Multi-base station cooperation target position and speed estimation method based on OTFS signal

Abstract

The invention discloses a multi-base-station cooperative target position and speed estimation method based on an OTFS signal. The method comprises the following steps: establishing a multi-base-station cooperative sensing signal model based on the OTFS signal; converting a parameter estimation problem into a sparse vector reconstruction problem based on the measurement matrix; designing a subspace tracking algorithm oriented to a multipath environment, and solving the sparse vector; performing target position and speed coarse estimation by using the sparse vector corresponding time delay and Doppler frequency; and establishing a parameter search space, and fusing the sensing information of multiple base stations to obtain the accurate position and speed of the target. According to the method, the target position and speed estimation under the multipath interference is realized by utilizing the characteristics of the time delay-Doppler domain of the OTFS signal and the multi-base-station cooperative gain, and the method has the advantages of high estimation precision, low complexity, multipath interference resistance and the like, and has a wide prospect in practical applications such as a communication sensing integrated system and multi-base-station cooperative sensing.

Description

Technical Field

[0001] This invention belongs to the field of mobile communication technology, specifically relating to a method for estimating the location and velocity of a target through multi-base station cooperation based on OTFS signals. Background Technology

[0002] As a key technology of 6G (6th Generation Mobile Communication) systems, integrated communication and sensing empowers mobile communication base stations to simultaneously perceive their surrounding environment and targets in real time while serving communication users. With the continuous maturation of 6G technology, applications such as vehicle-to-everything (V2X), smart cities, and drone surveillance are rapidly developing. These scenarios not only rely on highly reliable communication capabilities but also place higher demands on accurate target perception. However, single base stations are limited by their limited coverage and single observation perspective, resulting in significant bottlenecks in sensing range and positioning accuracy. By leveraging the advantages of networked deployment of mobile communication base stations, multi-base station collaborative sensing technology integrates the sensing information from multiple base stations, effectively overcoming the inherent limitations of single base stations, such as sensing blind spots and low sensing accuracy. This significantly improves the overall sensing performance of the system, providing crucial support for achieving high-precision and high-reliability target perception.

[0003] Currently, mobile communication systems primarily employ OFDM (Orthogonal Frequency Division Multiplexing) waveforms. However, in high-speed mobile scenarios, Doppler frequency shift disrupts the orthogonality between OFDM subcarriers, leading to inter-carrier interference. This not only reduces communication reliability but also limits the system's sensing accuracy. To address this issue, OTFS (Orthogonal Time Frequency Space) modulation has been proposed as an emerging multi-carrier modulation technique. OTFS transforms the signal from the traditional time-frequency domain to the DD (Delay-Doppler) domain, thereby converting the time-frequency selective channel into a time-invariant channel in the DD domain. This effectively counteracts the Doppler frequency shift caused by high-speed movement, achieving better sensing performance than OFDM. Therefore, applying OTFS waveforms to multi-base station cooperative sensing systems is expected to fully leverage its stable and accurate signal processing advantages in high-speed and complex scenarios, effectively meeting the higher requirements for accurate target sensing in future applications such as vehicle-to-everything (V2X) and low-altitude drone surveillance.

[0004] Currently, several studies have designed sensing algorithms based on OTFS. For example, Chinese invention patent application CN120446945A proposes a high-precision target parameter estimation method for a sensing-integrated system based on OTFS signals. First, it estimates time delay and Doppler frequency shift using mesh refinement technology, and then suppresses inter-path interference through recursive optimization. Chinese invention patent application CN120090902A proposes a fractional-order time delay and Doppler frequency shift estimation method based on OTFS. It uses maximum likelihood estimation and a parameter search method based on the Fibonacci sequence to search for time delay and Doppler frequency shift, and uses continuous interference elimination to suppress interference from the estimated path. Furthermore, the paper "On the Effectiveness of OTFS for Joint Radar Parameter Estimation and Communication" (published in IEEE Transactions on Wireless Communications, vol. 19, no. 9, pp. 5951-5965, 2020) proposes a parameter estimation method based on maximum likelihood, which decomposes the log-likelihood function into two parts: the desired signal and path interference. Parameter estimation is then performed using an iterative optimization algorithm, effectively suppressing inter-path interference. However, the above studies are all limited to single-base station sensing scenarios and do not consider the spatial diversity gain brought about by multi-base station cooperation, thus exhibiting shortcomings in terms of sensing range and estimation accuracy.

[0005] To overcome the limitations of single-base station sensing, existing work has focused on the design of multi-base station cooperative sensing algorithms. The paper "Symbol-Level Integrated Sensing and Communication Enabled Multiple BaseStations Cooperative Sensing" (published in IEEE Transactions on Vehicular Technology, vol. 73, no.1, pp.724-738, 2024) extracts distance and velocity phase features from the received signals along the LoS (Line of Sight) path at each base station based on OFDM waveforms, and then sends the feature information to the fusion center for symbol-level fusion. The paper "Multistatic Integrated Sensing and Communication System in Cellular Networks" (published in IEEE Globecom Workshops, pp.123-128, 2023) performs preliminary parameter estimation based on two-dimensional discrete Fourier transform on the sensed echoes along the LoS path at each receiving base station, and then sends the results to the fusion center for further processing. The paper "Multistatic Parameter Estimation in the Near / Far Field for Integrated Sensing and Communication" (published in IEEE Transactions on Vehicular Technology, vol. 73, no.1, pp.724-738, 2024) further explores this approach. Wireless Communications (vol.23, no.12, pp.17929-17944, 2024) uses the generalized likelihood ratio detection method to estimate the parameters of the Loss of Path Echo (LoS) at each base station, and finally completes the decision at the fusion center.

[0006] Although the above-mentioned work achieves higher sensing accuracy and sensing range compared with single base station through multi-base station cooperative sensing architecture, it still has the following limitations: First, each base station needs to transmit its local processing results to the fusion center for joint processing, which brings additional data transmission and hardware resource overhead; Second, the algorithm only considers the Loss of Sight (LoS) path and does not effectively handle inter-path interference caused by Non-Line of Sight (NLoS) path; Third, the above methods are all based on OFDM signal design, which faces inter-carrier interference problems in high-speed mobile scenarios, resulting in a decrease in sensing performance. Summary of the Invention

[0007] In view of the above, the present invention provides a multi-base station cooperative target position and velocity estimation method based on OTFS signal. It utilizes the time delay-Doppler domain characteristics of OTFS signal and multi-base station cooperative gain to realize target position and velocity estimation under multipath interference, and has the advantages of high estimation accuracy, low complexity and resistance to multipath interference.

[0008] A method for estimating the location and velocity of a target through multi-base station cooperation based on OTFS signals includes the following steps: (1) Establish a multi-base station cooperative sensing signal model based on OTFS signals under multipath interference environment; (2) Based on the above model, construct the measurement matrix and sparse vector, thereby transforming the parameter estimation problem into a sparse vector reconstruction problem and obtaining a new vector expression of the received signal in the DD domain; (3) Based on the new vector representation of the received signal in the DD domain, a subspace tracking algorithm for multipath environments is designed to solve sparse vectors; (4) Based on the time delay and Doppler frequency corresponding to the sparse vector and the location of each base station, make a preliminary estimate of the target's position and velocity; (5) Establish a parameter search space based on the preliminary estimated target position and target velocity, and fuse the multi-base station sensing information based on the maximum likelihood algorithm combined with NLoS path information to obtain the precise position and velocity of the target.

[0009] Furthermore, the specific implementation of step (1) is as follows: S11: For any base station, the transmitted signal of the base station in the DD domain is obtained by inverse symptotic Fourier transform. Convert to time-frequency domain signal Then, the time-frequency domain signal is transformed using the Heisenberg transform. Convert to time domain signal time domain signal The signal is transmitted directionally towards the approximate location of the target via the uniform array antenna of the base station; S12: The main base station receives the sensing echoes from its own signals and those transmitted by other base stations, which are reflected multiple times by buildings or directly illuminated by the target, and obtains the time-domain received signal. ; S13: Receive signal in the time domain Matched filtering and Wigner transform are performed to obtain the time-frequency domain signal. ,right Discrete time-frequency domain signals are obtained by performing discrete sampling. Then, through the symmetric Fourier transform, Convert to DD domain received signal Its vector expression is as follows (i.e., the multi-base station cooperative sensing signal model): in: Indicates the delay as Doppler frequency is The equivalent channel, for The vector form, For equivalent channel coefficients and , It is Gaussian white noise. , and They represent the first i From the first base station to the target and then to the main base station p The path delay, Doppler frequency, and channel coefficients, For the first i The number of paths per base station, For the number of base stations, J The imaginary unit, Indicates the angle and In the case of the first i The first base station to the target p The emission guide vector of the path, Indicates the angle and Under these circumstances, the receive steering vector at the main base station and They represent the first i The first base station to the target p The true orientation angle and true pitch angle of the path. and These represent the true direction of arrival (AOI) and true elevation of arrival (PEA) at the main base station, respectively. For the first i Transmit beamforming vectors of each base station The receive beamforming vector of the main base station. , , Indicates the angle In the case of the first i The first base station to the target p The emission guide vector of the path, Indicates the angle Under these circumstances, the receive steering vector at the main base station They represent the first i The first base station to the target p The transmit beamforming direction angle and elevation angle of the path, These represent the receiving beamforming direction angle and elevation angle at the main base station, respectively. Indicates the firsti The transmit power of each base station, This indicates the number of transmit or receive array elements in the base station; the superscript H indicates conjugate transpose. This is the generalized time delay-Doppler crosstalk coefficient matrix.

[0010] Furthermore, the specific implementation of step (2) is as follows: S21: Let the propagation distance of the signal from each base station to the target and then reflected back to the main base station be... The target speed range is ; S22: Take uniformly from the propagation distance interval Points are taken uniformly from the target speed range. These points are formed by combining them in pairs. Q Time delay - Doppler pair , q For delay-Doppler index and , ; S23: Construct a measurement matrix based on the multi-base station cooperative sensing signal model and the time delay-Doppler pair. Sparse vectors : in: For a binary indicator, when hour, ,otherwise The superscript T indicates transpose; S24: According to the measurement matrix Sparse vectors The new vector expression for the received signal in the DD domain is established as follows: Furthermore, the subspace tracking algorithm for multipath environments in step (3) first initializes the residual signal. = index set Then, the following iterative calculations are performed: S31: In the j In the round of iteration, the first j Wheel residual signal Projected onto the measurement matrix The projection vector is obtained from the above. From this projection vector, select the indices of the first few maximum values, while ensuring that the selected indices satisfy the following constraints with all previously selected indices: When the spacing of the time delay axis is greater than When the Doppler frequency axis spacing is greater than ,in M Indicates the number of delay taps. N Indicates the number of Doppler taps, Δ f Indicates the subcarrier spacing. T Indicates the duration of the symbol; S32: Store all indexes that meet the above conditions in the index set. In, and will Compared with the previous set of indexes Merge to obtain the index set ; S33: Using index sets The elements in the measurement matrix are column indices. Select the corresponding column vectors to construct a new matrix. ; S34: Calculate the matrix Channel coefficients corresponding to each column vector Paths whose channel coefficient modulus is greater than a given threshold are considered valid paths. ; S35: Update the index set for Index of elements in the matrix, update the matrix From Extract column indexes in The new matrix formed by the columns in the matrix updates the residual signal. ; S36: Perform the next round of iterations based on steps S31~S35 until the residual signal converges, then convert the effective path of the current round. As the final sparse vector solution.

[0011] Furthermore, the specific implementation of step (4) is as follows: S41: Sparse vectors The element with the largest median modulus value is considered as the channel coefficient of the Loss-of-Stake (LoS) path, thus initially determining the delay of the base station's LoS path. and Doppler frequency ; S42: Combine the LoS path delay equations of each base station and solve them to obtain a preliminary estimate of the target location; S43: Combine and solve the LoS path Doppler frequency shift equations of each base station to obtain a preliminary estimate of the target velocity.

[0012] Furthermore, the specific implementation of step (5) is as follows: S51: Perform uniform sampling around the initially estimated target position and target velocity to obtain a sample containing... Distance search set of points and contain Speed ​​search set of points ; S52: Search set based on distance The points in the equation are used to calculate the time delay search set corresponding to each base station. Search set based on speed The Doppler frequency search set corresponding to each base station is calculated using the Doppler frequency shift equation. ; S53: The DD domain received signal is represented as: in: Let G be the estimation error of the NLoS path, and let G be the composite matrix composed of the observation matrices of all base stations. i For the first i The observation matrix of each base station, where h represents the composite vector consisting of the complex channel coefficient vectors of all base stations. For the first i The complex channel coefficient vector of each base station sparse vector The first in One non-zero element, The number of non-zero elements in the sparse vector. and Measurement matrix Zhongyu The index corresponds to the delay and Doppler frequency. and The first i The latency and Doppler frequency of the LosS path for each base station, For the first i Equivalent channel coefficients of each base station's LosS path; S54: Based on the above signal representation, establish the following objective function and solve it to obtain accurate estimates of the LoS path delay and Doppler frequency of each base station; in: , ; S55: Based on the LoS path delay estimation results of each base station In the distance search set The precise location of the target is determined by the correspondence in the data, based on the LoS path Doppler frequency estimation results of each base station. In the speed search set The correspondence in the data determines the target's precise speed.

[0013] Preferably, in step S54, the maximum likelihood algorithm is used to solve the objective function. This algorithm expands the objective function and ignores constant terms, thereby transforming it into the following optimization objective and solving it: in: This is the maximum likelihood estimation function.

[0014] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described multi-base station cooperative target position and velocity estimation method based on OTFS signals.

[0015] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for estimating the target location and velocity based on OTFS signals through multi-base station cooperative operation.

[0016] This invention transforms the received signal into the DD domain, constructs a measurement matrix and corresponding sparse vectors, and converts the time delay and Doppler frequency estimation problem into a sparse vector reconstruction problem. Based on the guard interval, a subspace tracking algorithm that effectively suppresses inter-path interference is designed. According to the estimated LosS path delay and Doppler frequency, the initial position and velocity of the target are obtained. Based on this, a position search interval and a velocity search interval are constructed. Within the search interval, the maximum likelihood algorithm is used in conjunction with NLoS path information to fuse multi-base station sensing information, thereby obtaining the precise position and velocity of the target. Therefore, this invention has the following advantages compared to existing technologies: 1. This invention utilizes the sparsity of the time-delay-Doppler domain channel to transform the target time delay and Doppler frequency estimation problem into a sparse vector reconstruction problem, and further designs a subspace tracking algorithm for multipath interference scenarios, which can realize the estimation of time delay and Doppler frequency for multiple paths.

[0017] 2. This invention leverages the advantages of multi-base station cooperative sensing, combining the maximum likelihood algorithm with NLoS path information to achieve multi-base station sensing information fusion, suppress NLoS path interference, and achieve higher target delay and Doppler frequency estimation accuracy. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the application scenario targeted by the present invention.

[0019] Figure 2 This is a flowchart illustrating the overall process of the multi-base station cooperative target location and velocity estimation method of the present invention.

[0020] Figure 3 This is a schematic diagram of the subspace tracking algorithm for multipath scenarios in this invention.

[0021] Figure 4 This is a schematic diagram of the protection interval in this invention.

[0022] Figure 5 This is a schematic diagram of the location search range in this invention.

[0023] Figure 6 This is a schematic diagram of the speed search interval in this invention.

[0024] Figure 7 This diagram illustrates the performance comparison of the root mean square error of position and velocity estimation as a function of signal-to-noise ratio under different numbers of base stations.

[0025] Figure 8 This is a schematic diagram comparing the performance of the method of the present invention in terms of the root mean square error of position and velocity estimation as a function of signal-to-noise ratio under different numbers of paths. Detailed Implementation

[0026] To describe the present invention in more detail, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] For the application of target position and velocity estimation in practical scenarios such as urban low-altitude drone surveillance and next-generation vehicle-to-everything (V2X) networks, existing target position and velocity estimation methods based on OTFS signals are mostly for single-base station sensing scenarios, facing challenges such as poor sensing accuracy and short sensing distance. Existing multi-base station collaborative target position and velocity estimation methods process signals in the time-frequency domain, facing problems such as high Doppler frequency shift and multipath interference.

[0028] To address the above problems and challenges, this implementation method considers the following application scenarios: Figure 1 As shown, in this scene there are One base station, recorded The primary base station is the base station, and the other base stations are secondary base stations. The system receives the transmitted signals from itself and other base stations, which are reflected multiple times by buildings or directly illuminated by the target. Each base station is equipped with a uniform transmitting array and a uniform receiving array, and the approximate location of the target is known.

[0029] To address this scenario, this implementation transforms the parameter estimation problem into a sparse signal reconstruction problem by constructing a measurement matrix and sparse vectors. A compressed sensing algorithm that effectively suppresses inter-path interference is designed based on the guard interval. Then, based on the maximum likelihood algorithm combined with NLoS path information, multi-base station sensing information fusion is achieved, proposing a multi-base station cooperative target position and velocity estimation method based on OTFS signals in multi-path interference scenarios. The specific implementation steps are as follows: Figure 2 As shown: Step 1: Establish a multi-base station cooperative sensing signal model based on OTFS signals under multipath interference environment.

[0030] Record No. base stations The signal transmitted in the DD domain is ,in, Indicates the base station index. Indicates Doppler index, Indicates the number of Doppler taps. Indicates a delay index. Indicates the number of delay taps; for The time-frequency domain signal is obtained by performing an inverse symplectic Fourier transform. It is represented as: in: For symbol indexing, For subcarrier indexing.

[0031] Then, the time-frequency domain signal is transformed using the Heisenberg transform. Transform to the time domain to obtain the time domain signal Its expression is: in: Indicates the subcarrier spacing. Indicates the duration of the symbol. Indicates amplitude Duration is The transmitted waveform.

[0032] Time domain signal The signal is launched directionally towards the approximate location of the target using a uniform array antenna. The system receives the sensing echoes from its own signals and those from other base stations, which are reflected multiple times by buildings or directly onto the target, to obtain the time-domain received signal. Its expression is: in: Indicates the Loss path index. Indicates the NLoS path index. Indicates the first The number of paths to each base station; They represent the first From the first base station to the target and then to the main base station The path delay, Doppler frequency, and channel coefficients; They represent the first The first base station to the target The true orientation angle and true pitch angle of the path; They represent the main base station. The true heading angle and true pitch angle at the location, For the first i The first base station to the target p The emission guide vector of the path, The receiving steering vector at the main base station. , The superscript T indicates transpose. , Indicates the horizontal matrix element index, Indicates the vertical matrix element index. Indicates the horizontal element spacing. Indicates the vertical element spacing; and The first Individual base stations and main base stations The transmit beamforming vector and receive beamforming vector are generated based on the approximate location of the target obtained from prior information. , , here They represent the first The first base station to the target The transmit beamforming direction angle and elevation angle of the path; They represent the main base station. The receiving beamforming direction angle and elevation angle at the location, Indicates the first The transmit power of each base station, This indicates the number of base station transmit or receive array elements, and the superscript H indicates conjugate transpose.

[0033] For time-domain received signals Matched filtering and Wigner transform are performed to obtain the time-frequency domain signal. : in: It receives the shaping pulse, superscript This indicates a conjugate operation.

[0034] Furthermore, on Discrete time-frequency domain signals are obtained by performing discrete sampling. Then, the DD domain signal is obtained through the symmetric Fourier transform. Its expression is: Simplifying the above equation, we obtain the DD-domain input-output relationship in vector form as follows: in: These are the equivalent channel coefficients. It is Gaussian white noise. yes The vector form, It is the equivalent channel, and its expression is: in: The index represents the mutual ambiguity function between the received and transmitted waveforms. , , and and , , and The definitions are the same.

[0035] Step 2: Based on the measurement matrix, the parameter estimation problem is transformed into a sparse vector reconstruction problem.

[0036] Let the propagation distance of the signal from each base station to the target and then reflected back to the main base station be . The target speed range is ; take uniformly in the distance interval and velocity interval respectively and Each point is combined in pairs to form Time delay - Doppler pair , For delay-Doppler pair indexing.

[0037] Define a binary indicator ,when hour, ,otherwise At this point, the received signal can be represented as: Constructing the measurement matrix sparse vectors Then the received signal in the DD domain can be expressed as: Therefore, the received signal is re-represented using a measurement matrix and a sparse vector. Only the non-zero elements in the sparse vector need to be estimated, and the time delay and Doppler frequency of the corresponding path can be obtained from the index of the non-zero elements.

[0038] Step 3: Design a subspace tracking algorithm for multipath environments, and solve for sparse vectors. The specific algorithm flow is as follows: Figure 3 As shown: First, initialize the residual signal. = index set Then, the following iterative algorithm is performed, at the th... In the round of iterative calculation, the first Wheel residual signal Projected onto the measurement matrix The projection vector is obtained from the above. Select the front from the projection vector The index of the maximum value.

[0039] Meanwhile, to prevent strong path echoes from drowning out weak path echoes during the selection of the maximum value index, we propose a method based on spectral characteristics with a time delay axis width of [missing information]. At a Doppler axis width of The protection interval, such as Figure 4 As shown, ensure that the selected index is outside the protection interval of all selected indexes; store all indexes that meet the above conditions in the index set. In the middle, and merged with the index set of the previous round, that is .

[0040] With index set The elements in the measurement matrix are column indices. Select the corresponding column vectors to construct a new matrix. ;calculate Channel coefficients corresponding to each column vector Paths whose channel coefficient modulus is greater than a given threshold are considered valid paths. ,in The effective path threshold coefficient, express The Middle Each element.

[0041] Finally, update the index set. for Element index, update From Extract column indexes in The new matrix formed by the columns in the matrix updates the residual signal. .

[0042] Repeat the above process until the maximum number of iterations is reached or ,in This represents the convergence threshold, which yields the final output. .

[0043] Step 4: Use the sparse vector corresponding to the time delay, Doppler, and the location of each base station to make a rough estimate of the target's location and velocity.

[0044] remember The number of non-zero elements is ,for Each non-zero element ( ),according to The index in the measurement matrix Find the delay corresponding to the index and Doppler frequency ;Will The element with the largest modulus value is considered the channel coefficient of the Loss of Path (LoS) path, and its channel coefficient, delay, and Doppler frequency are denoted as follows: , and .

[0045] For the initial location estimation of the target, the following system of equations is established based on the LoS path delay of each base station: in: Indicates the speed of electromagnetic wave propagation. Indicates the target and the first The distance between base stations, here Indicates the first The location of each base station. This represents the target location to be estimated; solving the above system of equations yields a preliminary estimate of the target location. .

[0046] For the initial velocity estimation of the target, the following system of equations is established based on the LoS path Doppler frequency shift of each base station: in: Indicates the carrier frequency. Indicates the target to the number The unit steering vector of each base station, This represents the target velocity to be estimated; solving the above system of equations yields a preliminary estimate of the target velocity. .

[0047] Step 5: Establish a parameter search space. Based on the maximum likelihood algorithm combined with NLoS path information, fuse the sensing information from multiple base stations to obtain the precise location and velocity of the target.

[0048] like Figure 5 and Figure 6 As shown, around the initially estimated target location and target speed Uniform sampling is performed separately to obtain samples containing Distance search set of points and contain Speed ​​search set of points Furthermore, using the time delay equation And Doppler frequency shift equation Calculate the time delay search set corresponding to each base station. And Doppler frequency search set .

[0049] Let the estimation error of the NLoS path obtained by compressed sensing be . Its expression is: Therefore, the time-delay-Doppler domain received signal can be represented as: Then, based on the maximum likelihood algorithm, the LosS path delay and Doppler frequency of each base station are estimated: Expanding the problem and ignoring the constant term, we get: That is, in the time-delay search set And Doppler frequency search set In the middle, make The largest point represents the target's precise time delay and Doppler frequency. Finally, the search set is determined based on the distance. Speed ​​search set With delay search set And Doppler frequency search set The correspondence between these parameters yields the target's precise position and velocity.

[0050] Below, we will verify the effectiveness of the method of this invention through simulation examples. Simulation parameter settings: number of delay taps. Doppler tap count The subcarrier spacing is 250kHz, the carrier frequency is 30GHz, and the number of transmit and receive antenna array elements is... The number of horizontal array elements Vertical array elements Effective path threshold coefficient =0.092, convergence threshold The main base station is located at the origin. The base stations are arranged in a square The target is a polygon with an adjacent base station spacing of 200 meters. Within the sensing range of all base stations, a target moves in any direction at a speed of 60-80 m / s, and the distance between the target and each base station is no less than 50 meters.

[0051] Simulation Example 1: Setting the Number of Paths for Each Base Station This consists of one Loss-of-Sight (LoS) path and two Non-LoS (NLoS) paths, with a Rice coefficient of 8 dB. We use the DDIPIC algorithm, a single-base station parameter estimation algorithm proposed in the paper "Channel and Radar Parameter Estimation With Fractional Delay-Doppler Using OTFS" (published in IEEE Communications Letters, vol. 27, no. 5, pp.1392-1396, May 2023), as a comparison algorithm. This algorithm can effectively suppress multipath interference while achieving high-precision parameter estimation. The performance comparison of the root mean square error (RMSE) of target position and velocity estimation between the proposed method and the DDIPIC algorithm under different signal-to-noise ratios (SNRs), and the variation of the RMS error of target position and velocity estimation with SNR for different numbers of base stations, are shown below. Figure 7 As shown. By Figure 7 It is evident that the root mean square error of the method of this invention is superior to that of the comparative algorithms, and the accuracy of position and velocity estimation also improves with the increase of the number of base stations.

[0052] Simulation Example 2: Setting the Number of Base Stations The Rice coefficient is 8 dB. Since there is currently no research on the design of multi-base station cooperative sensing algorithms based on OTFS, we use the relevant literature on OFDM-based multi-base station cooperative sensing algorithm design, *Symbol-Level Integrated Sensing and Communication Enabled MultipleBase Stations Cooperative Sensing* (published in IEEE Transactions on Vehicular Technology, vol. 73, no. 1, pp. 724-738, Jan. 2024), as a comparative work. The performance of the root mean square error (RMSE) of target position and velocity estimation of the proposed method compared with existing OFDM multi-base station cooperative sensing methods under different signal-to-noise ratios is shown, as well as the variation of the RMS error of target position and velocity estimation of the proposed method with different numbers of paths, as follows: Figure 8 As shown. By Figure 8 It is evident that the root mean square error of the method of the present invention is superior to that of the existing OFDM multi-base station cooperative sensing method, and its sensing performance does not decrease significantly with the increase of the number of paths.

[0053] In summary, this invention suppresses inter-path interference during sparse vector reconstruction by introducing a guard interval. Then, it utilizes NLoS path information and performs multi-base station sensing information fusion based on the maximum likelihood algorithm, effectively suppressing NLoS path interference. It features high accuracy and robustness in position and velocity estimation, and has broad prospects for practical applications such as integrated communication and sensing systems, multi-base station collaborative low-altitude UAV sensing, and next-generation vehicle-to-everything (V2X) networks.

[0054] The above description of the embodiments is provided to enable those skilled in the art to understand and apply the present invention. Those skilled in the art can readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made to the present invention by those skilled in the art based on the disclosure thereof should be within the scope of protection of the present invention.

Claims

1. A method for estimating the location and velocity of a target through multi-base station cooperation based on OTFS signals, comprising the following steps: (1) Establish a multi-base station cooperative sensing signal model based on OTFS signals under multipath interference environment; (2) Based on the above model, construct the measurement matrix and sparse vector, thereby transforming the parameter estimation problem into a sparse vector reconstruction problem and obtaining a new vector expression of the received signal in the DD domain; (3) Based on the new vector representation of the received signal in the DD domain, a subspace tracking algorithm for multipath environments is designed to solve sparse vectors; (4) Based on the time delay and Doppler frequency corresponding to the sparse vector and the location of each base station, make a preliminary estimate of the target's position and velocity; (5) Establish a parameter search space based on the preliminary estimated target position and target velocity, and fuse the multi-base station sensing information based on the maximum likelihood algorithm combined with NLoS path information to obtain the precise position and velocity of the target.

2. The multi-base station cooperative target location and velocity estimation method based on OTFS signals according to claim 1, characterized in that, The specific implementation method of step (1) is as follows: S11: For any base station, the transmitted signal of the base station in the DD domain is obtained by inverse symptotic Fourier transform. Convert to time-frequency domain signal Then, the time-frequency domain signal is transformed using the Heisenberg transform. Convert to time domain signal time domain signal The signal is transmitted directionally towards the approximate location of the target via the uniform array antenna of the base station; S12: The main base station receives the sensing echoes from its own signals and those transmitted by other base stations, which are reflected multiple times by buildings or directly illuminated by the target, and obtains the time-domain received signal. ; S13: Receive signal in the time domain Matched filtering and Wigner transform are performed to obtain the time-frequency domain signal. ,right Discrete time-frequency domain signals are obtained by performing discrete sampling. Then, through the symmetric Fourier transform, Convert to DD domain received signal Its vector expression is as follows: ; ; in: Indicates the delay as Doppler frequency is The equivalent channel, for vector form, For equivalent channel coefficients and , It is Gaussian white noise. , and They represent the first i From the first base station to the target and then to the main base station p The path delay, Doppler frequency, and channel coefficients. For the first i The number of paths per base station, For the number of base stations, J The imaginary unit, Indicates the angle and In the case of the first i The first base station to the target p The launch guidance vector of the path, Indicates the angle and Under these circumstances, the receiving steering vector at the main base station and They represent the first i The first base station to the target p The true orientation angle and true pitch angle of the path, and These represent the true direction of arrival (DOA) and the true elevation of arrival (PEA) at the main base station, respectively. For the first i Transmit beamforming vectors of each base station The receive beamforming vector of the main base station. , , Indicates the angle In the case of the first i The first base station to the target p The launch guidance vector of the path, Indicates the angle Under these circumstances, the receiving steering vector at the main base station They represent the first i The first base station to the target p The transmit beamforming direction angle and elevation angle of the path, These represent the receiving beamforming direction angle and elevation angle at the main base station, respectively. Indicates the first i The transmit power of each base station, This indicates the number of transmit or receive array elements in the base station; the superscript H indicates conjugate transpose. This is the generalized time delay-Doppler crosstalk coefficient matrix.

3. The multi-base station cooperative target location and velocity estimation method based on OTFS signals according to claim 2, characterized in that, The specific implementation method of step (2) is as follows: S21: Let the propagation distance of the signal from each base station to the target and then reflected back to the main base station be... The target speed range is ; S22: Take uniformly from the propagation distance interval Points are taken uniformly from the target velocity range. These points are formed by combining them in pairs. Q Time delay - Doppler pair , q For delay-Doppler index and , ; S23: Construct a measurement matrix based on the multi-base station cooperative sensing signal model and the time delay-Doppler pair. Sparse vectors : ; ; in: For a binary indicator, when hour, ,otherwise The superscript T indicates transpose; S24: According to the measurement matrix Sparse vectors The new vector expression for the received signal in the DD domain is established as follows: 。 4. The multi-base station cooperative target location and velocity estimation method based on OTFS signals according to claim 3, characterized in that: The subspace tracking algorithm for multipath environments in step (3) first initializes the residual signal. = index set Then, the following iterative calculations are performed: S31: In the j In the round of iteration, the first j Wheel residual signal Projected onto the measurement matrix The projection vector is obtained from the above. From this projection vector, select the indices of the first few maximum values, while ensuring that the selected indices satisfy the following constraints with all previously selected indices: When the spacing of the time delay axis is greater than When the Doppler frequency axis spacing is greater than ,in M Indicates the number of delay taps. N Indicates the number of Doppler taps, Δ f Indicates the subcarrier spacing. T Indicates the duration of the symbol; S32: Store all indexes that meet the above conditions in the index set. In, and will Compared with the previous set of indexes Merge to obtain the index set ; S33: Using index sets The elements in the measurement matrix are column indices. Select the corresponding column vectors to construct a new matrix. ; S34: Calculate the matrix Channel coefficients corresponding to each column vector Paths whose channel coefficient modulus is greater than a given threshold are considered valid paths. ; S35: Update the index set for Index of elements in the matrix, update the matrix From Extract column indexes in The new matrix formed by the columns in the matrix updates the residual signal. ; S36: Perform the next round of iterations based on steps S31~S35 until the residual signal converges, then convert the effective path of the current round. As the final sparse vector solution.

5. The multi-base station cooperative target position and velocity estimation method based on OTFS signals according to claim 4, characterized in that, The specific implementation method of step (4) is as follows: S41: Sparse vectors The element with the largest median modulus value is considered as the channel coefficient of the Loss-of-Stake (LoS) path, thus initially determining the delay of the base station's LoS path. and Doppler frequency ; S42: Combine the LoS path delay equations of each base station and solve them to obtain a preliminary estimate of the target location; S43: Combine and solve the LoS path Doppler frequency shift equations of each base station to obtain a preliminary estimate of the target velocity.

6. The multi-base station cooperative target position and velocity estimation method based on OTFS signals according to claim 4, characterized in that, The specific implementation method of step (5) is as follows: S51: Perform uniform sampling around the initially estimated target position and target velocity to obtain a sample containing... Distance search set of points and contain Speed ​​search set of points ; S52: Search set based on distance The points in the equation are used to calculate the time delay search set corresponding to each base station. Search set based on speed The Doppler frequency search set corresponding to each base station is calculated using the Doppler frequency shift equation. ; S53: The DD domain received signal is represented as: ; ; ; ; ; ; in: Let G be the estimation error of the NLoS path, and let G be the composite matrix composed of the observation matrices of all base stations. i For the first i The observation matrix of each base station, where h represents the composite vector consisting of the complex channel coefficient vectors of all base stations. For the first i The complex channel coefficient vector of each base station sparse vector The first in One non-zero element, The number of non-zero elements in the sparse vector. and Measurement matrix Zhongyu The index corresponds to the delay and Doppler frequency. and The first i The latency and Doppler frequency of the LosS path for each base station, For the first i Equivalent channel coefficients of each base station's LosS path; S54: Based on the above signal representation, establish the following objective function and solve it to obtain accurate estimates of the LoS path delay and Doppler frequency of each base station; ; in: , ; S55: Based on the LoS path delay estimation results of each base station In the distance search set The precise location of the target is determined by the correspondence in the data, based on the LoS path Doppler frequency estimation results of each base station. In the speed search set The correspondence in the data determines the target's precise speed.

7. The multi-base station cooperative target position and velocity estimation method based on OTFS signals according to claim 6, characterized in that: In step S54, the maximum likelihood algorithm is used to solve the objective function. This algorithm expands the objective function and ignores the constant term, thereby transforming it into the following optimization objective, which is then solved: ; in: This is the maximum likelihood estimation function.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: The processor is used to execute the computer program to implement the multi-base station cooperative target location and velocity estimation method based on OTFS signals as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the multi-base station cooperative target location and velocity estimation method based on OTFS signals as described in any one of claims 1 to 7.

Images