Same-frequency mixed signal demodulation method based on optimal sampling and related equipment

Abstract

The invention provides a same-frequency mixed signal demodulation method based on optimal sampling and related equipment, and relates to the technical field of wireless communication. The method comprises the following steps: down-converting a received mixed signal into a baseband signal so as to estimate a channel parameter of each path of signal; according to the estimated channel parameters of each path of signals, constructing sufficient statistics about the received mixed signals; and constructing a transfer metric according to the sufficient statistics, thereby performing sequence search according to a Viterbi recursion relationship, finding a path with a maximum accumulated metric value, and backtracking a decision symbol according to the path to obtain a decision symbol sequence. A PCMA signal test result shows that compared with an existing method, the performance of the method is greatly improved and approximates to theoretical performance.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, in particular to the field of signal demodulation technology, and specifically to a co-frequency mixed signal demodulation method based on optimal sampling and related equipment. Background Art

[0002] With the rapid growth of communication needs and the widespread application of complex communication technologies, the electromagnetic space has become increasingly complex and crowded, which has led to more and more attention to the problem of single channel blind source separation (SCBSS), such as separation of co-frequency aliased signals, paired carrier multiple access (PCMA) signal monitoring, and space-based automatic identification system (AIS) signal detection. Single channel blind source separation refers to the process of recovering the source signal using only the observation signal received by a single sensor when the prior information of the source signal and the channel transmission parameters are unknown.

[0003] In the 2.4GHz frequency band, the simultaneous operation of wireless LAN and Bluetooth devices often leads to serious time-frequency aliasing interference. In mobile communication systems, user equipment (hereinafter referred to as UE) generally accesses the network through a public random access channel (RACH) and uses ALOHA-type protocols (a group of multiple access control protocols used in wireless communication networks, which allow multiple UEs to transmit data on the same channel) to achieve the purpose of accessing the base station. When multiple UEs in the same cell simultaneously occupy the RACH channel, a single-channel aliasing signal will appear. In traditional methods, the above situation is called a collision, which cannot be reliably received and processed. Therefore, when a collision is detected, the UE adopts a random backoff strategy and re-initiates access on the RACH channel. Obviously, if some collision signals can be reliably received and processed, it is possible to access multiple UEs at the same time, which will help to greatly improve access efficiency.

[0004] In order to separate the multi-signal symbol sequence of co-frequency mixed signals, Deng Wen et al. summarized the current mainstream separation methods, mainly including particle filtering (PF), per-survivor processing (PSP) and deep learning methods. Both PF and PSP are approximate maximum likelihood sequence detection algorithms with high complexity. In addition, since both simplify the signal transmission model, according to the transmission and reception processing structure of a single-channel non-aliased signal, a per-symbol sampling model is obtained for analysis, and the obtained performance needs to be further improved. For deep learning methods, the training of the network depends on the input data, but real data is difficult to obtain, which limits its practical application. Summary of the invention

[0005] In view of the problem of insufficient performance of existing methods for separating mixed signals at the same frequency, the present invention provides a method for demodulating mixed signals at the same frequency based on optimal sampling and related equipment.

[0006] In a first aspect, the present invention provides a method for demodulating a mixed signal at the same frequency based on optimal sampling, comprising:

[0007] Down-converting the received mixed signal to a baseband signal to estimate the channel parameters of each signal;

[0008] Constructing sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal;

[0009] A transfer metric is constructed according to the sufficient statistics, so as to perform a sequence search according to the Viterbi recursive relationship to find a path with the largest cumulative metric value, and then the decision symbol is traced back according to the path to obtain a decision symbol sequence.

[0010] Furthermore, the estimating of channel parameters of each signal specifically includes:

[0011]

[0012] in, and They represent the estimated value of the amplitude of the i-th signal in the received mixed signal, the estimated value of the channel phase influence and the estimated value of the delay, i=1,2; P0=(q i (0)) 2 ,q i (t) is the shaped waveform of the normalized energy of the i-th signal, that is, y k is the sampled value of the received mixed signal after passing through the matched filter at time kT, k is an integer, T represents the measured value of the symbol period of the received mixed signal, j is an imaginary unit, Δf i is the frequency deviation of the i-th signal, represents the p-times spectrum of the received mixed signal.

[0013] Furthermore, the sufficient statistic y is constructed according to the following formula n :

[0014]

[0015] in, represents the convolution operation, r(t) represents the received mixed signal, and * represents taking the conjugate.

[0016] Furthermore, constructing a transfer metric based on the sufficient statistics specifically includes:

[0017] Define the state S at time k k =(I k-L+2 ,I k-L+3 ,…,I k ), then input symbol I k The state is S k-1 Transfer to S k The transfer metric λ k for:

[0018]

[0019] Among them, Re represents the real part, L is the channel memory length, y n is a sufficient statistic, Δτ represents the time delay difference between the two signals.

[0020] Furthermore, a sequence search is performed according to the Viterbi recursive relationship to find the path with the largest cumulative metric value, which specifically includes:

[0021] Initialize g 0:L-1 , and define λ l (S′, S) is the branch metric from the state S′ at time l-1 to the state S at time l; define Γ l-1 (S′) is the cumulative measure of the surviving state S′ at time l-1; for all states S′, Γ0(S′) = 0 is initialized; define Γ l (S′, S) is the provisional cumulative metric of the path from state S′ at time l-1 to state S at time l, Γ l (S′,S)=Γ l-1 (S′)+λ l (S′,S);

[0022] Given n time instants, for each time instant l = 1, 2, ..., n, iteratively perform the following steps (a) to (d):

[0023] (a) Calculate the possible branch metric λ according to the given transfer metric formula l (S′,S);

[0024] (b) Calculate the tentative cumulative metric Γ for possible branches l (S′,S)=Γ l-1 (S′)+λ l (S′,S);

[0025] (c) For each state S at time l, from the metric Γ l (S′,S) Select and record the path with the maximum metric;

[0026] (d) Update the cumulative metric of state S to Γ l (S)=maxS′ {Γ l (S′,S)}

[0027] After executing the above iterative process, find the maximum cumulative metric Γ from the n cumulative metrics N (S).

[0028] In a second aspect, the present invention provides a co-frequency mixed signal demodulation device based on optimal sampling, comprising:

[0029] A channel parameter estimation module is used to down-convert the received mixed signal to a baseband signal to estimate the channel parameters of each signal;

[0030] A sufficient statistics construction module, used to construct sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal;

[0031] The sequence search module is used to construct a transfer metric according to the sufficient statistics, thereby performing a sequence search according to the Viterbi recursive relationship to find a path with the maximum cumulative metric value, and backtrack the decision symbol according to the path to obtain a decision symbol sequence.

[0032] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0033] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in the first aspect.

[0034] Beneficial effects of the present invention:

[0035] The present invention utilizes the Viterbi algorithm to select the most likely state transfer path, which is an optimal sampling method, to achieve effective demodulation of the same-frequency mixed signal.

[0036] The test results of PCMA signals show that the performance of the present invention is greatly improved compared with the existing methods and is close to the theoretical performance. It can also be extended to different co-frequency mixed signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic diagram of a flow chart of a method for demodulating a same-frequency mixed signal based on optimal sampling provided by an embodiment of the present invention;

[0038] Figure 2 The performance of the best receiver of the present invention for PCMA signals provided by the embodiment of the present invention, the theoretical performance and the double oversampling performance of the PSP algorithm are compared;

[0039] Figure 3 A schematic diagram of the structure of a same-frequency aliasing signal demodulation device based on optimal sampling provided by an embodiment of the present invention;

[0040] Figure 4 This is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, an embodiment of the present invention provides a method for demodulating a mixed signal at the same frequency based on optimal sampling, comprising the following steps:

[0043] S101: down-converting the received mixed signal into a baseband signal to estimate the channel parameters of each signal;

[0044] Specifically, down-conversion refers to converting the received high-frequency signal into a lower-frequency signal for subsequent processing and analysis. After down-conversion to the baseband signal, the baseband signal at this time usually contains useful signals and out-of-band interference. Therefore, it is necessary to remove the out-of-band interference through a low-pass filter, and then estimate the channel parameters of each path of the mixed signal. The low-pass filter allows signals below a specific cutoff frequency to pass, while suppressing signals above the cutoff frequency.

[0045] S102: construct sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal;

[0046] Specifically, statistics are used to provide information about parameters given data, and if a statistic can contain all information about the parameters in the sample data, it is called a sufficient statistic. This step constructs sufficient statistics of the received signal based on the channel parameters of the received signal.

[0047] S103: constructing a transfer metric according to the sufficient statistics, thereby performing a sequence search according to the Viterbi recursive relationship to find a path with the maximum cumulative metric value, and backtracking the decision symbol according to the path to obtain a decision symbol sequence.

[0048] Specifically, the transition metric is used to measure the probability of a state from the previous moment to a state at the current moment. Considering that the output of the channel depends not only on the current input symbol but also on the past input symbol, this step constructs the transition metric of the input symbol to the state change by using sufficient statistics, and then finds the path with the largest cumulative metric value through the Viterbi recursive relationship. The symbol sequence represented by this path is the decision symbol sequence.

[0049] The optimal sampling-based demodulation method for mixed signals at the same frequency provided by the embodiment of the present invention uses the Viterbi algorithm to select the most likely state transition path, which is an optimal sampling method, to achieve effective demodulation of mixed signals at the same frequency.

[0050] On the basis of the above embodiments, in order to intuitively reflect the performance of the demodulation method of the present invention, taking a mixed signal containing two signals as an example, the same-frequency mixed signal demodulation method based on optimal sampling of the present invention is further described, which specifically includes the following steps:

[0051] S201: Channel parameter estimation

[0052] In this embodiment, the signal model of the received signal is set to

[0053]

[0054] Among them, A 1 and A 2 Represents the amplitude of two signals, θ 1 and θ 2 Indicates the channel phase impact on the two signals, T 1 and T 2 Represents the symbol period of the two signals, τ 1 and τ 2 represents the delay between two signals. represents the nth symbol of the i-th signal, q i (t) is the shaped waveform of the normalized energy of the i-th signal, that is, n(t) is Gaussian white noise that is independent of the signal, and j is an imaginary unit.

[0055] In order to maximize the signal-to-noise ratio of the received signal and improve the signal quality of the received signal for subsequent demodulation and recovery, in practical applications, it is necessary to first pass the received signal r(t) through a matched filter; then, sample the signal after the matched filter to calculate the channel parameters of each signal. For ease of description, the present invention will record the sampled value obtained by sampling the signal after the matched filter at time kT as y k Among them, the matched filter can be designed by measuring the bandwidth and symbol period of the received signal.

[0056] Assuming that the symbols of the i-th signal are independent of each other and have an equal probability distribution, and that the symbols of any two signals are also independent of each other, the amplitude, phase, and delay parameters of each signal are obtained by the following formula:

[0057]

[0058] Where P0=(q i (0)) 2 ,y k is the sampled value of the received signal r(t) at time kT after passing through the matched filter, where k is an integer, T represents the symbol period of the measured received signal, and Δf i It represents the frequency deviation of the i-th signal after down-conversion.

[0059] In some embodiments, the value of p can be 2, 4, 8, etc. In order to obtain a better demodulation effect, in this embodiment, p=4, is the 4-fold spectrum of the received signal r(t), through which the two-way frequency deviation Δf can be directly measured i , and measure the corresponding delay difference.

[0060] S202: Constructing Demodulated Sufficient Statistics

[0061] After estimating the channel parameters, sufficient statistics for each channel can be constructed. The received signal needs to be passed through the matched filter of each channel to obtain the signal y i (t); then at the optimal sampling point (in this embodiment, nT i -τ i The moment is the best sampling point) to sample and obtain the sufficient statistics y for demodulation n .

[0062]

[0063] in represents the convolution operation, * Stands for conjugation.

[0064] S203: Viterbi recursive search

[0065] This step completes the symbol detection of the same frequency mixed signal through sequence search. Assume that the state at time k is S k , S k =(I k-L+2 ,I k-L+3 ,…,I k), L is the channel memory length. In wireless communications, the channel memory length refers to how long the channel influence can last, or how many previous symbols still have a significant impact on the current symbol received. In the scenario of co-frequency mixed signal demodulation, L is usually 4, 5 or a larger value.

[0066] Input symbol I k Status by S k-1 Transfer to S k , the output is the transfer metric:

[0067]

[0068] The transfer process can be expressed as in:

[0069]

[0070] When the shaping filter is selected as a root raised cosine filter, g(t) is a raised cosine signal, and Δτ represents the time delay difference between the two signals.

[0071] The measurement of the entire sequence can be equivalent to the following Viterbi recursive relation:

[0072]

[0073] According to formula (9), the demodulation algorithm of the same-frequency mixed signal can be obtained, as shown in Table 1.

[0074] Table 1

[0075]

[0076]

[0077] In order to verify the effectiveness of the scheme of the present invention, the scheme of the present invention is verified by demodulating a typical co-frequency mixed signal PCMA signal. PCMA (Paired Carrier Multiple Access) signal is a paired carrier multiple access signal, which is a frequency reuse technology applied to a two-way satellite communication system.

[0078] Assume that the amplitudes of the two signals are the same, the frequency deviation and phase deviation are 0, and the delay difference between the two paths is Δτ=0.2T. Perform shaping filtering and matching filtering according to the crosstalk length of 7. Assume that the equivalent channel truncation length L=4 (equivalent to calculating the impact of inter-code crosstalk of the current and the 7 symbols before and after), and both signals are BPSK modulated. The PSP algorithm is selected as a reference for the traditional method. The truncation length of the PSP algorithm is 7, and it is oversampled twice. The performance and theoretical bit error rate of the receiver (hereinafter referred to as the best receiver) using the demodulation method of the present invention and the twice oversampling performance of the PSP algorithm are shown as follows: Figure 2 shown.

[0079] Depend on Figure 2 It can be seen that the performance of the best receiver has approached the theoretical performance boundary curve. Especially under high signal-to-noise ratio conditions, the two are basically consistent, verifying the best performance of the best receiving structure. The performance of the best receiver is about 1-2dB better than the PSP with twice the oversampling. It can be seen that even if the PSP algorithm uses an oversampling structure, there is still a significant performance loss compared to the best receiver structure.

[0080] In addition, in order to prove the rationality of the optimal receiver of the present invention, the present invention also provides the following proof process.

[0081] Since the transmitted symbols are a priori equally likely, the minimum error probability detection of the symbol sequence vector can be obtained based on the maximum likelihood criterion. For the received waveform sample r(t) of the mixed signal, the number of transmitted symbols of a single signal is defined as N, and the symbol sequences of the two channels are and The joint likelihood function of can be expressed as

[0082]

[0083] in * represents the conjugate, and assumes that r(t) is obtained by matching the two signals to obtain the signal y 1 (t) and y 2 (t):

[0084] in represents the convolution operation. When the shaping is done using root raised cosine, g k (t) is the raised cosine, g k (0) = 1, definition

[0085]

[0086] y n Represents signal y 1 (t) and y 2 (t) Sampling at the optimal sampling point of the two paths according to the symbol rate and multiplying by the corresponding amplitude. Let Δτ = τ 1 -τ 2 ,definition

[0087]

[0088] g n-mis a 2×2 inter-symbol crosstalk coefficient matrix, where the element in the i-th row and j-th column is the shaped waveform of the j-th signal, which is correlated with the shaped waveform corresponding to the i-th signal, and then multiplied by the corresponding amplitude coefficient. This element is equivalent to the shaped waveform of the j-th signal passing through the matched filter corresponding to the i-th signal, sampling at the corresponding time, and multiplying the sampled value by the corresponding amplitude coefficient. k (t) and the two-way channel parameters A 1 ,A 2 ,θ 1 ,θ 2 When the delay difference Δτ is determined, g n-m Is a fixed value.

[0089] The purpose of the maximum likelihood receiver is to search for Maximizing the information sequence Formula (10) involves |r(t) 2 The integral of is a constant and can be ignored. represents the vector of symbol pairs sent at time n. Equation (10) can be replaced by an equivalent metric:

[0090]

[0091] According to formula (15), the sequence y 0:N Contains all the information used to decide the transmitted symbol, thus constituting the sequence I used to decide the transmitted symbol 0:N Sufficient statistics of . By traversing and searching all sequences, we can find the sequence that maximizes formula (15) This is the maximum likelihood estimation result. Expanding equation (15) yields equation (9), which means that we can recursively perform the Viterbi algorithm based on equation (9) to obtain the maximum likelihood sequence estimation result of the sequence.

[0092] Corresponding to the above method, such as Figure 3 As shown, an embodiment of the present invention further provides a co-frequency mixed signal demodulation device based on optimal sampling, including a channel parameter estimation module, a sufficient statistics construction module and a sequence search module.

[0093] Among them, the channel parameter estimation module is used to down-convert the received mixed signal to a baseband signal to estimate the channel parameters of each signal; the sufficient statistics construction module is used to construct sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal; the sequence search module is used to construct a transfer metric based on the sufficient statistics, so as to perform a sequence search according to the Viterbi recursive relationship, find the path with the largest cumulative metric value, and trace back the decision symbol according to the path to obtain a decision symbol sequence.

[0094] It should be noted that the demodulation device provided in the embodiment of the present invention is to implement the above-mentioned demodulation method. Its specific functions can be referred to the above-mentioned method embodiments, which will not be repeated here.

[0095] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 401, a communication interface 402, a memory 403 and a communication bus 404, wherein the processor 401, the communication interface 402 and the memory 403 communicate with each other through the communication bus 404. The processor 401 may call the logic instructions in the memory 403 to execute the same-frequency mixed signal demodulation method, which includes: down-converting the received mixed signal to a baseband signal to estimate the channel parameters of each signal; constructing sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal; constructing a transfer metric according to the sufficient statistics, so as to perform a sequence search according to the Viterbi recursive relationship, find the path with the largest cumulative metric value, and trace back the decision symbol according to the path to obtain a decision symbol sequence.

[0096] In addition, when the logic instructions in the above-mentioned memory 403 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0097] An embodiment of the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the same-frequency mixed signal demodulation method provided by the above-mentioned method embodiments.

[0098] An embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the same-frequency mixed signal demodulation method provided by the above-mentioned method embodiments is implemented.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A co-frequency mixed signal demodulation method based on optimal sampling, characterized in that: include: Down-converting the received mixed signal to a baseband signal to estimate the channel parameters of each signal; Constructing sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal; A transfer metric is constructed according to the sufficient statistics, so as to perform a sequence search according to the Viterbi recursive relationship to find a path with the largest cumulative metric value, and then the decision symbol is traced back according to the path to obtain a decision symbol sequence.

2. The method for demodulating mixed signals at the same frequency based on optimal sampling according to claim 1, characterized in that: The estimating of the channel parameters of each signal specifically includes: in, and They represent the estimated value of the amplitude of the i-th signal in the received mixed signal, the estimated value of the channel phase influence and the estimated value of the delay, i=1,2; P0=(q i (0)) 2 ,q i (t) is the shaped waveform of the normalized energy of the i-th signal, that is, y k is the sampled value of the received mixed signal after passing through the matched filter at time kT, k is an integer, T represents the measured value of the symbol period of the received mixed signal, j is an imaginary unit, Δf i is the frequency deviation of the i-th signal, represents the p-fold spectrum of the received mixed signal.

3. The method for demodulating mixed signals at the same frequency based on optimal sampling according to claim 2, characterized in that: The sufficient statistic y is constructed according to the following formula n : in, represents the convolution operation, r(t) represents the received mixed signal, and * represents taking the conjugate.

4. The method for demodulating mixed signals at the same frequency based on optimal sampling according to claim 3, characterized in that: Constructing a transfer metric based on the sufficient statistics includes: Define the state S at time k k =(I k-L+2 ,I k-L+3 ,…,I k ), then input symbol I k The state is S k-1 Transfer to S k The transfer metric λ k for: Among them, Re represents the real part, L is the channel memory length, y n is a sufficient statistic, Δτ represents the time delay difference between the two signals.

5. The method for demodulating mixed signals at the same frequency based on optimal sampling according to claim 4, characterized in that: Perform sequence search according to the Viterbi recursive relationship to find the path with the largest cumulative metric value, including: Initialize g 0:L-1 , and define λ l (S′, S) is the branch metric from the state S′ at time l-1 to the state S at time l; define Γ l-1 (S′) is the cumulative measure of the surviving state S′ at time l-1; for all states S′, Γ0(S′) = 0 is initialized; define Γ l (S′, S) is the provisional cumulative metric of the path from state S′ at time l-1 to state S at time l, Γ l (S′,S)=Γ l-1 (S′)+λ l (S′,S); Given n time instants, for each time instant l = 1, 2, ..., n, iteratively perform the following steps (a) to (d): (a) Calculate the possible branch metric λ according to the given transfer metric formula l (S′,S); (b) Calculate the tentative cumulative metric Γ for possible branches l (S′,S)=Γ l-1 (S′)+λ l (S′,S); (c) For each state S at time l, from the metric Γ l (S′,S) Select and record the path with the maximum metric; (d) Update the cumulative metric of state S to Γ l (S)=max S′ {Γ l (S′,S)} After executing the above iterative process, find the maximum cumulative metric Γ from the n cumulative metrics N (S).

6. A co-frequency mixed signal demodulation device based on optimal sampling, characterized in that: include: A channel parameter estimation module is used to down-convert the received mixed signal to a baseband signal to estimate the channel parameters of each signal; A sufficient statistics construction module, used to construct sufficient statistics about the received mixed signal according to the estimated channel parameters of each signal; The sequence search module is used to construct a transfer metric according to the sufficient statistics, thereby performing a sequence search according to the Viterbi recursive relationship to find a path with the maximum cumulative metric value, and backtrack the decision symbol according to the path to obtain a decision symbol sequence.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images