A multi-station cooperative target perception method and device, electronic equipment and storage medium

Abstract

This invention discloses a multi-station cooperative target sensing method, device, electronic device, and storage medium, relating to the fields of wireless communication and radar sensing technology. The method includes: constructing an AFDM signal containing chirped subcarriers and adding a cyclic prefix before transmission; receiving the echo and performing down-conversion, prefix removal, sampling, and DAFT to obtain three-dimensional observations; performing two-dimensional spatial smoothing on the observations to construct an enhanced observation matrix; jointly estimating the azimuth and elevation angles using the characteristics of a uniform planar array; and recovering the time delay-Doppler parameters using the least squares method to obtain local parameter estimates; performing cross-station target matching based on the local parameter estimates; using geometric complementarity relationships, establishing an equation system through weighted Gauss-Newton iteration and combining multi-station Doppler observations to solve for the three-dimensional position; and constructing a geometric relationship model based on the three-dimensional position and combining Doppler observations to establish an equation system to obtain the three-dimensional velocity. This invention effectively solves the performance limitations of existing technologies in high-dynamic scenarios.

Description

Technical Field

[0001] This invention relates to the fields of wireless communication and radar sensing technology, and in particular to a multi-station cooperative target sensing method, device, electronic device, and storage medium. Background Technology

[0002] With the booming development of low-altitude economy, urban air traffic, unmanned system collaboration, and airspace safety supervision, the demand for continuous perception and precise positioning of low-altitude targets is becoming increasingly prominent, and has become key to ensuring airspace safety and improving traffic efficiency. However, low-altitude targets often have characteristics such as low flight altitude, high maneuverability, rapid changes in state, and complex and changeable propagation environment, which poses a severe challenge to traditional target perception and positioning technologies.

[0003] Traditional high-dynamic target sensing technologies, such as radar and communication systems based on orthogonal frequency division multiplexing (OFDM), are significantly affected by Doppler spread in high-dynamic scenarios, leading to increased inter-carrier interference. This, in turn, exacerbates the coupling problem between time delay and Doppler parameters, making it difficult to achieve stable and accurate parameter estimation. Furthermore, traditional methods generally face problems such as high computational complexity, insufficient real-time performance, or accuracy limitations imposed by grid partitioning when dealing with multi-parameter, multi-target, and large-scale array scenarios, making it difficult to meet the high-precision and high-efficiency requirements of low-altitude target sensing.

[0004] Therefore, there is an urgent need for an efficient, accurate, and robust multi-station cooperative target perception method that can effectively adapt to highly dynamic low-altitude target scenarios, achieve joint estimation of parameters such as azimuth, pitch, range, and Doppler, and further recover the target's global three-dimensional position and true three-dimensional velocity. Summary of the Invention

[0005] The present invention provides a multi-station collaborative target perception method to address the problems of unstable parameter estimation, high computational complexity, poor real-time performance, and difficulty in recovering the target's global three-dimensional position and true velocity in existing technologies. The technical solution is as follows: According to one aspect of the present invention, a multi-station cooperative target sensing method is provided, the method comprising: constructing an AFDM signal containing chirped subcarriers at a transmitting station and adding a cyclic prefix before up-conversion transmission; receiving target scattered echoes at each receiving station and performing down-conversion, deprecation, sampling, and DAFT to obtain three-dimensional observations; performing two-dimensional spatial smoothing on the three-dimensional observations; constructing multiple subarray observation blocks and fusing them into an enhanced observation matrix; jointly estimating the azimuth and elevation angles based on the enhanced observation matrix using the characteristics of a uniform planar array; recovering the time delay-Doppler parameters using the least squares method to obtain the results of each receiving station. The system first estimates the local parameters of the target at each station; then performs cross-station target matching based on the estimated local parameters; uses geometric complementarity to perform three-dimensional positioning of the target through weighted Gauss-Newton iteration; and establishes a system of linear velocity equations by combining multi-station Doppler observations, and solves the system of equations together to obtain the global three-dimensional position of the target; based on the global three-dimensional position, it constructs the geometric relationship between the transmitting station, the target, and the receiving station; and establishes a system of linear observation equations for the target velocity by combining the Doppler observations independently estimated by each receiving station, and solves the system of equations together to obtain the true three-dimensional velocity of the target in the global coordinate system.

[0006] In one embodiment, the upconversion transmission of the AFDM signal containing chirped subcarriers after adding a cyclic prefix at the transmitting station is achieved through the following steps: constructing an AFDM signal containing chirped subcarriers at the transmitting station, adding a chirped cyclic prefix (CPP) before each AFDM symbol, upconverting the AFDM signal with the added cyclic prefix, and transmitting it through an antenna; the chirped cyclic prefix (CPP) is used to suppress inter-symbol interference caused by multipath propagation.

[0007] In one embodiment, the three-dimensional observation is obtained by receiving the target's scattered echo at each receiving station and performing down-conversion, deprecation, sampling, and DAFT. This is achieved through the following steps: receiving the scattered echo signal from the target at each receiving station; performing down-conversion processing on the scattered echo signal to reduce the high-frequency signal to the baseband frequency; removing the chirped cyclic prefix from the scattered echo signal to restore the original AFDM symbol structure; discretely sampling the scattered echo signal to convert it into a digital signal; and transforming the scattered echo signal to the Discrete Affine Fourier Transform (DAFT) domain to construct the three-dimensional observation. The three-dimensional observation includes the array row dimension, the array column dimension, and the symbol and transform domain sampling stack dimension.

[0008] In one embodiment, the three-dimensional observations are smoothed in two-dimensional space, and multiple subarray observation blocks are constructed and fused into an enhanced observation matrix through the following steps: For the three-dimensional observations, continuous subarrays are truncated along the row and column dimensions of the array by each receiving station, and the original observations are smoothed to construct multiple subarray observation blocks. The subarray observation blocks are then fused into an enhanced observation matrix. The smoothing process is used to reduce the correlation between multi-target signals and improve the effective observation rank.

[0009] In one embodiment, the azimuth and elevation angles are jointly estimated based on the enhanced observation matrix using the characteristics of a uniform planar array, and the time delay-Doppler parameters are recovered using the least squares method. The local parameter estimates of the target for each receiving station are obtained through the following steps: The signal subspace is extracted based on the enhanced observation matrix using the characteristics of a uniform planar array, and the azimuth and elevation angle parameters of the target are jointly estimated using the shift invariance principle to obtain angle estimation results; the angle estimation results reflect the spatial orientation information of the target relative to the receiving station; a receiving array manifold is constructed based on the angle estimation results, and the time delay-Doppler structure vector corresponding to the target is recovered using the least squares method; the structure vector is then subjected to pulse compression and local continuous refinement processing to obtain the range and Doppler parameter estimation results of the target.

[0010] In one embodiment, cross-station target matching is performed based on the local parameter estimation. The target is then three-dimensionally located using a weighted Gauss-Newton iteration method based on geometric complementarity. A system of linear velocity equations is established by combining multi-station Doppler observations, and the global three-dimensional position of the target is obtained through joint solution. This is achieved through the following steps: cross-station target matching is performed based on the local parameter estimation results obtained from each receiving station. This cross-station target matching ensures that different receiving stations observe the same target. The local parameter estimation includes azimuth, elevation, range, and Doppler. The matched target is then jointly located using a weighted Gauss-Newton iteration method based on the geometric complementarity between multiple receiving stations. Finally, the global three-dimensional position of the target is obtained by iteratively solving a weighted nonlinear least squares problem based on the observation data and geometric layout information of each receiving station.

[0011] In one embodiment, the geometric relationship between the transmitting station, target, and receiving station is constructed based on the global three-dimensional position. A set of linear observation equations for the target velocity is established by combining the Doppler observations independently estimated by each receiving station. The true three-dimensional velocity of the target in the global coordinate system is obtained by jointly solving the set of equations. This is achieved through the following steps: constructing a geometric relationship model between the transmitting station, target, and receiving station based on the global three-dimensional position; establishing a set of linear observation equations for the target velocity by combining the Doppler observations independently estimated by each receiving station; the Doppler observations reflect the target's velocity projection relative to the receiving station; and obtaining the true three-dimensional velocity vector of the target in the global coordinate system by jointly solving the set of linear observation equations. The set of linear observation equations describes the relationship between the target's true velocity and the radial velocity observed by each receiving station.

[0012] According to one aspect of the present invention, a multi-station cooperative target sensing device includes: an AFDM signal transmission and processing module, configured to construct an AFDM signal containing chirped subcarriers at a transmitting station, add a cyclic prefix, and then perform up-conversion transmission; receive target scattered echoes at each receiving station and perform down-conversion, deprecation, sampling, and DAFT to obtain three-dimensional observations; and a three-dimensional observation and parameter prediction module, configured to perform two-dimensional spatial smoothing on the three-dimensional observations, construct multiple subarray observation blocks and fuse them into an enhanced observation matrix, jointly estimate the azimuth and elevation angles based on the enhanced observation matrix using the characteristics of a uniform planar array, recover the time delay-Doppler parameters using the least squares method, and obtain the three-dimensional observations from each receiving station. The system includes: a receiving station for estimating local parameters of the target; a multi-station collaborative 3D positioning module for cross-station target matching based on the estimated local parameters, using geometric complementarity to perform 3D positioning of the target through weighted Gauss-Newton iteration, and establishing a system of linear velocity equations by combining multi-station Doppler observations, and jointly solving the system to obtain the global 3D position of the target; and a global velocity analytical solution module for constructing the geometric relationship between the transmitting station, the target, and the receiving station based on the global 3D position, establishing a system of linear observation equations for the target velocity by combining the Doppler observations independently estimated by each receiving station, and jointly solving the system of equations to obtain the true 3D velocity of the target in the global coordinate system.

[0013] According to one aspect of the present invention, an electronic device includes at least one processor and at least one memory, wherein computer-readable instructions are stored on the memory; the computer-readable instructions are executed by one or more of the processors, causing the electronic device to implement the multi-station cooperative target perception method as described above.

[0014] According to one aspect of the invention, a storage medium stores computer-readable instructions thereon, which are executed by one or more processors to implement the multi-station cooperative target perception method as described above.

[0015] The beneficial effects of the technical solution provided by this invention are: In the above technical solution, this invention first constructs an AFDM signal with chirped subcarriers at the transmitting station and adds a cyclic prefix before transmission. The receiving station receives the echo and performs down-conversion, deprecation, sampling, and DAFT to obtain three-dimensional observations. Next, the observations are smoothed in two-dimensional space to construct an enhanced observation matrix. The azimuth and elevation angles are jointly estimated using the characteristics of a uniform planar array, and the time delay-Doppler parameters are recovered using the least squares method to obtain local parameter estimates. Then, cross-station target matching is performed based on the local parameter estimates. Three-dimensional positioning is achieved through weighted Gauss-Newton iteration using geometric complementarity relationships, and a system of linear velocity equations is established using multi-station Doppler observations to solve for the global three-dimensional position. Finally, a geometric relationship model is constructed based on the global three-dimensional position, and a system of linear observation equations for the target velocity is established using Doppler observations to obtain the target's true three-dimensional velocity in the global coordinate system. This method achieves high-precision three-dimensional positioning and velocity recovery for low-altitude targets, effectively solving the performance limitations of traditional methods in high-dynamic scenarios. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a multi-station collaborative target perception method according to an exemplary embodiment; Figure 2 This is a flowchart illustrating a multi-station collaborative target perception method in an application scenario. Figure 3 This is a schematic diagram of the system structure in a multi-station collaborative target perception method for an application scenario; Figure 4 yes Figure 3 Comparison charts of estimation performance for azimuth, pitch, distance, and velocity in corresponding application scenarios; Figure 5 yes Figure 3 Performance comparison chart of multi-station collaborative positioning in corresponding application scenarios; Figure 6 yes Figure 3 Performance curves as the number of base stations changes in the corresponding application scenarios; Figure 7 yes Figure 3 A schematic diagram of the actual 3D velocity analysis error in the corresponding application scenario; Figure 8 This is a block diagram illustrating a multi-station cooperative target sensing device according to an exemplary embodiment; Figure 9 This is a hardware structure diagram of an electronic device according to an exemplary embodiment; Figure 10 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0018] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0019] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this disclosure means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0020] This invention provides a multi-station cooperative target perception method. By coordinating the processing of AFDM signals containing chirped subcarriers at multiple stations, it achieves high-precision 3D positioning and true 3D velocity recovery of low-altitude targets, effectively solving the performance limitations of traditional methods in high-dynamic scenarios. This multi-station cooperative target perception method is applicable to multi-station cooperative target perception devices, which can be electronic devices. The multi-station cooperative target perception method in this invention can be applied to various scenarios, such as multi-station cooperative target perception, etc.

[0021] Please see Figure 1 This invention provides a multi-station collaborative target perception method, which is applicable to electronic devices.

[0022] In the following method embodiments, for ease of description, the execution subject of each step of the method is an electronic device, but this does not constitute a specific limitation.

[0023] like Figure 1As shown, the method may include the following steps: Step 110: Construct an AFDM signal containing chirped subcarriers at the transmitting station, add a cyclic prefix, and then perform up-conversion transmission. Receive the target scattered echoes through each receiving station and perform down-conversion, deprecation, sampling, and DAFT to obtain the three-dimensional observation.

[0024] In one possible implementation, an analog radio frequency multiplexing (AFDM) signal containing chirped subcarriers is constructed at the transmitting station. A chirped cyclic prefix (CPP) is added before each AFDM symbol. The AFDM signal with the added cyclic prefix is ​​up-converted and then transmitted through an antenna.

[0025] In one possible implementation, scattered echo signals from the target are received by each receiving station. The scattered echo signals are down-converted to reduce the high-frequency signal to the baseband frequency. The chirped cyclic prefix in the scattered echo signals is removed to restore the original AFDM symbol structure. The scattered echo signals are then discretely sampled and converted into digital signals. The scattered echo signals are then transformed into the Discrete Affine Fourier Transform (DAFT) domain to construct a three-dimensional observation.

[0026] The chirped cyclic prefix (CPP) is used to suppress inter-symbol interference caused by multipath propagation; the three-dimensional observations include array row dimension, array column dimension, and symbol and transform domain sampling stack dimension, etc., which are not limited here.

[0027] Specifically, an AFDM signal containing chirped subcarriers is constructed at the transmitting station. This signal design constructs chirped characteristic subcarriers through Discrete Affine Fourier Transform (DAFT), which enhances the signal's ability to structurally represent time delay and Doppler in the Discrete Affine Fourier (DAF) domain.

[0028] Furthermore, a chirped cyclic prefix (CPP) is added before each AFDM symbol to suppress inter-symbol interference caused by multipath propagation and ensure stable signal transmission in complex propagation environments. The AFDM signal with the cyclic prefix is ​​then up-converted and transmitted into the air via an antenna to cover the target area.

[0029] Furthermore, each receiving station receives the scattered echo signal from the target and sequentially performs down-conversion processing, removal of the chirped cyclic prefix, discrete sampling, and DAFT to finally construct a three-dimensional observation including the array row dimension, the array column dimension, and the symbol and transform domain sampling stack dimension.

[0030] In the above process, the embodiments of the present invention effectively enhance the anti-interference capability of the signal by constructing an AFDM signal with chirped characteristics and adding a cyclic prefix, enabling the signal to maintain stable transmission even in highly dynamic and complex propagation environments. Simultaneously, a detailed receiving and processing procedure ensures accurate reception and conversion of the echo signal, providing a high-quality data foundation for subsequent processing.

[0031] Step 120: Perform two-dimensional spatial smoothing on the three-dimensional observations, construct multiple subarray observation blocks and fuse them into an enhanced observation matrix, use the characteristics of a uniform planar array to jointly estimate the azimuth and elevation angles based on the enhanced observation matrix, recover the time delay-Doppler parameters using the least squares method, and obtain the local parameter estimates of the target from each receiving station.

[0032] In one possible implementation, for three-dimensional observations, each receiving station extracts continuous subarrays along the row and column dimensions of the array and smooths the original observations to construct multiple subarray observation blocks, which are then fused into an enhanced observation matrix.

[0033] In one possible implementation, the signal subspace is extracted based on the enhanced observation matrix by utilizing the characteristics of a uniform planar array. The azimuth and elevation angle parameters of the target are jointly estimated using the shift invariance principle to obtain the angle estimation result. The receiving array manifold is constructed based on the angle estimation result. The time delay-Doppler structure vector corresponding to the target is recovered by the least squares method. The structure vector is then subjected to pulse compression and local continuous refinement to obtain the range parameter and Doppler parameter estimation results of the target.

[0034] Among them, smoothing is used to reduce the correlation between multi-target signals and improve the effective observation rank; angle estimation results are used to reflect the spatial orientation information of the target relative to the receiving station.

[0035] Specifically, for three-dimensional observations, each receiving station extracts continuous subarrays along the row and column dimensions of the array, smooths the original observations, constructs multiple subarray observation blocks, and merges these subarray observation blocks into an enhanced observation matrix to reduce the correlation between multi-target signals and improve the effective observation rank.

[0036] Furthermore, utilizing the characteristics of a uniform planar array, the signal subspace is extracted based on the enhanced observation matrix, and the azimuth and elevation angle parameters of the target are jointly estimated using the shift invariance principle to obtain the spatial orientation information of the target relative to the receiving station. Based on the angle estimation results, a receiving array manifold is constructed, and the time-delay-Doppler structure vector corresponding to the target is recovered using the least squares method. This structure vector is then subjected to pulse compression and local continuous refinement to obtain the estimated range and Doppler parameters of the target.

[0037] In the above process, the embodiments of the present invention effectively improve the quality of the observation matrix through two-dimensional spatial smoothing, providing more reliable data support for subsequent parameter estimation. Simultaneously, by utilizing the characteristics of a uniform planar array for azimuth and elevation angle estimation, and by recovering time delay and Doppler parameters using the least squares method, high-precision parameter estimation is achieved, laying a solid foundation for subsequent three-dimensional positioning and velocity recovery.

[0038] Step 130: Cross-station target matching is performed based on local parameter estimation. The target is three-dimensionally located by using geometric complementarity through weighted Gauss-Newton iteration. A system of linear velocity equations is established by combining multi-station Doppler observations, and the global three-dimensional position of the target is obtained by solving the equations together.

[0039] In one possible implementation, cross-station target matching is performed based on the local parameter estimation results obtained from each receiving station. The geometric complementarity between multiple receiving stations is used to perform joint localization processing on the matched targets through a weighted Gauss-Newton iterative method. Based on the observation data and geometric layout information of each receiving station, a weighted nonlinear least squares problem is solved iteratively to obtain the global three-dimensional position of the target.

[0040] Cross-station target matching is used to ensure that different receiving stations observe the same target; local parameter estimation includes azimuth, elevation, range, and Doppler, etc., which are not limited here.

[0041] Specifically, based on the local parameter estimation results (including azimuth, elevation, range, and Doppler) obtained from each receiving station, cross-station target matching is performed to ensure that different receiving stations observe the same target. Utilizing the geometric complementarity between multiple receiving stations, a weighted Gauss-Newton iterative method is used to jointly locate the matched target. Based on the observation data and geometric layout information of each receiving station, a weighted nonlinear least squares problem is iteratively solved to obtain the target's global three-dimensional position.

[0042] In the above process, the embodiments of the present invention effectively solve the correlation problem between multi-station observation data through cross-station target matching processing, ensuring the accuracy and consistency of positioning results. Simultaneously, utilizing the geometric complementarity between multiple receiving stations for three-dimensional positioning significantly improves positioning accuracy and robustness, enabling stable and reliable three-dimensional positioning even in complex environments.

[0043] Step 140: Construct the geometric relationship between the transmitter, target and receiver based on the global three-dimensional position, establish a set of linear observation equations for the target velocity by combining the Doppler observations obtained independently by each receiver, and obtain the true three-dimensional velocity of the target in the global coordinate system by jointly solving the set of equations.

[0044] In one possible implementation, a geometric relationship model between the transmitter, target, and receiver is constructed based on the global three-dimensional position. A set of linear observation equations for the target velocity is established by combining the Doppler observations obtained independently by each receiver. The true three-dimensional velocity vector of the target in the global coordinate system is obtained by jointly solving the set of linear observation equations.

[0045] Among them, the Doppler observations reflect the projected velocity of the target relative to the receiving station; the linear observation equations describe the relationship between the target's true velocity and the radial velocity observed by each receiving station.

[0046] Specifically, based on the target's global three-dimensional position, a geometric relationship model between the transmitting station, the target, and the receiving station is constructed to clarify the relative positional relationships between each station. Combining the Doppler observations (reflecting the target's velocity projection relative to the receiving station) obtained independently by each receiving station, a set of linear observation equations for the target's velocity is established. This set of equations describes the relationship between the target's true velocity and the radial velocity observed by each receiving station. By jointly solving the linear observation equations, the target's true three-dimensional velocity vector in the global coordinate system is obtained, achieving comprehensive perception of the target's motion state.

[0047] In the above process, this embodiment of the invention clarifies the relative positional relationships between various stations by constructing a geometric relationship model, providing an accurate geometric basis for velocity analysis. Simultaneously, a linear velocity observation equation system is established by combining Doppler observations, and the true three-dimensional velocity vector is obtained through joint solution, achieving a comprehensive and accurate perception of the target's motion state. This step not only improves the accuracy of velocity estimation but also enhances the system's adaptability to complex motion states.

[0048] Through the above process, this embodiment of the invention achieves high-precision three-dimensional positioning and true three-dimensional velocity recovery of low-altitude targets by signal transmission and reception processing, three-dimensional observation processing and parameter estimation, cross-station target matching and three-dimensional positioning, and global three-dimensional velocity analysis. This method fully utilizes the structured advantages of AFDM signals in the DAF domain and combines geometrically complementary information between multiple receiving stations, significantly improving the accuracy and robustness of target perception, and providing effective technical support for fields such as low-altitude safety supervision and urban air traffic.

[0049] In one application scenario, the multi-station collaborative target perception method of the present invention is used for multi-station collaborative target perception.

[0050] like Figure 2 As shown, the specific steps may include: Step S1: AFDM received signal preprocessing (downconversion, CPP removal, sampling, DAFT).

[0051] Specifically, in the low-altitude target monitoring scenario, multiple receiving stations are deployed at different locations, with each station's antenna continuously receiving scattered echo signals from low-altitude targets. First, the receiving stations down-convert the received radio frequency signals from high-frequency signals to baseband signals for subsequent processing. Next, the chirped cyclic prefix (CPP) is removed from the signal to recover the original AFDM symbol structure. Then, the signal is discretely sampled and converted into a digital signal. Finally, the signal is transformed to the Discrete Affine Fourier Transform (DAFT) domain to obtain a three-dimensional observation including the array row dimension, array column dimension, and the stacked dimension of the symbol and transform domain samples.

[0052] In the above process, the embodiments of the present invention perform preprocessing operations such as down-conversion, CPP removal, sampling, and DAFT on the received signal to convert the original radio frequency signal into a three-dimensional observation in the DAF domain suitable for subsequent processing, providing basic data for subsequent parameter estimation and target localization, and ensuring the validity and accuracy of the data.

[0053] Step S2: The antenna domain is smoothed in two-dimensional space, and the azimuth and elevation angles are estimated using ESPRIT.

[0054] Specifically, for the three-dimensional observations obtained in step S1, each receiving station truncates a continuous subarray along the row and column dimensions of the array, and smooths the original observations. This two-dimensional spatial smoothing operation weakens the correlation between multi-target signals and improves the effective observation rank. Subsequently, utilizing the characteristics of a uniform planar array, the signal subspace is extracted based on the smoothed enhanced observation matrix, and the azimuth and elevation parameters of multiple low-altitude targets are jointly estimated using the shift invariance principle.

[0055] In the above process, the embodiments of the present invention effectively extract the spatial orientation information of the target from complex observation data through two-dimensional spatial smoothing and the ESPRIT algorithm, providing key angle parameters for subsequent time-delay Doppler parameter estimation and target localization, thereby improving the accuracy and stability of parameter estimation.

[0056] Step S3: Reconstruct the antenna array steering vector and recover the time delay-Doppler structure components. Complete the time delay Doppler estimation through "pulse coarse estimation + one-dimensional search fine estimation".

[0057] Specifically, based on the azimuth and elevation angles estimated in step S2, the antenna array steering vector is reconstructed. Next, the time delay-Doppler structure components are recovered, and a "pulse coarse estimation + one-dimensional search fine estimation" method is used for time delay Doppler estimation. First, a pulse coarse estimation is performed to roughly determine the range of time delay and Doppler parameters. Then, a one-dimensional search fine estimation is performed within this range to further refine the target's time delay and Doppler parameter estimates.

[0058] In the above process, the embodiments of the present invention effectively recovered the time delay-Doppler structure components of the target by reconstructing the antenna array steering vector and adopting the method of "pulse coarse estimation + one-dimensional search fine estimation", accurately estimated the time delay and Doppler parameters of the target, provided important parameter support for subsequent target positioning and velocity recovery, and improved the accuracy and efficiency of parameter estimation.

[0059] Step S4: Obtain coarse target localization results based on single-base station estimation parameters, and perform cross-base station target matching through clustering. Multi-base station collaborative localization is achieved through weighted Gauss-Newton algorithm.

[0060] Specifically, each receiving station obtains a coarse localization result of the target based on its own estimated azimuth, elevation, range, and Doppler parameters. Then, cluster analysis is performed on the coarse localization results from multiple receiving stations to complete cross-base station target matching, ensuring that different receiving stations observe the same target. Next, utilizing the geometric complementarity among multiple receiving stations, a weighted Gauss-Newton iterative method is used to perform joint localization processing on the matched target. This method, based on the observation data and geometric layout information of each receiving station, obtains the global three-dimensional position of the target by iteratively solving a weighted nonlinear least squares problem.

[0061] In the above process, the embodiments of the present invention fully utilize the observation data and geometric complementary information of multiple receiving stations through operations such as single-base station coarse positioning, cross-base station target matching and multi-base station cooperative positioning, thereby improving the accuracy and robustness of target positioning and enabling accurate positioning of low-altitude targets in complex environments.

[0062] Step S5: Based on the geometric mapping relationship of multiple base stations, recover the true three-dimensional velocity of the target through least squares.

[0063] Specifically, based on the target's global three-dimensional position obtained in step S4, a geometric relationship model between the transmitting station, the target, and the receiving station is constructed. Combining the Doppler observations independently estimated by each receiving station—which reflect the target's velocity projection relative to the receiving station—a set of linear observation equations for the target's velocity is established. This set of equations describes the relationship between the target's true velocity and the radial velocity observed by each receiving station. Finally, by jointly solving this set of equations, the target's true three-dimensional velocity vector in the global coordinate system is obtained.

[0064] In the above process, the embodiments of the present invention construct a geometric relationship model and establish a set of linear velocity observation equations, and solve them using the least squares method, thereby achieving accurate recovery of the true three-dimensional velocity of the target. This provides key information for comprehensively perceiving the motion state of the target and improves the completeness and accuracy of target monitoring.

[0065] Through the above process, this invention achieves accurate perception of low-altitude targets in specific application scenarios of low-altitude target monitoring through multi-station collaboration. First, each receiving station preprocesses the received AFDM signal to obtain three-dimensional observations. Next, the azimuth and elevation angles of the target are estimated using two-dimensional spatial smoothing and the ESPRIT algorithm. Then, the antenna array steering vector is reconstructed, and time-delay Doppler estimation is performed using a specific method. Subsequently, coarse positioning results are obtained based on single-base station parameters, and cross-base station target matching is performed. Multi-base station collaborative positioning is achieved using the weighted Gauss-Newton method. Finally, based on the geometric mapping relationship of the multiple base stations, the true three-dimensional velocity of the target is recovered using the least squares method. Through this series of steps, this invention effectively solves the problems of high-precision positioning and velocity recovery in low-altitude target monitoring, providing reliable technical support for fields such as low-altitude safety supervision.

[0066] In another application scenario, such as Figure 3 The diagram illustrates the architecture of a low-altitude target sensing system with a single transmitter and multiple receivers working in AFDM (Alternating Current Distance Management) coordination. The scenario includes N receiver base stations and K low-altitude targets.

[0067] Using this system for target perception may include the following steps: Step 1. Signal Preprocessing Stage. The transmitting end constructs an AFDM transmission signal and adds a chirped cyclic prefix before each symbol before transmission. In a scenario of active and passive cooperative sensing by a single transmitting base station and multiple receiving base stations, each receiving base station receives the scattered echo from low-altitude targets. The echo signal is down-converted, de-chirped cyclic prefixed, discretely sampled, and transformed to the DAF domain using DAFT. A three-dimensional observation is constructed, including the array row dimension, array column dimension, and the symbol and transform domain sampling stack dimension.

[0068] Step 2. Two-dimensional spatial smoothing enhancement stage. For the uniform planar array receiving structure, each receiving station truncates continuous subarrays along the row and column dimensions, performs two-dimensional spatial smoothing on the original observations, constructs multiple subarray observation blocks, and further forms an enhanced observation matrix to reduce the correlation of multi-target signals, improve the effective observation rank, and enhance the robustness of subsequent parameter estimation.

[0069] Step 3. Azimuth-Elevation Parameter Estimation Stage. At each receiving station, the signal subspace is extracted from the enhanced observation matrix, and the shift-invariant property of a uniform planar array in two-dimensional space is used to jointly estimate the azimuth and elevation parameters of multiple low-altitude targets, thereby obtaining the spatial orientation information of each target relative to the current receiving station.

[0070] Step 4. Delay-Doppler Parameter Estimation Stage. Based on the angle estimation results obtained in Step 3, a receiving array manifold is constructed, and the delay-Doppler structure vector corresponding to each target is recovered by least squares. The structure vector is then subjected to pulse compression and local continuous refinement to obtain the target range parameters and Doppler parameter estimation results.

[0071] Step 5. Cross-base station target matching and 3D fusion localization stage. Local parameters such as angle, distance, and Doppler obtained from each receiving station are uploaded to the fusion center. Combined with the local coarse local localization results, consistent matching of the same target is completed between different receiving stations. Based on this, the weighted Gauss-Newton iterative method is used to perform joint localization of the same target after matching, utilizing the geometric complementarity between multiple receiving stations, to obtain the global 3D position of the target.

[0072] Step 6. Three-dimensional velocity analysis stage. Based on the target fusion position obtained in Step 5, construct the geometric relationship between the transmitter, target, and receiver, and combine the Doppler observations obtained independently by each receiver to establish a set of linear observation equations for the target velocity; the true three-dimensional velocity of the target in the global coordinate system is obtained by jointly solving the equations.

[0073] Furthermore, in step 1, the i-th AFDM symbol received by base station n can be represented in the DAF domain as: .

[0074] in, The target dispersion coefficient, The array guide vector for a uniform area array. Send the vector to the DAF field of the i-th symbol. The noise matrix in the DAF domain. Let k be the equivalent AFDM sensing channel matrix corresponding to target k. Let be the normalized delay and normalized Doppler of the k-th target relative to the n-th receiving base station.

[0075] Furthermore, in step 2, after the nth receiving station completes CPP removal and DAFT / DAF transformation, it organizes the observations into a third-order tensor according to "row index, column index, and stacking index". The observation tensor of receiving station n can be written as: in, This represents the outer product of vectors.

[0076] In a two-dimensional spatial smoothing operation, the length of the submatrix is ​​selected in the row dimension. In Levi's selection of subarray length The corresponding row and column selection matrices are respectively and Therefore, for The slice The sub-observations corresponding to spatial smoothing are: .

[0077] Further, in step 3, the sub-observations obtained in step 2 are vectorized into... And concatenate all the sub-observation vectors into a matrix as follows: .

[0078] right Perform truncated singular value decomposition to obtain the K-dimensional signal space. .

[0079] Since uniform arrays satisfy shift invariance in both row and column antenna dimensions, the ESPRIT algorithm can be used to estimate the phase increments in both dimensions. and Thus, the angle estimate is obtained: .

[0080] Furthermore, in step 4, the results estimated in step 3 are combined with... and Reconstructing the guide vector and Reconstructing the array manifold The distance-Doppler structure matrix was recovered using the least squares method. .

[0081] in, Indicates a false reversal. The distance-Doppler component vector representing the target k is... Reorganized into a matrix of size "DAF domain sample number AFDM symbol number", pulse compression is performed on it to obtain a coarse estimate of integer time delay and integer Doppler. From this, a one-dimensional search for a precise estimate of the time delay dimension and Doppler dimension is completed, and then, based on physical relationships, the following is obtained: and .

[0082] Furthermore, in step 5, the estimated values ​​of each base station parameter from step 4 are combined. This process obtains the coarse localization results of the target from each base station, and then performs clustering to complete cross-base station target matching. Furthermore, it is assumed that the localization of target k is... Define the corresponding nonlinear observation mapping at the receiving base station n. Therefore, the target location fusion is achieved by solving the following weighted nonlinear least squares problem:

[0083] in, This is the location fusion weight matrix corresponding to the nth receiving base station. A weighted Gaussian weighting method is used. The Newton method iteratively solves the aforementioned weighted nonlinear least squares problem.

[0084] Furthermore, in step 6, for the nth receiving base station, the true three-dimensional velocity of the target is obtained by combining the multi-station fusion positioning results obtained in step 5, through the base station line-of-sight unit vectors of the transmission link and the receiving link, and listing the relationship between the target's true velocity and the receiving base station's observed velocity.

[0085] One of the AFDM symbols contains The transmit symbol vector in the DAF domain is denoted as: (The vector consists of several chirped subcarriers.) The time-domain baseband symbol is obtained after inverse discrete affine Fourier transform (IDAFT): in, express Point Discrete Fourier Transform (FFT) matrix, and The diagonal chirp matrix associated with the AFDM parameters is defined as follows: , This indicates that the elements are expanded into a diagonal matrix. To suppress inter-symbol interference caused by multipath propagation, a length of [length missing] is added before each AFDM symbol. The chirped cyclic prefix (CPP) is transmitted after up-conversion.

[0086] Multiple receiving stations respectively adopt A uniform planar array receives the target's scattered echo. n Each receiving station performs CPP removal, discrete sampling, and DAFT transform on the echo to obtain a third-order observation tensor organized along the "row element - column element - stacked index" path. The stacked index contains information about the symbol dimension and the DAF field sampling dimension.

[0087] Assume the time delay and Doppler effect of the k-th target relative to the n-th receiving base station are respectively Define the normalized time delay and normalized Doppler as follows: , .

[0088] in, For the sampling time, the received number of samples is... An AFDM symbol can be represented in the DAF domain as: .

[0089] in, For the first The DAF field transmission vector of a symbol, The noise matrix in the DAF domain. The equivalent AFDM sensing channel matrix corresponding to target k can be expressed as:

[0090] in, and Describe the phase rotation and cyclic shift effects corresponding to normalized Doppler and normalized time delay, respectively. The CPP-related compensation matrix is ​​used when the CPP length is greater than the maximum multipath delay in the scene. .

[0091] Furthermore, after completing CPP removal and DAFT / DAF transformation at the nth receiving station, the observations are organized into a third-order tensor according to "row index, column index, and stack index": , .

[0092] Where L is the Each AFDM symbol and each symbol The effective observation length after stacking DAF domain observations. The steering vector of the receiving base station n for the k-th target in the row and column antenna dimensions can be expressed as: .

[0093] in, and These represent the phase increments in the row and column antenna dimensions, respectively, and d represents the antenna spacing. For carrier wavelength, , Let represent the pitch and azimuth angles of target k. Therefore, the observation tensor can be written as:

[0094] in, Represents the outer product of vectors. Let be the target dispersion coefficient. .

[0095] definition The distance Doppler component for the target. Along Expanding on the third dimension: ,in This represents the Khatri-Rao product.

[0096] Furthermore, in step 2, the two-dimensional spatial smoothing selects the subarray length in the row dimension. In Levi's selection of subarray length The corresponding number of subarrays is .

[0097] definition:

[0098] Select elements from the p-th to the p-th row in the original row matrix. q-th to q-th consecutive array elements are selected from the original Levy array elements. A continuous array of elements. Therefore, for The slice The sub-observations corresponding to spatial smoothing are: .

[0099] Furthermore, in step 3, the sub-observations are vectorized into... And concatenate all the sub-observation vectors into a matrix as follows: .

[0100] right Perform truncated singular value decomposition to obtain a K-dimensional signal space: .

[0101] Since uniform arrays satisfy shift invariance in both row and column antenna dimensions, the ESPRIT algorithm can be used to estimate the phase increments in both dimensions. and Thus, the angle estimate is obtained: .

[0102] Furthermore, in step 4, the estimation obtained in step 3 is combined with... and Reconstructing the guide vector and Therefore, the array manifold matrix can be reconstructed as: .

[0103] Recovering the distance-Doppler structure matrix using the least squares method: .

[0104] in, Indicates a false reversal. The distance-Doppler component vector representing the target k is calculated by dividing each column vector according to... Size reshaping into two-dimensional observation blocks: .

[0105] in, Indicates the first The DAF field structure vector corresponding to each AFDM symbol.

[0106] The time delay and Doppler components are defined as integer and fractional parts: The corresponding DAF domain structure vector under ideal noise-free conditions is Then its m-th element can be written as:

[0107] in, This represents the effects of time delay and Doppler effect. If time delay and Doppler effect are integers, then... And define it within. Modulus Operations, It is determined by the structure of the AFDM signal.

[0108] Use the sent symbol vector right Perform one-dimensional pulse compression:

[0109] in, The peak index after pulse compression is... Therefore, the normalized delay and integer part of the normalized Doppler of target k can be recovered as: .

[0110] Using integer estimates as initial values, time delay and Doppler matching components are constructed through set values. One-dimensional searches are then performed alternately in the time delay and Doppler dimensions to complete precise estimations of time delay and Doppler performance. The specific process is as follows: Figure 3 As shown.

[0111] Furthermore, in step 5, the target parameters estimated by the receiving base station n are: The locations of the receiving and transmitting base stations are known, combined with the target parameters. The estimated distance, azimuth, and elevation angles are obtained. At a single base station, the target's location can be determined using geometric relationships, and the parameter information and single-station positioning information are then uploaded to the data processing center. The data processing center uses clustering, setting clustering thresholds based on the prior error information of each base station, to complete cross-base station matching of the target—that is, matching the positioning results of the same target on different receiving base stations.

[0112] Assume the location of target k is Define the corresponding nonlinear observation mapping at the receiving base station n:

[0113] in, express Norm. Therefore, the target location fusion problem can be formulated as a weighted nonlinear least squares problem:

[0114] in, Let be the location fusion weight matrix corresponding to the nth receiving base station. Considering that the impact of angle error on spatial positioning results amplifies with increasing station distance, the following geometric weighting form can be adopted:

[0115] in , This represents the distance from target k to receiving base station n. Small constants to prevent ambiguity.

[0116] The weighted Gauss-Newton method is used to iteratively solve the above weighted nonlinear least squares problem. The algorithm flow is as follows: Figure 4 As shown.

[0117] Furthermore, in step 6, for the nth receiving base station, the multi-station fusion positioning results are combined. ,definition

[0118] Let represent the unit line-of-sight vectors of the base station for the transmit link and the receive link, respectively. Assume the true velocity of target k is... Based on the speed geometry relationship between "transmitting base station - target - receiving base station", we have the following formula:

[0119] in To represent the local estimation error of a single receiving base station, the radial velocity observation equations of all receiving base stations can be stacked and written as:

[0120] in, Number of receiving base stations Furthermore, when the radial velocity observations from different receiving base stations are not linearly correlated, the true velocity of target k can be solved using the least squares method.

[0121] In one embodiment, the location of the transmitting base station is set as The carrier frequency is set to The receiving end uses A uniform planar array is used, with the element spacing set to half a wavelength. In the AFDM signal parameters, the number of chirped subcarriers is set to... The subcarrier spacing is set to Therefore, the effective symbol duration is The sampling interval is Each perception frame contains AFDM symbols, AFDM parameters are taken ,satisfy The signal-to-noise ratio range is set to... 200 independent Monte Carlo experiments were conducted at each signal-to-noise ratio point. Three main receiving base stations were set up. The coordinates and velocities of the three sensing targets are as follows: , , The simulation considers free-space path loss, i.e., electromagnetic wave power attenuation is inversely proportional to the square of the path length. To uniformly evaluate the sensing performance of different methods, the root mean square error (RMSE) is used as the main performance indicator.

[0122] like Figure 4 As shown, Figure 4 (a) Figure 4 (b) Figure 4 (c) and Figure 4 (d) The performance comparison results of the proposed method and the comparative method in terms of azimuth, elevation, range, and radial velocity estimation are presented respectively in a single-receiver base station scenario. Figure 4 (a) and Figure 4 (b) It can be seen that as the signal-to-noise ratio (SNR) increases, the angle estimation error of each method generally decreases. However, methods such as MUSIC and OMP, which are based on grid search or spectral peak search, gradually saturate in the medium-to-high SNR region, while the method of this invention continues to decrease across the entire SNR range. Figure 4 (c) and Figure 4 (d) It can be seen that, in terms of distance estimation and radial velocity estimation, the method of this invention outperforms the subspace search method and the two-dimensional grid OMP method across the entire signal-to-noise ratio range, and further approaches the theoretical performance lower bound in the medium-to-high signal-to-noise ratio region. This is because this invention does not rely solely on a fixed discrete grid for coarse-grained search, but rather introduces a local continuous refinement process after the coarse estimation, thereby effectively reducing the error caused by the discrete grid. In summary... Figure 4 The results show that the present invention has better estimation performance for four types of parameters—azimuth, elevation, distance, and radial velocity—in a single receiving base station scenario, and its advantages are more obvious in the medium-to-high signal-to-noise ratio region.

[0123] Figure 5Comparative results of target 3D localization performance in a multi-receiver base station collaborative scenario are presented. It can be seen that as the signal-to-noise ratio (SNR) increases, the target position estimation error continuously decreases, indicating that the improved accuracy of angle and distance parameter estimation by a single receiver base station can be further transferred to the multi-station fusion stage, thereby improving the global target position recovery result. Compared with the coarse localization result of a single receiver base station, the weighted Gauss-Newton position fusion method adopted in this invention can significantly reduce the target position error. This method comprehensively utilizes the geometrically complementary information of multiple receiver base stations in their spatial distribution, unifying distance and direction observations from different perspectives into the same nonlinear optimization framework. Iterative solutions achieve joint constraints on the target's 3D position, thereby effectively reducing the impact of single-station local errors on the final fused localization result.

[0124] Furthermore, by Figure 6 It can be seen that when the number of receiving stations increases from a single station to two stations, the cooperative positioning performance is significantly improved, indicating that the spatial diversity gain and redundant observation information brought by multi-station observations can effectively improve global positioning accuracy. When the number of receiving stations exceeds three, the positioning performance continues to improve, but the rate of improvement tends to plateau. This suggests that under a specific geometric layout, the cooperative gain of multiple stations does not increase linearly, but is closely related to the independence of observation constraints and the degree of geometric dilution. Therefore, in engineering implementation, the number and deployment of receiving stations can be rationally selected by comprehensively considering system complexity, deployment costs, and fusion benefits, while meeting positioning accuracy requirements.

[0125] Figure 7 The results of the analytical error of the target's true 3D velocity as a function of the signal-to-noise ratio are presented in a multi-receiver station collaborative scenario. It can be seen that after obtaining the fused position, combining the radial velocity observations from multiple receiving base stations can further recover the target's true 3D velocity in the global coordinate system. Unlike a single receiving base station, which can only obtain the radial velocity projection of the target relative to the transceiver link, multi-receiver station collaborative observation can provide velocity constraints in multiple different directions, thereby achieving a joint solution for the target's true 3D velocity.

[0126] Through the above process, simulation results show that the proposed method for low-altitude target 3D localization and 3D velocity multi-base station cooperative sensing based on AFDM signals can achieve better azimuth, elevation, range, and radial velocity estimation performance than MUSIC, OMP, and subspace search methods in a single-receiver base station scenario, and is closer to CRLB. In a multi-receiver base station cooperative scenario, it can significantly improve the target 3D localization accuracy by utilizing the geometric complementary information of multiple base stations, and further recover the target's true 3D velocity, verifying the effectiveness and superiority of the method in low-altitude high-dynamic target sensing scenarios.

[0127] The following are embodiments of the apparatus of the present invention, which can be used to execute the multi-station cooperative target perception method involved in the present invention. For details not disclosed in the embodiments of the apparatus of the present invention, please refer to the method embodiments of the multi-station cooperative target perception method involved in the present invention.

[0128] Please see Figure 8 This invention provides a multi-station collaborative target sensing device 800.

[0129] The multi-station cooperative target sensing device 800 includes, but is not limited to: an AFDM signal transmission and processing module 810, a three-dimensional observation and parameter prediction module 830, a multi-station cooperative three-dimensional positioning module 850, and a global velocity analytical solution module 870.

[0130] Among them, the AFDM signal transmission and processing module 810 is used to construct an AFDM signal containing chirped subcarriers at the transmitting station, add a cyclic prefix, and then perform up-conversion transmission. The target scattered echo is received by each receiving station and down-converted, deprecated, sampled, and DAFT is performed to obtain three-dimensional observations.

[0131] The 3D observation and parameter prediction module 830 is used to perform 2D spatial smoothing on the 3D observations, construct multiple subarray observation blocks and fuse them into an enhanced observation matrix, use the characteristics of a uniform planar array to jointly estimate the azimuth and elevation angles based on the enhanced observation matrix, recover the time delay-Doppler parameters through the least squares method, and obtain the local parameter estimates of the target from each receiving station.

[0132] The multi-station collaborative 3D positioning module 850 is used to perform cross-station target matching based on local parameter estimation, use geometric complementarity to perform 3D positioning of the target through weighted Gauss-Newton iteration, and combine multi-station Doppler observations to establish a system of linear velocity equations, and jointly solve to obtain the global 3D position of the target.

[0133] The global velocity analytical solution module 870 is used to construct the geometric relationship between the transmitter, target and receiver based on the global three-dimensional position, establish a set of linear observation equations for the target velocity by combining the Doppler observations obtained independently by each receiver, and obtain the true three-dimensional velocity of the target in the global coordinate system by jointly solving the set of equations.

[0134] It should be noted that the multi-station collaborative target perception provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed. That is, the internal structure of the multi-station collaborative target perception device will be divided into different functional modules to complete all or part of the functions described above.

[0135] Furthermore, the embodiments of the multi-station collaborative target perception device and the multi-station collaborative target perception method provided in the above embodiments belong to the same concept, and the specific way in which each module performs operations has been described in detail in the method embodiments, and will not be repeated here.

[0136] Figure 9 A schematic diagram of the structure of an electronic device according to an exemplary embodiment is shown.

[0137] It should be noted that this electronic device is merely an example adapted to the present invention and should not be construed as providing any limitation on the scope of use of the present invention. Furthermore, this electronic device should not be interpreted as requiring or depending on having... Figure 9 One or more components of the exemplary electronic device 2000 shown.

[0138] The hardware structure of electronic devices 2000 can vary significantly due to differences in configuration or performance, such as... Figure 9 As shown, the electronic device 2000 includes: a power supply 210, an interface 230, at least one memory 250, and at least one central processing unit (CPU) 270.

[0139] Specifically, power supply 210 is used to provide operating voltage for various hardware devices on electronic device 2000.

[0140] Interface 230 includes at least one wired or wireless network interface 231 for interacting with external devices. Of course, in other examples adapted to this invention, interface 230 may further include at least one serial-to-parallel conversion interface 233, at least one input / output interface 235, and at least one USB interface 237, etc. Figure 9 As shown, this does not constitute a specific limitation.

[0141] The memory 250 serves as a carrier for resource storage and can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored on it include the operating system 251, application programs 253, and data 255, etc., and the storage method can be temporary storage or permanent storage.

[0142] The operating system 251 is used to manage and control the various hardware devices and application programs 253 on the electronic device 2000, so as to enable the central processing unit 270 to perform calculations and processing on the massive data 255 in the memory 250. It can be Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0143] Application 253 is a computer-readable instruction based on operating system 251 that performs at least one specific task, and may include at least one module ( Figure 9 (Not shown), each module may contain computer-readable instructions for electronic device 2000. For example, a multi-station cooperative target sensing device can be considered as application program 253 deployed on electronic device 2000.

[0144] Data 255 may be signal information, etc., and is stored in memory 250.

[0145] The central processing unit 270 may include one or more processors and is configured to communicate with the memory 250 via at least one communication bus to read computer-readable instructions stored in the memory 250, thereby enabling the computation and processing of massive amounts of data 255 in the memory 250. For example, a multi-station cooperative target perception method may be implemented by the central processing unit 270 reading a series of computer-readable instructions stored in the memory 250.

[0146] Furthermore, the present invention can also be implemented through hardware circuits or a combination of hardware circuits and software. Therefore, the implementation of the present invention is not limited to any specific hardware circuit, software, or combination thereof.

[0147] Please see Figure 10 This invention provides an electronic device 4000, which may include: a desktop computer, a laptop computer, a server, etc., with sensor recognition capabilities.

[0148] exist Figure 10 In this context, the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.

[0149] The data interaction between the processor 4001 and the memory 4003 can be achieved through at least one communication bus 4002. This communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0150] Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0151] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0152] The memory 4003 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program instructions or code in the form of instructions or data structures and accessible by the electronic device 4000, but not limited thereto.

[0153] The memory 4003 stores computer-readable instructions, and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002.

[0154] The computer-readable instructions are executed by one or more processors 4001 to implement the multi-station cooperative target perception method in the above embodiments.

[0155] Furthermore, this embodiment of the invention provides a storage medium storing computer-readable instructions, which are executed by one or more processors to implement the multi-station collaborative target perception method described above.

[0156] This invention provides a computer program product, which includes computer-readable instructions stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, thereby enabling the electronic device to implement the multi-station cooperative target perception method as described above.

[0157] Compared with related technologies, the beneficial effects of the present invention are: 1. This invention enables precise positioning of low-altitude unmanned aerial vehicles (UAVs) in urban areas. By coordinating the transmission of AFDM signals containing chirped subcarriers through multiple stations, the receiving station performs detailed processing on the echoes, accurately estimates parameters using methods such as two-dimensional spatial smoothing and the ESPRIT algorithm, and then performs cross-station target matching and multi-base station collaborative positioning, effectively overcoming interference from complex urban environments and accurately determining the location of the UAV.

[0158] 2. This invention has high-precision time delay and Doppler parameter estimation capabilities; by reconstructing the antenna array steering vector and adopting a specific "pulse coarse estimation + one-dimensional search fine estimation" strategy, the parameter range is first roughly determined and then a precise search is performed, thereby accurately recovering the time delay-Doppler structure components and accurately obtaining the target's time delay and Doppler parameters.

[0159] 3. This invention can achieve accurate cross-station target matching; by obtaining rich local parameter estimates including azimuth, elevation, distance and Doppler from each receiving station, cluster analysis is performed based on these multi-dimensional parameters to effectively ensure that different receiving stations are observing the same target, laying the foundation for subsequent cooperative positioning.

[0160] 4. This invention has a strong multi-base station collaborative working capability; by having each receiving station independently process signals and estimate parameters, and then utilizing the geometric complementary relationship between multiple stations, a weighted Gauss-Newton iterative method is used for joint positioning, making full use of multi-station data and geometric information to improve positioning accuracy and robustness, and adapt to complex monitoring scenarios.

[0161] 5. This invention can accurately recover the true three-dimensional velocity of the target; by constructing a geometric relationship model based on the global three-dimensional position of the target, and establishing a linear observation equation system by combining the Doppler observation values ​​independently estimated by each receiving station, and solving it using the least squares method, the true three-dimensional velocity vector of the target in the global coordinate system can be accurately obtained, and the target's motion state can be fully understood.

[0162] 6. This invention has strong anti-interference capabilities; by adding a chirped cyclic prefix to the transmitted signal to suppress inter-symbol interference caused by multipath propagation, and by employing a variety of algorithms and strategies in the signal processing process, it effectively copes with interference in complex urban electromagnetic environments and multi-target scenarios, ensuring the stability and reliability of the monitoring system.

[0163] 7. This invention can provide comprehensive support for urban low-altitude airspace safety management; through functions such as precise positioning, speed recovery, and real-time monitoring and early warning, it can promptly detect and handle illegal drone flight behavior, safeguard public safety, aviation order, and personal privacy, and provide strong protection for the safe and orderly operation of urban low-altitude airspace.

[0164] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0165] The above description is only a partial embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A multi-station cooperative target perception method, characterized in that, The method includes: An AFDM signal containing chirped subcarriers is constructed at the transmitting station and a cyclic prefix is ​​added before up-conversion and transmission. The target scattered echo is received at each receiving station and down-converted, deprecated, sampled, and DAFT is performed to obtain a three-dimensional observation. The three-dimensional observations are smoothed in two-dimensional space, multiple subarray observation blocks are constructed and fused into an enhanced observation matrix, and the azimuth and elevation angles are jointly estimated based on the enhanced observation matrix using the characteristics of a uniform planar array. The time delay-Doppler parameters are recovered by the least squares method to obtain the local parameter estimates of the target for each receiving station. Cross-station target matching is performed based on the local parameter estimation. The target is three-dimensionally located by using the geometric complementarity relationship through weighted Gauss-Newton iteration. A system of linear velocity equations is established by combining multi-station Doppler observations, and the global three-dimensional position of the target is obtained by solving the equations together. Based on the global three-dimensional position, the geometric relationship between the transmitting station, the target, and the receiving station is constructed. A set of linear observation equations for the target velocity is established by combining the Doppler observations obtained independently by each receiving station. The true three-dimensional velocity of the target in the global coordinate system is obtained by jointly solving the set of equations.

2. The multi-lab cooperative target perception method of claim 1, wherein, The process of constructing an AFDM signal containing chirped subcarriers at the transmitting station, adding a cyclic prefix, and then up-converting and transmitting it includes: A simulated radio frequency division multiplexing (AFDM) signal containing chirped subcarriers is constructed at the transmitting station. A chirped cyclic prefix (CPP) is added before each AFDM symbol. The AFDM signal with the added cyclic prefix is ​​up-converted and transmitted through an antenna. The chirped cyclic prefix (CPP) is used to suppress inter-symbol interference caused by multipath propagation.

3. The multi-lab coordinated target perception method of claim 1, wherein, The process of receiving the target's scattered echoes at each receiving station and performing down-conversion, deprecation, sampling, and DAFT to obtain three-dimensional observations includes: The scattered echo signals from the target are received by each receiving station, and the scattered echo signals are down-converted to reduce the high-frequency signal to the baseband frequency. The chirped cyclic prefix in the scattered echo signals is removed to restore the original AFDM symbol structure. The scattered echo signal is discretely sampled and converted into a digital signal. The scattered echo signal is then transformed into the Discrete Affine Fourier Transform (DAFT) domain to construct a three-dimensional observation. The three-dimensional observation includes the array row dimension, the array column dimension, and the symbol and transform domain sampling stack dimension.

4. The multi-lab coordinated target awareness method of claim 1, wherein, The process of performing two-dimensional spatial smoothing on the three-dimensional observations, constructing multiple subarray observation blocks, and fusing them into an enhanced observation matrix includes: For the three-dimensional observations, each receiving station extracts continuous subarrays along the row and column dimensions of the array and smooths the original observations to construct multiple subarray observation blocks. The subarray observation blocks are then fused into an enhanced observation matrix. The smoothing process is used to reduce the correlation between multi-target signals and improve the effective observation rank.

5. The multi-lab coordinated target awareness method of claim 1, wherein, The method of jointly estimating the azimuth and elevation angles based on the enhanced observation matrix using the characteristics of a uniform planar array, and recovering the time delay-Doppler parameters using the least squares method, yields the local parameter estimates of the target for each receiving station, including: The signal subspace is extracted based on the enhanced observation matrix using the characteristics of a uniform planar array, and the azimuth and elevation parameters of the target are jointly estimated using the shift invariance principle to obtain the angle estimation result; the angle estimation result is used to reflect the spatial orientation information of the target relative to the receiving station. Based on the angle estimation results, a receiving array manifold is constructed, and the time delay-Doppler structure vector corresponding to the target is recovered by the least squares method. The structure vector is then subjected to pulse compression and local continuous refinement to obtain the range parameters and Doppler parameter estimation results of the target.

6. The multi-lab coordinated target awareness method of claim 1, wherein, The process of cross-station target matching based on the local parameter estimation, using geometric complementarity to perform three-dimensional target localization through weighted Gauss-Newton iteration, and establishing a system of linear velocity equations based on multi-station Doppler observations to jointly solve for the global three-dimensional position of the target includes: Cross-station target matching is performed based on the local parameter estimation results obtained by each receiving station; the cross-station target matching is used to ensure that different receiving stations observe the same target; the local parameter estimation includes azimuth, elevation, range, and Doppler. By utilizing the geometric complementarity among multiple receiving stations, a weighted Gauss-Newton iterative method is used to perform joint localization processing on the matched target. Based on the observation data and geometric layout information of each receiving station, a weighted nonlinear least squares problem is iteratively solved to obtain the global three-dimensional position of the target.

7. The multi-lab coordinated target awareness method of claim 1, wherein, The process of constructing the geometric relationship between the transmitting station, target, and receiving station based on the global three-dimensional position, establishing a linear observation equation system for the target velocity by combining the Doppler observation values ​​independently estimated by each receiving station, and obtaining the true three-dimensional velocity of the target in the global coordinate system by jointly solving the equation system includes: Based on the global three-dimensional position, a geometric relationship model between the transmitting station, the target, and the receiving station is constructed. A linear set of observation equations for the target velocity is established by combining the Doppler observation values ​​independently estimated by each receiving station. The Doppler observation values ​​reflect the target's velocity projection relative to the receiving station. The true three-dimensional velocity vector of the target in the global coordinate system is obtained by jointly solving the linear observation equations; the linear observation equations describe the relationship between the true velocity of the target and the radial velocity observed by each receiving station.

8. A multi-station cooperative target perception apparatus, characterized by, The device includes: The AFDM signal transmission and processing module is used to construct an AFDM signal containing chirped subcarriers at the transmitting station, add a cyclic prefix, and then perform up-conversion transmission. The target scattered echo is received at each receiving station and down-converted, deprecated, sampled, and DAFT is performed to obtain three-dimensional observations. The three-dimensional observation and parameter prediction module is used to perform two-dimensional spatial smoothing on the three-dimensional observations, construct multiple subarray observation blocks and fuse them into an enhanced observation matrix, use the characteristics of a uniform planar array to jointly estimate the azimuth and elevation angles based on the enhanced observation matrix, recover the time delay-Doppler parameters through the least squares method, and obtain the local parameter estimates of the target for each of the receiving stations. The multi-station collaborative 3D positioning module is used to perform cross-station target matching based on the local parameter estimation, use geometric complementarity to perform 3D positioning of the target through weighted Gauss-Newton iteration, and combine multi-station Doppler observations to establish a system of linear velocity equations, and jointly solve to obtain the global 3D position of the target. The global velocity analytical solution module is used to construct the geometric relationship between the transmitting station, the target, and the receiving station based on the global three-dimensional position, establish a set of linear observation equations for the target velocity by combining the Doppler observation values ​​independently estimated by each receiving station, and obtain the true three-dimensional velocity of the target in the global coordinate system by jointly solving the set of equations.

9. An electronic device, comprising: include: At least one processor and at least one memory, wherein, The memory stores computer-readable instructions; The computer-readable instructions are executed by one or more of the processors, causing the electronic device to implement the multi-station cooperative target perception method as described in any one of claims 1 to 7.

10. A storage medium having stored thereon computer readable instructions, characterized in that, The computer-readable instructions are executed by one or more processors to implement the multi-station cooperative target perception method as described in any one of claims 1 to 7.

Images