An airborne situational awareness system and method based on DME signals

Abstract

This invention provides an airborne situational awareness system and method based on DME signals, comprising: a signal receiving module for processing received DME signals to obtain digital baseband signals; a signal detection module for constructing an overcomplete sparse dictionary and using a hybrid OMP-IHT algorithm to detect pulse time positions in the digital baseband signals; a multipath time difference localization module for calculating the three-dimensional coordinates of each target by measuring the time difference between the direct wave and the reflected wave; a multi-target joint tracking module for establishing a dynamic state-space model containing the positions and velocities of all targets and using extended Kalman filtering to achieve continuous multi-target tracking; and a parameter adaptive optimization module for real-time evaluation of tracking quality and dynamic adjustment of parameters based on the evaluation results, feeding back to the signal detection module and the multi-target joint tracking module. This invention can detect and track surrounding non-cooperative targets without requiring additional ground facilities or airborne radar.

Description

Technical Field

[0001] This invention relates to the field of avionics technology, and in particular to an airborne situational awareness system and method based on DME signals, which is applicable to the detection and tracking of non-cooperative targets (such as UAVs and small aircraft) by civil aircraft without the need for additional ground facilities or airborne radar. Background Technology

[0002] Currently, aircraft primarily rely on airborne radar systems or data links to receive ground situational information for detecting non-cooperative targets in their surroundings. However, radar systems are expensive, resulting in most civilian aircraft not being equipped with them, creating surveillance blind spots.

[0003] The DME system is an aviation radio navigation system standardized by the International Civil Aviation Organization (ICAO), with over 2,000 ground stations worldwide, covering more than 95% of global airspace. Traditional DME receivers use analog processing, only detecting the peak value of the direct wave for ranging, treating multipath signals as noise, and failing to fully utilize the signal's potential.

[0004] In the prior art, patent CN111948619B proposes a target detection method based on external radiation sources. However, this solution is geared towards ground application scenarios and does not consider the Doppler frequency shift (typical value ≥50Hz) and three-dimensional spatial geometric dynamic changes caused by the movement of the aircraft platform, resulting in a decrease in target positioning accuracy of more than 60% in the aviation environment. Summary of the Invention

[0005] This application provides an airborne situational awareness system and method based on DME signals to solve the problems in the background art.

[0006] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.

[0007] According to a first aspect of the embodiments of this application, an airborne situational awareness system based on DME signals is provided, comprising: The signal receiving module is used to process the received DME signal to obtain a digital baseband signal; The signal detection module is used to construct an overcomplete sparse dictionary suitable for DME signals; and combined with the overcomplete sparse dictionary, a hybrid OMP-IHT algorithm is used to detect the pulse time position of direct waves and multipath reflection signals in digital baseband signals. The multipath time difference positioning module is used to calculate the three-dimensional coordinates of each target by measuring the time difference between the direct wave and the reflected wave, combined with the known location of the machine and the ground station. The multi-target joint tracking module is used to establish a dynamic state-space model that includes the position and velocity of all targets, and uses extended Kalman filtering to achieve continuous tracking of multiple targets; The parameter adaptive optimization module is used to evaluate the tracking quality in real time and dynamically adjust the parameters based on the evaluation results, feeding back to the signal detection module and the multi-target joint tracking module.

[0008] According to one embodiment of this application, the signal detection module uses a hybrid OMP-IHT algorithm to detect the pulse time positions of direct waves and multipath reflection signals in the digital baseband signal. The specific process is as follows: Overcomplete dictionary building units are used to construct dictionary atoms based on the DME Gaussian pulse pair model. The time offset of the dictionary is stepped by 0.5μs to form an overcomplete sparse dictionary. The OMP-IHT units are mixed, and the orthogonal matching pursuit algorithm is executed to obtain the initial sparse coefficients. Then, the hard thresholding algorithm is executed to iteratively optimize and obtain the converged sparse coefficients. Finally, the fused sparse coefficients are calculated based on the initial sparse coefficients and the converged sparse coefficients. The pulse position extraction unit is used to set a threshold for the fusion sparse coefficient and take the time position corresponding to the fusion sparse coefficient that is greater than the threshold as the pulse candidate time. The pulse pair verification unit detects whether the time interval between each pair of adjacent pulses in the pulse candidate time meets the standard interval requirement. If it does, it is retained as a valid pulse pair, and a list of the start times of all valid pulse pairs is obtained, which is the pulse time position of the direct wave and multipath reflection signal.

[0009] According to one embodiment of this application, the multipath time difference positioning module includes a positioning solution unit; the positioning solution unit constructs a positioning equation and uses a nonlinear least squares method to jointly solve for the target position based on multiple measurement values ​​obtained by the machine at different locations.

[0010] According to one embodiment of this application, the multi-target joint tracking module includes: The target state vector establishment unit is used to establish a state vector for each target, and the overall state vector is a combination of all target state vectors. The tracking unit is used to perform continuous multi-target tracking based on the extended Kalman filter algorithm, wherein the state transition matrix is... ,in T is the sampling period; The tracking quality detection unit is used to calculate statistics that characterize tracking quality. When the statistics exceed the pre-screening threshold, the tracking quality is considered to have deteriorated, triggering the resynchronization mechanism.

[0011] According to one embodiment of this application, the operation of the parameter adaptive optimization module includes: The system acquires the relative velocity of the monitored targets and the estimated radar cross-section in real time. It dynamically adjusts the process noise figure and sampling period by querying a preset parameter table, and feeds the adjusted parameters back to the signal detection module and the multi-target tracking unit.

[0012] According to a second aspect of the embodiments of this application, an airborne situational awareness method based on DME signals is provided, comprising: The received DME signal is down-converted and digitally sampled to obtain a digital baseband signal; An overcomplete sparse dictionary suitable for DME signals is constructed; and combined with the overcomplete sparse dictionary, a hybrid OMP-IHT algorithm is used to detect the pulse time position of direct waves and multipath reflection signals in digital baseband signals. By measuring the time difference between the direct wave and the reflected wave, and combining this with the known location of the aircraft and the ground station, the three-dimensional coordinates of each target are determined. A dynamic state-space model containing the position and velocity of all targets is established, and an extended Kalman filter is used to achieve continuous multi-target tracking. Real-time assessment of tracking quality, and dynamic adjustment of sampling and tracking parameters based on the assessment results.

[0013] According to one embodiment of this application, the step of combining an overcomplete sparse dictionary and employing a hybrid OMP-IHT algorithm to detect the pulse time positions of direct waves and multipath reflection signals in a digital baseband signal specifically includes: Acquire digital baseband signals, complete sparse dictionary, and sparsity; The orthogonal matching pursuit algorithm is executed to obtain the initial sparse coefficients; and the hard thresholding algorithm is executed iteratively to obtain the converged sparse coefficients using the initial sparse coefficients as the initial values; and the fused sparse coefficients are then calculated based on the initial sparse coefficients and the converged sparse coefficients. A threshold is set for the fusion sparsity coefficient, and the time position corresponding to the fusion sparsity coefficient that is greater than the threshold is used as the pulse candidate time. The time interval between each pair of adjacent pulses in the candidate pulse time is checked to see if it meets the standard interval requirement. If it does, it is retained as a valid pulse pair, and the start time list of all valid pulse pairs is obtained, which is the pulse time position of the direct wave and multipath reflection signal.

[0014] According to one embodiment of this application, the step of determining the three-dimensional coordinates of each target by measuring the time difference between the direct wave and the reflected wave, and combining this with the known location of the local machine and the location of the ground station, specifically includes: For any target, the three-dimensional coordinates are solved using the following procedure: The time difference between the target measured by the local machine at multiple locations constitutes a nonlinear least squares problem. The nonlinear least squares problem is solved iteratively using the Gauss-Newton method to obtain the three-dimensional coordinates of the target.

[0015] According to one embodiment of this application, the step of establishing a dynamic state-space model containing the positions and velocities of all targets and using extended Kalman filtering to achieve continuous multi-target tracking specifically includes: Establish a state vector for each target, and the overall state vector is the combination of the state vectors of all targets. Multi-target continuous tracking is performed based on the extended Kalman filter algorithm, and the overall state vector is updated; where the state transition matrix is... ,in T is the sampling period; Calculate a statistic that characterizes the tracking quality. When the statistic exceeds the pre-screening threshold, the tracking quality is considered to have deteriorated, triggering a resynchronization mechanism.

[0016] According to one embodiment of this application, the real-time evaluation of tracking quality and the dynamic adjustment of sampling and tracking parameters based on the evaluation results specifically include: The system acquires the relative velocity of monitored targets and the estimated radar cross-section in real time. It dynamically adjusts the process noise figure and sampling period by querying a preset parameter table, and applies the adjusted parameters to the signal detection and multi-target tracking process.

[0017] Compared with existing technologies, the beneficial effects of adopting the above technical solution are as follows: (1) Zero infrastructure modification: Directly utilizes 1000+ DME ground stations worldwide (ICAO Annex 10 standard), without the need for additional launch equipment, and complies with civil aviation airworthiness regulations; (2) Revolutionary cost reduction: The system cost requires only software upgrades (<5,000), which is 97.5% lower than that of airborne radar (>200,000); (3) Robustness in extreme environments: The detection rate reaches 87.5% at -10dB SNR (compared to only 42.3% for traditional methods), meeting the detection requirements in severe weather (thunderstorms); (4) Multi-target tracking capability: A single unit can support simultaneous tracking of ≥5 targets (TCAS standard only requires 2 targets), with a resynchronization time of ≤10s; (5) Adaptive closed-loop control: The detection strategy is dynamically switched according to the target RCS (radar cross-section) to reduce the false alarm rate to below 0.8%. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0019] Figure 1 This is a schematic diagram of the DME multipath reflection positioning principle and detection scenario in an embodiment of this application.

[0020] Figure 2 This is a schematic diagram illustrating the working principle of an airborne situational awareness system based on DME signals, as described in an embodiment of this application.

[0021] Figure 3 This is a flowchart of an airborne situational awareness method based on DME signals, according to an embodiment of this application.

[0022] Figure 4 This is a flowchart illustrating the signal detection process in an embodiment of this application.

[0023] Figure 5 This is a flowchart illustrating the multi-target tracking process according to an embodiment of this application.

[0024] Figure 6 This is a schematic diagram illustrating parameter adaptive optimization in an embodiment of this application. Detailed Implementation

[0025] The embodiments of this application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar modules or modules 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 this application, and should not be construed as limiting this application. Rather, the embodiments of this application include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.

[0026] Example 1 This application proposes an airborne situational awareness system based on DME signals, which aims to solve the following problems existing in the prior art: (1) Civil aircraft lack low-cost non-cooperative target detection methods; (2) Professional radar systems are expensive (more than 500,000 yuan per unit), making them difficult to popularize; (3) Traditional DME processing methods ignore the potential of multipath signals; (4) Signal detection performance deteriorates under low signal-to-noise ratio (SNR≤-10 dB).

[0027] Before introducing the airborne situational awareness system based on DME signals of this application, it is necessary to explain the positioning principle and detection scenarios in which the scheme of this application operates. Please refer to... Figure 1 The machine (A) is equipped with airborne DME equipment. The DME system operates as follows: the transmitter (TX) of this equipment sends an interrogation signal; the ground DME station (D), upon receiving the interrogation signal, responds with a response signal, such as... Figure 1 The response signal consists of a direct wave signal and a reflected wave signal. The onboard DME receiver on the machine (A) calculates the distance between the AD and the time difference between the direct wave response signal and the interrogation signal (this is the DME ranging principle). The DME receiver on the machine (A) calculates the distance difference between the reflected wave response signal and the direct wave response signal, and then calculates the target (Bi) position P through the spatiotemporal motion characteristics, and displays the traffic on the display.

[0028] Please refer to Figure 2 The airborne situational awareness system based on DME signals proposed in this application mainly includes a signal receiving module, a signal detection module, a multipath time difference localization module, a multi-target joint tracking module, and a parameter adaptive optimization module. The signal receiving module processes the received DME signal to obtain a digital baseband signal. The signal detection module constructs an overcomplete sparse dictionary suitable for DME signals and, combined with the overcomplete sparse dictionary, uses a hybrid OMP-IHT algorithm to detect the pulse time positions of the direct wave and multipath reflected signals in the digital baseband signal. The multipath time difference localization module calculates the three-dimensional coordinates of each target by measuring the time difference between the direct wave and the reflected wave, combined with the known location of the local machine and the ground station. The multi-target joint tracking module establishes a dynamic state-space model containing the positions and velocities of all targets and uses extended Kalman filtering to achieve continuous multi-target tracking. The parameter adaptive optimization module evaluates the tracking quality in real time and dynamically adjusts the parameters based on the evaluation results, feeding them back to the signal detection module and the multi-target joint tracking module.

[0029] To better understand the implementation process of this application, the following provides a detailed explanation of each module.

[0030] The core function of the signal detection module is to accurately extract the arrival times of direct and reflected waves from the digital baseband signal. This module mainly includes an overcomplete dictionary construction unit, a hybrid OMP-IHT unit, a pulse position extraction unit, and a pulse pair verification unit.

[0031] Specifically, the overcomplete dictionary building unit constructs an overcomplete sparse dictionary suitable for DME signals. DME signals consist of paired Gaussian pulses, and their mathematical expression is:

[0032] in, It is a Gaussian impulse function. Determine the pulse width (typical value makes the half-power width 10 ... ), The standard pulse interval for DME (e.g., X mode is 10 ... Y mode is ), To process the number of pulse pairs within the window, For time variables, Transmitting signals for DME is the start time of the k-th pulse pair.

[0033] Complete sparse dictionary Each column corresponds to a possible time offset. The dictionary atom is defined as:

[0034] And each atom is normalized to ensure Time offset The step size determines the time resolution. In this embodiment, the time offset step size of the dictionary is... .

[0035] Received signal It can be represented as ,in, These are sparse coefficients, meaning only a few non-zero elements correspond to the actual positions of pulse pairs. It is noise.

[0036] The hybrid OMP-IHT unit is used to execute the orthogonal matching pursuit algorithm to obtain the initial sparse coefficients; and with the initial sparse coefficients as the initial values, the hard threshold algorithm is executed to iteratively optimize and obtain the converged sparse coefficients; and then the fused sparse coefficients are calculated based on the initial sparse coefficients and the converged sparse coefficients.

[0037] The OMP (Orthogonal Matching Pursuit) algorithm is sensitive to initial values ​​and is prone to getting trapped in local optima at low signal-to-noise ratios, but it has good convergence stability. The IHT (Iterative Hard Thresholding) algorithm has a fast convergence speed, but its support set estimation accuracy is insufficient. This embodiment addresses the specific needs of DME signal detection—the need to accurately extract pulse time positions under low signal-to-noise ratio conditions while meeting the real-time requirements of airborne equipment—by combining the two algorithms. OMP is used to obtain a stable initial support set, which is then used as the initial value for IHT iterative optimization. Finally, a weighted fusion method is used to combine the advantages of both algorithms.

[0038] Specifically, the implementation steps of the hybrid OMP-IHT sparse recovery algorithm are as follows: 1. Input: Observation signal (Discretized vectors), dictionary sparsity (Number of DME pulse pairs contained within a processing window); 2. OMP Phase: Execute the Orthogonal Matching Pursuit (OMP) algorithm to obtain the initial sparse coefficients. OMP iteratively selects the atoms most relevant to the residuals and updates the coefficients; 3. IHT stage: Using the initial value, perform optimization using the Iterative Hard Thresholding (IHT) algorithm:

[0039] in For hard thresholding operators, retain the one with the largest magnitude. One portion, This is the step size factor (taken as 0.5). Iterate until convergence, and obtain... ; 4. Coefficient Fusion: Calculate the sparse coefficients for fusion.

[0040] in, The fusion weight ranges from 0.3 to 0.7. In this embodiment, the value is 0.5, but in practical applications, it can be adaptively adjusted according to the signal-to-noise ratio: when SNR > 15dB, We set it to 0.3 to emphasize the accuracy of IHT; when SNR < -5dB, We chose 0.7 to emphasize the stability of OMP. This indicates the absolute value. The fusion coefficient combines the stability of OMP and the accuracy of IHT for subsequent pulse detection. Experiments show that this fusion method improves the detection rate of DME signals by 15% and 22% compared to using OMP or IHT alone.

[0041] The pulse position extraction unit is used to set a threshold for the fused sparse coefficients and to select the time positions corresponding to fused sparse coefficients greater than the threshold as candidate pulse times. In this embodiment, a threshold is set. The fusion coefficient greater than The position is considered as the candidate time of the detected pulse. .

[0042] The pulse pair verification unit is used to detect whether the time interval between each pair of adjacent pulses in the pulse candidate time meets the standard interval requirement. If it does, it is retained as a valid pulse pair, resulting in a list of the start times of all valid pulse pairs, i.e., the pulse time positions of the direct wave and multipath reflection signals. In this embodiment, the pulse candidate times are sorted in ascending order, and adjacent pulses are paired sequentially (the first and second are a pair, the third and fourth are a pair, and so on). The time interval between each pair of pulses is calculated. .like (in If the pulse pair corresponds to the standard interval of the current DME ground station's operating mode, it is retained and considered a valid DME signal (possibly a direct wave or a reflected wave); otherwise, it is discarded. The final output is a list of the start times of all valid pulse pairs. .

[0043] Furthermore, the multipath time difference positioning module includes a positioning solution unit, which constructs a positioning equation and uses a nonlinear least squares method to jointly solve for the target position based on multiple measurement values ​​obtained by the machine at different locations.

[0044] Specifically, let's assume the location of machine A is... (Provided by GPS / INS), the location of ground station D is... (Given). The propagation time of the direct wave is:

[0045] The corresponding theoretical arrival time is the launch time plus... By comparing the detected pulse times, the pulse closest to the theoretical arrival time is marked as the direct wave, and the rest are marked as reflected waves. The target corresponding to each reflected wave The measurement equation is:

[0046] in , The time difference between the reflected wave and the direct wave. For measuring noise.

[0047] For a single measurement, the above equation describes a hyperboloid of revolution with the ground station and the machine as foci. To calculate the three-dimensional position, multiple measurement positions of the machine at different times are required (the machine's motion provides spatial diversity). Let the machine be at... Location Measure the distance difference corresponding to the same target This constitutes a nonlinear least squares problem:

[0048] in Weights (can be set according to measurement quality).

[0049] For the nonlinear least squares problem, the Gauss-Newton method is used for iterative solution: - Initial guess It can be taken as the location of the ground station plus the estimated distance; - Iterative updates: ,in For the residual vector, Let be a Jacobian matrix, and its th... Behavior:

[0050] -Convergence condition: .

[0051] Final output: target position estimate , which serves as the measurement input for the multi-target joint tracking module.

[0052] In this embodiment, the multi-target joint tracking module mainly includes a target state vector establishment unit, a tracking unit, and a tracking quality detection unit. The target state vector establishment unit is used to establish a state vector for each target; the overall state vector is a combination of the state vectors of all targets. For each target... Define the state vector:

[0053] The overall system state vector is a combination of all target states (the local state is known and may not be included):

[0054] The tracking unit is used to perform continuous multi-target tracking based on the extended Kalman filter algorithm. In this embodiment, it is assumed that each target moves independently, and a uniform motion model is adopted:

[0055] The state transition matrix is ​​a block diagonal matrix. Each block is:

[0056] The sampling period is For process noise, the process noise covariance matrix is... ,in The settings are based on the target's maneuverability characteristics (see Tables 1 and 2).

[0057] The measurement vector consists of the target position output by the positioning module:

[0058] The measurement equation can be considered linear (direct position measurement), that is... ,in To select the matrix, the positional components in the state are extracted. Therefore, a standard Kalman filter can be used for updating without linearization. The specific EKF loop process is as follows: predict:

[0059]

[0060] renew:

[0061]

[0062]

[0063] This is the Kalman gain matrix.

[0064] Innovation Sequence Detection: Defining the Innovation Vector Its covariance , To measure the noise covariance matrix.

[0065] The tracking quality detection unit calculates statistics characterizing tracking quality. When the statistics exceed a pre-screening threshold, the tracking quality is considered to have deteriorated, triggering a resynchronization mechanism (resetting the filter or increasing process noise). In this embodiment, the statistics are constructed as follows: ,like (Pick If the tracking quality is low, it is considered to be a decline.

[0066] In the parameter adaptive optimization module, the relative velocity of the monitored target and the estimated RCS are acquired in real time. Specifically, the relative velocity is calculated by comparing the target velocity estimated by the tracking unit with the local speed. The approximate value of RCS is derived by inversely calculating the reflected signal strength (through the detected pulse amplitude) and the radar equations.

[0067] Then, the process noise figure and sampling period are dynamically adjusted by querying the preset parameter table. In this embodiment, the parameter table includes a mapping table between relative speed and process noise / sampling interval, and a mapping table between RCS range and detection strategy. Details are as follows: Table 1: Mapping Table of Relative Velocity and Process Noise / Sampling Interval

[0068] Process noise covariance matrix Take as:

[0069] Table 2: Mapping Table of RCS Range and Detection Strategy

[0070] The adjusted parameters are fed back to the signal detection module in real time (affecting the number of dictionary atoms, IHT iteration count, and fusion coefficient). (values, etc.) and multi-target tracking unit (update) and sampling period This enables closed-loop adaptive optimization.

[0071] This invention enables civil aircraft to detect and track non-cooperative targets in the vicinity (such as drones and small aircraft) without the need for additional ground facilities or airborne radar.

[0072] Example 2 This embodiment provides an airborne situational awareness method based on DME signals. Please refer to [link / reference]. Figure 3 ,include: S100: The received DME signal is down-converted and digitally sampled to obtain a digital baseband signal.

[0073] In this embodiment, the received DME signal is down-converted and digitally sampled to obtain a digital baseband signal.

[0074] S200. Construct an overcomplete sparse dictionary suitable for DME signals; and combine the overcomplete sparse dictionary with a hybrid OMP-IHT algorithm to detect the pulse time position of direct waves and multipath reflection signals in digital baseband signals.

[0075] Please refer to Figure 4 The signal detection process in this embodiment includes: S201. Construct an overcomplete sparse dictionary suitable for DME signals. DME signals consist of paired Gaussian pulses, and their mathematical expression is:

[0076] in, , The standard pulse interval for DME (e.g., X mode is 10 ... Y mode is ), To handle the number of pulse pairs within a window. (Dictionary) Each column corresponds to a possible time offset. An atom is defined as:

[0077] And each atom is normalized to ensure Time offset The step size determines the time resolution. In this embodiment, the time offset step size of the dictionary is... .

[0078] Received signal It can be represented as ,in These are sparse coefficients, meaning only a few non-zero elements correspond to the actual positions of pulse pairs. It is noise.

[0079] S202. Execute the hybrid OMP-IHT algorithm, specifically including: 1. Input: Observation signal (Discretized vectors), dictionary sparsity (Number of DME pulse pairs contained within a processing window); 2. OMP Phase: Execute the Orthogonal Matching Pursuit (OMP) algorithm to obtain the initial sparse coefficients. OMP iteratively selects the atoms most relevant to the residuals and updates the coefficients; 3. IHT stage: Using the initial value, perform optimization using the Iterative Hard Thresholding (IHT) algorithm:

[0080] in For hard thresholding operators, retain the one with the largest magnitude. One portion, This is the step size factor (taken as 0.5). Iterate until convergence, and obtain... ; 4. Coefficient Fusion: Calculate the sparse coefficients for fusion.

[0081] in, The value is set to 0.5, but in practical applications it can be adaptively adjusted according to the signal-to-noise ratio: when SNR > 15dB, We set it to 0.3 to emphasize the accuracy of IHT; when SNR < -5dB, We chose 0.7 to emphasize the stability of OMP. This indicates the absolute value. The fusion coefficient combines the stability of OMP and the accuracy of IHT for subsequent pulse detection. Experiments show that this fusion method improves the detection rate of DME signals by 15% and 22% compared to using OMP or IHT alone.

[0082] S203. Set a threshold for the fusion sparsity coefficient, and use the time position corresponding to the fusion sparsity coefficient that is greater than the threshold as the pulse candidate time.

[0083] In this embodiment, a threshold is set. The fusion coefficient greater than The position is considered as the candidate time of the detected pulse. .

[0084] S204. Detect whether the time interval between each pair of adjacent pulses in the candidate pulse time meets the standard interval requirement. If it does, retain it as a valid pulse pair and obtain a list of the start times of all valid pulse pairs, that is, the pulse time positions of the direct wave and multipath reflection signal.

[0085] In this embodiment, the candidate pulse times are sorted in ascending order, and adjacent pulses are paired sequentially (the first and second are paired, the third and fourth are paired, and so on). The time interval for each pair of pulses is calculated. .like (in If the pulse pair corresponds to the standard interval of the current DME ground station's operating mode, it is retained and considered a valid DME signal (possibly a direct wave or a reflected wave); otherwise, it is discarded. The final output is a list of the start times of all valid pulse pairs. .

[0086] S300. By measuring the time difference between the direct wave and the reflected wave, and combining it with the known positions of the machine and the ground station, the three-dimensional coordinates of each target are solved. In this embodiment, we assume that the position of machine A is... (Provided by GPS / INS), the location of ground station D is... (Given). The propagation time of the direct wave is:

[0087] The corresponding theoretical arrival time is the launch time plus... By comparing the detected pulse times, the pulse closest to the theoretical arrival time is marked as the direct wave, and the rest are marked as reflected waves. The target corresponding to each reflected wave The measurement equation is:

[0088] in , The time difference between the reflected wave and the direct wave. For measuring noise.

[0089] For a single measurement, the above equation describes a hyperboloid of revolution with the ground station and the machine as foci. To calculate the three-dimensional position, multiple measurement positions of the machine at different times are required (the machine's motion provides spatial diversity). Let the machine be at... Location Measure the distance difference corresponding to the same target This constitutes a nonlinear least squares problem:

[0090] in Weights (can be set according to measurement quality).

[0091] For the nonlinear least squares problem, the Gauss-Newton method is used for iterative solution: - Initial guess It can be taken as the location of the ground station plus the estimated distance; - Iterative updates: ,in For the residual vector, Let be a Jacobian matrix, and its th... Behavior:

[0092] -Convergence condition: .

[0093] When the convergence condition is met, the final target location estimate is output. .

[0094] S400: Establish a dynamic state-space model that includes the position and velocity of all targets, and use extended Kalman filtering to achieve continuous multi-target tracking.

[0095] Please refer to Figure 5 The steps for implementing continuous multi-target tracking are as follows: S401. Establish a state vector for each target; the overall state vector is the combination of all target state vectors. Specifically, for each target... Define the state vector:

[0096] The overall system state vector is a combination of all target states (the local state is known and may not be included):

[0097] S402, Perform continuous multi-target tracking based on the extended Kalman filter algorithm.

[0098] In this embodiment, it is assumed that each target moves independently, and a uniform motion model is adopted:

[0099] The state transition matrix is ​​a block diagonal matrix. Each block is:

[0100] The sampling period is For process noise, the covariance matrix ,in The settings are based on the target's maneuverability characteristics (see Tables 1 and 2).

[0101] The measurement vector consists of the target position output by the positioning module:

[0102] The measurement equation can be considered linear (direct position measurement), that is... ,in To select the matrix, the positional components in the state are extracted. Therefore, a standard Kalman filter can be used for updating without linearization. The specific EKF loop process is as follows: predict:

[0103]

[0104] renew:

[0105]

[0106]

[0107] Innovation Sequence Detection: Defining the Innovation Vector Its covariance .

[0108] S403. Calculate the statistics characterizing the tracking quality. When the statistics exceed the pre-screening threshold, the tracking quality is considered to have deteriorated, triggering a resynchronization mechanism (resetting the filter or increasing process noise). In this embodiment, the statistics are constructed as follows: ,like (Pick If the tracking quality is low, it is considered to be a decline.

[0109] The S500 provides real-time assessment of tracking quality and dynamically adjusts sampling and tracking parameters based on the assessment results.

[0110] Please refer to Figure 6 In this embodiment, the parameter adaptive optimization process is as follows: S501, Real-time monitoring of target characteristics In this embodiment, the target characteristics include relative speed and RCS. Specifically, the relative speed is calculated by comparing the target speed estimated by the tracking unit with the local speed. The approximate value of RCS is derived by inversely calculating the reflected signal strength (through the detected pulse amplitude) and the radar equations.

[0111] S502. By querying the preset parameter table, the process noise figure and sampling period are dynamically adjusted, and the adjusted parameters are applied to the signal detection and multi-target tracking process to achieve adaptive parameter optimization.

[0112] For those skilled in the art, the specific meanings of the above terms in this invention can be understood according to the specific circumstances; the accompanying drawings in the embodiments are used to clearly and completely describe the technical solutions in the embodiments of this invention. Obviously, the described embodiments are some embodiments of this invention, but not all embodiments. Generally, the components of the embodiments of this invention described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0113] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. An airborne situational awareness system based on DME signals, characterized in that, include: The signal receiving module is used to process the received DME signal to obtain a digital baseband signal; The signal detection module is used to construct an overcomplete sparse dictionary suitable for DME signals; In conjunction with an overcomplete sparse dictionary, a hybrid OMP-IHT algorithm is used to detect the pulse timing positions of direct waves and multipath reflection signals in digital baseband signals; The multipath time difference positioning module is used to calculate the three-dimensional coordinates of each target by measuring the time difference between the direct wave and the reflected wave, combined with the known location of the machine and the ground station. The multi-target joint tracking module is used to establish a dynamic state-space model that includes the position and velocity of all targets, and uses extended Kalman filtering to achieve continuous tracking of multiple targets; The parameter adaptive optimization module is used to evaluate the tracking quality in real time and dynamically adjust the parameters based on the evaluation results, feeding back to the signal detection module and the multi-target joint tracking module.

2. The airborne situational awareness system based on DME signals according to claim 1, characterized in that, In the signal detection module, a hybrid OMP-IHT algorithm is used to detect the pulse time positions of direct waves and multipath reflection signals in the digital baseband signal. The specific process is as follows: Overcomplete dictionary building units are used to construct dictionary atoms based on the DME Gaussian pulse pair model. The time offset of the dictionary is stepped by 0.5μs to form an overcomplete sparse dictionary. The OMP-IHT unit is used to perform the orthogonal matching pursuit algorithm to obtain the initial sparse coefficients; and the hard thresholding algorithm is used to iteratively optimize and obtain the converged sparse coefficients using the initial sparse coefficients as the initial values. Then, the fused sparsity coefficients are calculated based on the initial sparsity coefficients and the converged sparsity coefficients; The pulse position extraction unit is used to set a threshold for the fusion sparse coefficient and take the time position corresponding to the fusion sparse coefficient that is greater than the threshold as the pulse candidate time. The pulse pair verification unit detects whether the time interval between each pair of adjacent pulses in the pulse candidate time meets the standard interval requirement. If it does, it is retained as a valid pulse pair, and a list of the start times of all valid pulse pairs is obtained, which is the pulse time position of the direct wave and multipath reflection signal.

3. The airborne situational awareness system based on DME signals according to claim 1, characterized in that, The multipath time difference positioning module includes a positioning solution unit; the positioning solution unit constructs a positioning equation and uses a nonlinear least squares method to jointly solve for the target position based on multiple measurement values ​​obtained by the machine at different locations.

4. The airborne situational awareness system based on DME signals according to claim 1, characterized in that, The multi-target joint tracking module includes: The target state vector establishment unit is used to establish a state vector for each target, and the overall state vector is a combination of all target state vectors. The tracking unit is used to perform continuous multi-target tracking based on the extended Kalman filter algorithm, wherein the state transition matrix is... ,in T is the sampling period; The tracking quality detection unit is used to calculate statistics that characterize tracking quality. When the statistics exceed the pre-screening threshold, the tracking quality is considered to have deteriorated, triggering the resynchronization mechanism.

5. The airborne situational awareness system based on DME signals according to claim 1, characterized in that, The working process of the parameter adaptive optimization module includes: The system acquires the relative velocity of the monitored targets and the estimated radar cross-section in real time. It dynamically adjusts the process noise figure and sampling period by querying a preset parameter table, and feeds the adjusted parameters back to the signal detection module and the multi-target tracking unit.

6. An airborne situational awareness method based on DME signals, characterized in that, include: The received DME signal is down-converted and digitally sampled to obtain a digital baseband signal; Construct an overcomplete sparse dictionary suitable for DME signals; In conjunction with an overcomplete sparse dictionary, a hybrid OMP-IHT algorithm is used to detect the pulse timing positions of direct waves and multipath reflection signals in digital baseband signals; By measuring the time difference between the direct wave and the reflected wave, and combining this with the known location of the aircraft and the ground station, the three-dimensional coordinates of each target are determined. A dynamic state-space model containing the position and velocity of all targets is established, and an extended Kalman filter is used to achieve continuous multi-target tracking. Real-time assessment of tracking quality, and dynamic adjustment of sampling and tracking parameters based on the assessment results.

7. The airborne situational awareness method based on DME signals according to claim 6, characterized in that, The method of combining an overcomplete sparse dictionary and employing a hybrid OMP-IHT algorithm to detect the pulse time positions of direct waves and multipath reflection signals in digital baseband signals specifically includes: Acquire digital baseband signals, complete sparse dictionary, and sparsity; The orthogonal matching pursuit algorithm is executed to obtain the initial sparse coefficients; and the hard thresholding algorithm is executed iteratively to obtain the converged sparse coefficients using the initial sparse coefficients as the initial values; and the fused sparse coefficients are then calculated based on the initial sparse coefficients and the converged sparse coefficients. A threshold is set for the fusion sparsity coefficient, and the time position corresponding to the fusion sparsity coefficient that is greater than the threshold is used as the pulse candidate time. The time interval between each pair of adjacent pulses in the candidate pulse time is checked to see if it meets the standard interval requirement. If it does, it is retained as a valid pulse pair, and the start time list of all valid pulse pairs is obtained, which is the pulse time position of the direct wave and multipath reflection signal.

8. The airborne situational awareness method based on DME signals according to claim 6, characterized in that, The method involves measuring the time difference between the direct wave and the reflected wave, and combining this with the known positions of the local machine and the ground station to determine the three-dimensional coordinates of each target. Specifically, this includes: For any target, the three-dimensional coordinates are solved using the following procedure: The time difference between the target measured by the local machine at multiple locations constitutes a nonlinear least squares problem. The nonlinear least squares problem is solved iteratively using the Gauss-Newton method to obtain the three-dimensional coordinates of the target.

9. The airborne situational awareness method based on DME signals according to claim 6, characterized in that, The establishment of a dynamic state-space model containing the positions and velocities of all targets, and the use of extended Kalman filtering to achieve continuous multi-target tracking, specifically includes: Establish a state vector for each target, and the overall state vector is the combination of the state vectors of all targets. Multi-target continuous tracking is performed based on the extended Kalman filter algorithm, and the overall state vector is updated; where the state transition matrix is... ,in T is the sampling period; Calculate a statistic that characterizes the tracking quality. When the statistic exceeds the pre-screening threshold, the tracking quality is considered to have deteriorated, triggering a resynchronization mechanism.

10. The airborne situational awareness method based on DME signals according to claim 6, characterized in that, The real-time evaluation of tracking quality and the dynamic adjustment of sampling and tracking parameters based on the evaluation results specifically include: The system acquires the relative velocity of monitored targets and the estimated radar cross-section in real time. It dynamically adjusts the process noise figure and sampling period by querying a preset parameter table, and applies the adjusted parameters to the signal detection and multi-target tracking process.

Images