baseband common mode algorithm model can be replaced with a local function corresponding to each baseband module to convert the baseband common mode algorithm model into a propagation function of each baseband module.

[0119] For each baseband module, its corresponding local function can be determined by the characteristics of the corresponding factor graph. The characteristics of the factor graph can include sparsity, signal value space, etc.

[0120] The evolution modes of the propagation functions of different baseband modules are described below.

[0121] In some embodiments, the function to be evolved in the common mode algorithm model can be replaced by a local functional function corresponding to the channel estimation module, thereby obtaining a propagation function of the channel estimation module.

[0122] Exemplarily, the local function function corresponding to the channel estimation module can be as shown in formula (10), using the local function function F corresponding to the channel estimation module i (h), replace the function to be evolved F in the baseband common mode algorithm model shown in formula (9) i (x), and then the GAMP is evolved. The propagation function of the channel estimation module is shown in formula (11):

[0123]

[0124] Where h is the output signal of the channel estimation module corresponding to the channel vector to be estimated; y is the signal to be processed obtained by the receiving end, corresponding to the input signal of the channel estimation module; σ 2 is the noise variance. ij (μk ) is the probability information r that the i-th function node is transmitted to the j-th variable node along the edge ij (μ k ) in LLR form, Q ji (μ k ) is the probability message q sent by the j-th variable node to the i-th function node ji (μ k ) in LLR form, x k is the output signal of the kth baseband module, k∈{1,2,...,M}, μ k The signal x to be processed j The value of .

[0125] Exemplarily, the propagation function of the channel estimation module can iterate the input signal multiple times. When the maximum number of iterations is reached, GAMP stops iterating and outputs channel information (probability message).

[0126] Exemplarily, the local function function corresponding to the MIMO detection module can be as shown in formula (12), using the local function function F corresponding to the MIMO detection module i (h), replace the function to be evolved F in the baseband common mode algorithm model shown in formula (9) i (x), and thus the BP detection algorithm is evolved, and the propagation function of the MIMO detection module is obtained as shown in formula (13):

[0127]

[0128] Wherein, x is the input signal of the function node to be solved. In formula (13), the function node is a MIMO detection module, and x can be the symbol vector to be estimated; y i is the i-th element of the output signal of baseband module k, corresponding to the output signal of the MIMO detection module; h i is the i-th row vector of the channel matrix, σ 2 is the noise variance. ij (μ k ) is the probability information r that the i-th function node is transmitted to the j-th variable node along the edge ij (μ k ) in LLR form, Q ji (μ k ) is the probability message q sent by the j-th variable node to the i-th function node ji (μ k ) in LLR form, x k is the output signal of the kth baseband module, k∈{1,2,...,M}, μ k The signal x to be processed j The value of .

[0129] Exemplarily, the propagation function of the MIMO detection module can be used to iterate the local signal to be processed (input signal) corresponding to the MIMO detection module for multiple times. When the maximum number of iterations is reached, the BP detection algorithm terminates the message update and outputs the symbol soft information (symbol LLR soft information). The calculation formula can be shown as formula (14):

[0130]

[0131] Among them, γ j (μ k ) is the input signal x of the function node to be solved j The value of μ k LLR soft information, N r is the number of receiving antennas, that is, the number of constraints in the mathematical model of the MIMO detection module.

[0132] Exemplarily, the local function corresponding to the modulation / demodulation module can be as shown in formula (15), and the local function corresponding to the modulation / demodulation module can substantially represent the mapping relationship between the modulation symbol and the bit. i (x), replace the function to be evolved F in the baseband common mode algorithm model shown in formula (9) i (x), thereby evolving the propagation function of the modulation / demodulation module, and the obtained propagation function of the modulation / demodulation module is shown in formula (16):

[0133] F i (h) = δ k-m (15)

[0134]

[0135] Among them, δ k-m μ k As the modulation symbol and the modulated bit sequence c m The mapping relationship indicates that if the modulation symbol μ k and the modulated bit sequence c m If the mapping relationship is satisfied, it is 1, otherwise it is 0; R j (n) is the LLR soft information of the nth bit of the jth modulation symbol, c n The bit sequence c that satisfies the mapping relationship m The nth bit in .

[0136] Exemplarily, if the channel decoding module is an LDPC decoding module, the local performance function corresponding to the channel decoding module can be as shown in formula (17). Using the local performance function F corresponding to the LDPC decoding modulei (h), replace the function to be evolved F in the baseband common mode algorithm model shown in formula (9) i (x), and thus the BP decoding algorithm is evolved, and the BP decoding algorithm of the LDPC decoding module is shown in formula (18). Assuming that the LDPC code is based on the bit domain expansion, and the message transmitted in the factor graph is a 0 / 1 type bit LLR, the propagation function of the LDPC decoding module can also be simplified to formula (19).

[0137]

[0138] Wherein, x is the input signal of the function node to be solved. In formula (19), the function node can be a decoding module, and x can be the LDPC coded bit, that is, the output signal of the channel decoding module; H(i,:) is the i-th row vector of the LDPC code check matrix; I j (μ k ) is the initialization prior information of the transmitted signal, i.e., the input signal of the channel decoding module, I j It is the bit LLR soft information input to the decoder.

[0139] Exemplarily, the propagation function of the LDPC decoding module is iterated multiple times, and the BP decoding algorithm of the LDPC decoding module terminates the message update and outputs the signal feedback information of each bit. The calculation formula thereof can be shown as formula (20):

[0140]

[0141] Among them, γ j is the 0 / 1 type bit LLR of the jth transmitted bit. T is the number of check formulas of the LDPC code, that is, the number of constraints in the mathematical model of the channel decoding module.

[0142] It should be noted that the embodiments of the present application do not limit the local functional functions and corresponding propagation functions corresponding to each baseband module. For example, Table 1 records the local functional functions and corresponding propagation functions corresponding to various baseband modules.

[0143] Table 1

[0144]

[0145] S203 , using the propagation functions corresponding to the baseband modules, processing the signal to be processed received by the receiving end to obtain soft bit information of the signal to be processed.

[0146] In this step, after the baseband common-mode algorithm model is converted into the propagation functions of each baseband module according to the local functional functions corresponding to each baseband module, the propagation functions corresponding to each baseband module can be used to process the signal to be processed received by the receiving end to obtain the bit soft information of the signal to be processed.

[0147] It should be understood that the embodiments of the present application do not limit how to use the propagation functions corresponding to each baseband module to process the signal to be processed received by the receiving end. In some embodiments, the soft information generated by the propagation functions of each baseband module can be iterated multiple times between the propagation functions of each baseband module until the early stopping condition is met or the number of cycles reaches the maximum number of iterations to obtain the bit soft information of the signal to be processed.

[0148] Among them, the signal to be processed is the input of the first propagation function in the first loop iteration process, and each loop iteration process includes forward message passing and backward message passing. The soft information generated by the propagation function of the baseband module corresponding to each transmission node in the forward message passing and backward message passing is used as the input of the propagation function of the baseband module corresponding to the next transmission node.

[0149] Exemplarily, the early stopping condition may be a cyclic redundancy check (CRC) condition, and the maximum number of iterations may be specifically set according to actual conditions, such as 10 times, 20 times, 50 times, etc.

[0150] In some embodiments, in the forward message transmission, the GAMP algorithm function can be first used to process the pilot reception signal or the signal feedback information of the previous cycle to generate the channel information of the current cycle. Secondly, the BP detection algorithm function is used to process the channel information of the current cycle and the signal to be processed to generate the symbol soft information of the current cycle. Thirdly, the soft modulation and demodulation algorithm is used to demodulate the symbol soft information of the current cycle to generate the demodulation information of the current cycle. Finally, the demodulation information of the current cycle is interleaved to generate the bit soft information of the current cycle.

[0151] For example, Figure 4 A schematic diagram of a receiver based on a baseband common mode algorithm model provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the baseband module in the receiving end includes a channel estimation module, a MIMO detection module, a modulation / demodulation module, an interleaver module and a channel decoding module. p is the pilot signal, and y is the signal to be processed received by the receiving end. There are four transmission steps in the forward message transmission.

[0152] In the first forward pass step, the channel estimation module receives the pilot signal y sent by the transmitter.p Or the signal feedback information A(x1) of the previous cycle is received, and the channel estimation module iteratively processes it through the GAMP algorithm to output the channel information E(x1) of the current cycle. In the second forward transmission step, the MIMO detection module uses the BP detection algorithm function to process the channel information E(x1) of the current cycle and the signal y to be processed, and outputs the symbol soft information E(x2) of the current cycle. In the third forward transmission step, the modulation / demodulation module uses the soft modulation and demodulation algorithm to demodulate the symbol soft information E(x2) of the current cycle to generate the demodulation information E(x3) of the current cycle. In the fourth forward transmission step, the interleaver module interleaves the demodulation information E(x3) of the current cycle to generate the bit soft information E(x4) of the current cycle.

[0153] It should be understood that the above-mentioned channel information E(x1), symbol soft information E(x2), demodulation information E(x3) and bit soft information E(x4) can all be soft information generated by the propagation function of each baseband module.

[0154] Exemplarily, the channel information E(x1) can be the output signal h of the channel estimation module corresponding to the channel vector to be estimated in formula (11), and the symbol soft information E(x2) can be the LLR soft information γ in formula (13). j (μ k ), the demodulation information E(x3) can be the bit soft information R j (n).

[0155] In some embodiments, in the backward message transmission, the BP decoding algorithm function can be used to process the bit information of the current cycle first to generate the bit prior information of the current cycle. Secondly, the bit prior information of the current cycle can be deinterleaved to generate the deinterleaved bit prior information of the current cycle. Thirdly, the soft modulation and demodulation algorithm can be used to modulate the deinterleaved bit prior information of the current cycle to generate the symbol prior soft information of the current cycle. Finally, the BP detection algorithm function can be used to process the channel information and the signal to be processed of the current cycle to generate the signal feedback information of the current cycle.

[0156] Continue to refer Figure 4 , the backward message passing also includes four delivery steps.

[0157] In the first backward transfer step, the channel decoding module can, based on the bit soft information R(x4) of the current cycle, iterate the BP decoding algorithm and then feed the bit prior information A(x4) of the current cycle to the interleaver. In the second backward transfer step, the interleaver module can deinterleave the bit prior information A(x4) of the current cycle to generate the deinterleaved bit prior information A(x3) of the current cycle. In the third backward transfer step, the deinterleaved bit prior information A(x3) of the current cycle is modulated by the modulation / demodulation module to generate the symbol prior soft information A(x2) of the current cycle. In the fourth backward transfer step, the MIMO detection module feeds back the signal feedback information A(x1) of the current cycle to the channel estimation module after iterating the BP detection algorithm based on the symbol prior soft information A(x2) of the current cycle. The signal feedback information A(x1) of the current cycle can iteratively update the channel information E(x1) through the GAMP algorithm in the next cycle.

[0158] It should be understood that the above-mentioned bit prior information A(x4), deinterleaving bit prior information A(x3), deinterleaving bit prior information A(x3) and signal feedback information A(x1) can all be soft information generated by the propagation function of each baseband module. Among them, the bit prior information A(v4) can be the LDPC coded bit x in formula (19).

[0159] For example, Figure 5 A schematic diagram of a factor graph provided in an embodiment of the present application is shown as follows: Figure 5 As shown, after the receiving end device receives the signal y to be processed, it can be processed by the function node f corresponding to the channel estimation module and the MIMO module. The function node f processes the signal y to be processed respectively to obtain variable nodes h and s. Subsequently, the modulation / demodulation module processes the corresponding variable node s to obtain the variable node c. Finally, the corresponding variable node v is obtained after passing through the interleaver and processed by the function node corresponding to the channel decoding module.

[0160] It should be noted that the signal to be processed may include multiple orthogonal frequency division multiplexing (OFDM) symbols, and each OFDM symbol transmits different information on a subcarrier. In the present application, channel estimation, MOMO detection, modulation or demodulation can be performed on multiple OFDM symbols in parallel to obtain soft information of multiple OFDM symbols processed in parallel, and then the soft information of multiple OFDM symbols processed in parallel is sent to the interleaver for interleaving processing and subsequent channel decoding processing.

[0161] S204: Determine the original transmission symbol corresponding to the signal to be processed according to the bit soft information.

[0162] In this step, when the propagation function corresponding to each baseband module is used to process the signal to be processed received by the receiving end and the bit soft information of the signal to be processed is obtained, the original transmission symbol corresponding to the signal to be processed can be determined according to the bit soft information.

[0163] It should be understood that the embodiments of the present application do not limit how to determine the original transmitted symbol corresponding to the signal to be processed based on the bit soft information. In some embodiments, a hard decision can be made on the bit soft information to restore the original transmitted symbol after recovery.

[0164] It should be understood that hard decision is a process of simplifying bit soft information, that is, converting each bit soft information into the closest binary value. For example, if the bit soft information is greater than 0, the hard decision result is 1; if the bit soft information is less than 0, the hard decision result is 0. Restoring the original transmitted symbol through hard decision is simple, direct and efficient.

[0165] In this application, bit soft information is obtained after the early stopping condition is met or the number of cycles reaches the maximum number of iterations, and then a hard decision is made to restore the original transmitted symbol. Compared with directly using hard decisions, iterative decoding combined with bit soft information can provide better bit error rate performance when channel conditions are poor.

[0166] The signal processing method provided in the embodiment of the present application first obtains the baseband common-mode algorithm model of the wireless communication and the local function corresponding to each baseband module in the receiving end. Secondly, according to the local function corresponding to each baseband module, the baseband common-mode algorithm model is converted into the propagation function of each baseband module respectively. Thirdly, the propagation function corresponding to each baseband module is used to process the signal to be processed received by the receiving end to obtain the bit soft information of the signal to be processed. Finally, according to the bit soft information, the original transmitted symbol corresponding to the signal to be processed is determined. Since the propagation function corresponding to each baseband module is evolved and generated through the baseband common-mode algorithm model, the loss of useful information caused by the different message dimensions on which the algorithms corresponding to each baseband module are based can be avoided, thereby improving the system bit error rate performance.

[0167] The following describes how to generate a baseband common-mode algorithm model. Figure 6 A flow chart of another signal processing method provided in an embodiment of the present application is shown as follows: Figure 6 As shown, the system includes S301-S309:

[0168] S301. Obtain a baseband signal receiving model.

[0169] The baseband signal receiving model may be a general mathematical model corresponding to the signal to be processed received by the receiving end after the bit signal to be transmitted is processed and transmitted by each baseband module of the transmitting end device.

[0170] S302: Decompose the baseband signal receiving model according to the Bayesian probability formula to obtain a system-level baseband algorithm model.

[0171] The system-level baseband algorithm model is used to solve the estimated value of the original transmitted symbol according to the signal to be processed.

[0172] S303 , splitting the signal to be processed into local signals to be processed corresponding to each baseband module, and converting the system-level baseband algorithm model into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module.

[0173] S304, splitting the local signal to be processed into input signals of corresponding functional nodes in the factor graph, and converting the module-level baseband algorithm model into a node-level baseband algorithm model according to the input signals of corresponding functional nodes in the factor graph.

[0174] S305 . Convert the node-level baseband algorithm model into a baseband common-mode algorithm model according to the probability information transmitted along the edge between the function node in the factor graph and the variable node in the factor graph.

[0175] Among them, the variable node is used to characterize the local signal to be processed of the corresponding baseband module, and the functional node is used to characterize the constraint conditions of the corresponding baseband module on the local signal to be processed; the system-level baseband algorithm model includes at least one of the following: maximum a posteriori probability MAP algorithm model, message passing algorithm model, and linear transformation algorithm model.

[0176] S306: Acquire a baseband common mode algorithm model of wireless communication and local functional functions corresponding to each baseband module in the receiving end.

[0177] S307 . According to the local functional functions corresponding to the baseband modules, the baseband common mode algorithm model is converted into the propagation functions of the baseband modules respectively.

[0178] S308 . Process the signal to be processed received by the receiving end using the propagation function corresponding to each baseband module to obtain soft bit information of the signal to be processed.

[0179] S309: Determine the original transmission symbol corresponding to the signal to be processed according to the bit soft information.

[0180] The signal processing method provided in the embodiment of the present application first obtains the baseband common-mode algorithm model of the wireless communication and the local function corresponding to each baseband module in the receiving end. Secondly, according to the local function corresponding to each baseband module, the baseband common-mode algorithm model is converted into the propagation function of each baseband module respectively. Thirdly, the propagation function corresponding to each baseband module is used to process the signal to be processed received by the receiving end to obtain the bit soft information of the signal to be processed. Finally, according to the bit soft information, the original transmitted symbol corresponding to the signal to be processed is determined. Since the propagation function corresponding to each baseband module is evolved and generated through the baseband common-mode algorithm model, the loss of useful information caused by the different message dimensions on which the algorithms corresponding to each baseband module are based can be avoided, thereby improving the system bit error rate performance.

[0181] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0182] Based on the same inventive concept, the embodiment of the present application also provides a signal processing device for implementing the signal processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more signal processing device embodiments provided below can refer to the limitations on the signal processing method above, and will not be repeated here.

[0183] In an exemplary embodiment, Figure 7 As shown, a signal processing device 400 is provided, comprising: an acquisition module 401, a conversion module 402 and a processing module 403, wherein:

[0184] The acquisition module 401 acquires the baseband common mode algorithm model of wireless communication and the local functional functions corresponding to each baseband module in the receiving end.

[0185] The conversion module 402 is used to convert the baseband common mode algorithm model into the propagation function of each baseband module according to the local functional function corresponding to each baseband module.

[0186] The processing module 403 is used to process the signal to be processed received by the receiving end using the propagation function corresponding to each baseband module to obtain the bit soft information of the signal to be processed; and determine the original transmission symbol corresponding to the signal to be processed according to the bit soft information.

[0187] In one embodiment, the signal processing device 400 further includes:

[0188] Model generation module 401 is used to obtain a baseband signal receiving model; decompose the baseband signal receiving model according to the Bayesian probability formula to obtain a system-level baseband algorithm model, and the system-level baseband algorithm model is used to solve the estimated value of the original transmitted symbol according to the signal to be processed; split the signal to be processed into local signals to be processed corresponding to each baseband module, and convert the system-level baseband algorithm model into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module; split the local signals to be processed into input signals of corresponding functional nodes in the factor graph, and convert the module-level baseband algorithm model into a node-level baseband algorithm model according to the input signals of corresponding functional nodes in the factor graph; convert the node-level baseband algorithm model into a baseband common-mode algorithm model according to the probability information transmitted along the edge between the functional nodes in the factor graph and the variable nodes in the factor graph;

[0189] Among them, the variable node is used to characterize the local signal to be processed of the corresponding baseband module, and the functional node is used to characterize the constraint conditions of the corresponding baseband module on the local signal to be processed; the system-level baseband algorithm model includes at least one of the following: maximum a posteriori probability MAP algorithm model, message passing algorithm model, and linear transformation algorithm model.

[0190] In one of the embodiments, the baseband common-mode algorithm model includes functions to be evolved; the conversion module 402 is also used to replace the functions to be evolved in the baseband common-mode algorithm model with local functional functions corresponding to each baseband module, so as to convert the baseband common-mode algorithm model into propagation functions of each baseband module.

[0191] In one embodiment, the processing module 403 is also used to perform multiple loop iterative processing on the soft information generated by the propagation functions of each baseband module between the propagation functions of each baseband module until the early stopping condition is met or the number of loops reaches the maximum number of iterations, so as to obtain the bit soft information of the signal to be processed.

[0192] Among them, the signal to be processed is the input of the first propagation function in the first loop iteration process, and each loop iteration process includes forward message passing and backward message passing. The soft information generated by the propagation function of the baseband module corresponding to each transmission node in the forward message passing and backward message passing is used as the input of the propagation function of the baseband module corresponding to the next transmission node.

[0193] In one of the embodiments, the propagation function includes at least one of the following: Gaussian approximate message propagation GAMP algorithm function, belief propagation BP detection algorithm function, soft modulation and demodulation algorithm, BP decoding algorithm function.

[0194] In one embodiment, forward messaging includes:

[0195] Use the GAMP algorithm function to process the pilot reception signal or the signal feedback information of the previous cycle to generate the channel information of the current cycle; use the BP detection algorithm function to process the channel information of the current cycle and the signal to be processed to generate the symbol soft information of the current cycle; use the soft modulation and demodulation algorithm to demodulate the symbol soft information of the current cycle to generate the demodulation information of the current cycle; interleave the demodulation information of the current cycle to generate the bit soft information of the current cycle.

[0196] In one embodiment, the backward message passing includes:

[0197] The bit information of the current cycle is processed using the BP decoding algorithm function to generate the bit prior information of the current cycle; the bit prior information of the current cycle is deinterleaved to generate the deinterleaved bit prior information of the current cycle; the deinterleaved bit prior information of the current cycle is modulated using the soft modulation and demodulation algorithm to generate the symbol prior soft information of the current cycle; the channel information of the current cycle and the signal to be processed are processed using the BP detection algorithm function to generate the signal feedback information of the current cycle.

[0198] Each module in the above signal processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0199] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal or a network device. The internal structure diagram thereof may be as follows: Figure 8As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, a near field communication (Near Field Communication, NFC) or other technologies. When the computer program is executed by the processor, a signal processing method is implemented.

[0200] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0201] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the above signal processing method when executing the computer program.

[0202] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned signal processing method is implemented.

[0203] In one embodiment, a computer program product is provided, comprising a computer program, which implements the above signal processing method when executed by a processor.

[0204] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0205] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0206] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A signal processing method, characterized in that: The method comprises: Obtaining a baseband common mode algorithm model for wireless communication and local function functions corresponding to each baseband module in a receiving end; According to the local functional functions corresponding to the baseband modules, the baseband common mode algorithm model is converted into the propagation functions of the baseband modules respectively; Using the propagation functions corresponding to the baseband modules, the signal to be processed received by the receiving end is processed to obtain the bit soft information of the signal to be processed; An original transmitted symbol corresponding to the signal to be processed is determined according to the bit soft information.

2. The method according to claim 1, characterized in that Before acquiring the baseband common mode algorithm model of wireless communication and the local function corresponding to each baseband module in the receiving end, the method further includes: Obtain a baseband signal receiving model; Decomposing the baseband signal receiving model according to the Bayesian probability formula to obtain a system-level baseband algorithm model, wherein the system-level baseband algorithm model is used to solve the estimated value of the original transmitted symbol according to the signal to be processed; Splitting the signal to be processed into local signals to be processed corresponding to each baseband module, and converting the system-level baseband algorithm model into a module-level baseband algorithm model according to the local signals to be processed corresponding to each baseband module; Splitting the local signal to be processed into input signals of corresponding functional nodes in the factor graph, and converting the module-level baseband algorithm model into a node-level baseband algorithm model according to the input signals of corresponding functional nodes in the factor graph; Converting the node-level baseband algorithm model to the baseband common-mode algorithm model according to probability information transmitted along the edge between the function node in the factor graph and the variable node in the factor graph; Among them, the variable node is used to characterize the local signal to be processed of the corresponding baseband module, and the functional node is used to characterize the constraint conditions of the corresponding baseband module on the local signal to be processed; the system-level baseband algorithm model includes at least one of the following: maximum a posteriori probability MAP algorithm model, message passing algorithm model, and linear transformation algorithm model.

3. The method according to claim 1, characterized in that The baseband common mode algorithm model includes a function to be evolved; and according to the local function functions corresponding to the respective baseband modules, converting the baseband common mode algorithm model into propagation functions of the respective baseband modules respectively includes: The functions to be evolved in the baseband common-mode algorithm model are replaced with the local functional functions corresponding to the respective baseband modules, so as to convert the baseband common-mode algorithm model into the propagation functions of the respective baseband modules.

4. The method according to any one of claims 1 to 3, characterized in that: The using the propagation functions of the respective baseband modules to process the signal to be processed received by the receiving end to obtain the target bit soft information of the signal to be processed includes: The soft information generated by the propagation functions of the baseband modules is subjected to multiple loop iterations between the propagation functions of the baseband modules until an early stopping condition is met or the number of loops reaches a maximum number of iterations, so as to obtain the bit soft information of the signal to be processed; Among them, the signal to be processed is the input of the first propagation function in the first loop iteration process, and each loop iteration process includes forward message passing and backward message passing. The soft information generated by the propagation function of the baseband module corresponding to each transmission node in the forward message passing and the backward message passing is used as the input of the propagation function of the baseband module corresponding to the next transmission node.

5. The method according to claim 4, characterized in that The propagation function includes at least one of the following: Gaussian approximate message propagation GAMP algorithm function, belief propagation BP detection algorithm function, soft modulation and demodulation algorithm, BP decoding algorithm function.

6. The method according to claim 5, characterized in that The forward message transmission includes: Use the GAMP algorithm function to process the pilot reception signal or the signal feedback information of the previous cycle to generate the channel information of the current cycle; Using the BP detection algorithm function, processing the channel information of the current cycle and the signal to be processed to generate symbol soft information of the current cycle; Using a soft modulation and demodulation algorithm, demodulating the symbol soft information of the current cycle to generate demodulation information of the current cycle; The demodulation information of the current cycle is interleaved to generate bit soft information of the current cycle.

7. The method according to claim 5 or 6, characterized in that: The backward message transmission includes: Use the BP decoding algorithm function to process the bit information of the current cycle to generate the bit prior information of the current cycle; Performing deinterleaving processing on the bit prior information of the current cycle to generate deinterleaved bit prior information of the current cycle; Using a soft modulation and demodulation algorithm, modulating the deinterleaved bit prior information of the current cycle to generate symbol prior soft information of the current cycle; The BP detection algorithm function is used to process the channel information of the current cycle and the signal to be processed to generate signal feedback information of the current cycle.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images