Message selective transmission-based OTFS signal detection method

Abstract

The invention relates to an orthogonal time-frequency-space (OTFS) signal detection method based on message selective transmission, which is suitable for a high-speed mobile communication scene. According to the method, firstly, on the basis of an original MP detection algorithm, calculation of a symbol log-likelihood ratio (LLR) is added, the accuracy of message passing is improved, and by introducing an edge grouping strategy, edges with different channel coefficients are grouped to optimize a factor graph structure, and the influence of a loop effect on detection performance is reduced. And then a damping factor is introduced in the iteration process, the message oscillation amplitude in each round of iteration is controlled, and the algorithm convergence speed is increased. In the detection process, a damping message selective delivery (D-MsP) detector further improves the detection performance and convergence efficiency of the OTFS signal in the multipath channel by using a sparsity principle. The D-MsP detector realizes good balance between performance and computation complexity, computation overhead is remarkably reduced while high detection precision is guaranteed, and the method is particularly suitable for a high-speed mobile communication scene.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communications, and in particular to an OTFS signal detection method based on selective message transmission. Background Art

[0002] With the rapid development of wireless communication technology, the demand for efficient and stable signal detection methods is increasing. In multipath and high mobility environments, traditional orthogonal frequency division multiplexing (OFDM) systems will encounter performance bottlenecks due to frequency selective fading. To solve this problem, orthogonal time-frequency space modulation (OTFS) came into being. OTFS significantly improves the robustness of transmitted signals in multipath environments by evenly distributing information symbols in the time-frequency domain. However, with the widespread application of OTFS technology, the complexity of signal detection has also increased significantly. In OTFS systems, the channel matrix is ​​complex and contains a large amount of path gain and noise interference. Traditional detection methods, such as the linear minimum mean square error (LMMSE) detector, can estimate the prior probability of transmitted symbols by optimizing weights, reduce computational complexity and improve bit error performance. However, the LMMSE detector only performs well under certain conditions. When the channel complexity increases, its detection performance will decrease. Therefore, how to further improve the accuracy and convergence speed of signal detection has become a research focus.

[0003] In recent years, the message passing algorithm based on factor graph has been widely used in signal detection in wireless communications. The message passing algorithm achieves efficient symbol detection by iteratively passing information on the factor graph. However, in the OTFS system, the sparsity of the channel matrix makes the direct use of the traditional message passing algorithm computationally intensive and may lead to information oscillation problems.

[0004] The present invention proposes an OTFS signal detection method based on selective message transmission, which combines the initial estimation capability of the LMMSE detector and the iterative optimization characteristics of the factor graph message transmission algorithm. Through the grouping strategy of path gain, the message transmission path is divided into "selected edge set" and "simplified edge set", and the symbol messages in different sets are treated differently, and the damping factor is introduced in the iterative process to further stabilize the message transmission. Efficient and stable OTFS signal detection is achieved, which is suitable for communication systems in complex multipath propagation and high-speed mobile scenarios. Summary of the invention

[0005] The present invention aims to solve the above problems of the prior art. A method for detecting OTFS signals based on selective message transmission is proposed. The technical solution of the present invention is as follows:

[0006] An OTFS signal detection method based on selective message transmission comprises the following steps:

[0007] Step 1: According to the received signal and the channel matrix, the LMMSE detector is used to initialize the prior probability of the transmitted symbol to obtain the initialized symbol prior information;

[0008] Step 2: Calculate the initial message using the log-likelihood ratio (LLR) formula. The initial conditional probability based on the symbol provides a basis for subsequent iterations.

[0009] Step 3: Divide the transmitted symbols associated with each received symbol into a "selected edge set" and a "simplified edge set" based on the path gain;

[0010] Step 4: Construct a sparse factor graph, and perform selective message transmission iteration on this basis;

[0011] Step 5: Introduce a damping factor in each iteration to control the oscillation amplitude of message transmission;

[0012] Step 6: Repeat the iteration until the maximum number is reached or the convergence condition is met, and output the final LLR result of the symbol.

[0013] 2. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that the step 1 specifically comprises: using the received signal y, the channel matrix H, the noise power σ 2 Building transmission symbol vector for:

[0014]

[0015] in, represents an estimated value, is the transmission symbol vector estimation vector, and H is the NM×NM dimensional equivalent channel matrix in the DD domain. Let the indexes of the rows and columns of the equivalent channel matrix be d and c respectively, that is, H={h dc ,d∈[1,MN],c∈[1,MN]}, and use I d and J c denotes the index set of the non-zero elements in the dth row and cth column of the matrix, and |I d |=|J c |=P, that is, there are only P non-zero elements in each row and column of the matrix H. The subscript MMSE stands for the abbreviation of Minimum Mean Square Error, and P is the number of multipaths. (!) H represents the conjugate transpose operation, (·) -1 represents the inverse operation, I represents the MN×MN dimensional unit matrix, M is the number of subcarriers, and N is the number of symbols.

[0016] according to Calculation symbol a k The initial prior probability of is:

[0017]

[0018] Among them, p j (a k ), j∈{1,2,...,MN} is the jth symbol a k The probability of represents the kth modulation symbol in the constellation, is the constellation cardinality, exp(·) is the exponential function, ||·|| 2 Represents the square of the Euclidean distance.

[0019] 3. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that the step 2 specifically comprises: the calculation of the initial prior probability from the symbol node To the function node f yi The log-likelihood ratio message is

[0020]

[0021] in(·) (l) refers to the lth iteration, The kth modulation symbol is obtained from the symbol node To the function node The log-likelihood ratio message transmitted, log(·) is the logarithmic function, p (l) (x j =a k |y,H) is the symbol x j Given the received signal y and the channel matrix H, it is estimated as a in the lth iteration k The probability, p (l) (x j =a1|y,H) is the symbol x j Given the received signal y and the channel matrix H, the probability that it is estimated to be a1 at the lth iteration.

[0022] 4. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that the step 3 specifically comprises: dividing the related transmission symbols of the received symbols into two sets according to the path gain. Specifically, the non-zero elements of the related rows in the channel matrix are sorted, and the symbols corresponding to the index are divided into a selected edge set and a simplified edge set. The symbol messages in the selected edge set remain unchanged, and the symbol messages in the simplified edge set only retain the maximum value.

[0023] 5. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that the step 4 is specifically: the sparse factor graph is established according to the set partitioning in step 3, and the message selective transmission iteration is performed on this basis, and the function node transmits the message to the symbol node as follows:

[0024]

[0025] For the kth modulation symbol from the function node To symbol node The log-likelihood ratio message transmitted, max(·) is the maximization objective function, Reason A 1×P-dimensional row vector, Reason A P×1-dimensional column vector, is a constraint condition, indicating that the symbol x is sent j Must be a k , is the index set of the positions of the non-zero elements in the tth column, For symbolic nodes To the function node The log-likelihood ratio message delivered at the l-1th iteration, For the kth modulation symbol from the function node To symbol node The log-likelihood ratio message delivered at iteration l-1.

[0026] The message that the symbol node passes to the function node is:

[0027]

[0028] Where J(j) is the index set of the positions of the non-zero elements in the jth column.

[0029] 6. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that the step 5 specifically comprises: introducing a damping factor Δ in each iteration of step 4 μ , specifically:

[0030]

[0031] where Δ μ ∈[0,1) is a damping factor that improves the performance by controlling the convergence speed.

[0032] 7. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that the step 6 specifically comprises: determining whether the maximum number of iterations I is reached.max , if satisfied, stop the iteration and calculate the LLR of the symbol:

[0033]

[0034] Among them, γ j (k) is a symbolic node for a k The total amount of messages.

[0035] Given a set of symbolic LLRs, according to the function node and the associated bit node c q The factor graph of (q∈{1,2,…,Q}) can be used to derive its associated coded bits and Q modulation order. Finally, using the sum-product algorithm and the maximum logarithm approximation, the LLR of the coded bits can be calculated as:

[0036]

[0037] in, is the LLR calculation result of the q-th bit code, is the maximum value of the coded bit state 1, The maximum value of the coded bit state is 0.

[0038] The advantages and beneficial effects of the present invention are as follows:

[0039] The present invention provides an OTFS signal detection method based on selective message transmission, aiming to improve the detection performance of OTFS signals in complex multipath and high-speed mobile environments. The method first uses the LMMSE detector to make a preliminary estimate of the received signal, and by initializing the prior probability of the symbol, the computational complexity is reduced and the detection accuracy is improved. Then, based on the difference in symbol path gain, the symbol is divided into a selected edge set and a simplified edge set, and a selective message transmission algorithm is implemented on the factor graph to ensure efficient detection of symbol messages under complex channel conditions.

[0040] This method controls the oscillation amplitude of message transmission through the damping factor, thereby improving the stability and convergence speed of the algorithm. Specifically, in each round of iteration, the symbol message in the simplified edge set is optimized to retain the maximum value, thereby reducing redundant calculations and improving system efficiency. At the same time, when the maximum number of iterations is reached or the preset convergence conditions are met, the method outputs the log-likelihood ratio (LLR) of the symbol as the final detection result to ensure the reliability and accuracy of the detection.

[0041] Relying on the combination of LMMSE detection and factor graph message passing, the present invention realizes efficient demodulation of the received signal and reduces the interference between symbols through the symbol grouping strategy of path gain. This method performs well in complex multipath propagation environments and is suitable for application scenarios that require high reliability and low latency, such as 5G / 6G mobile communications, Internet of Vehicles, and drone communications.

[0042] Through the above technical design, the present invention not only improves the detection accuracy of OTFS signals, but also reduces the required computing resources. The introduction of path gain division and damping factor effectively controls the instability and oscillation problems in the calculation process. In different channel environments, this method can ensure the stability and efficiency of detection, providing strong support for high-speed mobile communications and applications under complex channel conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a block diagram of an OTFS system for high-speed mobile scenarios of the present invention;

[0044] Figure 2 It is a flow chart of a low-complexity OTFS signal detection method for high-speed mobile scenarios of the present invention;

[0045] Figure 3 is a factor graph model diagram of MP detection provided by the present invention; DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] The present invention considers a single-input single-output orthogonal time-frequency space system, such as Figure 1 As shown, OTFS maps the data stream to the DD domain for data transmission. The OTFS system block diagram is shown in Figure 1 As shown. Using the DD domain symbols, Q is the system modulation order, and the symbol letter Data symbol Construct an OTFS signal, where k∈[0,N-1] and l∈[0,M-1] represent the Doppler frequency shift index and delay index, respectively.

[0048] The symbol x[k,l] is first converted to the TF domain symbol X[n,m] using the inverse symplectic finite Fourier transform (ISFFT), which is

[0049]

[0050] Where n=0,...,N-1, m=0,...,M-1 are time index and frequency index respectively.

[0051] Then, the TF domain symbol X[n,m] is transformed into a continuous time signal s(t) using the Heisenberg Transform, expressed as:

[0052]

[0053] Where g tx (t) represents the transmit pulse shaping, Δf and represent the subcarrier spacing and symbol duration respectively.

[0054] Then the time domain signal s(t) is transmitted through the dual selection channel, and the received signal can be written as

[0055] r(t)=∫∫h(τ,ν)e j2πν(t-τ) s(t-τ)dτdν+n(t)

[0056] where ∫∫(·)dτdv is the double integral of the function, h(τ,v) is the complex channel impulse response, which represents the channel response to the Doppler ν and delay τ of the pulse. The sparse representation of the channel h(τ,v) in the DD domain satisfies

[0057]

[0058] Where P is the number of propagation paths, h i , τ i and ν i are the channel gain, delay and Doppler shift associated with the ith path, respectively, and δ(·) is the impulse response function.

[0059] At the receiving end, we use g rx (t) is used as the received pulse, and the received signal r(t) is transformed into the TF domain using the Wigner Transform, which is expressed as

[0060]

[0061] Y[n,m]=Y(t,f) t＝nT,f＝mΔf

[0062] The input-output relationship in the TF domain can be simplified by adopting ideal shaping

[0063] Y[n,m]=H[n,m]X[n,m]+V[n,m]

[0064] Where V[n,m] is a zero-mean, variance σ 2 The equivalent channel H[n,m] in the TF domain is the additive white Gaussian noise (AWGN) of

[0065] H[n,m]=∫∫h(τ,ν)e j2πνnT e -j2π(ν+mΔf)τ dνdτ

[0066] Finally, the TF signal Y[n,m] is mapped back to the DD domain received signal using the symplectic finite Fourier transform (SFFT), which is expressed as

[0067]

[0068] Using the above equations, the input-output relationship in the DD domain can be derived as

[0069]

[0070] in It is the sampling form of the circular convolution of the channel response and the window function w(τ,v).

[0071] Further, we stack y[k,l] into a vector y, that is:

[0072] y=Hx+n

[0073] In the formula, is the receiving vector, x k+Nl =x[k,l] is the transmitted symbol vector, is the AWGN vector, represents the space of MN×1-dimensional complex matrices, and H is the NM=NM-dimensional equivalent channel matrix in the DD domain. Let the indexes of the rows and columns of the equivalent channel matrix be d and c respectively, that is, H={h dc ,d∈[1,MN],c∈[1,MN]}, and use I d and J c denotes the index set of the non-zero elements in the dth row and cth column of the matrix, and |I d |=|J c |=P, that is, there are only P non-zero elements in each row and column of the matrix H.

[0074] The principle and implementation process of the method of the present invention are described in detail below. Specifically: Figure 2 As shown,

[0075] The OTFS signal detection method based on selective message transmission is characterized in that the method comprises the following steps:

[0076] 1. An OTFS signal detection method based on selective message transmission, characterized in that the method comprises the following steps:

[0077] Step 1: According to the received signal and the channel matrix, the LMMSE detector is used to initialize the prior probability of the transmitted symbol to obtain the initialized symbol prior information;

[0078] Using the received signal y, channel matrix H, noise power σ 2 Building transmission symbol vector for:

[0079]

[0080] in, represents an estimated value, is the transmission symbol vector estimation vector, and H is the NM×NM dimensional equivalent channel matrix in the DD domain. Let the indexes of the rows and columns of the equivalent channel matrix be d and c respectively, that is, H={h dc ,d∈[1,MN],c∈[1,MN]}, and use I d and J c denotes the index set of the non-zero elements in the dth row and cth column of the matrix, and |I d |=|J c |=P, that is, there are only P non-zero elements in each row and column of the matrix H. The subscript MMSE stands for the abbreviation of Minimum Mean Square Error, and P is the number of multipaths. (·) H represents the conjugate transpose operation, (·) -1 represents the inverse operation, I represents the MN×MN dimensional unit matrix, M is the number of subcarriers, and N is the number of symbols.

[0081] according to Calculation symbol a k The initial prior probability of is:

[0082]

[0083] Among them, p j (a k ), j∈{1,2,...,MN} is the jth symbol a k The probability of represents the kth modulation symbol in the constellation, is the constellation cardinality, exp(·) is the exponential function, ||·|| 2 Represents the square of the Euclidean distance.

[0084] Step 2: Calculate the initial message using the log-likelihood ratio (LLR) formula. The initial conditional probability based on the symbol provides a basis for subsequent iterations.

[0085] The initial prior probability calculated from the symbol node in step 1 To the function node The log-likelihood ratio message is

[0086]

[0087] in(·) (l) refers to the lth iteration, The kth modulation symbol is obtained from the symbol node To the function node The log-likelihood ratio message transmitted, log(·) is the logarithmic function, p (l) (x j =a k |y,H) is the symbol x j Given the received signal y and the channel matrix H, it is estimated as a in the lth iteration k The probability, p (l) (x j =a1|y,H) is the symbol x j The probability that the received signal y and the channel matrix H are estimated to be a1 at the lth iteration.

[0088] Step 3: Divide the transmitted symbols associated with each received symbol into a "selected edge set" and a "simplified edge set" based on the path gain;

[0089] The related transmitted symbols of the received symbols are divided into two sets according to the path gain. Specifically, the non-zero elements of the related rows in the channel matrix are sorted, and the symbols corresponding to the index are divided into a selected edge set and a simplified edge set. The symbol messages in the selected edge set remain unchanged, and the symbol messages in the simplified edge set only retain the maximum value.

[0090] Step 4: Construct a sparse factor graph, and perform selective message transmission iteration on this basis;

[0091] The sparse factor graph is established according to the set partitioning in step 3, and the message selective transmission iteration is performed on this basis, and the message transmitted from the function node to the symbol node is:

[0092]

[0093] For the kth modulation symbol from the function node To symbol node The log-likelihood ratio message transmitted, max(·) is the maximization objective function, Reason A 1×P-dimensional row vector, Reason A P×1-dimensional column vector, is a constraint condition, indicating that the symbol x is sent j Must be a k , is the index set of the positions of the non-zero elements in the tth column, For symbolic nodes To the function node The log-likelihood ratio message delivered at the l-1th iteration, For the kth modulation symbol from the function node To symbol node The log-likelihood ratio message delivered at iteration l-1.

[0094] The message that the symbol node passes to the function node is:

[0095]

[0096] Where J(j) is the index set of the positions of the non-zero elements in the jth column.

[0097] Step 5: Introduce a damping factor in each iteration to control the oscillation amplitude of message transmission;

[0098] In each iteration of step 4, a damping factor Δ is introduced μ , specifically:

[0099]

[0100] where Δ μ ∈[0,1) is a damping factor that improves the performance by controlling the convergence speed.

[0101] Step 6: Repeat the iteration until the maximum number is reached or the convergence condition is met, and output the final LLR result of the symbol.

[0102] Determine whether the maximum number of iterations I has been reached max , if satisfied, stop the iteration and calculate the LLR of the symbol:

[0103]

[0104] γ j (k) is a symbolic node for a k The total amount of messages.

[0105] Given a set of symbolic LLRs, according to the function node and the associated bit node c q The factor graph of (q∈{1,2,...,Q}) can be used to derive its associated coded bits, where Q is the modulation order. Finally, using the sum-product algorithm and the maximum logarithm approximation, the LLR of the coded bits can be calculated as:

[0106]

[0107] in, is the LLR calculation result of the q-th bit code, is the maximum value of the coded bit state 1, The maximum value of the coded bit state is 0.

[0108] As described above, the present invention can be well implemented. Other contents not described in detail in the present invention specification belong to the well-known technology of professionals in the field and will not be described in detail.

[0109] The following is a convergence and performance comparison between the low-complexity OTFS signal detection method for high-speed mobile scenarios of the present invention and the algorithm in the prior art. Through the comparison, it can be seen that compared with the prior art, the present invention not only has an advantage in convergence speed, but also still has good performance in high-speed mobile scenarios.

[0110] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. An OTFS signal detection method based on selective message transmission, characterized in that: The method comprises the following steps: Step 1: According to the received signal and the channel matrix, the LMMSE detector is used to initialize the prior probability of the transmitted symbol to obtain the initialized symbol prior information; Step 2: Calculate the initial message using the log-likelihood ratio (LLR) formula. The initial conditional probability based on the symbol provides a basis for subsequent iterations. Step 3: Divide the transmitted symbols associated with each received symbol into a "selected edge set" and a "simplified edge set" based on the path gain; Step 4: Construct a sparse factor graph, and perform selective message transmission iteration on this basis; Step 5: Introduce a damping factor in each iteration to control the oscillation amplitude of message transmission; Step 6: Repeat the iteration until the maximum number is reached or the convergence condition is met, and output the final LLR result of the symbol.

2. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that: The step 1 is specifically as follows: using the received signal y, the channel matrix H, the noise power σ 2 Constructing transmission symbol vector for: in, represents an estimated value, is the transmission symbol vector estimation vector, and H is the NM×NM dimensional equivalent channel matrix in the DD domain. Let the indexes of the rows and columns of the equivalent channel matrix be d and c respectively, that is, H={h dc ,d∈[1,MN],c∈[1,MN]}, and use I d and J c denotes the index set of the non-zero elements in the dth row and cth column of the matrix, and |I d |=|J c |=P, that is, there are only P non-zero elements in each row and column of the matrix H. The subscript MMSE stands for the abbreviation of Minimum Mean Square Error, and P is the number of multipaths. (·) H represents the conjugate transpose operation, (·) -1 represents the inverse operation, I represents the MN×MN dimensional unit matrix, M is the number of subcarriers, and N is the number of symbols. according to Calculation symbol a k The initial prior probability of is: Among them, p j (a k ), j∈{1,2,...,MN} is the jth symbol a k The probability of represents the kth modulation symbol in the constellation, is the constellation cardinality, exp(·) is the exponential function, ||·|| 2 Represents the square of the Euclidean distance.

3. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that: The step 2 is specifically as follows: the calculation of the initial prior probability from the symbol node in step 1 To the function node The log-likelihood ratio message is in(·) (l) refers to the lth iteration, For the kth modulation symbol from symbol node To the function node f yi The log-likelihood ratio message transmitted, log(·) is the logarithmic function, p (l) (x j =a k |y,H) is the symbol x j Given the received signal y and the channel matrix H, it is estimated as a in the lth iteration k The probability, p (l) (x j =a1|y,H) is the symbol x j The probability that the received signal y and the channel matrix H are estimated to be a1 at the lth iteration.

4. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that: The step 3 is specifically: dividing the related transmission symbols of the received symbols into two sets according to the path gain. Specifically: sorting the non-zero elements of the related rows in the channel matrix, and dividing the symbols corresponding to the index into a selected edge set and a simplified edge set. The symbol message in the selected edge set remains unchanged, and the symbol message in the simplified edge set only retains the maximum value.

5. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that: The step 4 is specifically as follows: a sparse factor graph is established according to the set partitioning in step 3, and on this basis, message selective transmission iteration is performed, and the message transmitted from the function node to the symbol node is: For the kth modulation symbol from the function node To symbol node The log-likelihood ratio message transmitted, max(·) is the maximization objective function, Reason A 1×P-dimensional row vector, Reason A P×1-dimensional column vector, is a constraint condition, indicating that the symbol x is sent j Must be a k , is the index set of the positions of the non-zero elements in the tth column, For symbolic nodes To the function node The log-likelihood ratio message delivered at the l-1th iteration, For the kth modulation symbol from the function node To symbol node The log-likelihood ratio message delivered at iteration l-1. The message that the symbol node passes to the function node is: Where J(j) is the index set of the positions of the non-zero elements in the jth column.

6. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that: The step 5 is specifically as follows: in each iteration of the step 4, a damping factor Δ is introduced. μ , specifically: where Δ μ ∈[0,1) is a damping factor that improves the performance by controlling the convergence speed.

7. The OTFS signal detection method based on selective message transmission according to claim 1, characterized in that: The step 6 is specifically as follows: determining whether the maximum number of iterations has been reached. max , if satisfied, stop the iteration and calculate the LLR of the symbol: γ j (k) is a symbolic node for a k The total amount of messages. Given a set of symbolic LLRs, according to the function node and the associated bit node c q The factor graph of (qv{1,2,...,Q}) can be used to derive its associated coded bits and Q modulation order. Finally, using the sum-product algorithm and the maximum logarithm approximation, the LLR of the coded bits can be calculated as: in, is the LLR calculation result of the q-th bit code, is the maximum value of the coded bit state 1, The maximum value of the coded bit state is 0.

Images