A radio detection method and system for a drone

Abstract

This application provides a radio detection method and system for unmanned aerial vehicles (UAVs), belonging to the field of UAV technology. The method includes: dividing the airspace to be detected into multiple scanning areas based on the refresh rate requirements of the airspace; receiving full-band radio frequency (RF) signals from the multiple scanning areas using an antenna array to obtain raw RF signals; cyclically switching the raw RF signals using an RF switch array to obtain a time-sequentially processed signal; preprocessing and extracting frequency domain features from the time-sequentially processed signal to obtain target RF features; adjusting the cyclic switching period and gain parameters of the RF switch array based on the target RF features using an intelligent switching strategy constructed based on signal strength feedback, and calculating the direction of arrival and distance of the detected target based on an amplitude ratio direction-finding algorithm; and determining the position and trajectory of the detected target by analyzing the target RF features over a preset continuous period. This application improves the accuracy of UAV detection.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a radio detection method and system for UAVs. Background Technology

[0002] With the popularization of the drone industry and the continuous expansion of its application scenarios, incidents of illegal drone flights without legal approval and without regulatory oversight are frequent, posing a serious threat to low-altitude public safety, aviation operation order, and the protection of important areas. The need for low-altitude security and drone control is becoming increasingly urgent. Radio detection technology, unaffected by lighting, weather, or terrain conditions, has become the mainstream technology for passive drone detection. However, traditional detection systems suffer from limitations such as insufficient sensitivity in detecting low-altitude, small, and slow-moving drone targets; difficulty in covering mainstream drone communication and remote control signals with a single frequency band; inability to adapt to the differentiated refresh rate requirements of areas with different security levels; high link loss in radio frequency switching; and poor timing consistency. Furthermore, traditional amplitude comparison and direction finding are significantly affected by channel errors and environmental attenuation, failing to meet the demands for high-precision, high-real-time drone detection. Summary of the Invention

[0003] To address the aforementioned technical problems, this application provides a radio detection method and system for unmanned aerial vehicles (UAVs).

[0004] A first aspect of this application provides a radio detection method for an unmanned aerial vehicle (UAV), comprising: The space to be detected is divided into multiple scanning areas based on the different refresh rate requirements of the space to be detected. The original radio frequency signal is obtained by receiving full-band radio frequency signals in multiple scanning areas through an antenna array; The original radio frequency signal is cyclically switched using a radio frequency switch array to obtain a time-sequential serial processed signal. The time-series serialized signal is preprocessed and frequency domain features are extracted to obtain target radio frequency features that characterize the radio frequency signal attributes of the detection target. Based on the target's radio frequency characteristics, an intelligent switching strategy based on signal strength feedback is used to adjust the cyclic switching period and gain parameters of the radio frequency switch array, and the direction of arrival and distance of the detected target are calculated based on the amplitude ratio direction finding algorithm. By analyzing the target's radio frequency characteristics over a preset continuous period, the real-time position and flight trajectory of the target are obtained through fitting.

[0005] A second aspect of this application provides a radio detection system for an unmanned aerial vehicle (UAV), comprising: The airspace partitioning planning module is used to divide the airspace to be detected into multiple scanning areas based on different refresh rate requirements of the airspace to be detected. The radio frequency signal receiving module is used to receive full-band radio frequency signals in multiple scanning areas through an antenna array to obtain the original radio frequency signal; The signal switching and processing module is used to perform cyclic switching processing on the original radio frequency signal using a radio frequency switch array to obtain a time-sequential serial processed signal. The feature extraction and processing module is used to perform signal preprocessing and frequency domain feature extraction on the time-series serial processing signal to obtain target radio frequency features that characterize the radio frequency signal attributes of the detection target; The parameter adjustment and ranging module is used to adjust the cyclic switching period and gain parameters of the radio frequency switch array based on the target radio frequency characteristics and an intelligent switching strategy based on signal strength feedback, and to calculate the direction of arrival and distance value of the detected target based on the amplitude ratio direction finding algorithm. The position trajectory fitting module is used to fit the real-time position and flight trajectory of the target by analyzing the target's radio frequency characteristics over a preset continuous period.

[0006] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described radio detection method for a drone.

[0007] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the radio detection method for unmanned aerial vehicles described above.

[0008] The beneficial effects of the radio detection method and system for UAVs provided in this application are as follows: This application achieves differentiated priority scanning by dividing the airspace to be detected according to different refresh rate requirements, thereby improving the detection response speed and resource utilization efficiency in sensitive areas; and based on the cyclic switching processing of antenna array and RF switch array, while ensuring full-band signal reception coverage, it converts parallel signals into time-sequential serial processing signals, reducing hardware complexity and cost; by using target RF features extracted through signal preprocessing and frequency domain feature extraction, and an intelligent switching strategy based on signal strength feedback, the cyclic switching period and gain parameters of the RF switch array are adjusted, improving the system's adaptability to complex electromagnetic environments and target signal changes; and based on the amplitude ratio direction finding algorithm, the direction of arrival and distance values ​​are accurately calculated, effectively improving the real-time performance, accuracy, and environmental adaptability of the detection system, further enhancing the accuracy of UAV detection. Attached Figure Description

[0009] Figure 1A schematic flowchart of a radio detection method for an unmanned aerial vehicle provided in an embodiment of this application; Figure 2 A structural block diagram of a radio detection system for an unmanned aerial vehicle (UAV) provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0010] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0011] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1-3 The following is an explanation using specific examples.

[0012] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a radio detection method for an unmanned aerial vehicle (UAV) according to an embodiment of this application. The method includes: S101: The airspace to be detected is divided into multiple scanning areas based on the different refresh rate requirements of the airspace to be detected.

[0013] In this embodiment, firstly, a refresh rate threshold for the airspace to be detected is determined based on a preset security level and airspace sensitivity. Secondly, based on the refresh rate threshold, the beamwidth of the antenna array, and the scanning step angle, the airspace to be detected is discretized to obtain several spatial grid cells. Finally, adjacent spatial grid cells with the same refresh rate or a difference of less than a preset first threshold are aggregated to obtain multiple scanning regions with different scanning priorities.

[0014] One method in this embodiment further includes: statistically analyzing the target access probability and average radial motion velocity of each spatial grid unit at different time periods based on historical detection data to construct a spatiotemporal distribution map of the airspace situation; adjusting the detection refresh rate weights of each spatial grid unit using a reinforcement learning algorithm based on the spatiotemporal distribution map of the airspace situation; clustering and fusing grid units whose differences between detection refresh rate weights are less than a preset refresh rate weight threshold and which are spatially adjacent to each other to generate a scanning area division scheme that adapts to time; and re-dividing the airspace to be detected based on the scanning area division scheme to obtain multiple scanning areas. Among them, historical detection data refers to the target signals, positions, times, and motion states that have been actually detected and recorded over a period of time; spatial grid cells are the smallest spatial units into which the entire airspace to be detected is discretized according to dimensions such as angle and distance; target entry probability refers to the likelihood of a detection target such as a UAV appearing in a certain spatial grid cell within a certain time period; average radial velocity refers to the average velocity of the detection target relative to the detection system in the direction of approach or departure; detection refresh rate weight represents the numerical value of the detection importance and scanning frequency priority of the corresponding spatial grid cell; refresh rate weight threshold is a pre-set value used to determine whether the detection refresh rate weights of two grids are close enough to be merged into the same scanning area.

[0015] This embodiment uses a reinforcement learning algorithm to dynamically and intelligently optimize the detection refresh rate weights of each spatial grid cell based on the real-time airspace situation. The algorithm takes the constructed spatiotemporal distribution map of the airspace situation as input, the detection refresh rate weights of each spatial grid cell as adjustable actions, and aims to obtain higher detection priority for areas with high target entry probability and high average radial velocity, while achieving optimal overall detection resource utilization. Through continuous interaction and trial and error with historical detection data and real-time airspace situation, it autonomously learns and outputs the optimal weight allocation strategy.

[0016] S102: Receives full-band radio frequency signals from multiple scanning areas through an antenna array to obtain the original radio frequency signal.

[0017] In this embodiment, an antenna array is used to acquire signals across multiple scanning areas, achieving synchronous coverage of areas with different priorities. The antenna array can operate simultaneously on the frequency bands commonly used for UAV remote control, image transmission, and data link, without frequent switching of operating frequency bands. Through this multi-area parallel reception method, electromagnetic signals within all scanning areas can be acquired at once.

[0018] When the antenna array is in operation, it continuously receives radio frequency (RF) signals emitted by drones and other radiation sources in different scanning areas, converting electromagnetic waves in space into electrical signals to form raw RF signals. These raw RF signals include key information such as the amplitude, frequency, and phase of the target signal, but are also mixed with environmental noise, electromagnetic interference, and background signals. The antenna array itself is a receiving array composed of multiple antenna elements, operating in frequency bands commonly used for drone remote control and image transmission, and can simultaneously receive RF signals from multiple directions and scanning areas. Frequency band RF signals refer to all RF signals used in remote control, communication, and image transmission during operation, and are the primary targets for target detection and identification.

[0019] S103: The original radio frequency signal is cyclically switched using an RF switch array to obtain a time-sequential serial processed signal.

[0020] In this embodiment, an RF switch array is used to orderly switch multiple parallel signals, achieving the integration and scheduling of multi-channel signals. The antenna array includes multiple receiving channels. Connecting all of them to the back-end processing unit would increase hardware complexity and cost. Therefore, an RF switch array is needed to select the path of multiple raw RF signals in turn, simplifying the receiving architecture while ensuring signal integrity. The cyclic switching process selects different antenna channels sequentially according to a preset timing sequence, arranging the multiple parallel signals in time. The preset timing sequence is a cyclic scanning timing sequence pre-planned by allocating dwell time sequentially according to priority from high to low, dynamically judging the remaining time of the cycle, and performing time slot interpolation on the area to be inserted.

[0021] The RF switch array switches channels at a fixed or adaptively adjusted period, selecting only one original RF signal at each switching moment to enter the processing link. This periodic polling process covers all channels. This process converts spatially parallel multi-channel signals into a single data stream arranged in time sequence. After cyclic switching, the originally parallel multiple original RF signals are transformed into a time-sequential serial processed signal with a clear time segmentation and channel correspondence. This time-sequential serial processed signal can be connected to a single processing link for digitization and analysis, effectively reducing the requirement for the number of hardware channels and improving resource utilization.

[0022] S104: Perform signal preprocessing and frequency domain feature extraction on the time-series serially processed signal to obtain target radio frequency features that characterize the radio frequency signal attributes of the detection target.

[0023] In this embodiment, the time-series serial signal is preprocessed and frequency domain features are extracted to obtain target RF features that can fully characterize the RF signal attributes of the target being detected. For example, firstly, based on the region number, channel number, and timestamp information carried in the signal, the continuous time-series serial data stream is deserialized, segmented, and aligned according to the scanning area and antenna channel, and the independent time-domain IQ signal for each region and channel is recovered. Abnormalities such as missing or truncated sampling points are filled using linear interpolation or edge-preserving methods. Subsequently, the segmented signal is preprocessed, sequentially performing DC offset elimination, bandpass filtering, and automatic gain control to suppress zero-frequency components and out-of-band interference, ensuring the signal amplitude is within the optimal processing range. For noise and random interference introduced by the RF receiving link, a wavelet thresholding method is used. The signal is decomposed into three levels using a db4 wavelet basis, and an adaptive threshold is calculated based on the noise standard deviation and signal length. The high-frequency noise components are then soft-thresholded and the signal is reconstructed, improving the signal-to-noise ratio while preserving the amplitude and phase characteristics of the target signal. Then, using the pre-calibrated antenna array amplitude and phase calibration coefficients, amplitude compensation and phase correction are performed on the signals of each channel to eliminate system errors caused by channel inconsistency.

[0024] After preprocessing, the calibrated time-domain signal is windowed to reduce spectral leakage, and then converted to the frequency domain using a Fast Fourier Transform (FFT) to obtain the signal's amplitude and phase spectra. Based on the frequency domain results, target radio frequency (RF) features are extracted, including center frequency, signal bandwidth, peak power, average power, signal-to-noise ratio (SNR), and amplitude and phase values ​​of each antenna channel at the signal frequency, forming a multi-dimensional feature vector. To ensure feature stability, the frequency domain results of multiple consecutive frames are subjected to moving averages and outlier removal to eliminate transient interference and outlier effects. Finally, the extracted frequency domain features, channel amplitude and phase information, region number, and timestamp are jointly encapsulated to form the target RF feature. The target in this case refers to a drone.

[0025] S105: Based on the target's RF characteristics, an intelligent switching strategy based on signal strength feedback is used to adjust the cyclic switching period and gain parameters of the RF switch array, and the direction of arrival and distance of the detected target are calculated based on the amplitude ratio direction finding algorithm.

[0026] In this embodiment, the received power value corresponding to the signal strength indication value is obtained, and path loss is calculated on the received power value to obtain the effective received power. Based on the direction of arrival, a multipath attenuation correction coefficient corresponding to the direction of arrival is obtained by matching with a preset spatial environment attenuation factor database. The effective received power is then corrected by environmental compensation using the multipath attenuation correction coefficient to obtain the corrected received power. The corrected received power and the typical transmit power range of the detection target are input into a pre-trained logarithmic distance path loss model to obtain the path loss value. The path loss value is then substituted into the propagation loss equation to obtain the distance between the detection target and the detection system.

[0027] S106: By analyzing the target radio frequency characteristics of a preset continuous period, the real-time position and flight trajectory of the detected target are obtained through fitting.

[0028] In this embodiment, real-time target position calculation and flight trajectory fitting are achieved by performing time-series analysis on the target radio frequency features within a preset continuous period. Specifically, firstly, based on the timestamp, the target radio frequency features within multiple consecutive scanning periods are correlated and matched. Target identity is confirmed based on stable features such as center frequency, signal bandwidth, and amplitude distribution, different detected targets are distinguished, and false points and isolated interference are eliminated, ensuring that the data involved in trajectory calculation all come from the same real target. Based on this, the target arrival direction and distance values ​​calculated in each period are converted into three-dimensional spatial coordinates in a unified coordinate system, forming a discrete set of target position points at continuous moments. The preset continuous period is determined comprehensively based on the system-set scanning period, the target's minimum tracking frame rate, the typical UAV movement speed, and trajectory smoothness requirements. Simultaneously, it is adaptively adjusted based on the current signal-to-noise ratio and target distance to ensure trajectory continuity and reasonable use of computational resources.

[0029] To improve position accuracy and trajectory smoothness, filtering algorithms are used to optimize discrete position points. For example, Kalman filtering is used to suppress measurement noise, and a uniform or uniformly accelerated motion model is established based on the target's motion characteristics to predict and correct the target position, resulting in a smoother position sequence that more closely resembles the actual motion state. Subsequently, based on multiple consecutive valid position points, polynomial fitting or least squares fitting methods are used to construct the target's motion trajectory, and the target motion trajectory parameters are updated in real time using a sliding window approach, allowing the output target motion trajectory to dynamically follow changes in the target's attitude and motion state.

[0030] Finally, the system integrates the real-time calculated position information with the fitted target trajectory and outputs the target's real-time coordinates, velocity, heading angle, and predicted trajectory simultaneously.

[0031] As one possible embodiment, the radio detection method for UAVs is applied to a radio detection device for UAVs, which includes: a high-gain directional antenna array, a radio frequency switch array, a signal processing unit, a control unit, and a power management unit; wherein, the control unit controls the radio frequency switch array to cyclically switch different antenna elements, so that each antenna element in the high-gain directional antenna array takes turns connecting to the signal processing unit, thereby improving the detection speed and detection range of the system, while reducing the size of the device.

[0032] A high-gain directional antenna array, consisting of multiple high-gain directional antenna elements, is used to receive radio frequency signals transmitted by low-altitude drones; An RF switch array, connected to a high-gain directional antenna array, is used to cyclically switch the various antenna elements in the high-gain directional antenna array. The signal processing unit is connected to the radio frequency switch array and is used to process the radio frequency signals output by the radio frequency switch array; The control unit, connected to the radio frequency switch array and the signal processing unit, is used to control the switching timing of the radio frequency switch array and the operating parameters of the signal processing unit; it is also used to execute a radio detection method for a drone.

[0033] The power management unit provides a stable power supply to all parts of the system.

[0034] Specifically, the control unit switches the RF switch array according to a preset timing sequence, allowing each antenna element in the high-gain directional antenna array to take turns accessing the receiving channel. The switching timing can employ a polling method or an intelligent switching method based on signal strength. Within each switching cycle, the signal processing unit processes the RF signal received by the currently operating antenna, including low-noise amplification, filtering, down-conversion, and A / D conversion. The processed digital signal is converted into a frequency domain signal using a Fast Fourier Transform (FFT), and spectral analysis and signal detection are performed based on the frequency domain signal. Based on the signal detection results, the control unit adjusts the switching strategy and gain control parameters in real time. If a drone signal is detected, the switching speed is adjusted based on the signal strength to increase the detection frequency, or the gain of the relevant antennas is adjusted to enhance the signal strength. Simultaneously, by comparing the signal strengths received by different antennas, the direction of arrival of the drone can be determined.

[0035] In this embodiment, the RF switch array is connected to the high-gain directional antenna array for cyclically switching between the antenna elements in the high-gain directional antenna array. The RF switch array adopts an N×M matrix structure, where N is the number of antennas and M is the number of receive channels. For example, the RF switch array adopts a 12×2 matrix structure, consisting of two 12-to-1 RF switches and a switching control circuit. Each RF switch uses GaAsFET technology, with an insertion loss of less than 0.3dB and an isolation greater than 50dB in the operating frequency band. The switching speed is 50 nanoseconds, enabling fast switching. The RF switch array also integrates a 12-channel broadband low-noise amplifier, with each channel providing 20dB of gain.

[0036] The signal processing unit is connected to the RF switch array and is used to process the RF signals output by the RF switch array. The signal processing unit includes a dual-channel receiver and a digital signal processing module. Each channel of the dual-channel receiver consists of a wideband low-noise amplifier, a filter, a mixer, an intermediate frequency amplifier, and a frequency synthesizer. The intermediate frequency is set to 312.5MHz, and the intermediate frequency bandwidth is 80MHz. The digital signal processing module adopts an FPGA+DSP architecture. The FPGA is responsible for real-time signal processing such as A / D acquisition, digital down-conversion, and FFT calculation, while the DSP is responsible for signal detection, parameter estimation, and target recognition.

[0037] The control unit connects to the RF switch array and signal processing unit, controlling the switching timing of the RF switch array and the operating parameters of the signal processing unit. The control unit uses an ARM processor and is responsible for the timing control, parameter adjustment, and data processing of the entire system. The control unit communicates with the RF switch array via an SPI interface to achieve antenna switching control. Simultaneously, the control unit is also responsible for communication with the host computer, sending the detection results to the monitoring center via a network port or 4G module.

[0038] In this embodiment, the antenna array is a high-gain directional antenna array composed of multiple high-gain directional antennas, connected to an RF switch array via RF cables. The high-gain directional antenna array employs log-periodic antennas or microstrip patch antennas, operating at frequencies from 70MHz to 6GHz to cover the frequency bands used by various drones. The antenna array adopts a circular or square layout, with each antenna element having a gain of over 14dB and a beamwidth of approximately 3dB at around 30°. By cyclically switching different antenna elements through the RF switch array, a 360° omnidirectional coverage detection effect is achieved. For example, the high-gain directional antenna array uses 12 log-periodic antenna elements.

[0039] Upon receiving the start command, this embodiment enters detection mode and performs cyclic detection according to a preset workflow. The specific workflow includes: The control unit controls the RF switch array according to a preset timing sequence, sequentially connecting the 12 antenna elements to the receiving channel. The switching period is set to 10 milliseconds, meaning each antenna element operates for 10 milliseconds. The RF signal received by the currently operating antenna passes through the RF switch array and enters the dual-channel receiver. The signal is first amplified by a low-noise amplifier, then filtered by a bandpass filter to remove out-of-band interference, then down-converted to an intermediate frequency (IF) by a mixer, and finally amplified by the IF before being sent to an A / D converter. The A / D converter has a sampling rate of 500MHz, converting the analog signal to a digital signal. The digital signal first undergoes digital down-conversion and decimation filtering to reduce the data rate. Then, an FFT transform is performed to convert the time-domain signal to a frequency-domain signal. Spectrum analysis is used to detect the presence of a drone signal. If a signal is detected, further analysis is performed on characteristic parameters such as the carrier frequency, pulse width, and repetition frequency. Based on the differences in signal strength received by different antennas, an amplitude-based direction-finding algorithm is used to estimate the drone's direction of arrival. Simultaneously, the drone's distance is estimated based on the signal strength and antenna gain information. By analyzing signals over multiple consecutive cycles, the speed and flight trajectory of the UAV can be estimated. The detection results, including target type, location, and speed information, are packaged into a standard format and sent to the monitoring center via Ethernet or a 4G module. Simultaneously, real-time detection results, including spectrum graphs and target trajectories, are displayed on a local screen. After processing one antenna element, the system immediately switches to the next antenna element and repeats the detection process, achieving continuous scanning and detection of the 360° airspace.

[0040] This embodiment adopts a modular design concept, dividing the entire system into multiple functional modules such as an antenna array module, an RF switch array module, a signal processing module, and a control module. The modules are connected via standard interfaces, facilitating production, debugging, and maintenance.

[0041] A multi-layer PCB design integrates RF circuits, digital circuits, and control circuits onto a single circuit board. Through rational layout and routing, signal transmission paths are reduced, minimizing electromagnetic interference. Simultaneously, surface mount technology is employed to improve integration and reduce circuit board size.

[0042] The system incorporates heat sinks in high-power components such as power amplifiers and FPGAs, dissipating heat through natural convection or forced air cooling. Additionally, ventilation holes are designed into the system casing to ensure timely heat dissipation.

[0043] The system employs a shielded design, with shielding covers installed at critical locations to prevent mutual interference between internal circuits. Simultaneously, input and output interfaces are filtered to prevent external interference from entering the system.

[0044] As can be seen from the above, this application achieves differentiated priority scanning by dividing the airspace to be detected according to different refresh rate requirements, thereby improving the detection response speed and resource utilization efficiency in sensitive areas. Furthermore, the cyclic switching processing based on the antenna array and RF switch array ensures full-band signal reception coverage while converting parallel signals into time-sequential serial processing signals, reducing hardware complexity and cost. By using target RF features extracted through signal preprocessing and frequency domain feature extraction, and an intelligent switching strategy based on signal strength feedback, the cyclic switching period and gain parameters of the RF switch array are adjusted, improving the system's adaptability to complex electromagnetic environments and changing target signals. Finally, the accurate calculation of the direction of arrival and distance values ​​based on the amplitude ratio direction finding algorithm effectively improves the real-time performance, accuracy, and environmental adaptability of the detection system, further enhancing the accuracy of UAV detection.

[0045] In one embodiment of this application, the spatial domain to be detected is divided based on different refresh rate requirements to obtain multiple scanning regions, including: Based on the preset security level and airspace sensitivity, determine the detection refresh rate threshold of the airspace to be detected; Based on the refresh rate threshold, the beamwidth of the antenna array, and the scanning step angle, the spatial domain to be detected is discretized to obtain several spatial grid cells. Adjacent spatial grid cells with the same or a difference of less than a preset first threshold are aggregated to obtain multiple scanning regions with different scanning priorities.

[0046] In this embodiment, a hierarchical mapping mechanism is first established based on preset security levels and airspace sensitivity to determine the detection refresh rate threshold for the airspace to be detected. Security levels are divided into high, medium, and low tiers. Airspace sensitivity is quantified and assigned values ​​based on airspace functional attributes, control requirements, and the distribution of key protected areas, forming a sensitivity-refresh rate correspondence table to determine different refresh rates for areas with different sensitivity levels. The preset security levels are categorized into high, medium, and low based on the importance of the protected area, control level, and risk level, and are fixed in configuration. Airspace sensitivity is quantified and assigned values ​​based on airspace function, no-fly zone requirements, population density, distribution of key targets, and historical intrusion risk. The corresponding detection refresh rate threshold is determined by the pre-established hierarchical mapping table. The hierarchical mapping table establishes a one-to-one correspondence between security level, airspace sensitivity, and refresh rate threshold by dividing security levels into high, medium, and low tiers and airspace sensitivity into high, medium, and low levels based on the distribution of key targets, no-fly zone levels, population density, and risk level.

[0047] Based on the refresh rate threshold, the antenna array beamwidth, and the scanning step angle, the spatial domain to be detected is regularly discretized to generate a series of spatial grid cells. During the partitioning, it is ensured that the coverage area of ​​each spatial grid matches the antenna beamwidth, and the scanning step angle is less than or equal to the beamwidth at half power.

[0048] Spatial grid cells that are spatially adjacent and have the same refresh rate or a difference less than a preset first threshold are aggregated into several continuous scanning regions, and assigned corresponding scanning priorities based on their refresh rates. During the aggregation of adjacent spatial grid cells, a region growing method is used, gradually merging adjacent grids of the same level using high refresh rate grids as seeds, ensuring consistent refresh rates and regular boundaries within each scanning region. This ultimately results in multiple scanning regions with different scanning priorities. High refresh rates correspond to high security levels and high spatial sensitivity, representing core areas that the system needs to prioritize and continuously track. Growing outwards from these core areas ensures the integrity of these key regions and prevents them from being fragmented or scattered by low-priority grids. The preset first threshold is determined through experimental calibration and simulation optimization based on the system's allowable refresh rate deviation, scanning timing uniformity, and trajectory continuity requirements after region aggregation.

[0049] From the above, it can be concluded that this embodiment, by dynamically setting the detection refresh rate threshold based on security level and airspace sensitivity, can reasonably allocate detection resources according to actual protection needs. Furthermore, by discretizing the airspace into a grid based on antenna beamwidth and scanning step angle, it ensures uniform airspace coverage and eliminates detection blind spots, while simultaneously improving the accuracy of matching the detection performance of each grid unit with hardware capabilities. By aggregating adjacent grids with similar refresh rates to form scanning areas with differentiated scanning priorities, it reduces the overhead of frequent switching and enables high-frequency, high-reliability monitoring of key areas. This improves overall detection efficiency while enhancing adaptability to complex security scenarios, resource utilization, and target tracking continuity.

[0050] In one embodiment of this application, a radio detection method for a drone further includes: Based on the refresh rate requirements of each scanning area, calculate the percentage of theoretical scans for each scanning area in a single complete scanning cycle, and determine the dwell time weight based on the percentage of theoretical scans. The minimum dwell time unit for each scanning region is obtained based on the beam switching time of the antenna array, the setup time of the RF switch array, and the Fourier transform period of the signal processing unit. Based on the dwell time weight and the minimum dwell time unit, calculate the dwell time required for each scan region in one scan cycle; Based on the priority order of each scanning region, the required dwell time of each scanning region within the preset scanning cycle is sequentially filled into the scanning timing table to generate a cyclically executed scanning timing.

[0051] In this embodiment, the theoretical percentage of scans for each scan region within a single complete scan cycle is first calculated based on the refresh rate requirements of each scan region. This percentage serves as the basis for allocating dwell time weights. Specifically, the refresh rate of each scan region is divided by the sum of the refresh rates of all regions to obtain the theoretical percentage of scans for the corresponding region. This percentage is then normalized and used as the dwell time weight.

[0052] Based on this, the minimum dwell time unit for each scanning region is determined by taking the maximum value of the beam switching time of the antenna array, the channel establishment and stabilization time of the RF switch array, and the minimum period required for the signal processing unit to complete one Fourier transform. Using the minimum dwell time unit as the basic time slice, and according to the dwell time weight of each region and the total scanning period, the dwell time that should be allocated to each scanning region in one period is calculated.

[0053] Finally, based on the scanning area priority from high to low, the required dwell time for each area is sequentially filled into the pre-built scanning timing table. The time slices are then arranged and spliced ​​in an orderly manner to form a regular scanning timing sequence that can be executed cyclically. A certain amount of redundant time intervals is reserved during the timing allocation process to cope with channel switching jitter, sudden signal anomalies, and real-time parameter adjustments.

[0054] From the above, it can be concluded that this embodiment achieves on-demand allocation of detection resources by calculating the proportion of scan times and dwell time weights according to refresh rate requirements; by determining the minimum dwell time unit based on the hardware delays of the antenna, RF switch, and signal processing, the integrity and reliability of the signal reception and processing process can be guaranteed, avoiding data anomalies caused by excessively short timing; by calculating the dwell time of each region based on the dwell time weights and the minimum time unit, and generating a cyclic scanning timing sequence according to priority, it can not only meet the refresh rate requirements of different regions, but also optimize the timing arrangement and reduce switching overhead and timing conflicts.

[0055] In one embodiment of this application, based on the priority order of each scanning region, the required dwell time of each scanning region within a preset scanning cycle is sequentially filled into a scanning timing table to generate a cyclically executed scanning timing, including: All scanned areas are sorted in descending order of refresh rate requirements to generate a priority list; Initialize a blank scan timing table and set the current time pointer to the beginning of the timing table; Select the highest priority scan region from the priority list in sequence, and determine whether the remaining time of the current complete scan cycle is greater than the dwell time required to insert the highest priority scan region. If the judgment result is yes, then fill the dwell time required for the highest priority scanning area into the current time pointer position, and move the time pointer forward by the corresponding dwell time length; If the result is negative, the remaining unscheduled time of the current highest priority scan area is marked as pending insertion, and the current highest priority scan area is skipped, and the next priority scan area is processed. After completing one round of priority list traversal, if there are still scan areas waiting to be inserted, the beam switching idle interval of each scan area is compressed, or the scan area is inserted in the reserved time slot of the next scan cycle, until the dwell time of all scan areas is allocated and the scan timing is generated.

[0056] In this embodiment, each scanning region is first sorted from highest to lowest refresh rate requirement, with higher refresh rates indicating higher priority, thus generating an ordered priority list. Then, a blank scanning timing table with a length equal to a preset complete scanning cycle is initialized, and the time pointer is initialized to the starting position of the timing table as the starting point for dwell time allocation, establishing a unified time reference for subsequent region-by-region and time-slot-by-time timing arrangement. The preset complete scanning cycle is determined by the refresh rate of the highest priority scanning region, the minimum switching cycle of the RF switch, and the minimum signal processing time, taking the minimum common cycle that satisfies the refresh rate constraints of all regions.

[0057] The system sequentially selects the highest-priority scan regions from the head of the priority list, calculates the remaining available time within the current scan cycle in real time, and compares this remaining available time with the required dwell time for the highest-priority scan region. For example, if the remaining time is sufficient to accommodate the required dwell time for the highest-priority scan region, the corresponding dwell time is entered into the current position of the time pointer, and the time pointer is moved forward by the corresponding length, completing the timing allocation for this region. If the remaining time is insufficient, the unallocated remaining available time for that region is marked as pending insertion, temporarily skipped, and the process moves to the next priority region.

[0058] In this embodiment, after completing a full priority list traversal, it checks whether there are still scan regions waiting to be inserted. If there are unallocated regions, available time slots are explored by compressing the idle interval during beam switching and optimizing the reserved time slots for switch switching delays, and these slots are preferentially inserted into the remaining time slots of the current cycle. If there are no available time slots in the current cycle, the time waiting to be inserted is accumulated to the next scan cycle, and the insertion is completed using the reserved protection time slots, until the dwell time of all scan regions meets the allocation requirements. The reserved protection time slot is a redundant period of time specifically reserved within the complete scan cycle to compensate for unexpected time consumption such as beam switching jitter, timing insertion, and parameter adaptive adjustment, and to avoid timing overflow or insufficient region dwell time.

[0059] From the above, it can be concluded that this embodiment, by constructing a priority list based on refresh rate from high to low, can ensure that high-priority, high-security scanning areas receive timing resources first, guaranteeing the real-time detection of key airspaces. By using a time pointer and the remaining available time of the cycle to determine dwell time allocation, time waste and conflicts are avoided. For areas that cannot be scheduled in the current time slot, a pending insertion mark is used, and the remaining time allocation is completed by compressing and switching idle intervals or inter-cycle insertion. This ensures that all scanning areas can obtain dwell time that meets the refresh rate requirements, and also improves the utilization rate of the scanning cycle and the robustness of timing scheduling.

[0060] In one embodiment of this application, based on the target radio frequency characteristics, an intelligent switching strategy using signal strength feedback is employed to adjust the cyclic switching period and gain parameters of the radio frequency switch array, and the direction of arrival and distance of the detected target are calculated based on an amplitude ratio direction finding algorithm, including: Extract the signal strength indication value from the target radio frequency characteristics, and compare the signal strength indication value with a preset second threshold and a preset third threshold; When the signal strength indication value is greater than the second threshold, the gain parameter of the RF switch array is reduced through an intelligent switching strategy, and the cycle switching period is shortened accordingly. When the signal strength indication value is less than the third threshold, the gain parameter of the RF switch array is increased through an intelligent switching strategy, and the cyclic switching period is extended accordingly. Using the adjusted cycle switching period and gain parameters, the timing signal amplitude corresponding to each channel of the RF switch array is obtained; Based on the time-series signal amplitude corresponding to each channel, the amplitude comparison direction finding algorithm is adopted. By comparing the signal amplitude difference between adjacent antenna channels, the algorithm searches the preset antenna pattern database and calculates the direction of arrival of the target wave. The future wave direction and signal strength indication value are integrated, and the distance between the target and the detection system is obtained based on the signal propagation attenuation model.

[0061] In this embodiment, a signal strength indicator value is first extracted from the target radio frequency characteristics. This signal strength indicator value is calculated jointly by the average signal power of multiple channels and the signal-to-noise ratio, and instantaneous fluctuations are removed by continuous multi-frame moving average. The smoothed signal strength indicator value is then compared in segments with preset second thresholds (upper limit for strong signals) and third thresholds (lower limit for weak signals) to form the basis for adaptive adjustment of gain and switching cycle.

[0062] When the signal strength indication value is greater than the second threshold, it is determined to be a strong signal scenario. The intelligent switching strategy automatically reduces the gain parameters of the RF switch array and the receiving link to suppress link saturation and shortens the cycle switching period proportionally to improve the area refresh rate and target tracking real-time performance. When the signal strength indication value is less than the third threshold, it is determined to be a weak signal scenario. The receiving gain is increased accordingly to improve receiving sensitivity, while the cycle switching period is appropriately extended. The second and third thresholds are determined through experimental calibration based on the linear dynamic range of the receiving link, the saturation power point, the noise floor, and the minimum detectable signal power required by the system. The second threshold is set as the upper limit power value of the linear operating range of the receiving link to prevent signal saturation distortion. The third threshold is set as the minimum reliable signal power value that is greater than the system noise floor and meets the detection sensitivity requirements to ensure effective detection and direction finding under weak signal conditions.

[0063] The RF switch array is redriven using the adjusted cycle switching period and gain parameters. The timing-sequential signal amplitudes of each antenna channel at the same time and under the same direction of arrival are acquired sequentially according to the scanning timing sequence, and amplitude and phase error compensation is performed. Based on the calibrated multi-channel amplitude values, an amplitude comparison direction-finding algorithm is used to calculate the signal amplitude difference and amplitude ratio between adjacent antenna channels. The measured results are matched against a pre-calibrated antenna pattern database. The azimuth and elevation angles of the target are calculated through minimum error interpolation, thus achieving accurate estimation of the direction of arrival.

[0064] Finally, based on the fusion of the incoming wave direction and the real-time signal strength indication value, and based on the signal center frequency, antenna gain, feeder loss and free space propagation attenuation model, a mapping relationship from received power to target distance is constructed. After eliminating interference and multipath effects, the radial distance value between the target and the detection system is calculated, and the distance output is smoothed by continuous frame filtering.

[0065] From the above, it can be concluded that this embodiment achieves adaptive adjustment of the RF switch array gain and cycle switching period by extracting the signal strength indication value and comparing it with multi-level thresholds. This can avoid receiving link saturation caused by strong signals and improve receiving sensitivity and stability. Based on the adjusted parameters, the timing signal amplitude of each channel is obtained, and the incoming wave direction is calculated according to the amplitude comparison direction finding algorithm and the antenna pattern database, which effectively improves the direction finding accuracy and anti-interference capability. Finally, the incoming wave direction and signal strength are fused and the distance is calculated using the signal propagation attenuation model, realizing the joint estimation of target angle and radial distance, which improves the overall adaptability, accuracy and reliability of detection.

[0066] In one embodiment of this application, the future wave direction and signal strength indication value are integrated, and the distance between the detection target and the detection system is obtained based on a signal propagation attenuation model, including: Obtain the received power value corresponding to the signal strength indication value, and calculate the path loss of the received power value to obtain the effective received power; Based on the matching between the direction of arrival and the preset spatial environment attenuation factor database, the multipath attenuation correction coefficient corresponding to the direction of arrival is obtained. The effective received power is corrected by environmental compensation using a multipath attenuation correction coefficient to obtain the corrected received power. The corrected received power and the typical transmit power range of the target are input into the pre-trained logarithmic distance path loss model to obtain the path loss value. Substituting the path loss value into the propagation loss equation, we obtain the distance between the target and the detection system.

[0067] In this embodiment, the signal strength indication value is first obtained from the target RF characteristics and converted into a corresponding received power value. By performing mean filtering and outlier removal on multiple frames of continuous data, transient interference and noise fluctuations are suppressed to obtain a stable and reliable received power estimate. Then, the received power value is calibrated according to parameters such as antenna gain, feeder loss, and filter insertion loss. Fixed losses generated by the hardware link are deducted to obtain the effective received power. The method for calibrating the received power value includes: first converting the signal strength indication value into a nominal received power value, then successively subtracting the fixed loss values ​​such as antenna feeder loss, filter insertion loss, cable attenuation, and connector loss (factory calibration or actual measurement), and adding the received gain of the corresponding antenna element to finally obtain the effective received power.

[0068] Based on the direction of arrival of the target wave, azimuth and elevation angles are matched in a pre-defined airspace environment attenuation factor database. Prior information such as terrain obstruction, reflection intensity, and multipath component intensity in that direction is queried to obtain the corresponding multipath attenuation correction coefficient. This airspace environment attenuation factor database is generated through pre-determined real-world calibration or scenario simulation and dynamically updated according to different security airspaces and altitudes, thus representing the attenuation effect of the actual environment on radio wave propagation. The multipath attenuation correction coefficient is then used to perform environmental compensation correction on the effective received power, eliminating power deviations caused by non-free-space propagation factors such as multipath, reflection, and obstruction, resulting in a received power more closely approximating real propagation conditions. Specifically, the airspace environment attenuation factor database is established through real-world calibration or scenario simulation, using a multi-dimensional index based on azimuth, elevation, and range segments to store attenuation factors such as terrain obstruction, multipath reflection intensity, building attenuation, and ground / wall reflection coefficients for each direction. Furthermore, this database can be updated through online learning. For the same direction of arrival, the average value of historical multi-frame statistics is used as the multipath attenuation correction coefficient to compensate for power deviations caused by non-free-space propagation.

[0069] The corrected received power and the typical transmit power range corresponding to the target are input together into the pre-trained logarithmic distance path loss model to obtain the path loss value. The formula for calculating the path loss value is:

[0070] Where PL(d) is the path loss from the target to the detection system, in dB. d is the distance between the target and the system, in meters (m). d0 is the reference distance (taken as 1 m). PL(d0) is the path loss at the reference distance, in dB. λ is the signal carrier wavelength, λ = c / f, unit: m; f is the carrier frequency, unit: Hz; c is the speed of light, c = 3 × 10⁸ m / s; γ is the path loss value (dimensionless, 2–5); X σ This is an environmental correction factor.

[0071] Finally, the path loss value is substituted into the standard radio wave propagation loss equation. Based on known parameters such as frequency, wavelength, and system gain, the radial distance between the target and the detection system is calculated by inversion. Through continuous multi-frame sliding filtering and outlier removal, the distance output is further smoothed and the error caused by measurement fluctuations is suppressed, so that the obtained distance value has higher accuracy, stability and environmental adaptability.

[0072] This embodiment constructs a logarithmic distance path loss model based on measured data and scene simulation data. It completes the model by fitting and training path loss samples, corresponding signal carrier frequencies, reference distances, and multipath attenuation data under different environments. Parameter settings are hierarchically divided. The basic layer contains fixed parameters, including the reference distance d0 (default 1m) and the speed of light c = 3 × 10⁸ m / s, determined by physical characteristics and industry standards. The core adjustment layer contains scene adaptation parameters, with the path loss value γ determined based on scene type (open / sparsely populated: 2–2.5; suburban / slightly obstructed: 3–3.5; urban / densely built-up: 3.5–4.5; severely obstructed: 4–5). The environmental correction factor X... σ The multipath attenuation correction coefficient corresponding to the direction of arrival is obtained from the preset airspace environment attenuation factor database; the input layer parameters are dynamically adapted parameters, including the corrected received power after hardware link calibration (deducting fixed losses such as feeders, filters, and connectors and superimposing antenna receiving gain) and environmental compensation, as well as the confidence interval mean of the typical transmit power range of the detected target. The logarithmic distance path loss model calculates the output path loss value through the collaborative calculation of parameters at each level.

[0073] As can be seen from the above, this embodiment effectively eliminates the influence of environmental factors such as hardware link errors, terrain occlusion, and multipath reflection on distance calculation by combining received power calibration, multipath attenuation environmental compensation, pre-trained logarithmic distance path loss model and propagation loss equation, further improving the ranging accuracy and reliability in complex security airspace. At the same time, by matching the environmental attenuation factor according to the direction of arrival, environmentally adaptive distance calculation is realized, enabling stable and accurate target distance output under different directions and terrain conditions, providing a more reliable measurement basis for subsequent target positioning, trajectory fitting and security early warning.

[0074] In one embodiment of this application, a radio detection method for a drone further includes: The distance value is used as a new feedback parameter and input into the intelligent switching strategy for signal strength feedback; Based on the distance value, adjust the cyclic switching period and the adjustment step size of the gain parameter in the intelligent switching strategy.

[0075] In this embodiment, the target distance value is added as a new closed-loop feedback parameter, which, together with the original signal strength indication value, is input into the intelligent handover strategy based on signal strength feedback, forming a two-dimensional adaptive adjustment mechanism of signal strength + distance. Within the strategy, the distance value is first subjected to sliding filtering and outlier removal to eliminate outliers and abrupt interference, resulting in a stable and reliable distance estimation result.

[0076] Based on the distance value, adjust the cyclic switching period and the adjustment step size of the gain parameter in the intelligent switching strategy.

[0077] As can be seen from the above, this embodiment, by inputting the distance value as a new feedback parameter into the intelligent switching strategy, forms a two-dimensional closed-loop adjustment mechanism of signal strength and distance. This effectively avoids the problems of single signal strength feedback being susceptible to environmental interference and insufficient adjustment accuracy, while improving the stability and rationality of parameter adjustment. The distance value after filtering can eliminate abnormal interference, making adaptive control more accurate and reliable, and enhancing the tracking continuity, detection sensitivity, and overall environmental adaptability of UAV targets.

[0078] In one embodiment of this application, adjusting the cyclic switching period and the adjustment step size of the gain parameter in the intelligent handover strategy according to the distance value includes: Extract the Doppler frequency shift value from the target's radio frequency characteristics, and calculate the radial flight velocity of the detected target based on the Doppler frequency shift value; Obtain environmental characteristic parameters of the current detection scene, including electromagnetic interference intensity, signal-to-noise ratio, or spectrum occupancy. Based on radial flight speed and environmental characteristic parameters, an adjustment step size lookup table mapping to distance intervals is constructed; The target distance interval is obtained by determining the interval to which the distance value belongs based on the preset distance interval division rules; Based on the target distance interval and coefficient lookup table, the target period adjustment step size coefficient and target gain adjustment step size coefficient corresponding to the target distance interval are obtained; Multiply the current cycle adjustment step size by the target cycle adjustment step size coefficient to obtain the updated cycle adjustment step size; Multiply the current gain adjustment step size by the target gain adjustment step size coefficient to obtain the updated gain adjustment step size.

[0079] In this embodiment, the Doppler frequency shift value is first extracted from the target's radio frequency characteristics. By performing least-squares fitting on the time-series change of the signal center frequency, frequency drift and noise interference are removed to obtain a stable Doppler frequency shift estimate. Based on the carrier wavelength of the radio frequency signal, the radial flight velocity of the target relative to the detection system is calculated. This radial flight velocity represents the target's speed and trend of motion. Simultaneously, environmental characteristic parameters of the current detection scenario are acquired, including electromagnetic interference intensity, received signal-to-noise ratio, and spectrum occupancy. Quantitative environmental quality indicators are obtained through multi-channel statistics and long-term moving averages, accurately characterizing the electromagnetic environment complexity and signal reception reliability of the current airspace.

[0080] Based on radial flight speed and environmental characteristic parameters, and according to the detection requirements and system operating characteristics of different distance ranges, an adjustment step size coefficient lookup table is constructed, mapping one-to-one with the distance range. The near-distance range corresponds to rapid target movement changes and significant electromagnetic environmental influences, so a larger step size coefficient is configured to improve response speed. The far-distance range corresponds to stable movement and susceptibility to signal interference, so a smaller step size coefficient is configured to ensure operational stability, ensuring that the coefficient values ​​simultaneously adapt to both target movement states and actual environmental conditions. Based on preset distance range division rules, filtered and smoothed distance values ​​are categorized into the corresponding target distance ranges. The distance ranges are divided into three levels—near-distance, medium-distance, and far-distance—based on the UAV's flight distance, system detection range, signal attenuation characteristics, and security level requirements. The near-distance range is 0–500m, corresponding to rapid target movement changes and high signal strength; the medium-distance range is 500–2000m, corresponding to stable signal and movement states; and the far-distance range is above 2000m, corresponding to weak signals and susceptibility to interference. The boundary points of each range are determined comprehensively through measured signal attenuation curves and system sensitivity indicators.

[0081] The target distance interval is matched and queried in the coefficient lookup table to obtain the target period adjustment step size coefficient and the target gain adjustment step size coefficient that are suitable for the current target state and the detection scenario. Then, the currently used period adjustment step size and gain adjustment step size are multiplied by the corresponding coefficients to obtain the updated period adjustment step size and gain adjustment step size.

[0082] As can be seen from the above, this embodiment calculates the radial flight velocity by extracting the Doppler frequency shift, and constructs an adjustment step size coefficient lookup table that maps to the range interval based on environmental characteristic parameters such as electromagnetic interference intensity, signal-to-noise ratio, and spectrum occupancy. This allows for dynamic optimization of the adjustment step size of the cycle and gain from multiple dimensions, including the target's motion state, the detection environment, and the distance, making the parameter adjustment more in line with the actual scenario and the target's changing patterns. Furthermore, by adaptively obtaining the corresponding step size coefficient based on the range interval and updating it in real time, the response speed, adjustment accuracy, and operational stability of the intelligent switching strategy are effectively improved.

[0083] In one embodiment of this application, an amplitude comparison direction finding algorithm is employed. By comparing the signal amplitude difference between adjacent antenna channels and searching a preset antenna pattern database, the direction of arrival of the detected target is calculated, including: When the preset calibration trigger conditions are met, a calibration signal with known characteristics is injected into the receiving channel to perform initial calibration of the amplitude and phase consistency of the multi-antenna channels. The calibration trigger conditions include system initialization, ambient temperature change greater than a preset temperature threshold, or time since the last calibration greater than a preset calibration interval. During the normal detection phase, the target signal from the same source is extracted from the received multi-channel signal, the residual amplitude and phase error between channels is estimated in real time based on the signal subspace projection method, and the antenna pattern database is corrected according to the residual amplitude and phase error. The corrected radiation pattern database is matched with the current multi-channel signal amplitude, and multidimensional interpolation and weighted centroid algorithm are used to calculate the instantaneous direction of arrival of the target with sub-wavelength level accuracy.

[0084] In this embodiment, when preset calibration trigger conditions are met, including power-on initialization, ambient temperature change exceeding a preset temperature threshold, or the time since the last calibration exceeding a preset calibration interval, the channel amplitude and phase consistency calibration process is initiated. Specifically, a standard calibration signal with known amplitude, phase, and frequency is injected into each receiving channel through a built-in calibration source. Under the same conditions, the amplitude response and phase offset of multiple antenna channels are measured, and the amplitude and phase compensation values ​​of each channel relative to the reference channel are calculated and stored, achieving initial calibration of amplitude and phase consistency between channels. The preset temperature threshold is determined based on the temperature drift characteristic parameters of the receiving channel devices (after factory testing and calibration) and is set to 5℃~10℃. The preset calibration interval is based on the aging rate of the receiving channel devices, environmental stability, and system detection accuracy requirements, and is set through actual measurement verification and simulation optimization, with a value ranging from 2 to 24 hours. The calculation method for amplitude and phase compensation values ​​includes: taking any receiving channel as the reference channel, after injecting the same calibration signal, measuring the amplitude and phase values ​​of the output signal of each channel respectively, subtracting the amplitude and phase of each channel from the reference channel to obtain the amplitude compensation value = reference channel amplitude - current channel amplitude, and the phase compensation value = reference channel phase - current channel phase. The amplitude compensation value and phase compensation value corresponding to each group of channel numbers are then stored in the system register.

[0085] During normal detection operation, a target signal with high signal-to-noise ratio and stable characteristics is selected from the signals received by multiple antenna channels. Based on the signal subspace projection and orthogonal matching method, the residual amplitude and phase errors caused by device drift and environmental changes between channels are obtained. The residual error is then used to correct the antenna pattern database so that the antenna pattern database always matches the current channel state and working environment.

[0086] The modified antenna pattern database is matched with the amplitude of the current multi-channel measured signal. An angular resolution is improved based on a multi-dimensional interpolation algorithm, and a weighted centroid algorithm is used to suppress noise and interference, achieving sub-wavelength-level accuracy in determining the instantaneous direction of arrival (ROA) of the target. Specifically, the multi-dimensional interpolation algorithm uses the amplitude difference of the measured multi-channel signal as input to the modified antenna pattern database. It performs fine calculations between adjacent angular sample points using linear or quadratic interpolation, generating a denser virtual angular response between discrete angular sampling points. This overcomes the hardware angular step size limitation and distinguishes minute angular differences at the sub-wavelength level. The weighted centroid algorithm uses multiple angular samples from the antenna pattern database that are closest to the measured multi-channel amplitude vector. It uses the angular value of each sample as position coordinates and the matching degree between the sample and the measured signal as weights. Multiple angular coordinates are weighted, summed, and normalized to obtain the final target ROA angle. The influence of outliers is suppressed by weighting noise and interference samples, achieving high-precision ROA at the sub-wavelength level.

[0087] From the above, it can be concluded that this embodiment effectively eliminates the direction-finding errors caused by antenna channel inconsistency, temperature drift, and device slow drift by performing channel amplitude and phase consistency calibration under preset calibration trigger conditions and using the same source target signal and signal subspace projection method to correct residual amplitude and phase errors and antenna pattern database in real time during normal detection. The subwavelength-level high-precision direction of arrival calculation is achieved by using multidimensional interpolation and weighted centroid algorithm, which effectively improves the direction-finding stability, accuracy, and reliability of the system under long-term operation and complex environments.

[0088] Corresponding to the radio detection method for drones in the above embodiments, Figure 2 This is a structural block diagram of a radio detection system for a drone according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The radio detection system 20 for UAVs includes: an airspace partitioning planning module 21, an RF signal receiving module 22, a signal switching processing module 23, a feature extraction processing module 24, a parameter adjustment and ranging module 25, and a position trajectory fitting module 26.

[0089] Among them, the airspace partitioning planning module 21 is used to divide the airspace to be detected based on different refresh rate requirements of the airspace to be detected, so as to obtain multiple scanning areas; The radio frequency signal receiving module 22 is used to receive full-band radio frequency signals in multiple scanning areas through the antenna array to obtain the original radio frequency signal; Signal switching processing module 23 is used to perform cyclic switching processing on the original radio frequency signal using a radio frequency switch array to obtain a time-sequential serial processing signal; Feature extraction and processing module 24 is used to perform signal preprocessing and frequency domain feature extraction on the time-series serially processed signal to obtain target radio frequency features that characterize the radio frequency signal attributes of the detection target; The parameter adjustment and ranging module 25 is used to adjust the cyclic switching period and gain parameters of the radio frequency switch array based on the target radio frequency characteristics and adopts an intelligent switching strategy based on signal strength feedback, and calculates the direction of arrival and distance value of the detected target based on the amplitude ratio direction finding algorithm. The position trajectory fitting module 26 is used to fit the real-time position and flight trajectory of the target by analyzing the target radio frequency characteristics of a preset continuous period.

[0090] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the shown spatial partitioning planning module 21, radio frequency signal receiving module 22, signal switching processing module 23, feature extraction processing module 24, parameter adjustment and ranging module 25, and position trajectory fitting module 26 are illustrated.

[0091] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0092] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0093] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0094] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in any embodiment of the radio detection method for UAVs provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0095] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0096] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0097] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0098] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0099] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0101] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0102] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A radio detection method for unmanned aerial vehicles (UAVs), characterized in that, include: The space to be detected is divided into multiple scanning areas based on the different refresh rate requirements of the space to be detected. The original radio frequency signal is obtained by receiving full-band radio frequency signals in multiple scanning areas through an antenna array; The original radio frequency signal is cyclically switched using a radio frequency switch array to obtain a time-sequential serial processed signal. The time-series serialized signal is preprocessed and frequency domain features are extracted to obtain target radio frequency features that characterize the radio frequency signal attributes of the detection target. Based on the target's radio frequency characteristics, an intelligent switching strategy based on signal strength feedback is used to adjust the cyclic switching period and gain parameters of the radio frequency switch array, and the direction of arrival and distance of the detected target are calculated based on the amplitude ratio direction finding algorithm. By analyzing the target's radio frequency characteristics over a preset continuous period, the real-time position and flight trajectory of the target are obtained through fitting.

2. The radio detection method for unmanned aerial vehicles according to claim 1, characterized in that, The spatial domain to be detected is divided into multiple scanning regions based on different refresh rate requirements, including: Based on the preset security level and airspace sensitivity, the refresh rate threshold of the airspace to be detected is determined; Based on the refresh rate threshold and the beamwidth and scanning step angle of the antenna array, the spatial domain to be detected is discretized to obtain several spatial grid cells. Adjacent spatial grid cells with the same or a difference of less than a preset first threshold are aggregated to obtain multiple scanning regions with different scanning priorities.

3. The radio detection method for unmanned aerial vehicles according to claim 2, characterized in that, Also includes: Based on the refresh rate requirements of each scanning area, calculate the percentage of theoretical scans for each scanning area in a single complete scanning cycle, and determine the dwell time weight based on the percentage of theoretical scans. The minimum dwell time unit for each scanning region is obtained based on the beam switching time of the antenna array, the setup time of the RF switch array, and the Fourier transform period of the signal processing unit. Based on the dwell time weight and the minimum dwell time unit, calculate the dwell time required for each scanning region in one scanning cycle; Based on the priority order of each scanning region, the required dwell time of each scanning region within the preset scanning cycle is sequentially filled into the scanning timing table to generate a cyclically executed scanning timing.

4. The radio detection method for unmanned aerial vehicles according to claim 3, characterized in that, Based on the priority order of each scanning region, the required dwell time of each scanning region within a preset scanning cycle is sequentially filled into the scanning timing table to generate a cyclically executed scanning timing, including: All scanned areas are sorted in descending order of refresh rate requirements to generate a priority list; Initialize a blank scan timing table and set the current time pointer to the beginning of the timing table; Select the highest priority scanning region from the priority list in sequence, and determine whether the remaining time of the current complete scanning cycle is greater than the dwell time required to insert the highest priority scanning region. If the judgment result is yes, then fill the dwell time required for the current highest priority scanning area into the current time pointer position, and move the time pointer backward by the corresponding dwell time length; If the determination result is negative, then the remaining unscheduled time of the current highest priority scan area is marked as pending insertion, and the current highest priority scan area is skipped, and the next priority scan area is processed. After completing one round of priority list traversal, if there are still scan areas waiting to be inserted, the beam switching idle interval of each scan area is compressed, or the scan area is inserted in the reserved time slot of the next scan cycle, until the dwell time of all scan areas is allocated, and the scan timing sequence is generated.

5. A radio detection method for unmanned aerial vehicles according to claim 1, characterized in that, The method involves adjusting the cyclic switching period and gain parameters of the RF switch array based on the target's RF characteristics using a signal strength feedback intelligent switching strategy, and calculating the target's direction of arrival and distance based on an amplitude ratio direction finding algorithm, including: Extract the signal strength indication value from the target radio frequency feature, and compare the signal strength indication value with a preset second threshold and a preset third threshold; When the signal strength indication value is greater than the second threshold, the gain parameter of the RF switch array is reduced by the intelligent switching strategy, and the cyclic switching cycle is shortened accordingly. When the signal strength indication value is less than the third threshold, the gain parameter of the RF switch array is increased by the intelligent switching strategy, and the cyclic switching cycle is extended accordingly. Using the adjusted cycle switching period and gain parameters, the timing signal amplitude corresponding to each channel of the RF switch array is obtained; Based on the time-series signal amplitudes corresponding to each channel, an amplitude comparison direction finding algorithm is used to calculate the direction of arrival of the target by comparing the signal amplitude differences between adjacent antenna channels, searching a preset antenna pattern database. By integrating the direction of arrival of the wave with the signal strength indication value, and based on the signal propagation attenuation model, the distance between the target and the detection system is obtained.

6. A radio detection method for unmanned aerial vehicles according to claim 5, characterized in that, The step of integrating the direction of arrival of the wave with the signal strength indication value, and obtaining the distance between the target and the detection system based on the signal propagation attenuation model, includes: Obtain the received power value corresponding to the signal strength indication value, and perform path loss calculation on the received power value to obtain the effective received power; Based on the direction of arrival of the wave, a multipath attenuation correction coefficient corresponding to the direction of arrival of the wave is obtained by matching it with a preset spatial environment attenuation factor database. The effective received power is then corrected by environmental compensation using the multipath attenuation correction coefficient to obtain the corrected received power. The corrected received power and the typical transmit power range of the target are input into a pre-trained logarithmic distance path loss model to obtain the path loss value. Substituting the path loss value into the propagation loss equation, the distance between the target and the detection system is obtained.

7. A radio detection method for unmanned aerial vehicles according to claim 6, characterized in that, Also includes: The distance value is used as a new feedback parameter and input into the intelligent switching strategy for signal strength feedback; Based on the distance value, the adjustment step size of the cyclic switching period and gain parameter in the intelligent switching strategy is adjusted.

8. A radio detection method for a UAV according to claim 7, characterized in that, The step of adjusting the cyclic switching period and the adjustment step size of the gain parameter in the intelligent switching strategy according to the distance value includes: The Doppler frequency shift value is extracted from the target's radio frequency characteristics, and the radial flight velocity of the target is calculated based on the Doppler frequency shift value; Obtain environmental characteristic parameters of the current detection scene, including electromagnetic interference intensity, signal-to-noise ratio, or spectral occupancy. Based on the radial flight speed and the environmental characteristic parameters, an adjustment step size lookup table mapping to the distance interval is constructed; The target distance interval is obtained by determining the interval to which the distance value belongs based on the preset distance interval division rules; Based on the target distance interval and the coefficient lookup table, the target period adjustment step size coefficient and the target gain adjustment step size coefficient corresponding to the target distance interval are obtained; Multiply the current cycle adjustment step size by the target cycle adjustment step size coefficient to obtain the updated cycle adjustment step size; Multiply the current gain adjustment step size by the target gain adjustment step size coefficient to obtain the updated gain adjustment step size.

9. A radio detection method for an unmanned aerial vehicle according to claim 5, characterized in that, The amplitude comparison direction finding algorithm is used to calculate the direction of arrival of the target by comparing the signal amplitude difference between adjacent antenna channels, searching a preset antenna pattern database, including: When the preset calibration trigger conditions are met, a calibration signal with known characteristics is injected into the receiving channel to perform initial calibration of the amplitude and phase consistency of the multi-antenna channels; the calibration trigger conditions include system initialization, ambient temperature change greater than a preset temperature threshold, or time since the last calibration greater than a preset calibration interval. During the normal detection phase, the target signal from the same source is extracted from the received multi-channel signal, the residual amplitude and phase error between channels is estimated in real time based on the signal subspace projection method, and the antenna pattern database is corrected according to the residual amplitude and phase error. The corrected radiation pattern database is matched with the current multi-channel signal amplitude, and multidimensional interpolation and weighted centroid algorithm are used to calculate the instantaneous direction of arrival of the target with sub-wavelength level accuracy.

10. A radio detection system for unmanned aerial vehicles (UAVs), characterized in that, include: The airspace partitioning planning module is used to divide the airspace to be detected into multiple scanning areas based on different refresh rate requirements of the airspace to be detected. The radio frequency signal receiving module is used to receive full-band radio frequency signals in multiple scanning areas through an antenna array to obtain the original radio frequency signal; The signal switching and processing module is used to perform cyclic switching processing on the original radio frequency signal using a radio frequency switch array to obtain a time-sequential serial processed signal. The feature extraction and processing module is used to perform signal preprocessing and frequency domain feature extraction on the time-series serial processing signal to obtain target radio frequency features that characterize the radio frequency signal attributes of the detection target; The parameter adjustment and ranging module is used to adjust the cyclic switching period and gain parameters of the radio frequency switch array based on the target radio frequency characteristics and an intelligent switching strategy based on signal strength feedback, and to calculate the direction of arrival and distance value of the detected target based on the amplitude ratio direction finding algorithm. The position trajectory fitting module is used to fit the real-time position and flight trajectory of the target by analyzing the target's radio frequency characteristics over a preset continuous period.

Images