Method for reducing out-of-time frequency switching (OTFS) transmission signal stopband energy and storage medium

Abstract

This invention belongs to the field of wireless communication technology and discloses a method and storage medium for reducing the stopband energy of the transmitted signal in an OTFS system. The method includes: S1: initial simulated weighting processing of filters on each subcarrier at the transmitting and receiving ends; S2: establishing a power spectral density expression for the OTFS signal after initial simulated weighting; S3: establishing a filter weight optimization problem based on the power spectral density expression with the objective of minimizing the signal stopband energy, and solving it to obtain the first optimized weight and the signal stopband energy value after the first optimized weighting; S4: establishing a mean square error expression for the OTFS system after initial simulated weighting; S5: establishing a weight optimization problem with the joint objective of reducing the mean square error and signal stopband energy of the OTFS system, and solving it to obtain the optimal filter weight coefficients. This invention reduces the stopband energy of the signal while also reducing transmission performance losses such as system mean square error, and improves the frequency domain focusing ability of the OTFS system signal.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, and more specifically, relates to a method for reducing the stopband energy of OTFS transmitted signals by adding appropriate weights to the filters on each subcarrier. Background Technology

[0002] Orthogonal Frequency Division Multiplexing (OFDM) is a classic multicarrier modulation technique widely used in communication systems such as Wi-Fi, 4G, and 5G. It has strong resistance to multipath fading in wireless channels, but it is highly sensitive to frequency offset. In the 6G era, high-quality communication in high-speed mobile scenarios will be crucial. However, high-speed movement leads to a strong Doppler effect, causing frequency offset, disrupting subcarrier orthogonality, and introducing inter-carrier interference. This will severely degrade the performance of OFDM systems.

[0003] In recent years, Orthogonal Time-Frequency Space (OTFS) modulation technology has attracted widespread attention. Unlike OFDM, OTFS introduces a time-delay-Doppler domain channel representation and signal modulation method, enabling it to adapt to time-varying channel characteristics under high-speed mobility. It transforms the highly time-varying time-frequency domain channel model into a sparse and slowly varying time-delay-Doppler domain channel. In the time-delay-Doppler domain, each symbol within a data frame experiences almost identical and unchanging fading. Therefore, in high-mobility scenarios, OTFS can achieve more significant performance gains compared to OFDM. OTFS modulation is expected to become one of the important air interface transmission technologies in future wireless communications. In practical applications, the same rectangular filter is typically used as both the shaping and receiving filters on each subcarrier at both the transmitting and receiving ends of an OTFS system. This ensures orthogonal transmission under ideal channel conditions, and rectangular filters are simple, efficient, and easy to implement in hardware. However, rectangular filter shaping causes slow attenuation of the signal's power spectrum sidelobes, resulting in out-of-band (OOB) radiation, causing spectral leakage, interfering with adjacent frequency bands, and thus affecting signal analysis and processing, reducing spectral efficiency. Therefore, how to reduce the stopband energy of the OTFS transmitted signal is a technical problem that urgently needs to be solved.

[0004] Existing solutions mainly fall into two categories: windowing the transmitted signal and using other filters. In the windowing method, the window function is directly multiplied by the transmitted symbol, which affects the information of the transmitted symbol, especially the edge regions. To enable the receiver to successfully reconstruct the symbol, a cyclic prefix is ​​usually added to each transmitted symbol, increasing the computational load. Furthermore, lengthening the symbol reduces spectral efficiency. While using filters with excellent time-frequency domain focusing performance to replace rectangular filters, such as IOTA filters and PHYDYS filters, can effectively reduce signal sidelobes, existing studies still use filters with identical energy for all subcarriers. This prevents the filters from being optimized more specifically based on the characteristics of the sidelobes of each subcarrier signal. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the prior art, the present invention provides a method for reducing the stopband energy of the transmitted signal in an OTFS system, thereby solving the technical problem that the existing methods for reducing signal stopband energy use identical filters for all subcarriers.

[0006] To achieve the above objectives, the present invention provides a method for reducing the stopband energy of OTFS transmitted signals, comprising:

[0007] S1. Initial simulated weighting processing of filters for each subcarrier on the transmitting and receiving ends of the OTFS system;

[0008] S2. Establish the power spectral density expression of the OTFS signal after initial simulated weighting of the filter;

[0009] S3. Based on the power spectral density expression in S2, establish a filter weight optimization problem with the goal of minimizing the signal stopband energy, and solve the optimization problem to obtain the first optimized weight coefficients and the signal stopband energy value after the first optimized weighting of the filter.

[0010] S4. Establish the mean square error expression for the OTFS system after initial simulated weighting of the filter;

[0011] S5. Using the signal stopband energy value described in S3 and the mean square error expression of the OTFS system in S4, establish a weight optimization problem with both as the joint objective, and solve for the optimal filter weight coefficients.

[0012] Preferably, in step S1, the same shaping filter with fixed frequency and bandwidth is used on each subcarrier of both the transmitting and receiving ends. The initial analog weighted shaping filter formula for the transmitting end is g. tx,m (t)=w m ·g tx (t), where w mFor the weights of the shaping filter corresponding to the m-th subcarrier at the transmitting end, g tx (t) is the expression for a shaping filter with fixed frequency and bandwidth, where the same g is used on each subcarrier at the transmitting end. tx (t); The formula for the receiver shaping filter after initial analog weighting is: in For the weights of the receive filter corresponding to the m-th subcarrier at the receiver, g rx (q) is the expression for a receiver filter with fixed frequency and bandwidth, where the same g is used on each subcarrier at the receiver. rx (q). Thus, the receiver filter and the transmitter filter can satisfy the conditions for orthogonal transmission of the system under an ideal channel.

[0013] Preferably, in step S2, the power spectral density expression is: Where m is the subcarrier index, M is the number of subcarriers, and w m Let be the weight of the filter on the m-th subcarrier, where The symbol x[k,l] represents the average power of the Doppler domain symbol to be transmitted, and T represents the symbol interval. Represents the shaping filter g tx The energy spectrum of (t), Represents the shaping filter g tx The average autocorrelation function of (t),

[0014] Preferably, the optimization problem model in step S3 is:

[0015] P1 model:

[0016] Constraints: |w m -1|≤θ,m=0,1,...,M-1,

[0017]

[0018] Where θ represents the fluctuation limit between the weight and 1. Let m represent the signal stopband range of the m-th subcarrier. With the total signal power remaining constant, the optimization problem is solved to obtain the first optimization weight coefficient and the signal stopband energy value after the first optimization weighting of the filter.

[0019] Preferably, θ is 0.3-0.6.

[0020] Preferably, the mean square error expression of the OTFS system in step S4 is: Where E[|u(n,m)| 2The ] represents the average power of the time-frequency domain equivalent noise u[n,m], where n is the time-domain index and m is the frequency-domain index. Where N is the number of symbols in each frame of signal, u(q) represents Gaussian white noise, and g rx (q) is the expression for a receiver filter with a fixed frequency and bandwidth.

[0021] Preferably, the optimization problem model in step S5 is:

[0022] P2 model:

[0023] Constraints:

[0024]

[0025] Among them, TH e The threshold value representing the signal stopband energy can be used to transform the optimization problem model into a function that represents the signal stopband energy. To optimize the convex problem model of the variables, the optimal weights and the system mean square error are obtained by solving the problem.

[0026] According to another aspect of the present invention, the present invention also provides a computer-readable storage medium having a communication program based on reducing the stopband energy of OTFS transmission signals, the communication program being executed to implement the above-described method for reducing the stopband energy of OTFS transmission signals.

[0027] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages:

[0028] The method of this invention optimizes the sidelobes of the signal by adding initial simulated weights to the filters on each subcarrier at the transmitting and receiving ends of the OTFS system, effectively reducing the stopband energy of the signal. Furthermore, considering that adding optimized weight coefficients will increase the mean square error between symbols at the transmitting and receiving ends and affect the system transmission performance, this invention performs secondary optimization of the weights by jointly considering both the system mean square error and the signal stopband energy. This can suppress the signal stopband energy while minimizing the signal transmission performance loss caused by the system mean square error, etc. Attached Figure Description

[0029] Figure 1 This is a schematic flowchart of a method for reducing the stopband energy of OTFS transmitted signals by adding appropriate weights to the filters on each subcarrier, as disclosed in Embodiment 1 of the present invention.

[0030] Figure 2 This is a schematic diagram comparing the power spectral density of the system's transmitted signal before and after filter weighting, provided in Embodiment 1 of the present invention.

[0031] Figure 3 This is a schematic diagram showing the performance comparison of the mean square error of the system before and after the filter weight optimization, which takes into account the system mean square error and the signal stopband energy in conjunction with the system mean square error and the signal stopband energy, as provided in Embodiment 1 of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0034] like Figure 1 The diagram shows a flowchart of a method disclosed in this invention for reducing the stopband energy of an OTFS transmitted signal by adding appropriate weights to the filters on each subcarrier at the transmitting and receiving ends of an OTFS system. Figure 1 The method shown includes:

[0035] S1. Initial simulation weighting is performed on the filters of each subcarrier at both the transmitting and receiving ends.

[0036] In some embodiments, step S1 specifically includes:

[0037] Weighting is applied to the shaping filters on each subcarrier at the transmitting end and the receiving filters on each subcarrier at the receiving end. Weighting adds appropriate weight coefficients to the filters on each subcarrier without changing the total power of the transmitted signal before and after filtering, thereby altering the energy of the filters on each subcarrier and relaxing the restriction of using identical filters for all subcarriers. Simultaneously, the weighted filters can be more effectively optimized based on the sidelobe characteristics of each subcarrier signal, further unlocking the potential of using filters to reduce signal sidelobe.

[0038] At the transmitting end, shaping filters with the same fixed frequency and bandwidth are used on each subcarrier. Sidelobes of the signal are optimized by adding weighting coefficients to each filter. Assume the weighting factor is w. m The parameter m corresponds to the frequency axis, representing the m-th subcarrier. Therefore, the weighted shaping filter on each subcarrier can be represented as g. tx,m (t)=wm ·g tx (t), where g tx (t) is the expression for a shaping filter with a fixed frequency and bandwidth, using the same g on each subcarrier. tx (t).

[0039] The same fixed-parameter receiving filters are used on each subcarrier at the receiving end. By adding weighting coefficients to the receiving filters, the receiving filters and shaping filters can satisfy the condition of orthogonal transmission in an ideal channel. The formula for the weighted receiving filter is as follows: in For the weights of the receive filter corresponding to the m-th subcarrier, g rx (q) is the expression for a receiver filter with a fixed frequency and bandwidth, using the same g on each subcarrier. rx (q).

[0040] S2. Obtain the power spectral density expression of the OTFS signal after weighting by the output filter;

[0041] In some embodiments, step S2 specifically includes:

[0042] Assume the OTFS system has M subcarriers in the time-frequency domain, with an interval of Δf between adjacent subcarriers. Each frame contains N symbols with a time interval of T, where T·Δf=1. The time-delay Doppler domain symbol to be transmitted is represented by x[k,l], where parameter k corresponds to the Doppler domain (k=0,1,...,N-1) and parameter l corresponds to the time-delay domain (l=0,1,...,M-1). The time-frequency domain transmitted symbol obtained after ISFFT transformation of x[k,l] is represented by X[n,m], where parameter n corresponds to the time axis, representing the nth symbol (n=0,1,...,N-1), and parameter m corresponds to the frequency axis, representing the mth subcarrier (m=0,1,...,M-1). X[n,m] is transformed by Heisenberg to obtain the time-domain transmitted signal. Where g tx,m (t-nT) is the expression for the weighted shaping filter on the m-th subcarrier. From S1, the formula for the weighted shaping filter on the m-th subcarrier is g. tx,m (t)=w m ·g tx (t), where w m For the weights of the shaping filter corresponding to the m-th subcarrier, g tx (t) is the expression for a shaping filter with a fixed frequency and bandwidth, using the same g on each subcarrier. tx (t).

[0043] First, the autocorrelation function R of the transmitted signal after filter weighting is derived. w (t,τ) is

[0044]

[0045] The average autocorrelation function of the signal after filter weighting is further obtained. in The average power of the symbol x[k,l] is represented. Represents the shaping filter g tx The average autocorrelation function of (t).

[0046] Thus, the power spectrum S of the filtered weighted signal is obtained. w (ω) is

[0047]

[0048] in, This indicates that the shaping filter g tx The energy spectrum of (t),

[0049] S3. Based on the power spectrum expression, establish a filter weight optimization problem with the goal of minimizing the signal stopband energy and solve it to obtain the optimized weight and the signal stopband energy value after filter weighting.

[0050] In some embodiments, step S3 specifically includes:

[0051] Without changing the total signal power, and constraining the fluctuation range of the weights, a filter weight optimization problem is established with the objective of minimizing the signal stopband energy. The problem model is as follows:

[0052] P1 model:

[0053] Constraints: |w m -1|≤θ,m=0,1,...,M-1,

[0054]

[0055] Where θ represents the fluctuation limit between the weight and 1. Let represent the signal stopband range of the m-th subcarrier, with the total signal power remaining constant. Solving the optimization problem yields the optimized weights and the signal stopband energy values ​​after filter weighting.

[0056] In some embodiments, θ can be a value between 0.3 and 0.6.

[0057] S4. Obtain the system mean square error expression after weighting by the output filter;

[0058] In some embodiments, step S4 specifically includes:

[0059] Assuming the channel is an additive white Gaussian noise (AWGN) channel, the signal reaches the receiver after passing through the channel, and the received signal can be represented as follows: Where u(q) represents Gaussian white noise.

[0060] The receiver uses a fixed-parameter receive filter g on each subcarrier. rx (t), receiving filter g rx (t) and shaping filter g tx (t) satisfies the condition for orthogonal transmission of the system under ideal channel conditions, and corresponds to the transmitter, adding weights to the receive filter on the m-th subcarrier.

[0061] The sign mean square error of the OTFS system with transmit and receive filter weights in the [k,l] domain, is derived under Gaussian white noise channel. Let the symbol signal-to-noise ratio be σ. s ,So The results show that when the symbol signal-to-noise ratio is constant, the mean square error is closely related to the weight allocation, and the symbol mean square error is independent of the indices k and l. Therefore, the error level of the system can be determined by any values ​​of k and l. To measure.

[0062] S5. Establish a filter weight optimization problem that jointly considers the system mean square error and the signal stopband energy, and solve it to obtain the optimal filter weight coefficients.

[0063] In an optional implementation, step S5 specifically includes:

[0064] To minimize the system mean square error while limiting the signal stopband energy, a weight optimization model is established. The problem is modeled as follows:

[0065] P2 model:

[0066] Constraints:

[0067]

[0068] Among them, TH e This represents the threshold value for the signal stopband energy. The optimization problem can be transformed into... To optimize the convex problem of variables, the optimal weighting coefficients and the system mean square error are obtained by solving the problem.

[0069] In some embodiments, S6 is also included: after re-adding the optimal weighting coefficients obtained in S5 to the filters on each subcarrier of the OTFS system transmitter and receiver, the OTFS communication system can be run again, which can reduce the stopband energy of the OTFS transmission signal and reduce the impact on the OTFS system transmission.

[0070] The following describes a method, apparatus, and system for reducing the stopband energy of OTFS transmitted signals through specific exemplary embodiments. These embodiments are only used to explain the technical solutions of the invention and are not intended to limit the scope of the invention.

[0071] Example 1

[0072] A method for reducing the stopband energy of OTFS transmitted signals by adding appropriate weights to filters on each subcarrier includes:

[0073] S1. Initial simulation weighting is performed on the filters of each subcarrier at both the transmitting and receiving ends.

[0074] In S1.1, the same shaping filters with fixed frequency and bandwidth are used on each subcarrier of the transmitter. Sidelobes of the signal are optimized by adding weighting coefficients to each filter. Assume the weighting factor is w. m The parameter m corresponds to the frequency axis, representing the m-th subcarrier, and its value ranges from m = 0, 1, ..., N-1. Then, the weighted shaping filter on each subcarrier can be represented as g. tx,m (t)=w m ·g tx (t), where g tx (t) is the expression for a shaping filter with a fixed frequency and bandwidth, using the same g on each subcarrier. tx (t).

[0075] The same fixed-parameter receiving filters are also used on each subcarrier of the S1.2 receiver. By adding weighting coefficients to the receiving filters, the receiving filters and shaping filters can satisfy the condition of orthogonal transmission of the system under ideal channel conditions. The formula for the weighted receiving filter is as follows: in For the weights of the receive filter corresponding to the m-th subcarrier, g rx (q) is the expression for a receiver filter with a fixed frequency and bandwidth, using the same g on each subcarrier. rx (q).

[0076] S2. Obtain the power spectral density expression of the OTFS signal after weighting by the output filter;

[0077] S2.1 Assume the OTFS system has M subcarriers in the time-frequency domain, with an interval of Δf between adjacent subcarriers. Each frame contains N symbols, with a symbol time interval of T, where T·Δf=1. The time-delay Doppler domain symbol to be transmitted is represented by x[k,l], where parameter k corresponds to the Doppler domain, k=0,1,...,N-1, and parameter l corresponds to the time-delay domain, l=0,1,...,M-1. The time-frequency domain transmitted symbol obtained after ISFFT transformation of x[k,l] is represented by X[n,m], where parameter n corresponds to the time axis, representing the nth symbol, with a value range of n=0,1,...,N-1, and parameter m corresponds to the frequency axis, representing the mth subcarrier, with a value range of m=0,1,...,M-1. X[n,m] is transformed by Heisenberg to obtain the time-domain transmitted signal s. w (t), Where g tx,m (t-nT) is the expression for the weighted shaping filter on the m-th subcarrier. From S1.1, the formula for the weighted shaping filter is g. tx,m (t)=w m ·g tx (t), where w m For the weights of the shaping filter corresponding to the m-th subcarrier, g tx (t) is the expression for a shaping filter with a fixed frequency and bandwidth, using the same g on each subcarrier. tx (t).

[0078] S2.2 derives the autocorrelation function R of the transmitted signal after filter weighting. w (t,τ) is

[0079]

[0080] The average autocorrelation function of the signal after filter weighting is further obtained. in The average power of the symbol x[k,l] is represented. Represents the shaping filter g tx The average autocorrelation function of (t).

[0081] S2.3 is derived from S2.2, which yields the power spectrum S of the filtered signal. w (ω) is

[0082]

[0083] in, This indicates that the shaping filter g tx The energy spectrum of (t),

[0084] S3, based on the power spectrum expression, establishes a filter weight optimization problem with the objective of minimizing the signal stopband energy and solves it to obtain the optimized weights and the weighted signal stopband energy values, including:

[0085] S3.1 Without changing the total signal power, constrain the fluctuation range of the weights and establish a filter weight optimization problem with the objective of minimizing the signal stopband energy. The problem model is as follows:

[0086] P1 model:

[0087] Constraints: |w m -1|≤θ,m=0,1,...,M-1,

[0088]

[0089] Where θ represents the fluctuation limit between the weight and 1. This represents the signal stopband range of the m-th subcarrier. The range can be defined by a fixed number of out-of-band sidelobes, while the total signal power remains constant.

[0090] S3.2 The fluctuation limits θ for the optimization problem are 0.3, 0.5, and 0.6. The signal stopband range is set as the energy of the six side lobes on one side of the main bandwidth. The optimized weights and the filter-weighted signal stopband energy values ​​are obtained by solving. Statistical analysis shows that the optimized weights obtained when the fluctuation limit θ is 0.3 reduce the transmitted signal stopband energy by 2.4337 dB; the optimized weights obtained when the fluctuation limit θ is 0.5 reduce the transmitted signal stopband energy by 4.4298 dB; and the optimized weights obtained when the fluctuation limit θ is 0.6 reduce the transmitted signal stopband energy by 5.5473 dB.

[0091] S4. Obtain the system mean square error expression after weighting by the output filter;

[0092] S4.1 Assuming the channel is an additive white Gaussian noise (AWGN) channel, the signal reaches the receiver after passing through the channel, and the received signal can be expressed as follows: Where u(q) represents Gaussian white noise.

[0093] From S1.2, we can obtain that the same fixed-parameter receiving filter g is used on each subcarrier at the receiving end. rx (t), receiving filter g rx (t) and shaping filter g tx (t) satisfies the condition for orthogonal transmission of the system under ideal channel conditions, and corresponds to the transmitter, adding weights to the receive filter on the m-th subcarrier.

[0094] S4.2 derives from S4.1 the sign mean square error of the OTFS system delay-Doppler domain index [k,l] weighted by the transmit and receive filters in a Gaussian white noise channel. Let the symbol signal-to-noise ratio be σ. s ,So The results show that when the symbol signal-to-noise ratio is constant, the mean square error is closely related to the weight allocation, and the symbol mean square error is independent of the indices k and l. Therefore, the error level of the system can be determined by any values ​​of k and l. To measure.

[0095] S5. Establish a filter weight optimization problem that jointly considers the system mean square error and the signal stopband energy, and solve for the filter weight coefficients:

[0096] To minimize the system mean square error while limiting the signal stopband energy, a weight optimization model is established. The problem is modeled as follows:

[0097] P2 model:

[0098] Constraints:

[0099]

[0100] Among them, TH e This represents the threshold value for the signal stopband energy. The optimization problem can be transformed into... To optimize the convex problem of variables, the optimized weights and system mean square error are obtained. After obtaining the above filter weight coefficients, they are added back to the filters on each subcarrier at the transmitting and receiving ends of the OTFS system before running the OTFS communication system. This reduces the stopband energy of the OTFS transmitted signal while minimizing the impact on OTFS system transmission.

[0101] Results and Discussion

[0102] To further illustrate the performance of the method provided in this invention for reducing the stopband energy of the OTFS transmitted signal by adding appropriate weights to the filters on each subcarrier, the following detailed description is provided in conjunction with specific experiments:

[0103] In the experiment, the number of subcarriers M in the OTFS system was set to 64, the number of symbols N in one frame of signal was 20, and the original shaping filter and the receiving filter with fixed frequency and bandwidth used rectangular filters. The filter length was the same as the length of one time-frequency symbol of the system. It is worth noting that other high-performance filters can also be used as the original filters. Figure 2The diagram illustrates the power spectral density of the system's transmitted signal after the first optimized weighting of the filter. The weights used for "Weighted Signal 1" are the optimized weights that minimize the signal's stopband energy when the fluctuation limit θ is 0.3; the weights used for "Weighted Signal 2" are the optimized weights that minimize the signal's stopband energy when the fluctuation limit θ is 0.5; and the weights used for "Weighted Signal 3" are the optimized weights that minimize the signal's stopband energy when the fluctuation limit θ is 0.6. Figure 2 As can be seen, compared to the unweighted transmitted signal, the filter weighting effectively reduces the sidelobes of the transmitted signal, and the larger the fluctuation range of the weights, the more significant the reduction effect on the sidelobes. Numerical statistics show that the first optimized weight, obtained when the fluctuation limit θ is 0.6, reduces the stopband energy of the transmitted signal by 5.5473 dB, achieving the best effect.

[0104] Figure 3 This demonstrates a comparison of the system's mean square error performance before and after optimizing the filter weights, taking into account both the system's mean square error and the signal stopband energy. "Weighting 2" uses a set of first optimized weights that reduces the signal stopband energy by 5.5473 dB, while "Weighting 3" uses a set of optimal weights optimized after considering both the system's mean square error and the signal stopband energy, resulting in a 6 dB reduction in the signal stopband energy. Figure 3 The comparison shows that by jointly considering both the system mean square error and the signal stopband energy and performing secondary optimization on the weights, the loss of transmission performance such as the system mean square error can be effectively reduced while ensuring the suppression of the signal stopband energy.

[0105] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for reducing the stopband energy of a transmitted signal in an OTFS system, characterized in that, Includes the following steps: S1. Initial simulated weighting processing of filters for each subcarrier on the transmitting and receiving ends of the OTFS system; S2. Establish the power spectral density expression for the OTFS signal after initial analog weighting by the filter; the power spectral density expression is: , where parameters m For subcarrier index, M For the number of subcarriers, For the first m The weights of the filters on the subcarrier, where Represents the Doppler field symbol for transmission delay average power, T Indicates the symbol interval, Represents a shaping filter The energy spectrum, , Represents a shaping filter The average autocorrelation function, ; S3. Based on the power spectral density expression described in S2, establish a filter weight optimization problem with the objective of minimizing the signal stopband energy, and solve the optimization problem to obtain the first optimized weight coefficients and the signal stopband energy value after the first optimized weighting of the filter; the optimization problem model is as follows: ， in, This represents the fluctuation limit between the weight and 1. Indicates the first Given the signal stopband range of the subcarrier and the total signal power remaining constant, the optimization problem is solved to obtain the first optimization weight coefficient and the signal stopband energy value after the first optimization weighting of the filter. S4. Establish the mean square error expression for the OTFS system after initial simulated weighting of the filter; the mean square error expression for the OTFS system is: ,in Represents the time-frequency domain equivalent noise average power, n For time-domain indexing, m For frequency domain indexing, ,in N The number of symbols contained in each frame of signal. This represents Gaussian white noise. This is the expression for a receiver filter with a fixed frequency and bandwidth; S5. Using the signal stopband energy value described in S3 and the OTFS system mean square error expression in S4, establish a weight optimization problem with both as the joint objective, and solve for the optimal filter weight coefficients; the optimization problem model is: ， in, The threshold value representing the signal stopband energy can be used to transform the optimization problem model into a function that represents the signal stopband energy. To optimize the convex problem model of the variables, the optimal weights and the system mean square error are obtained by solving the problem.

2. The method according to claim 1, characterized in that, In step S1, the same shaping filter with fixed frequency and bandwidth is used on each subcarrier at both the transmitting and receiving ends. The formula for the shaping filter at the transmitting end after initial analog weighting is as follows: ,in For the corresponding sending end m The weights of the shaping filter on the subcarrier, The expression for a shaping filter with fixed frequency and bandwidth is given, using the same expression on each subcarrier at the transmitting end. The formula for the receiver shaping filter after initial analog weighting is: ,in For the corresponding receiver end m The weights of the receiver filters on the subcarriers. The expression for a receiver filter with fixed frequency and bandwidth, using the same filter on each subcarrier at the receiver. .

3. The method according to claim 1, characterized in that, It is 0.3-0.

6.

4. A computer-readable storage medium having a communication program based on reducing the stopband energy of an OTFS transmission signal, the communication program being executed to implement the method for reducing the stopband energy of an OTFS transmission signal according to any one of claims 1-3.

Images