Target detection system during travel

Abstract

The invention relates to the field of target monitoring, and discloses an inter-travel target detection system. The inter-travel target detection system comprises a multi-station direction finding system, a signal detection module, a direction finding error correction module, a direction finding optimization module, a sensor optimization layout module, a target position estimation module and an intelligent alarm module. The invention further provides an inter-travel target detection method. High-precision target positioning is achieved through the steps of multi-station direction finding, signal detection, error correction, direction finding optimization, target position calculation and the like. In combination with the intelligent alarm module, the system can evaluate target threats in real time and automatically trigger alarms and countering measures according to conditions. According to the invention, through multi-station direction finding, dynamic optimization calculation, Bayesian estimation and a particle filtering algorithm, the target positioning precision, the real-time performance and the system robustness are improved; and in combination with the intelligent alarm module, the system can intelligently evaluate threats and adjust response measures, so that the efficiency and accuracy of real-time monitoring, target tracking and threat response are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of target monitoring, and in particular to a moving target detection system. Background Art

[0002] Existing target detection systems usually rely on a single measuring station or a relatively simple direction-finding method, resulting in insufficient measurement accuracy in dynamic environments. In traditional technologies, multi-station direction-finding systems have great limitations, especially when the target is moving at high speed or in complex environments. The reduction in measurement accuracy leads to large errors in the estimation of the target position. Most systems ignore the influence of the Doppler effect and signal interference on the direction-finding results, which in turn affects the accurate calculation of the target position. Therefore, the existing technology has poor application effect in complex scenarios such as multipath interference and rapid target movement, and cannot meet the requirements of high precision and high robustness.

[0003] In the prior art, although the measurement data undergoes simple error correction, the effect of error correction is often limited due to the lack of full consideration of environmental factors, optimization of measurement station layout and other factors; error correction in the prior art mostly relies on fixed environmental assumptions and cannot dynamically adapt to different measurement conditions and environmental changes; for example, weather, terrain, signal noise, etc. will affect the quality of the target signal, but many systems lack a dynamic adjustment mechanism, which can easily lead to the accumulation of measurement errors, thereby affecting the accuracy of subsequent target positioning.

[0004] The target positioning system in the prior art also lacks an intelligent early warning and response mechanism; the existing early warning system usually only determines whether to issue an alarm based on a simple target position, but does not have a sufficient evaluation and decision-making mechanism for the target's motion trajectory, threat level, etc.; especially in a complex dynamic environment, the alarms and countermeasures in the prior art often do not respond in a timely manner, and are difficult to flexibly adjust according to different threat levels; the lack of intelligent decision-making makes the existing system often appear to be inefficient and inaccurate when facing potential threats; therefore, the present invention proposes a moving target detection system to address the deficiencies of the prior art. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a moving target detection system, which solves the problems of multipath interference, rapid target movement, and dynamic correction of direction-finding errors, and significantly improves the accuracy, real-time nature, and system robustness of target positioning.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a moving target detection system, comprising: Multi-station direction-finding system, which is used to deploy multiple direction-finding stations to obtain the arrival angle information of radio signals and improve positioning accuracy through the coordinated work of each direction-finding station; The signal detection module is connected to the multi-station direction finding system and is used to detect the radio signal of the target and extract the arrival angle information; The direction finding error correction module is connected with the signal detection module and the multi-station direction finding system, analyzes the frequency spectrum characteristics of the target signal based on the short-time Fourier transform, and corrects the direction finding error caused by the Doppler effect; The direction finding optimization module is connected to the direction finding error correction module, uses Bayesian estimation to establish a direction finding error model, and dynamically optimizes the error in combination with the real-time environmental status; The sensor layout optimization module is connected to the direction finding optimization module, calculates the lower bound of the direction finding error based on the Cramer-Rao bound, adjusts the position of the direction finding station, and optimizes the layout of the direction finding station; The target position estimation module is connected to the direction finding optimization module, and integrates the corrected direction finding angle information to accurately calculate the target position through the maximum a posteriori estimation method; The intelligent warning module is connected to the target position estimation module, performs threat assessment based on the calculated target position and situational awareness, automatically generates warning information and executes corresponding countermeasures.

[0007] Preferably, the signal detection module includes: A spectrum scanning unit for scanning target radio signals; Short-time Fourier transform unit, used to analyze the spectral characteristics of the signal; The Doppler frequency shift detection unit is used to calculate the Doppler frequency shift of the target.

[0008] Preferably, the direction finding error correction module comprises: Doppler compensation unit, used to calculate the relative speed of the target; An angle correction unit, used to correct the measured arrival angle; Particle filter unit, used to dynamically correct direction finding errors.

[0009] Preferably, the direction finding optimization module includes: A measurement likelihood calculation unit, used for calculating a measurement likelihood function based on a direction finding error; An environment status analysis unit, used for classifying a multipath interference environment; A high-order Bayesian estimation unit is used to optimize the target direction finding error by combining environmental factors.

[0010] Preferably, the target position estimation module includes: A position calculation unit, used to calculate the optimal position of the target; The optimal estimation operator is used to optimize the target position based on maximum a posteriori estimation; The direction finding error compensation unit is used to correct the target position deviation based on Bayesian estimation.

[0011] Preferably, the direction finding optimization module adopts a maximum a posteriori estimation algorithm, and the posterior probability distribution of the target position satisfies: Target location Depend on The arrival angle of the direction-finding station Sure; The formula for calculating the posterior probability of the target location is: ; in: is the posterior probability of the target location; is the measurement likelihood function; is the prior probability of the target location; is the normalization constant.

[0012] Preferably, the direction finding error correction module calculates Doppler frequency shift based on short-time Fourier transform, satisfying: Assume the target signal is , then the short-time Fourier transform is defined as: ; in, For the signal at time and frequency The time-frequency representation of ; is the input signal; is the local analysis window function, making the signal The vicinity of moments is weighted; is the Fourier basis function; is the time-integrated variable.

[0013] Preferably, the Doppler shift detection unit calculates the Doppler shift of the target ,satisfy: ; in: is the Doppler shift; is the radial velocity of the target; is the speed of light; is the carrier frequency.

[0014] Preferably, the sensor optimization layout module optimizes the direction finding station layout based on the Cramer-Rao bound, satisfying: The theoretical lower bound of the direction finding error is calculated by the Cramer-Rao bound: ; in: is the direction finding error variance; is the Fisher information amount.

[0015] The present invention also provides a moving target detection method, comprising the following steps: S1. Initialize the multi-station direction finding system, deploy multiple direction finding stations, and determine the spatial coordinates of each direction finding station; S2, target radio signal detection, scanning the target radio signal through the signal detection module to extract the arrival angle information; S3, Doppler frequency shift adaptive correction, based on short-time Fourier transform analysis of the target signal's spectrum characteristics, calculate the Doppler frequency shift and correct the direction finding error; S4, direction finding optimization calculation, using Bayesian estimation to establish a direction finding error model, and dynamically optimize the error in combination with the environmental state; S5, optimize the layout of direction finding stations, calculate the lower bound of direction finding error based on Cramer-Rao bound, and adjust the location of direction finding stations to optimize direction finding accuracy; S6, target position calculation, integrating the corrected direction finding angle information, and using maximum a posteriori estimation to calculate the optimal target position; S7, intelligent warning and decision-making, based on the calculated target position, determines whether the target enters the warning area and executes corresponding warning and countermeasures.

[0016] The present invention provides a moving target detection system, which has the following beneficial effects: 1. The present invention adopts multi-station direction finding and dynamic optimization calculation technology, combined with Bayesian estimation and Doppler frequency shift correction method, to achieve the technical effect of high-precision direction finding in complex environments; compared with the traditional direction finding system using a single measurement station in the prior art, the present invention significantly improves the system's positioning accuracy of the target through the collaborative work of multiple direction finding stations, especially in the case of rapid target movement or multipath interference, and can effectively reduce errors.

[0017] 2. The present invention solves the problem that the traditional system cannot adapt to changes in a dynamic environment by optimizing the layout of the direction-finding station and dynamically compensating for environmental factors. By optimizing the location of the direction-finding station using the Cramer-Rao bound and combining it with real-time environmental status analysis, the system can maintain high accuracy and stability in various environments, significantly improving the robustness and adaptability of the system.

[0018] 3. By combining the maximum a posteriori estimation (MAP) and particle filtering algorithms, the present invention realizes real-time estimation of the target position; compared with the positioning scheme based on static data in the prior art, the present invention can dynamically update the target position and provide efficient positioning response when the target moves quickly; this real-time optimization enables the system to respond quickly and provide accurate position information in complex real-time monitoring tasks.

[0019] 4. The present invention evaluates the threat level of the target through an intelligent alarm module, providing a more intelligent response mechanism; compared with the system that only relies on fixed alarms in traditional technologies, the present invention can flexibly adjust the alarm level and response strategy according to information such as the type, location and speed of the target; the system can adjust the response measures in real time according to the target threat, reducing unnecessary alarms and reaction time, and improving the system's response efficiency to potential threats. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a system architecture diagram of the present invention; Figure 2 This is a signal detection module architecture diagram of the present invention; Figure 3 It is a diagram of the architecture of the direction finding error correction module of the present invention; Figure 4 This is a diagram of the direction finding optimization module architecture of the present invention; Figure 5 This is a diagram of the target position estimation module architecture of the present invention; Figure 6 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1-Figure 5 , an embodiment of the present invention provides a moving target detection system, comprising: Multi-station direction-finding system, which is used to deploy multiple direction-finding stations to obtain the arrival angle information of radio signals and improve positioning accuracy through the coordinated work of each direction-finding station; In the moving target detection system, the multi-station direction finding system is one of the core basic modules of the whole system. This module is mainly responsible for receiving the radio signal of the target and calculating the angle of arrival (AOA) of the target based on the spatial distribution of multiple direction finding stations. The direction finding accuracy and data integrity of this module directly determine the accuracy of subsequent modules such as direction finding error correction, direction finding optimization, and target position estimation. Therefore, it is necessary to fully consider factors such as the layout of the direction finding stations, the stability of signal reception, the direction finding calculation method, and the influence of direction finding errors. In general, the direction finding system needs to meet the basic requirements of high precision and low error, and at the same time have the ability to adapt to changes in different environments.

[0023] As an option, a multi-station direction-finding system usually adopts an arrangement of more than three stations to form a stable geometric distribution and improve the robustness of the direction-finding results. Specifically, in the moving target detection system, this module mainly uses direction-finding stations arranged at multiple points in space to synchronously receive radio signals and calculate the arrival angle of the target signal in combination with the angle measurement method.

[0024] In this embodiment, the multi-station direction finding system is composed of multiple fixed or mobile direction finding stations, and the position of each direction finding station is: ; in: Indicates The location of the direction-finding stations; Indicates The abscissa of each direction-finding station in the horizontal coordinate system; Indicates The vertical coordinate of each direction-finding station in the horizontal coordinate system; is the number of the direction finding station, the value range is ; is the total number of direction finding stations, which is usually not less than 3 to ensure the stability and reliability of direction finding calculation.

[0025] In some embodiments, each direction-finding station is equipped with a high-gain antenna and an independent signal receiving unit to capture the target's radio signal; generally, the antenna of the direction-finding station can adopt a multi-element array antenna, such as a uniform circular array (UCA) or a uniform linear array (ULA), to improve the angular resolution; in an urban environment, a highly directional array antenna can be selected to reduce multipath interference, and in an open environment, an omnidirectional antenna can be used to enhance the signal coverage capability.

[0026] Specifically, the target signal detected by each direction-finding station contains different arrival angle information, which can be expressed by the following formula: ; in: For the The angle of arrival measured by the direction-finding stations; is the unknown coordinates of the target; For the The known coordinates of the direction-finding stations; is the direction finding error, including the error caused by measurement noise, signal multipath propagation and Doppler effect; is the direction finding station number, the value range is ,in is the total number of direction finding stations, usually not less than 3.

[0027] In one possible implementation, the direction finding station adopts coherent direction finding technology, that is, the local oscillation sources of all direction finding stations are synchronized to ensure the phase consistency of the direction finding data, thereby reducing the direction finding error; in some embodiments, a time synchronization protocol (such as IEEE1588) can also be used to perform precise signal time calibration to further improve the direction finding stability of the system.

[0028] In this embodiment, in order to improve the direction finding accuracy, the system adopts a multi-station joint direction finding method, that is, the data of multiple direction finding stations are fused and processed to calculate the optimal direction finding angle of the target; specifically, assuming that the arrival angle information of the target signal is known, the possible position of the target can be solved by the following joint estimation algorithm: ; in: is the estimated target position; is the variable of the target position; For the The weight coefficient of each direction-finding station; For the The angle of arrival measured by the direction-finding stations; For the The known coordinates of the direction-finding stations; is the total number of direction-finding stations; Represents the index of the direction finding station, the value range is .

[0029] In general, the weight coefficient A weighting method based on the signal-to-noise ratio (SNR) can be used, namely: ; in: For the The weighting coefficients of the direction-finding stations; is the signal-to-noise ratio of the direction-finding station; is the total number of direction-finding stations; It is the index of the direction finding station, used to traverse all direction finding stations.

[0030] In a possible implementation, the optimization of direction finding error can also be combined with Kalman filter (KF) or particle filter (PF) methods to further improve the stability of direction finding data; for example, on a mobile platform, since the relative position of the direction finding station is constantly changing, the extended Kalman filter (EKF) method can be used to model the time series of the direction finding data and correct the target position through the state update formula.

[0031] In some embodiments, considering that the distribution of direction finding stations in the actual environment may be uneven, the system can introduce an adaptive direction finding station optimization strategy, that is, dynamically adjust the data weight of the direction finding station according to the motion state of the target; for example, when the target is close to a certain direction finding station, the data weight of the direction finding station can be increased to reduce the measurement error caused by the long-distance direction finding station; in addition, in a high dynamic environment, the system can also combine the inertial navigation (IMU) data to compensate for the direction finding results, thereby further improving the direction finding accuracy.

[0032] As an option, in a multi-station direction finding system, if the direction finding station is a movable carrier (such as a drone, vehicle, etc.), the system can dynamically adjust the position of the direction finding station according to the target's motion trajectory, thereby optimizing the direction finding geometric distribution and improving the stability of the direction finding solution; in some high-precision application scenarios, such as military reconnaissance systems, it can also be combined with differential GPS (DGPS) or Beidou high-precision positioning technology to perform real-time calibration of the position of the direction finding station to further reduce the direction finding error.

[0033] The signal detection module is connected to the multi-station direction finding system and is used to detect the radio signal of the target and extract the arrival angle information; In the moving target detection system, the signal detection module is used to receive and analyze the target's radio signal and is the starting point of the entire system data processing chain. The detection accuracy, anti-interference capability and real-time performance of this module directly affect the accuracy of subsequent direction-finding error correction, direction-finding optimization and target position estimation. In general, the signal detection module needs to support broadband signal reception, real-time spectrum analysis, and be able to cope with signal changes caused by environmental noise, multipath effects and target motion. As an option, short-time Fourier transform (STFT), wavelet transform and other methods can be used to extract features from the target signal to improve the detection capability of weak signals or frequency-hopping signals.

[0034] Specifically, the signal detection module of the present invention includes a spectrum scanning unit, a short-time Fourier transform unit, and a Doppler frequency shift detection unit; the spectrum scanning unit is responsible for searching for radio signals within a set frequency band, the short-time Fourier transform unit is used to perform time-frequency analysis on the signal, and the Doppler frequency shift detection unit calculates the relative speed of the target by analyzing the movement of the signal spectrum.

[0035] In this embodiment, the spectrum scanning unit adopts a broadband receiver structure, which can cover multiple working frequency bands and support adaptive scanning. In general, the signal scanning is performed in a preset step frequency manner to ensure that all frequency bands where the target may exist are covered within a limited time. The set scanning step frequency Depends on the bandwidth of the signal And the system's analytical capabilities to meet: ; in: is the total bandwidth scanned; is the number of scanning frequency points; is the single-step scanning interval.

[0036] As an option, the scanning process can be based on power threshold detection (Energy Detection), that is, when the signal power is higher than a certain threshold of background noise, the system determines that there may be a target signal; assuming that the background noise power is , then the detection threshold Set to: ; in: is the system noise power; is the detection threshold adjustment factor.

[0037] In some embodiments, if the scanning frequency band is relatively wide, in order to reduce the scanning time, the system may adopt a parallel spectrum analysis method, that is, simultaneously perform real-time analysis on multiple frequency bands to improve detection efficiency.

[0038] In this embodiment, the short-time Fourier transform unit is used to perform time-frequency analysis on the target signal to extract the temporal dynamic characteristics of the signal; specifically, the short-time Fourier transform (STFT) calculation formula is: ; in, The signal in time and frequency The time-frequency representation of ; is the input signal; is the local analysis window function, making the signal The vicinity of moments is weighted; is the Fourier basis function; is the time-integrated variable.

[0039] In general, window functions The selection of will affect the time-frequency resolution of STFT; as an option, if the system focuses on short-term signal mutations, a Gaussian window or a Hamming window can be used; if the spectrum accuracy is concerned, a rectangular window can be used to obtain better frequency resolution.

[0040] In some embodiments, in order to improve the ability to detect weak signals, wavelet transform (WaveletTransform) can be used for multi-scale analysis. The specific calculation formula is: ; in, is the wavelet transform coefficient, which indicates the signal at scale and time location The transformation result below: is the input signal, i.e., the radio signal received from the target; is the scale factor, controlling the resolution of the transformation, and the smaller Focus on high frequency components, larger Focus on low-frequency components; is the time offset, which determines the position of the transformation on the time axis; is the time variable; is the complex conjugate of the mother wavelet function.

[0041] In one possible implementation, if the target signal uses a frequency hopping communication method, the system can use a time-frequency tracking algorithm to model the change of the target frequency over time and predict the future frequency distribution of the signal, thereby improving the probability of target detection.

[0042] In this embodiment, the Doppler frequency shift detection unit is used to calculate the relative motion speed of the target and correct the frequency offset in the signal detection process; generally, the Doppler frequency shift of the target satisfy: ; in: is the radial velocity of the target; is the speed of light; is the carrier frequency.

[0043] In some embodiments, in order to improve the detection accuracy of Doppler frequency shift, a method of combining short-time Fourier transform with time-frequency peak tracking may be used, namely: Calculate the STFT result of the signal; Identify the dominant frequency components of a signal ; Calculate frequency offset through time series analysis , and by Calculate the target speed.

[0044] In one possible implementation, if the Doppler frequency shift of the signal is large, it may cause spectrum broadening of the signal during reception. In this case, a broadband matched filtering method may be used to correct the frequency distortion caused by the frequency shift and improve the stability of signal detection.

[0045] In some embodiments, if there is a change in the acceleration of the target, that is, If it changes with time, the second-order Doppler correction method can be used, that is: ; in: is the corrected Doppler shift; is the radial acceleration of the target; is the signal propagation time; is the carrier frequency.

[0046] In practical applications, the signal detection module needs to take into account detection accuracy, computational complexity, and real-time performance; therefore, in certain scenarios with high computational load, such as drone detection, the system can use deep learning methods for signal classification, such as convolutional neural networks (CNN) or long short-term memory networks (LSTM), to identify different types of radio signal characteristics.

[0047] In another possible implementation, if the signal environment is complex, the system can introduce adaptive threshold detection (Adaptive Thresholding), that is, dynamically adjust the signal detection threshold according to the real-time noise level to reduce the false detection rate and improve the detection sensitivity.

[0048] The direction finding error correction module is connected with the signal detection module and the multi-station direction finding system, analyzes the frequency spectrum characteristics of the target signal based on the short-time Fourier transform, and corrects the direction finding error caused by the Doppler effect; In the moving target detection system, the direction finding error correction module is a key component, which aims to compensate for the measurement errors caused by factors such as the Doppler effect and signal multipath propagation. The aforementioned signal detection module has extracted the spectral characteristics of the target signal and calculated the frequency shift information of the signal. The next task is to use this data to correct the measured angle of arrival (AOA) to ensure the accuracy of the final target positioning.

[0049] In this embodiment, the direction finding error correction module includes a Doppler compensation unit, an angle correction unit and a particle filter unit, and each unit performs correction processing according to different error sources.

[0050] The Doppler compensation unit is responsible for calculating the frequency offset caused by the relative velocity of the target and then correcting the measurement angle. During the target movement, the radial velocity of the target relative to the receiver This causes the signal frequency to shift, which in turn affects the angular accuracy of the measurement results; specifically, based on the Doppler frequency shift provided by the signal detection module , the corrected arrival angle can be calculated : ; in: is the measured original arrival angle; is the corrected arrival angle; is the Doppler shift of the target; is the speed of light; is the carrier frequency; is the radial velocity of the target.

[0051] In this calculation, the radial velocity of the target It is calculated through the frequency shift of the signal, and then the original arrival angle is corrected to compensate for the frequency deviation caused by movement.

[0052] The angle correction unit is responsible for processing the direction-finding error caused by multipath effects and signal noise. In some embodiments, multipath propagation of signals can cause time delays and phase changes in the signals, thereby causing direction-finding errors. To solve this problem, the angle correction unit uses an adaptive algorithm to perform weighted processing on the signals received by the direction-finding station to reduce the impact of the multipath effect.

[0053] One way to implement angle correction is to eliminate the error by taking a weighted average of the measurement results. The weighted average formula is as follows: ; in: is the corrected final arrival angle; For the The angle of arrival measured by the direction-finding stations; For the The weighting coefficients of the direction-finding stations; is the total number of direction finding stations.

[0054] In general, the weight Weighting is performed according to the signal quality and the reliability of the direction-finding station so that the measurement results with a high signal-to-noise ratio contribute more to the final correction result.

[0055] The particle filter unit is mainly used in the system to dynamically correct the direction-finding error and estimate the target's motion trajectory. Particle filtering is a nonlinear estimation algorithm based on the Monte Carlo method, which is suitable for processing complex noise and non-Gaussian distribution. In this embodiment, the particle filter unit estimates the position and speed of the target by simulating the states of multiple particles, and updates the weights of the particles according to the measurement results.

[0056] The basic process of particle filtering includes: Prediction stage: Predict the next position of the target based on the motion model. The state equation is as follows: ; in: It is at the moment The state vector represents the position, velocity or other relevant dynamic characteristics of the target; It is at the moment The predicted status of is the state transfer matrix, representing the transition from time To time The state update relationship depends on the target's motion model; It represents the random disturbance caused by factors such as incomplete model and sensor noise during the target motion.

[0057] Update phase: based on the latest measurements , update the particle weight: ; in: is the current measurement value (i.e. the corrected arrival angle); is the observation matrix; To measure noise.

[0058] Resampling stage: according to the weight of each particle , resample the particles to increase the proportion of high-weight particles, thereby optimizing the estimation.

[0059] Through these steps, the particle filter unit can dynamically correct the target position and continuously track the trajectory of the target, greatly improving the accuracy of direction finding correction.

[0060] The direction finding optimization module is connected to the direction finding error correction module, uses Bayesian estimation to establish a direction finding error model, and dynamically optimizes the error in combination with the real-time environmental status; In the moving target detection system, the direction finding optimization module is mainly responsible for improving the direction finding accuracy of the system by optimizing the measured angle of arrival (AOA) information; the aforementioned signal detection module has extracted the basic characteristics of the target signal through spectrum scanning, short-time Fourier transform and other methods, and the direction finding error correction module compensates for the measurement errors caused by Doppler effect, signal interference, etc.; the next task is to further optimize the measurement results in order to provide more accurate input data for the target position estimation module, thereby achieving accurate target positioning.

[0061] In this embodiment, the direction finding optimization module uses the maximum a posteriori estimation (MAP) and high-order Bayesian estimation (HBE) methods to optimize the measured arrival angle data; the purpose of the optimization is to reduce the direction finding deviation caused by signal noise, environmental changes and dynamic errors, and to improve the estimation accuracy of the target position.

[0062] The maximum a posteriori estimation (MAP) method is used to estimate the optimal position of the target based on the measurement results; in general, the posterior probability distribution of the target position is calculated by the Bayesian formula; specifically, given the measurement data and target location , the posterior probability of the target location can be expressed as: ; in: is the posterior probability of the target location; is the measurement likelihood function; is the prior probability of the target location; is the normalization constant.

[0063] In one possible implementation, the measurement data is the arrival angle information provided by multiple direction-finding stations, and the measurement likelihood function It can be modeled by assuming that the measurement error follows a Gaussian distribution, namely: ; in: The target location Given measurement data The likelihood function of ; For the The measurement results of the direction-finding stations (raw measured arrival angles); For the The corrected angle of arrival of each direction-finding station; For the The error variance of the measurement is The noise level of each measurement; is the number of measuring stations, that is, the number of sites participating in the measurement.

[0064] In some cases, in order to further improve the accuracy of the measurement results, the system can use a weighted method to enhance the reliance on high signal-to-noise ratio (SNR) measurements, such as optimizing the estimate of the target position by weighted least squares (WLS).

[0065] The higher-order Bayesian estimation (HBE) method is proposed to overcome the shortcomings of traditional Bayesian estimation in dealing with non-Gaussian noise and dynamic environments. Compared with traditional methods, HBE not only considers the prior distribution of the target position and the statistical characteristics of the measurement error, but also introduces the environmental state variable To model the impact of the external environment on the measurement results; specifically, the posterior probability of the target location is: ; in: In a known environment In the case of given measurement data Rear target position The posterior probability of is the prior probability of the environmental state; In the environmental state The prior distribution of the target location; For a given target location and environmental status Under the conditions, the measured data The likelihood function of .

[0066] In some embodiments, the environmental state Dynamic adjustments can be made by modeling multipath effects, meteorological factors (such as wind speed, air pressure), or other factors that affect the measurement results; these factors may change over time, affecting the signal propagation path of the target and the final measurement results.

[0067] In this embodiment, in order to further improve the direction finding accuracy, the direction finding optimization module dynamically adjusts the weight coefficient of each direction finding station. , so that high-quality measurements (i.e., measurements with a high signal-to-noise ratio) contribute more to the final target position estimate; the adjustment of the weight coefficient is based on the signal-to-noise ratio (SNR) of each direction-finding station and the reliability of the measurement; specifically, the calculation of the weight takes into account the uncertainty of the measurement and is set by the following formula: ; in: For the The weighting coefficients of the direction-finding stations; is an adjustment factor used to control the impact of the signal-to-noise ratio; For the The signal-to-noise ratio of a direction-finding station reflects the reliability of the measurement.

[0068] In this formula, the signal-to-noise ratio The higher the weight coefficient The larger the signal-to-noise ratio, the greater the impact of high-quality measurement results on the final target position estimate. Conversely, if the signal-to-noise ratio is low, the weight of the measurement result will be smaller, thereby reducing the interference of low-quality data on the final position estimate.

[0069] Through this signal-to-noise ratio weighted strategy, the system can effectively filter out low-quality measurement data, reduce the impact of environmental noise on target position estimation, and improve the accuracy and reliability of target measurement.

[0070] The sensor layout optimization module is connected to the direction finding optimization module, calculates the lower bound of the direction finding error based on the Cramer-Rao bound, adjusts the position of the direction finding station, and optimizes the layout of the direction finding station; In the moving target detection system, the sensor optimization layout module is an important module to improve the direction finding accuracy and system stability; the aforementioned signal detection module and direction finding error correction module respectively complete the signal extraction, analysis and error correction, but the direction finding accuracy of the system is still limited by the location distribution of the direction finding stations and their relative geometric relationship; therefore, in order to further improve the measurement accuracy, optimizing the layout of the sensors and reasonably configuring the location of the direction finding stations become essential steps; by optimizing the layout of the direction finding stations, the comprehensiveness of the measurement data can be ensured, the measurement errors can be reduced, and the system's estimation accuracy of the target position can be improved.

[0071] In this embodiment, the sensor optimization layout module calculates the theoretical lower bound of the direction finding error by combining the Cramer-Rao bound (CRLB) to optimize the layout of the direction finding stations to ensure that the measurement error is minimized during actual use; the optimization goal of this module is to improve the direction finding accuracy and coverage of the system by reasonably arranging the positions of the direction finding stations, thereby providing more accurate and reliable data support for subsequent direction finding optimization and target positioning.

[0072] First, the Cramer-Rao bound (CRLB) provides a theoretically optimal lower bound for the error, which can provide a limit for the measurement accuracy; in general, the minimum variance of the direction finding error can be expressed by the following formula: ; in: is the variance of the measured angle, representing the degree of measurement error; is the Fisher information quantity, defined as: ; in: For the given data Likelihood function under ; is the measured angle, usually the angle of arrival of the target.

[0073] In this formula, the Fisher information It is the key factor that determines the measurement accuracy; in general, It depends on the location distribution of the direction finding stations, that is, the geometric distribution of the direction finding stations will directly affect the measurement error.

[0074] In the sensor optimization layout module, the goal is to optimize the layout of the direction finding stations so that the Fisher information Maximize, thereby minimizing the measurement error; specifically, the optimization algorithm analyzes multiple candidate layout schemes, calculates the lower bound of the error under each layout, and selects the layout that optimizes the measurement accuracy.

[0075] In general, the location of the direction-finding station The optimization layout will be performed in the target area; the optimization problem can be expressed by the following objective function: ; in, is the variance of the measured angle, indicating the degree of measurement error; is the number of direction finding stations, that is, the total number of direction finding stations participating in the measurement in the system; For the The weight coefficient of each direction-finding station is usually set according to the signal-to-noise ratio (SNR) or other reliability parameters of the measuring station; For the The Fisher information of a DF station represents the contribution of the measurement data at this DF station to the final angle estimate; is the arrival angle of the target.

[0076] In one possible implementation, the weight coefficient The signal-to-noise ratio (SNR) of the measuring stations can be dynamically adjusted so that the measuring stations with higher SNRs account for a larger proportion in the optimization process.

[0077] Through optimization algorithms such as gradient descent or genetic algorithms, the system can quickly search for the optimal layout; specifically, the system tries multiple layout schemes, calculates the measurement error of each scheme, and finally selects the layout with the smallest error; this optimization process can be carried out through the following steps: Initialization layout: The system randomly or evenly distributes the locations of the initial direction-finding stations according to the characteristics of the target area; Calculate the amount of information: Based on the position of each direction finding station, calculate the Fisher information of each measuring station , and according to its weight coefficient Calculate the total amount of information ; Optimize layout: adjust the location of direction finding stations through optimization algorithms (such as gradient descent or genetic algorithm) to minimize measurement errors; Output optimal layout: After a certain number of iterations, output the direction finding station layout plan with the smallest error.

[0078] In some embodiments, in order to improve the computational efficiency of the algorithm, the system may also adopt a distributed optimization method, that is, dividing the optimization task distributed by the direction finding stations into multiple subtasks, which are calculated in parallel on multiple processing units, thereby shortening the optimization time.

[0079] In addition, the optimization scheme in this embodiment also takes into account the impact of the dynamic environment; for example, in the application scenario of a mobile platform such as a drone or a robot, the position of the direction-finding station may change with the movement of the platform; therefore, the system can adopt a real-time feedback mechanism to dynamically adjust the layout of the direction-finding station according to the real-time measurement data and the system to adapt to environmental changes.

[0080] The target position estimation module is connected to the direction finding optimization module, and integrates the corrected direction finding angle information to accurately calculate the target position through the maximum a posteriori estimation method; In the moving target detection system, the main task of the target position estimation module is to calculate the precise position of the target based on the data provided by the aforementioned signal detection module, direction finding error correction module and direction finding optimization module; the previous modules have already processed the signal extraction, error correction and optimization, and the target position estimation module is responsible for combining the measurement data provided by multiple direction finding stations to finally derive the spatial position of the target.

[0081] In this embodiment, the target position estimation module adopts a method that combines maximum a posteriori estimation (MAP) and particle filtering (PF); the maximum a posteriori estimation method relies on measurement data and prior knowledge of the target to estimate the optimal solution for the target position, while the particle filtering method tracks the target position in real time in a dynamic environment to provide a more accurate estimate.

[0082] The maximum a posteriori estimation (MAP) method is used to estimate the most likely target location by optimizing the posterior probability of the target location. Using multi-station measurement data, the MAP method considers the prior distribution of the target location and the current measurement data, and obtains the optimal location of the target by calculating the posterior probability. In general, the posterior probability distribution of the target location is It is calculated based on the Bayesian formula, in which the prior distribution and measurement likelihood function reflect the relationship between the target's historical motion information and the measurement data.

[0083] By maximizing the posterior probability, the system can obtain the best estimate of the target position, especially when there is noise or uncertainty in the measurement results. The MAP method can better combine prior knowledge to compensate for these uncertainties.

[0084] In the particle filter (PF) method, the estimation of the target position depends on the state estimation of multiple particles; particle filtering is a Monte Carlo method that estimates the target position through three stages: prediction, update, and resampling; each particle represents a possible target state, and the probability of the target state is reflected by dynamically updating the particle weight.

[0085] The process of particle filtering is as follows: Prediction stage: predict the position of the target at the next moment based on the target's motion model; the state of each particle is updated through the target's motion characteristics such as speed and acceleration.

[0086] Update phase: Combined with the current measurement data, update the weight of each particle; the weight is calculated based on the probability of the particle generating a measurement result in the current state.

[0087] Resampling stage: Resample the particles according to their weights, retain the particles with high weights, and remove the particles with low weights, so as to improve the accuracy of target position estimation.

[0088] Through this process, the particle filter can continuously optimize the estimation of the target position during the target movement and adapt to the changes in the environment.

[0089] In order to improve the accuracy of position estimation, this embodiment can also combine the motion model of the target, such as linear motion, uniformly accelerated motion, etc., to further constrain the prediction range of the particles, thereby reducing the computational complexity and improving real-time performance; at the same time, the system can dynamically adjust the number and weights of particles according to changes in the actual environment to adapt to different measurement conditions and environmental noise.

[0090] In some embodiments, the particle filter algorithm can further improve the estimation accuracy by a weighted particle filter method, which performs weighted calculation on the weight of the particle according to the matching degree between the predicted value of the particle and the measured data.

[0091] The intelligent warning module is connected to the target position estimation module, performs threat assessment based on the calculated target position and situational awareness, automatically generates warning information and executes corresponding countermeasures; In the moving target detection system, the intelligent warning module is the key module to achieve system response, which aims to generate warning information according to the target position and system status and take corresponding response measures; the aforementioned signal detection module, direction finding error correction module, direction finding optimization module and target position estimation module have completed the tasks of target signal processing, error correction and position estimation. The system can now provide real-time situational awareness through the intelligent warning module; this module can not only issue an alarm when the target approaches a specific area, but also execute countermeasures according to the threat level of the target, such as activating a jamming system or executing a target interception mission.

[0092] In this embodiment, the intelligent alarm module adopts an adaptive alarm strategy based on the target's location information, speed, target type and other characteristics, and triggers corresponding response measures by judging the threat level of the target; the core function of the intelligent alarm module is to judge the threat level according to the situation of the target entering the warning area and generate alarm information.

[0093] The target monitoring and early warning mechanism is the basis of the intelligent warning module. Specifically, the system sets a warning area (usually a circular area) and monitors in real time whether the target enters the area. The real-time position of the target is provided by the target position estimation module. Assume that the position of the target is The center of the warning area is , the radius is ; When the target enters the warning area, the system triggers an early warning; the mathematical expression is: ; in: is the distance from the current position of the target to the center of the warning area; is the current position of the target; It is the center of the warning area; is the radius of the warning area.

[0094] when When the target enters the warning area, the system triggers an alarm. Specifically, the system determines whether the target threatens a specific area based on the target's current position, speed and other information, and generates an alarm message.

[0095] Threat assessment and response decision-making is another key function in the intelligent warning module; based on the target's approach speed, type (for example, whether it is a drone, aircraft, ship, etc.), preset safety standards, and the target's historical trajectory, the system can assess the target's threat level; generally, threat assessment is based on the distance between the target and the warning area. , the speed of the target And the target trajectory prediction for comprehensive analysis; target threat level It can be estimated by the following formula: ; in: is the distance between the target and the warning area; is the radial velocity of the target, reflecting the speed at which the target approaches; Classification information for target types, which may include drones, aircraft, potential threat types, etc.

[0096] In this way, the system is able to dynamically adjust the threat level , and take corresponding early warning or countermeasures according to different threat levels.

[0097] The response mechanism performs corresponding actions according to the threat level of the target; generally, targets with lower threat levels may only trigger information recording or alarms, targets with medium threat levels will trigger alarms and enable emergency response mechanisms, and targets with higher threat levels may trigger system countermeasures, such as radio interference, drone interception, etc.; Specifically, the decision of the response mechanism can be expressed by the following formula: ; in: Indicates the system's Possible responses include: To record the alarm, no immediate action is required; To trigger a warning, inform relevant personnel to pay attention to the existence of the target and take precautions; To initiate countermeasures and execute countermeasures such as wireless jamming or target interception; For the goal The threat level reflects the potential threat level of the target to the system. and They are respectively the preset low threat and high threat thresholds, which are used to divide different threat level intervals.

[0098] In specific implementations, the threat level threshold can be adjusted according to the actual application scenario; for example, in military reconnaissance missions, targets with higher threats may directly trigger target interception measures; while in civilian scenarios, targets with lower threat levels may only trigger information recording and alarms.

[0099] The dissemination and visualization of warning information is also part of the intelligent warning module. In some embodiments, the system notifies relevant personnel through a display interface or an alarm system, informing them of the specific location of the target, the threat level, and the corresponding treatment suggestions. In addition, the warning information can also be transmitted to other systems or personnel through the network for further response measures. The warning information includes but is not limited to: Target type; The current position and movement trajectory of the target; Threat level; Suggested countermeasures.

[0100] Generally, the system transmits alarm information to the equipment of relevant personnel or the command center through a wireless communication network to ensure timely delivery and response of information.

[0101] See also Figure 6 The present invention also provides a moving target detection method, comprising the following steps: S1. Initialize the multi-station direction finding system, deploy multiple direction finding stations, and determine the spatial coordinates of each direction finding station; First, the system needs to initialize the multi-station direction finding system, deploy multiple direction finding stations, and determine the spatial coordinates of each direction finding station; by reasonably distributing the positions of the direction finding stations, ensure that the target area is covered and accurate direction finding data is provided; the layout design of the direction finding stations should take into account the geographical environment, target motion characteristics, and the accuracy required by the system; in some embodiments, the positions of the direction finding stations can be laid out based on the Cramer-Rao Bound (CRLB) optimization algorithm to ensure that the measurement error is minimized.

[0102] S2, target radio signal detection, scanning the target radio signal through the signal detection module to extract the arrival angle information; During the target detection phase, the radio signal emitted by the target is scanned through the signal detection module; this module not only extracts the spectral characteristics of the target signal, but also identifies the angle of arrival (AOA) of the target signal, providing key data for subsequent direction-finding calculations; usually, signal detection uses multi-band scanning, short-time Fourier transform and other methods to accurately analyze the target signal, thereby improving the ability to adapt to different targets and environmental interference.

[0103] S3, Doppler frequency shift adaptive correction, based on short-time Fourier transform analysis of the target signal's spectrum characteristics, calculate the Doppler frequency shift and correct the direction finding error; Due to the movement of the target, the signal will be affected by the Doppler effect during the reception process, resulting in frequency offset; the spectral characteristics of the target signal are analyzed by short-time Fourier transform (STFT), and the Doppler frequency shift adaptive correction module calculates the Doppler frequency shift of the target and corrects the direction-finding error based on the calculation result; this correction step ensures that when the target moves at high speed, the system can accurately measure the arrival angle and eliminate the error caused by frequency shift.

[0104] S4, direction finding optimization calculation, using Bayesian estimation to establish a direction finding error model, and dynamically optimize the error in combination with the environmental state; After obtaining the corrected direction finding angle information, the direction finding optimization module uses Bayesian estimation to establish a direction finding error model, and dynamically optimizes the measurement error in combination with environmental state analysis; by analyzing factors such as multipath interference and noise in the environment, the system can adjust the weight of the measurement results in real time, thereby improving the accuracy of direction finding; in some embodiments, the environmental state includes factors such as weather changes and terrain interference, which can affect the quality of signal propagation and measurement results.

[0105] S5, optimize the layout of direction finding stations, calculate the lower bound of direction finding error based on Cramer-Rao bound, and adjust the location of direction finding stations to optimize direction finding accuracy; In order to further optimize the direction-finding accuracy, the system calculates the theoretical lower bound of the direction-finding error based on the Cramer-Rao bound (CRLB) and uses an optimization algorithm to adjust the position of the direction-finding station. By optimizing the layout of the direction-finding stations, the measurement accuracy of the system can be maximized and the measurement error can be reduced. The optimization process takes into account the geometric distribution of the direction-finding stations, the signal-to-noise ratio, and the characteristics of the target motion. The optimized layout of the direction-finding stations can provide higher-quality data support for subsequent target position estimation.

[0106] S6, target position calculation, integrating the corrected direction finding angle information, and using maximum a posteriori estimation to calculate the optimal target position; Based on the corrected direction-finding angle information of each direction-finding station, the system calculates the optimal position of the target through the maximum a posteriori estimation (MAP) method; at this time, the estimation of the target position not only takes into account the current measurement data, but also combines the motion model and prior information of the target; by maximizing the posterior probability of the target position, the system can give the optimal estimate of the target position; in some embodiments, in order to improve the accuracy of position estimation, methods such as particle filtering can be used for dynamic optimization.

[0107] S7, intelligent warning and decision-making, based on the calculated target position, determines whether the target has entered the warning area and executes corresponding warning and countermeasures; After the target position is accurately calculated, the intelligent alarm module makes a judgment based on whether the target enters the warning area; if the target enters the predetermined safety area, the system will generate an alarm and execute corresponding countermeasures, such as activating the jamming system or guiding the interception equipment; the intelligent alarm module dynamically determines the alarm level based on the target's threat level, speed, type and other factors, and sends real-time alarm information to relevant personnel, thereby ensuring timely response to potential threats.

[0108] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Moving target detection system, characterized in that: include: Multi-station direction-finding system, which is used to deploy multiple direction-finding stations to obtain the arrival angle information of radio signals and improve positioning accuracy through the coordinated work of each direction-finding station; The signal detection module is connected to the multi-station direction finding system and is used to detect the radio signal of the target and extract the arrival angle information; The direction finding error correction module is connected with the signal detection module and the multi-station direction finding system, analyzes the frequency spectrum characteristics of the target signal based on the short-time Fourier transform, and corrects the direction finding error caused by the Doppler effect; The direction finding optimization module is connected to the direction finding error correction module, uses Bayesian estimation to establish a direction finding error model, and dynamically optimizes the error in combination with the real-time environmental status; The sensor layout optimization module is connected to the direction finding optimization module, calculates the lower bound of the direction finding error based on the Cramer-Rao bound, adjusts the position of the direction finding station, and optimizes the layout of the direction finding station; The target position estimation module is connected to the direction finding optimization module, and integrates the corrected direction finding angle information to accurately calculate the target position through the maximum a posteriori estimation method; The intelligent warning module is connected to the target position estimation module, performs threat assessment based on the calculated target position and situational awareness, automatically generates warning information and executes corresponding countermeasures.

2. The moving target detection system according to claim 1, characterized in that: The signal detection module comprises: A spectrum scanning unit for scanning target radio signals; Short-time Fourier transform unit, used to analyze the spectral characteristics of the signal; The Doppler frequency shift detection unit is used to calculate the Doppler frequency shift of the target.

3. The moving target detection system according to claim 1, characterized in that: The direction finding error correction module comprises: Doppler compensation unit, used to calculate the relative speed of the target; An angle correction unit, used to correct the measured arrival angle; Particle filter unit, used to dynamically correct direction finding errors.

4. The moving target detection system according to claim 1, characterized in that: The direction finding optimization module comprises: A measurement likelihood calculation unit, used for calculating a measurement likelihood function based on a direction finding error; An environment status analysis unit, used for classifying a multipath interference environment; A high-order Bayesian estimation unit is used to optimize the target direction finding error by combining environmental factors.

5. The moving target detection system according to claim 1, characterized in that: The target position estimation module comprises: A position calculation unit, used to calculate the optimal position of the target; The optimal estimation operator is used to optimize the target position based on maximum a posteriori estimation; The direction finding error compensation unit is used to correct the target position deviation based on Bayesian estimation.

6. The moving target detection system according to claim 1, characterized in that: The direction finding optimization module adopts the maximum a posteriori estimation algorithm, and the posterior probability distribution of the target position satisfies: Target location Depend on The arrival angle of the direction-finding station Sure; The formula for calculating the posterior probability of the target location is: ； in: is the posterior probability of the target location; is the measurement likelihood function; is the prior probability of the target location; is the normalization constant.

7. The moving target detection system according to claim 1, characterized in that: The direction finding error correction module calculates the Doppler frequency shift based on short-time Fourier transform, satisfying: Assume the target signal is , then the short-time Fourier transform is defined as: ； in, For the signal at time and frequency The time-frequency representation of ; is the input signal; is the local analysis window function, making the signal The vicinity of moments is weighted; is the Fourier basis function; is the time-integrated variable.

8. The moving target detection system according to claim 2, characterized in that: The Doppler shift detection unit calculates the Doppler shift of the target ,satisfy: ； in: is the Doppler shift; is the radial velocity of the target; is the speed of light; is the carrier frequency.

9. The moving target detection system according to claim 1, characterized in that: The sensor optimization layout module optimizes the direction finding station layout based on the Cramer-Rao boundary, satisfying: The theoretical lower bound of the direction finding error is calculated by the Cramer-Rao bound: ； in: is the direction finding error variance; is the Fisher information amount.

10. A moving target detection method, applied to a moving target detection system according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. Initialize the multi-station direction finding system, deploy multiple direction finding stations, and determine the spatial coordinates of each direction finding station; S2, target radio signal detection, scanning the target radio signal through the signal detection module to extract the arrival angle information; S3, Doppler frequency shift adaptive correction, based on short-time Fourier transform analysis of the target signal's spectrum characteristics, calculate the Doppler frequency shift and correct the direction finding error; S4, direction finding optimization calculation, using Bayesian estimation to establish a direction finding error model, and dynamically optimize the error in combination with the environmental state; S5, optimize the layout of direction finding stations, calculate the lower bound of direction finding error based on Cramer-Rao bound, and adjust the location of direction finding stations to optimize direction finding accuracy; S6, target position calculation, integrating the corrected direction finding angle information, and using maximum a posteriori estimation to calculate the optimal target position; S7, intelligent warning and decision-making, based on the calculated target position, determines whether the target enters the warning area and executes corresponding warning and countermeasures.

Images