An active defense system and method for unmanned aerial vehicles based on multi-sensor fusion

Abstract

The present invention discloses an active defense system and method for unmanned aerial vehicles based on multi-sensor fusion, including a perception layer, a processing layer and a defense layer. The perception layer obtains target features through radar and electromagnetic sensing; the processing layer adopts a dual-channel parallel architecture for feature extraction, and performs selective state updates based on frame interval thresholds; the defense layer selects interference methods according to signal strength and optimizes power output in combination with a propagation loss model. The present invention solves technical problems such as low data processing efficiency, high target state update overhead, and poor interference defense effect in existing unmanned aerial vehicle defense systems. The system overhead is reduced through a selective state update mechanism, data processing efficiency is improved through dual-channel parallel processing, and precise interference is achieved through an innovative combination of defense strategies, ultimately achieving an efficient and real-time defense effect.

Description

Technical Field

[0001] The present invention relates to the field of drone defense technology, and in particular to a drone active defense system and method based on multi-sensor fusion. Background Art

[0002] With the rapid development and widespread application of drone technology, the security threat of illegal or malicious use of drones has become increasingly prominent, and effective defense against drones has become an important issue that needs to be solved urgently. At present, drone defense systems mainly use a combination of multi-sensor detection and electronic jamming to achieve identification, positioning and defense of target drones through multi-source information fusion.

[0003] However, the existing drone defense systems have the following main shortcomings:

[0004] Traditional defense systems use a single-channel serial processing architecture, which is inefficient when processing multi-source sensor data and cannot fully utilize the advantages of multi-sensor collaborative detection. For example, when processing radar spatial information and electromagnetic signature information, the shared data processing channel can easily cause data congestion, affecting the real-time performance of the system.

[0005] Existing systems generally use a fixed-cycle state update mechanism, which updates the target state regardless of the size of the change. This mechanism has obvious defects in system resource utilization and real-time response. When the target is relatively stationary or moving slowly, frequent state updates will cause unnecessary system overhead.

[0006] In terms of defense strategies, existing systems often use preset fixed interference methods and lack the ability to adaptively adjust to different target characteristics. At the same time, the control of interference power is relatively extensive and fails to fully consider the spatial propagation characteristics, which not only affects the defense effect, but also easily causes energy waste and electromagnetic environmental pollution.

[0007] Most defense systems suffer from insufficient coordination at the data fusion level and fail to effectively integrate spatial and electromagnetic feature information, resulting in incomplete and inaccurate perception of target status, affecting the accuracy of defense decisions.

[0008] Therefore, it is urgent to propose a new drone defense system to achieve efficient multi-source data parallel processing, intelligent state update mechanism and accurate defense strategy to improve the real-time performance, accuracy and defense effectiveness of the system. At the same time, the system also needs to have the ability to adaptively adjust the interference mode and power according to the target characteristics to achieve a more accurate and efficient defense effect. Summary of the invention

[0009] The purpose of the present invention is to propose an active defense system and method for unmanned aerial vehicles based on multi-sensor fusion, so as to solve the problems of low data processing efficiency, rigid state update mechanism, single defense strategy, etc. in the existing system, and to improve the real-time performance and defense effect of the system.

[0010] To achieve the above purpose, the technical solution provided by the present invention is as follows:

[0011] An active defense system for unmanned aerial vehicles based on multi-sensor fusion, including a perception layer, a processing layer and a defense layer connected in sequence:

[0012] The perception layer includes a radar sensor module and an electromagnetic sensor module, wherein the radar sensor module is used to obtain a target's distance and azimuth data stream, and the electromagnetic sensor module is used to obtain a target's working frequency band data stream;

[0013] The processing layer includes a parallel data processing module and a feature fusion module. The parallel data processing module uses a dual-channel parallel architecture to extract features from the acquired data stream. The feature fusion module selectively updates the target state based on a frame interval threshold to generate a real-time distance state quantity and a spectrum feature state quantity of the target.

[0014] The defense layer is provided with a global defense module and a power regulation module. The global defense module selects an interference mode according to the target signal strength and generates a corresponding interference response signal. The power regulation module optimizes and regulates the output power of the interference response signal based on a spatial propagation loss model.

[0015] Furthermore, the parallel data processing module includes a data pipeline unit and a parallel processing unit. The data pipeline unit sets independent FIFO queues for sequential access to the distance and azimuth data streams of the radar sensor and the working frequency band data streams of the electromagnetic sensor. The parallel processing unit correspondingly sets two independent processing channels. The first processing channel performs coordinate conversion and normalization processing on the distance and azimuth data streams, and the second processing channel performs spectrum analysis on the working frequency band data stream, extracts frequency band characteristics and signal strength, and outputs standardized distance and azimuth data and standardized frequency band characteristic data, respectively.

[0016] Furthermore, the parallel data processing module includes a data pipeline unit and a parallel processing unit. The data pipeline unit sets independent FIFO queues for sequential access to the distance and azimuth data streams of the radar sensor and the working frequency band data streams of the electromagnetic sensor. The parallel processing unit correspondingly sets two independent processing channels. The first processing channel performs coordinate conversion and normalization processing on the distance and azimuth data streams, and the second processing channel performs spectrum analysis on the working frequency band data stream, extracts frequency band characteristics and signal strength, and outputs standardized distance and azimuth data and standardized frequency band characteristic data, respectively.

[0017] Furthermore, the feature fusion module includes a data cache unit and an incremental update unit, the data cache unit stores the standardized distance and azimuth data and the standardized frequency band feature data at a preset frame interval, the incremental update unit calculates the distance difference and spectrum feature difference between each new sampling frame and the current state frame, updates the target state when the difference exceeds a preset threshold, outputs the real-time updated target distance state quantity and spectrum feature state quantity, and realizes the selective update of the target state.

[0018] Furthermore, the global defense module includes a signal generating unit, a real-time response unit and a defense strategy unit. The signal generating unit extracts the target working frequency band information in the spectrum characteristic state quantity and generates a swept frequency interference signal within the upper and lower ranges of the frequency band. The real-time response unit outputs a corresponding interference signal according to the selection of the defense strategy unit. The defense strategy unit selects a point frequency interference mode when the target signal strength is greater than a preset threshold, and selects a bandwidth interference mode when the target signal strength is less than the preset threshold.

[0019] Furthermore, the power regulation module receives the target real-time distance state quantity, calculates the interference required power based on the free space propagation loss model, sets the power margin value on this basis, and optimizes the transmission power of the interference response signal in real time by using a fixed power step-by-step increasing method.

[0020] The present invention also provides a UAV active defense method based on multi-sensor fusion, comprising the following steps:

[0021] S1: Acquire target features through multiple sensors, including obtaining target distance and azimuth data through radar sensors, and obtaining target operating frequency band data through electromagnetic sensors;

[0022] S2: using a dual-channel parallel architecture to perform data preprocessing, assigning the distance and azimuth data to a first processing channel for target space feature extraction, and assigning the working frequency band data to a second processing channel for electromagnetic feature extraction;

[0023] S3: Perform data fusion based on a preset frame interval, calculate feature differences between adjacent frames, update the target state when the feature difference exceeds a preset threshold, and generate real-time spatial state quantities and spectral state quantities;

[0024] S4: Select a defense strategy and a corresponding interference method according to the target signal strength. When the target signal strength is greater than a preset threshold, a point frequency interference method is used. When the target signal strength is less than a preset threshold, a bandwidth interference method is used. A corresponding interference response signal is generated within the target working frequency band.

[0025] S5: Calculate the required interference power in the current scenario based on the spatial propagation loss model, and optimize and adjust the output power of the interference response signal in real time by increasing the fixed power step size step by step.

[0026] The beneficial effects of the present invention are as follows:

[0027] (1) The present invention adopts a dual-channel parallel architecture for data processing, realizes independent parallel processing of spatial features and electromagnetic features, and significantly improves the data processing efficiency and real-time performance of the system. By configuring an independent FIFO queue for each channel, data congestion problems are effectively avoided.

[0028] (2) The present invention innovatively proposes a selective state update mechanism based on a frame interval threshold, which determines whether to update the target state by calculating the feature difference between the new sampling frame and the current state frame, effectively reducing the system computing overhead and improving the system resource utilization efficiency.

[0029] (3) The present invention designs an adaptive defense strategy, dynamically selects point-frequency interference or bandwidth interference according to the target signal strength, and optimizes and adjusts the interference power in combination with the spatial propagation loss model, thereby achieving a more accurate and efficient defense effect while avoiding unnecessary energy consumption.

[0030] (4) The present invention realizes comprehensive perception and accurate assessment of the target status through effective fusion and coordination of multi-sensor information, provides a reliable basis for defense decision-making, and improves the defense effectiveness and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is the architecture diagram of the drone active defense system.

[0032] Figure 2 It is the process layer flow chart.

[0033] Figure 3 Select a flowchart for your defense strategy.

[0034] Figure 4 Flowchart optimized for power. DETAILED DESCRIPTION

[0035] The following is a clear and complete description of the technical solutions in the embodiments of the invention in conjunction with the drawings in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments of the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0036] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0037] like Figure 1 As shown, the present invention discloses an active defense system for unmanned aerial vehicles based on multi-sensor fusion, including a perception layer, a processing layer and a defense layer. The system adopts a layered architecture design, and the signal connection between each layer is carried out through a standardized interface to achieve modular design and flexible expansion.

[0038] As a preferred embodiment of the present invention, the specific implementation of the perception layer is as follows:

[0039] The radar sensor module uses a phased array radar to detect the spatial characteristics of the target, with an operating frequency of 8-12GHz, and obtains the distance and azimuth data stream of the target in real time through the transmitting / receiving antenna array. Among them, the distance data stream contains the real-time distance value from the target to the system, which can reach an accuracy of 0.5m within the maximum detection range of 5km; the azimuth data stream contains the horizontal azimuth of the target relative to the system, and realizes 360° omnidirectional scanning with a resolution of 1°. The data stream is output at an update frequency of 100Hz to ensure real-time tracking of the target's motion state.

[0040] The electromagnetic sensing module obtains the target's working frequency band data stream through a broadband antenna array and a high-speed sampling unit. The module covers the 100MHz-6GHz frequency band and can simultaneously monitor multiple working frequency bands such as remote control and image transmission of the drone. A high-speed sampling rate of 2GSps is used to accurately capture the target working frequency band, and the sensitivity reaches -100dBm to ensure effective detection of weak signal targets. The working frequency band data stream contains characteristic parameters such as center frequency, bandwidth, signal strength, etc., and is output with an update cycle of 10ms to ensure timely response to changes in spectrum characteristics. Through the collaborative work of the two modules of the perception layer, the system can simultaneously obtain the spatial position information and electromagnetic characteristic information of the target.

[0041] As a preferred embodiment of the present invention, Figure 2 As shown, the specific implementation of the processing layer is as follows:

[0042] The parallel data processing module adopts a dual-channel parallel architecture to achieve efficient feature extraction. Among them, the data pipeline unit sets up independent FIFO queues to manage the data streams from the radar sensor and the electromagnetic sensor. Each queue is configured with a 32KB deep cache space, and the first-in-first-out principle is adopted to ensure the sequential access of data. The parallel processing unit sets up two independent processing channels: the first processing channel receives the distance and azimuth data streams of the radar sensor, realizes spatial positioning through the conversion of polar coordinates to rectangular coordinates, and uses the maximum and minimum value normalization method to standardize the distance data to the [0, 1] interval, and the azimuth data to the [-π, π] interval, and finally outputs the standardized distance and azimuth data; the second processing channel processes the working frequency band data stream of the electromagnetic sensor, uses 1024-point FFT transformation for spectrum analysis, extracts characteristic parameters such as signal bandwidth, center frequency and signal strength, and outputs standardized frequency band feature data after logarithmic normalization.

[0043] The feature fusion module includes a data cache unit and an incremental update unit, which realizes the selective update of the target state through a mechanism based on a frame interval threshold. The data cache unit temporarily stores the standardized distance and azimuth data and frequency band feature data at a preset frame interval of 10ms, and supports a cache capacity of up to 1000 frames of data. The incremental update unit makes a state update judgment by calculating the feature difference between each new sampling frame and the current state frame in real time: for spatial features, the Euclidean distance is calculated as the distance difference, and the threshold is set to 5m; for spectral features, the mean square error of the power spectrum density is calculated as the spectral feature difference, and the threshold is set to 3dB. When any difference exceeds the corresponding preset threshold, the state update mechanism is triggered, and the updated target real-time distance state quantity and spectral feature state quantity are generated and output based on the new sampling frame, and the shortest update cycle is not less than 50ms. This selective update mechanism can effectively reduce the system computing load, optimize the system resource utilization efficiency while ensuring real-time response capability.

[0044] As a preferred embodiment of the present invention, the specific implementation of the defense layer is as follows:

[0045] Reference Figure 3, the defense layer is equipped with a global defense module and a power regulation module to achieve adaptive interference defense against drone targets. The global defense module includes a signal generation unit, a real-time response unit, and a defense strategy unit. Among them, the signal generation unit can generate a sweep interference signal within the range of ±20MHz of the frequency band by extracting the target working frequency band information from the spectrum characteristic state quantity, thereby achieving effective interference to the target communication link. The real-time response unit is responsible for timely outputting the corresponding interference signal according to the selection of the defense strategy unit, and the response delay is less than 1ms to ensure the real-time nature of the defense. The defense strategy unit dynamically selects the interference mode based on the target signal strength: when the target signal strength is greater than the preset threshold -60dBm, the point frequency interference mode is selected, and the interference bandwidth is 1 / 10 of the target signal bandwidth to achieve precise strikes; when the signal strength is less than or equal to the preset threshold, the bandwidth interference mode is selected, and the interference bandwidth is twice the target signal bandwidth to ensure interference coverage.

[0046] Reference Figure 4 The power regulation module optimizes the output power of the interference response signal by receiving the real-time distance state of the target. This module calculates the required power for interference based on the free space propagation loss model. The formula is:

[0047] PL(d)=32.45+20log(f)+20log(d)

[0048] Among them, f is the working frequency band (the target working frequency in the range of 100MHz-6GHz), and d is the real-time distance of the target detected by the radar sensor (in the range of 0.5m-5km). Based on the calculated theoretical power, a 3dB power margin value is set to compensate for the additional loss in the actual propagation environment. A fixed power step of 0.5dB is used for step-by-step adjustment, and the maximum output power is limited to 50W, which not only ensures the interference effect but also avoids power waste. Through this real-time optimization mechanism, the system can dynamically adjust the interference power according to the change of the target distance to achieve accurate and efficient defense effect.

Claims

1. An active defense system for unmanned aerial vehicles based on multi-sensor fusion, characterized in that: It includes a perception layer, a processing layer, and a defense layer that are sequentially signal-connected; The perception layer includes a radar sensor module and an electromagnetic sensor module. The radar sensor module is used to obtain the target's distance and azimuth data stream, and the electromagnetic sensor module is used to obtain the target's working frequency band data stream; the processing layer includes a parallel data processing module and a feature fusion module. The parallel data processing module uses a dual-channel parallel architecture to extract features from the acquired data stream, and the feature fusion module selectively updates the target state based on the frame interval threshold to generate the target's real-time distance state quantity and spectrum feature state quantity; The defense layer is provided with a global defense module and a power regulation module. The global defense module selects an interference mode according to the target signal strength and generates a corresponding interference response signal. The power regulation module optimizes and regulates the output power of the interference response signal based on a spatial propagation loss model.

2. The UAV active defense system based on multi-sensor fusion according to claim 1 is characterized in that: The parallel data processing module includes a data pipeline unit and a parallel processing unit. The data pipeline unit sets independent FIFO queues for sequential access to the distance and azimuth data streams of the radar sensor and the working frequency band data streams of the electromagnetic sensor. The parallel processing unit correspondingly sets two independent processing channels. The first processing channel performs coordinate conversion and normalization processing on the distance and azimuth data streams, and the second processing channel performs spectrum analysis on the working frequency band data stream, extracts frequency band characteristics and signal strength, and outputs standardized distance and azimuth data and standardized frequency band characteristic data respectively.

3. The UAV active defense system based on multi-sensor fusion according to claim 1 is characterized in that: The feature fusion module includes a data cache unit and an incremental update unit. The data cache unit stores the standardized distance and azimuth data and the standardized frequency band feature data at a preset frame interval. The incremental update unit calculates the distance difference and spectrum feature difference between each new sampling frame and the current state frame. When the difference exceeds a preset threshold, the target state is updated and the real-time updated target distance state quantity and spectrum feature state quantity are output to achieve selective update of the target state.

4. The UAV active defense system based on multi-sensor fusion according to claim 1 is characterized in that: The global defense module includes a signal generating unit, a real-time response unit and a defense strategy unit. The signal generating unit extracts the target working frequency band information in the spectrum characteristic state quantity and generates a swept frequency interference signal within the upper and lower ranges of the frequency band. The real-time response unit outputs a corresponding interference signal according to the selection of the defense strategy unit. The defense strategy unit selects a point frequency interference mode when the target signal strength is greater than a preset threshold, and selects a bandwidth interference mode when the target signal strength is less than the preset threshold.

5. The UAV active defense system based on multi-sensor fusion according to claim 1 is characterized in that: The power regulation module receives the target real-time distance state quantity, calculates the interference required power based on the free space propagation loss model, sets the power margin value on this basis, and optimizes the transmission power of the interference response signal in real time by using a fixed power step-by-step increasing method.

6. An active defense method for drones based on multi-sensor fusion, characterized in that: The following steps are involved: S1: Acquire target features through multiple sensors, including obtaining target distance and azimuth data through radar sensors, and obtaining target operating frequency band data through electromagnetic sensors; S2: using a dual-channel parallel architecture to perform data preprocessing, allocating the distance and azimuth data to a first processing channel for target space feature extraction, and allocating the working frequency band data to a second processing channel for electromagnetic feature extraction; S3: Perform data fusion based on a preset frame interval, calculate feature differences between adjacent frames, update the target state when the feature difference exceeds a preset threshold, and generate real-time spatial state quantities and spectral state quantities; S4: Select a defense strategy and a corresponding interference method according to the target signal strength. When the target signal strength is greater than a preset threshold, a point frequency interference method is used. When the target signal strength is less than a preset threshold, a bandwidth interference method is used. A corresponding interference response signal is generated within the target working frequency band. S5: Calculate the required interference power in the current scenario based on the spatial propagation loss model, and optimize and adjust the output power of the interference response signal in real time by increasing the fixed power step size step by step.

Images