Unmanned aerial vehicle detection method and system based on pilot frequency synchronization sequence matching

Abstract

The invention relates to an unmanned aerial vehicle detection method and system based on pilot frequency synchronization sequence matching, and belongs to the technical field of unmanned aerial vehicle detection. Receiving a to-be-detected signal, and intercepting a signal segment of which the length is a preset point number from the to-be-detected signal; performing fast Fourier transform on the signal segment to obtain frequency domain representation of the to-be-detected signal; performing fast Fourier transform on the general reference sequence according to a preset point number to obtain a frequency domain general reference sequence; performing conjugate operation on the frequency domain general reference sequence to obtain a frequency domain reference signal; performing convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; and performing constant false alarm rate detection on the frequency domain convolution result to obtain an unmanned aerial vehicle detection result. According to the invention, the requirement of rapid and efficient detection of the unmanned aerial vehicle in a complex signal environment can be met.

Description

Technical Field

[0001] The present application relates to the technical field of drone detection, and in particular to a drone detection method and system based on pilot synchronization sequence matching. Background Art

[0002] In recent years, with the rapid development of drone technology, the use of consumer drones has gradually become popular, and drones have been widely used in various industries. However, the rapid popularization of this technology has also brought many safety risks and management difficulties. In key areas such as airports, energy facilities, and large-scale event venues, the illegal intrusion of drones may pose a serious threat to public safety. Therefore, the detection of drones has become an important technical means to protect public safety and privacy.

[0003] At present, common drone detection technologies mainly include frequency band detection technology and machine learning recognition technology based on spectrum waterfall diagram. Among them, frequency band detection technology needs to scan each sub-band one by one, the detection time is long and the process is complicated, and it is difficult to cope with multi-target detection under dynamic frequency bands; while the recognition technology signal based on spectrum waterfall diagram has problems such as long acquisition time and lack of real-time performance. Therefore, how to improve real-time performance and detection accuracy to meet the needs of fast and efficient detection of drones in complex signal environments has become a problem that needs to be solved urgently. Summary of the invention

[0004] In order to meet the demand for fast and efficient detection of drones in complex signal environments, the present application provides a drone detection method and system based on pilot synchronization sequence matching.

[0005] In the first aspect, the present application provides a drone detection method based on pilot synchronization sequence matching, which adopts the following technical solution: A drone detection method based on pilot synchronization sequence matching, the detection method comprising: Constructing a universal reference sequence; receiving a signal to be detected, and extracting a signal segment having a length of a preset number of points from the signal to be detected; Performing a fast Fourier transform on the signal segment to obtain a frequency domain representation of the signal to be detected; Performing a fast Fourier transform on the universal reference sequence according to the preset number of points to obtain a frequency domain universal reference sequence; Performing a conjugate operation on the frequency domain universal reference sequence to obtain a frequency domain reference signal; Performing a convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; A constant false alarm rate test is performed on the frequency domain convolution result to obtain a drone detection result.

[0006] By adopting the above technical solution, a universal reference sequence that adapts to different bandwidths is constructed, and combined with frequency domain transformation, convolution operation and constant false alarm rate detection, efficient and accurate detection of drone signals is achieved. Utilizing the characteristics of the pilot synchronization sequence in the drone image transmission signal / drone remote control signal, this method performs precise matching in the frequency domain, greatly improves the signal-to-noise ratio, and effectively reduces false detection and missed detection through dynamic threshold detection. The technical solution of this application has the advantages of strong real-time performance, high detection accuracy and low computational cost. It is suitable for the safety monitoring needs of drones in complex signal environments and provides reliable technical support for the security of airports, energy facilities and key areas.

[0007] Optionally, the drone detection result is used to determine whether there is a drone image transmission signal in the signal to be detected; the step of constructing a universal reference sequence includes: According to the bandwidth characteristics of the UAV image transmission signal, a pilot synchronization sequence of the UAV image transmission signal is obtained, where the pilot synchronization sequence is composed of a ZC sequence; Construct an all-zero sequence whose pilot synchronization sequence points match the sampling rate of the drone image transmission signal, and embed the pilot synchronization sequences corresponding to different bandwidths into the specified positions of the all-zero sequence to obtain a frequency domain pilot synchronization sequence; wherein the preset number of points is an integer multiple n of the number of points of the pilot synchronization sequence, and n≥2; The frequency domain pilot synchronization sequence is converted into the time domain through inverse fast Fourier transform to obtain the universal reference sequence.

[0008] By adopting the above technical solution, a universal reference sequence suitable for signals of different bandwidths is constructed based on the pilot synchronization sequence of the drone image transmission signal characteristics. During the construction process, by matching the bandwidth points, embedding the all-zero sequence and performing the inverse fast Fourier transform, the spectral characteristics of the universal reference sequence are guaranteed to be consistent with the drone signal. The ultimately generated universal reference sequence can not only cover drone signals of different bandwidths, but also effectively improve the accuracy of signal detection and the signal-to-noise ratio in subsequent matching operations.

[0009] Optionally, the step of performing constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result includes: Analyzing peak features in the frequency domain convolution result; Determining whether the peak characteristic meets a preset threshold condition; If so, it is determined that there is a drone image transmission signal in the signal to be detected, that is, the drone is detected; if not, it is determined that there is no drone image transmission signal in the signal to be detected, that is, the drone is not detected.

[0010] By adopting the above technical solution, it is possible to effectively extract and identify drone image transmission signals in complex signal environments. The use of dynamic threshold strategies, combined with the amplitude and position characteristics of the peak, greatly reduces the possibility of noise interference and false alarms, while ensuring reliable detection of weak target signals. This technical solution has the characteristics of high accuracy, high robustness and real-time performance, and can provide a fast and accurate solution for drone detection, which is suitable for the security needs of airports, energy facilities and other key areas.

[0011] Optionally, the drone detection result is used to determine whether there is a drone remote control signal in the signal to be detected; the step of constructing a universal reference sequence includes: According to the bandwidth characteristics of the UAV remote control signal, a pilot synchronization sequence of the UAV remote control signal is obtained, where the pilot synchronization sequence is composed of a ZC sequence; Construct an all-zero sequence whose pilot synchronization sequence points match the sampling rate of the drone remote control signal, and embed the pilot synchronization sequences corresponding to different bandwidths into the specified positions of the all-zero sequence to obtain a frequency domain pilot synchronization sequence; wherein the preset number of points is an integer multiple n of the number of points of the pilot synchronization sequence, and n≥2; The frequency domain pilot synchronization sequence is converted into the time domain through inverse fast Fourier transform to obtain the universal reference sequence.

[0012] By adopting the above technical solution, combined with bandwidth characteristics, frequency domain embedding and time domain signal generation, the solution can generate a reference template consistent with the actual remote control signal, ensuring the matching and robustness of subsequent detection.

[0013] Optionally, the step of performing constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result includes: Analyzing peak features in the frequency domain convolution result (including the position and amplitude of the peak); Determining whether the peak characteristic meets a preset threshold condition; If so, it is determined that the drone remote control signal exists in the signal to be detected, that is, the drone is detected; if not, it is determined that the drone remote control signal does not exist in the signal to be detected, that is, the drone is not detected.

[0014] By adopting the above technical solution, analyzing the peak characteristics (including position and amplitude) in the frequency domain convolution results, combined with dynamic preset threshold conditions (number of protection units, number of reference units and detection threshold), high-precision detection of drone remote control signals is achieved. When the peak characteristics meet the threshold conditions, the system can reliably determine the presence of drone signals; otherwise, the interference of background noise or other non-target signals can be eliminated.

[0015] Optionally, the constant false alarm rate detection includes a bilateral average constant false alarm rate strategy, a two-side unit average selection strategy, or a two-side unit average selection strategy.

[0016] Optionally, the preset threshold condition includes a preset number of protection units, a preset number of reference units and a preset detection threshold.

[0017] In the second aspect, the present application provides a drone detection system based on pilot synchronization sequence matching, which adopts the following technical solutions: A drone detection system based on pilot synchronization sequence matching, the detection system comprising: Sequence building module, used to build universal reference sequences; A signal receiving module, used for receiving a signal to be detected; A signal segment interception module, used for intercepting a signal segment with a length of a preset number of points from the signal to be detected; A signal segment processing module, used for performing a fast Fourier transform on the signal segment to obtain a frequency domain representation of the signal to be detected; A frequency domain conversion module, used for performing a fast Fourier transform on the universal reference sequence according to the preset number of points to obtain a frequency domain universal reference sequence; A conjugate operation module, used for performing a conjugate operation on the frequency domain universal reference sequence to obtain a frequency domain reference signal; A convolution module, used for performing a convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; The constant false alarm rate detection module is used to perform constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result.

[0018] In a third aspect, the present application provides a computer device, which adopts the following technical solution: A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect.

[0020] To summarize, the present application includes at least one of the following beneficial technical effects: the technical solution of the present application has the advantages of strong real-time performance, high detection accuracy, and low computing cost. It is suitable for the safety monitoring needs of drones in complex signal environments, and provides reliable technical support for the security of airports, energy facilities, and key areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a first flow chart of a drone detection method according to one of the embodiments of the present application.

[0022] Figure 2 It is a frequency domain diagram of a signal segment intercepted from a signal to be detected in one of the embodiments of the present application.

[0023] Figure 3 It is a second flow chart of the drone detection method according to one of the embodiments of the present application.

[0024] Figure 4 It is a frequency domain schematic diagram of a universal reference sequence constructed in one of the embodiments of the present application.

[0025] Figure 5 It is a time domain schematic diagram of a universal reference sequence constructed in one of the embodiments of the present application.

[0026] Figure 6 This is a third flow chart of the drone detection method according to one of the embodiments of the present application.

[0027] Figure 7 It is a schematic diagram of the distribution of detection peaks in the frequency domain convolution result of one of the embodiments of the present application.

[0028] Figure 8 It is a schematic diagram of the distribution of detection peaks in the frequency domain convolution results of another embodiment of the present application.

[0029] Fig. 9 It is a schematic diagram of the distribution of detection peaks in the frequency domain convolution result of yet another embodiment of the present application.

[0030] Fig.10 This is a fourth flow chart of the drone detection method according to one of the embodiments of the present application.

[0031] Fig.11 This is a fifth flow chart of a drone detection method according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-11 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0033] At present, common drone detection technologies are mainly concentrated in the following two ways: Frequency band detection technology: detects the radio frequency band where the drone communication signal is located. For example, the sub-bands in a specific frequency band list are detected one by one to determine whether the drone exists. This technical solution usually requires scanning each sub-band step by step, which takes a long time to detect. It is difficult to deal with drones that communicate in multiple frequency bands in complex environments. In addition, the detection process is complicated. When the frequency of the drone communication signal changes, it is easy to cause missed detection or false detection.

[0034] Machine learning recognition technology based on spectrum waterfall diagram: By collecting spectrum waterfall diagrams of drone signals to build a sample data set, and using convolutional neural network (CNN) for training and classification to identify drone signals, although this technical solution can theoretically achieve high recognition accuracy, it also has the following shortcomings: the process of obtaining spectrum waterfall diagrams in actual detection is time-consuming, especially in complex environments, the delay in signal feature extraction is large, and it cannot meet real-time requirements; machine learning models have high requirements for computing resources, and the training and implementation costs are relatively expensive, which is not conducive to large-scale promotion and application.

[0035] Therefore, the above-mentioned existing technologies have certain limitations when dealing with efficient and accurate detection of multi-band drone signals in complex environments.

[0036] Based on this, an embodiment of the present application discloses a drone detection method based on pilot synchronization sequence matching.

[0037] Reference Figure 1 , a drone detection method based on pilot synchronization sequence matching, the detection method comprising: Step S101, constructing a universal reference sequence; Among them, the drone image transmission signal / drone remote control signal emitted by the drone during the communication process contains specific pilot synchronization sequences. These sequences have strong signal characteristics, such as being composed of Zadoff-Chu sequences, with low cross-correlation and good periodic correlation, and are related to the signal bandwidth. Therefore, for drone image transmission signals / drone remote control signals of different bandwidths, a universal reference sequence in the frequency domain can be constructed. The universal reference sequence is formed by embedding pilot synchronization sequences of different bandwidths into specific positions of the all-zero sequence, and then converted to the time domain through inverse fast Fourier transform (IFFT) to ensure that it is suitable for drone signals of different bandwidths.

[0038] Specifically, the pilot synchronization sequence of the drone image transmission signal / drone remote control signal is composed of a ZC sequence, and the ZC sequence with a length of N can be expressed as:

[0039] In the above formula, is the root index of the ZC sequence.

[0040] Usually, when drones communicate, they construct a ZC sequence of length N in the frequency domain and place it in the center of an all-zero sequence with M points. Then, an inverse Fourier transform (IFFT) is performed to obtain the pilot synchronization sequence in the time domain. M can be calculated by the following formula:

[0041] In the above formula, It represents rounding up, M is the length of the all-zero sequence, and N is the number of points in the pilot synchronization sequence.

[0042] It can be understood that the constructed universal reference sequence can effectively match drone image transmission signals / drone remote control signals of different bandwidths, provide a standard reference for subsequent signal detection, and improve the adaptability and accuracy of detection.

[0043] Step S102, receiving a signal to be detected, and extracting a signal segment having a length of a preset number of points from the signal to be detected; Among them, in a complex signal environment, drone signals are usually buried in a lot of background noise. Therefore, by receiving the signal in the radio frequency band and intercepting a signal segment of a specific length, it can be ensured that each intercepted signal segment contains a complete pilot synchronization sequence.

[0044] Reference Figure 2 As shown, it is a frequency domain diagram of a signal segment of 4096 points of the signal to be detected in one of the embodiments of the present application. Specifically, the length of the preset number of points can be configured as an integer multiple n of the number of pilot synchronization sequence points, and n≥2. For example, an 18MHz bandwidth corresponds to a signal length of 2048 points, and the corresponding interception length can be configured to 4096 points to ensure the periodicity of the signal and the integrity of the spectrum distribution. Through reasonable signal interception, it is possible to ensure that the pilot synchronization sequence to be detected is completely retained, while reducing the amount of redundant data and reducing the computational complexity of subsequent processing.

[0045] Step S103, performing fast Fourier transform on the signal segment to obtain a frequency domain representation of the signal to be detected; Among them, Fast Fourier Transform (FFT) is an efficient signal transformation method that can convert time domain signals to frequency domain. In the frequency domain, the pilot synchronization sequence of the drone signal has obvious spectral characteristics, and its low correlation and high peak characteristics facilitate subsequent detection. At the same time, converting the signal to the frequency domain can also reduce the interference of accumulated noise in the time domain on the signal characteristics. Through FFT transformation, the complex signal processing in the time domain is simplified to the processing of feature points in the frequency domain, which improves the efficiency and accuracy of signal detection.

[0046] In one embodiment of the present application, the received signal to be detected that may contain a drone signal is recorded as s UAVAt this time, it is impossible to determine whether there is a drone signal in the signal to be detected. To ensure that the drone pilot synchronization sequence that may exist in each detection can be completely included in the intercepted signal, a signal with a length of 4096 points is extracted from the received signal to be detected every 2048 points, and the signal is transformed into the frequency domain by FFT, which is recorded as S UAV The present application does not need to divide the frequency band to be detected into sub-bands, but directly detects the signal to be detected transformed into the frequency domain, that is, detects the full frequency range of the frequency band to be detected, so the process is simpler and easier to implement.

[0047] Step S104, performing fast Fourier transform on the universal reference sequence according to a preset number of points to obtain a frequency domain universal reference sequence; In order to achieve signal matching in the frequency domain, the universal reference sequence needs to be subjected to a fast Fourier transform of the same length as the signal to be detected so that it has the same frequency domain feature representation. The generation of the frequency domain universal reference sequence is a process of converting the time domain universal reference sequence to the frequency domain, and its characteristics include the position and energy distribution of the pilot synchronization sequence.

[0048] Specifically, after obtaining the frequency domain universal reference sequence, it can be matched with the frequency domain representation of the signal to be detected, laying the foundation for subsequent convolution operations.

[0049] Step S105, performing a conjugate operation on the frequency domain universal reference sequence to obtain a frequency domain reference signal; Before performing convolution operations in the frequency domain, the universal reference sequence needs to be conjugated to enhance matching performance. The conjugation operation can change the phase of the complex signal, so that the convolution result produces a higher peak characteristic when matching, while reducing the impact of background noise on the convolution result. The frequency domain reference signal generated by the conjugation operation further improves the convolution peak characteristic performance.

[0050] In the embodiment of the present application, a 4096-point FFT is performed on the constructed universal reference sequence and transformed into the frequency domain, which is recorded as S ref , and then to S ref Take the conjugate, denoted by .

[0051] Step S106, performing a convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; Among them, convolution operation is the core step of signal matching. In the frequency domain, the convolution operation can highlight the pilot synchronization sequence characteristics of the drone signal by multiplying and accumulating the frequency domain characteristics of the signal to be detected and the reference signal point by point. When there is a drone signal in the signal to be detected, there will be an obvious peak in the convolution result, and when it does not exist, the convolution result tends to be smoothly distributed.

[0052] Specifically, the signal to be detected is intercepted and transformed into the frequency domain and the universal reference sequence transformed to the frequency domain and conjugated To do convolution, the calculation formula is:

[0053] in, represents the convolution operation, is the result after convolution, is the reference signal after conjugation operation.

[0054] In the embodiments of the present application, a higher signal-to-noise ratio (SNR) improvement can be obtained through convolution operation. For example, the 9MHz drone image transmission signal can be improved by 28dB, the 18MHz drone image transmission signal can be improved by 31dB, and the 2.2Mhz drone remote control signal can be improved by 22db. This allows better detection results to be obtained even when the signal-to-noise ratio of the original signal to be detected is relatively low, greatly improving the detection performance of the drone.

[0055] Step S107, performing constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result.

[0056] Among them, constant false alarm rate (CFAR) detection is a commonly used target detection technology, which determines the presence or absence of a target in a signal by setting a preset threshold condition. In the embodiment of the present application, CFAR detection dynamically adjusts the detection threshold based on the peak characteristics (such as amplitude and position) of the convolution result and the statistical characteristics of the background noise. By presetting the preset threshold conditions such as the number of protection units, the number of reference units, and the detection threshold, the influence of the target point on the threshold calculation is avoided. When there is a peak value exceeding the detection threshold in the convolution result, it is considered that a drone signal is detected.

[0057] Exemplarily, the number of protection units may be set to 16, the number of reference units may be set to 32, and the detection threshold may be set to 12 dB. The above parameters may be flexibly adjusted and configured according to actual conditions.

[0058] In some embodiments, a bilateral averaging (CA-CFAR) strategy may be adopted. This strategy calculates the average value of the background noise on both sides of the target and is suitable for scenarios with uniform background noise distribution. A large average selection strategy (GO-CFAR) may also be adopted. This strategy selects the larger average value among the noise units on both sides of the target and is suitable for high noise environments, which can effectively avoid false alarms. A small average selection strategy (SO-CFAR) may also be adopted. This strategy selects the smaller average value among the noise units on both sides of the target and is suitable for scenarios with low noise or sparse targets, which can enhance the detection capability of weak targets.

[0059] In the above implementation, a universal reference sequence that adapts to different bandwidths is constructed, and frequency domain transformation, convolution operation and constant false alarm rate detection are combined to achieve efficient and accurate detection of drone signals. Utilizing the characteristics of the pilot synchronization sequence in the drone image transmission signal / drone remote control signal, this method performs precise matching in the frequency domain, greatly improves the signal-to-noise ratio, and effectively reduces false detections and missed detections through dynamic threshold detection. The technical solution of this application has the advantages of strong real-time performance, high detection accuracy and low computational cost. It is suitable for drone safety monitoring needs in complex signal environments and provides reliable technical support for the security of airports, energy facilities and key areas.

[0060] As one of the embodiments of the present application, the present application can apply the detection of drone image transmission signals to realize drone detection, that is, the drone detection result is used to determine whether there is a drone image transmission signal in the signal to be detected.

[0061] Reference Figure 3 When drone image transmission signal detection is applied, the step of constructing a universal reference sequence in step S101 includes: Step S201, obtaining a pilot synchronization sequence of the drone image transmission signal according to the bandwidth characteristics of the drone image transmission signal; wherein the pilot synchronization sequence is composed of a ZC sequence; Among them, the image transmission signal of the drone usually contains a pilot synchronization sequence during transmission. This is a fixed sequence designed for communication synchronization, which has the characteristics of strong periodic correlation and small noise influence. The pilot synchronization sequence is composed of the Zadoff-Chu (ZC) sequence. The mathematical characteristics of the ZC sequence include constant amplitude characteristics (all elements of the ZC sequence have the same amplitude, which helps to reduce the peak-to-average ratio), low cross-correlation (the cross-correlation between different ZC sequences is extremely low, and the signals can be effectively distinguished even in complex communication environments) and bandwidth dependence (the length and spectral characteristics of the ZC sequence are directly related to the bandwidth and sampling rate of the signal).

[0062] In some embodiments, the UAV image transmission signal is usually transmitted with a specific bandwidth (such as 9MHz or 18MHz), and its system sampling rate is fixed. Therefore, the corresponding pilot synchronization sequence can be extracted for these known bandwidth characteristics. For example, for a 9MHz image transmission signal, the system sampling rate is 15.36MHz, the number of pilot synchronization sequence points is 1024 points, including a ZC sequence with a length of 601 points and a root index of 600 (denoted as ). When the 9MHz image transmission signal pilot synchronization sequence is generated, a 1024-point all-zero sequence is first constructed, and then the 213th to 813th points are equal to , that is, the pilot synchronization sequence of the 9MHz image transmission signal in the frequency domain is obtained, which is recorded as ;right Perform IFFT to obtain the pilot synchronization sequence of the 9MHz image transmission signal in the time domain. .

[0063] For the 18MHz image transmission signal, the system sampling rate is 30.72MHz, the number of pilot synchronization sequence points is 2048 points, including a ZC sequence with a length of 1201 points and a root index of 1200 (denoted as ). When the 18MHz image transmission signal pilot synchronization sequence is generated, a 2048-point all-zero sequence is first constructed, and then the 425th to 1625th points are equal to , that is, the pilot synchronization sequence of the 18MHz image transmission signal in the frequency domain is obtained, which is recorded as ;right Perform IFFT to obtain the pilot synchronization sequence of the 18MHz image transmission signal in the time domain. .

[0064] It can be understood that by extracting the bandwidth-specific pilot synchronization sequence, the unique characteristics of the drone image transmission signal can be obtained, providing a basis for the subsequent construction of a universal reference sequence.

[0065] Step S202, constructing an all-zero sequence whose pilot synchronization sequence points match the sampling rate of the drone image transmission signal, and embedding pilot synchronization sequences corresponding to different bandwidths into the specified positions of the all-zero sequence to obtain a frequency domain pilot synchronization sequence; wherein the preset number of points is an integer multiple n of the number of pilot synchronization sequence points, n≥2; In order to make the pilot synchronization sequence have the correct position distribution in the frequency domain, it needs to be embedded in an all-zero sequence that matches the sampling rate of the drone image transmission signal. The key to the embedding process is that the number of points in the all-zero sequence is consistent with the sampling rate of the image transmission signal to ensure that the generated frequency domain signal can accurately reflect the spectral characteristics of the drone signal. For example, for image transmission signals of different bandwidths, the number of points in the all-zero sequence corresponding to a 9MHz bandwidth is 1024, while the number of points corresponding to an 18MHz bandwidth is 2048.

[0066] Step S203: convert the frequency domain pilot synchronization sequence into the time domain through inverse fast Fourier transform to obtain a universal reference sequence.

[0067] Among them, the inverse fast Fourier transform (IFFT) converts the frequency domain signal into a time domain signal, which is a complex representation of the frequency domain signal on the time axis. By performing IFFT on the frequency domain pilot synchronization sequence, a time domain signal is generated, which is the universal reference sequence and can reflect the characteristics of the drone image transmission signal in time, such as periodicity and amplitude distribution.

[0068] In the above implementation, a universal reference sequence suitable for signals of different bandwidths is constructed based on the pilot synchronization sequence of the drone image transmission signal characteristics. During the construction process, by matching the bandwidth points, embedding the all-zero sequence and performing the inverse fast Fourier transform, the spectrum characteristics of the universal reference sequence are guaranteed to be consistent with the drone signal. The ultimately generated universal reference sequence can not only cover drone signals of different bandwidths, but also effectively improve the accuracy of signal detection and the signal-to-noise ratio in subsequent matching operations.

[0069] Reference Figure 4 , Figure 5 In one embodiment of the present application, two all-zero sequences with 2048 points can be constructed based on the 18MHz image transmission signal, which are respectively recorded as and ,make The 725th to 1325th points are equal to ,make The 425th to 1625th points are equal to , and then and Point-to-point addition is performed to obtain a common reference sequence in the frequency domain (such as Figure 3 As shown); Perform IFFT on the universal reference sequence in the frequency domain to obtain the universal reference sequence in the time domain (like Figure 4 as shown).

[0070] Reference Figure 6 When drone image transmission signal detection is applied, the step of performing constant false alarm rate detection on the frequency domain convolution result in step S107 to obtain the drone detection result includes: Step S301, analyzing the peak features in the frequency domain convolution result; Among them, the peak characteristics include the position and amplitude of the peak; Specifically, in the frequency domain, the peak characteristics of the convolution result are the main criterion for the presence of drone signals. Convolution is essentially a mathematical operation to evaluate the correlation between the signal to be detected and the reference signal. When there is a pilot synchronization sequence in the signal to be detected, a peak will appear in the convolution result. The characteristics of these peaks include position and amplitude. The position of the peak reflects the time offset or spectrum distribution of the pilot synchronization sequence in the signal. If the peak position is consistent with the characteristic position of the reference signal, it means that the signal may contain a drone image transmission signal; the peak amplitude reflects the strength of the match. A higher amplitude usually indicates that the signal contains a stronger pilot synchronization sequence, while a lower amplitude may be noise or a weak signal.

[0071] Reference Figure 7 , Figure 8 The figures are schematic diagrams of the distribution of detection peaks in the frequency domain convolution results of two different embodiments of the present application. Figure 7 There is a peak result and the peak amplitude is high, which reflects a stronger pilot synchronization sequence; Figure 8 There is no peak result in the , indicating that there is a mismatch between the signal to be detected and the frequency domain reference signal.

[0072] Step S302, determine whether the peak feature meets the preset threshold condition; if so, jump to step S303; if not, jump to step S304; The preset threshold conditions include the preset number of protection units, the preset number of reference units and the preset detection threshold. The number of protection units is used to prevent the target signal itself from affecting the background noise statistics; the number of reference units is the number of units used to count the characteristics of the background noise. The more the number, the more stable the statistical results; the detection threshold is the dynamic threshold value calculated based on the background noise intensity counted by the reference unit. The threshold value will be adjusted as the background noise intensity changes, and is used to distinguish the target signal from the noise.

[0073] Reference Fig. 9 As shown, it is a schematic diagram of the frequency domain convolution result of one of the embodiments of the present application meeting the preset threshold condition. It can be clearly observed that the peak amplitude in the frequency domain convolution result exceeds the threshold value and the position matches the expectation.

[0074] Step S303, determining that there is a drone image transmission signal in the signal to be detected, that is, the drone is detected; Step S304: determine that there is no drone image transmission signal in the signal to be detected, that is, no drone is detected.

[0075] Specifically, when the peak amplitude in the convolution result exceeds the threshold value and the position matches the expected value, it can be determined that there is a drone image transmission signal in the signal to be detected. If the peak amplitude in the convolution result does not exceed the dynamic threshold, or the peak position does not match the expected characteristics, it can be determined that there is no drone image transmission signal in the signal to be detected. This judgment is based on the statistical characteristics of CFAR detection and can effectively distinguish random noise or other non-correlated signals from target signals.

[0076] In the above implementation, the dynamic threshold strategy is used, combined with the amplitude and position characteristics of the peak, to greatly reduce the possibility of noise interference and false alarms, while ensuring reliable detection of weak target signals, and can effectively extract and identify drone image transmission signals in complex signal environments. This technical solution has the characteristics of high accuracy, high robustness and real-time performance, and can provide a fast and accurate solution for drone detection, which is suitable for the security needs of airports, energy facilities and other key areas.

[0077] As another embodiment of the present application, the present application can apply the detection of drone remote control signals to realize drone detection, that is, the drone detection result is used to determine whether there is a drone remote control signal in the signal to be detected.

[0078] Reference Fig.10 When the detection of the drone remote control signal is applied, the step of constructing the universal reference sequence in step S101 includes: Step S401, obtaining a pilot synchronization sequence of the drone remote control signal according to the bandwidth characteristics of the drone remote control signal; specifically, the pilot synchronization sequence is composed of a ZC sequence; Among them, the pilot synchronization sequence of the drone remote control signal is an important part of the specific communication protocol for achieving signal synchronization. Its core characteristics are stability and low correlation. The pilot synchronization sequence is usually composed of a Zadoff-Chu (ZC) sequence. The length of the pilot synchronization sequence directly depends on the bandwidth characteristics and sampling rate of the signal.

[0079] In some embodiments, the common UAV remote control signal bandwidth is 2.2MHz. For a 2.2MHz remote control signal, the system sampling rate is 15.36MHz, the number of pilot synchronization sequence points is 1024 points, including a ZC sequence with a length of 145 points and a root index of 121 (denoted as ). When the 2.2MHz remote control signal pilot synchronization sequence is generated, a 1024-point all-zero sequence is first constructed, and then the 441st to 585th points are equal to , that is, the pilot synchronization sequence of the 2.2MHz remote control signal in the frequency domain is obtained, which is recorded as S2; performing IFFT on S2, the pilot synchronization sequence S2 of the 2.2MHz remote control signal in the time domain can be obtained.

[0080] Step S402, constructing an all-zero sequence whose pilot synchronization sequence points match the sampling rate of the drone remote control signal, and embedding pilot synchronization sequences corresponding to different bandwidths into the specified positions of the all-zero sequence to obtain a frequency domain pilot synchronization sequence; wherein the preset number of points is an integer multiple n of the number of pilot synchronization sequence points, n≥2; Specifically, the number of points in the all-zero sequence needs to match the sampling rate and time domain length of the remote control signal to ensure that the frequency domain characteristics can be fully mapped. For example, for a remote control signal with a bandwidth of 2.2MHz, the sampling rate is 15.36MHz, and the number of points in the all-zero sequence should be 1024 points corresponding to the sampling rate. Then, according to the modulation method and frequency domain characteristics of the drone remote control signal, the pilot synchronization sequence is embedded in the center or specified position of the all-zero sequence to ensure that its signal spectrum position in the frequency domain is consistent with the actual spectrum range of the remote control signal.

[0081] Step S403: convert the frequency domain pilot synchronization sequence into the time domain through inverse fast Fourier transform to obtain a universal reference sequence.

[0082] Among them, the inverse fast Fourier transform (IFFT) is the core algorithm for converting frequency domain signals to time domain. Its function is to generate corresponding time domain signals from frequency domain features.

[0083] In the above implementation, by combining bandwidth characteristics, frequency domain embedding and time domain signal generation, the solution can generate a reference template consistent with the actual remote control signal, ensuring the matching degree and robustness of subsequent detection.

[0084] Reference Fig.11 When the detection of the drone remote control signal is applied, the step of performing constant false alarm rate detection on the frequency domain convolution result in step S107 to obtain the drone detection result includes: Step S501, analyzing the peak features in the frequency domain convolution result; Among them, in the process of detecting the remote control signal of the UAV, the peak characteristics of the frequency domain convolution result are an important basis for judging whether the target signal exists. The convolution operation concentrates the matching characteristics on the peak by matching the signal to be detected with the universal reference signal. In the convolution result, the peak position corresponds to the position of the pilot synchronization sequence in the signal to be detected, reflecting the offset characteristics of the target signal in the frequency domain, which can be used for subsequent signal time synchronization or positioning; the peak amplitude indicates the degree of matching between the signal to be detected and the reference signal. The higher the amplitude, the stronger the matching correlation, and the more likely the existence of the UAV remote control signal.

[0085] Step S502, determine whether the peak feature meets the preset threshold condition; if so, jump to step S503; if not, jump to step S504; Among them, the preset threshold condition is the key basis for distinguishing the real signal from the background noise, which usually includes the number of protection units, the number of reference units and the detection threshold.

[0086] Step S503, determining that there is a drone remote control signal in the signal to be detected, that is, detecting the drone; Step S504: determine that there is no drone remote control signal in the signals to be detected, that is, no drone is detected.

[0087] Among them, when the peak feature meets the preset threshold condition, it can be determined that there is a drone remote control signal in the signal to be detected. The basis for the determination is the high correlation characteristic of the pilot synchronization sequence, that is, the pilot synchronization sequence can only match the reference signal to produce a significant peak when the drone signal exists.

[0088] In the above implementation, the peak characteristics (including position and amplitude) in the frequency domain convolution result are analyzed, and combined with the dynamic preset threshold conditions (number of protection units, number of reference units and detection threshold), high-precision detection of drone remote control signals is achieved. When the peak characteristics meet the threshold conditions, the system can reliably determine the presence of drone signals; otherwise, the interference of background noise or other non-target signals can be eliminated.

[0089] An embodiment of the present application also discloses a drone detection system based on pilot synchronization sequence matching.

[0090] A drone detection system based on pilot synchronization sequence matching, the detection system comprising: Sequence building module, used to build universal reference sequences; A signal receiving module, used for receiving a signal to be detected; A signal segment interception module, used to intercept a signal segment with a length of a preset number of points from the signal to be detected; A signal segment processing module is used to perform fast Fourier transform on the signal segment to obtain a frequency domain representation of the signal to be detected; A frequency domain conversion module, used for performing a fast Fourier transform on the universal reference sequence according to a preset number of points to obtain a frequency domain universal reference sequence; A conjugate operation module, used for performing a conjugate operation on a frequency domain universal reference sequence to obtain a frequency domain reference signal; A convolution module, used to perform a convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; The constant false alarm rate detection module is used to perform constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result.

[0091] A drone detection system based on pilot synchronization sequence matching in an embodiment of the present application can implement any of the above-mentioned drone detection methods, and the specific working process of each module in the drone detection system can refer to the corresponding process in the above-mentioned method embodiment.

[0092] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0093] The embodiment of the present application also discloses a computer device.

[0094] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a drone detection method based on pilot synchronization sequence matching as described above is implemented.

[0095] The embodiment of the present application also discloses a computer-readable storage medium.

[0096] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the above-mentioned drone detection methods based on pilot synchronization sequence matching.

[0097] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0098] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0099] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A drone detection method based on pilot synchronization sequence matching, characterized in that: The detection method comprises: Constructing a universal reference sequence; receiving a signal to be detected, and extracting a signal segment having a length of a preset number of points from the signal to be detected; Performing a fast Fourier transform on the signal segment to obtain a frequency domain representation of the signal to be detected; Performing a fast Fourier transform on the universal reference sequence according to the preset number of points to obtain a frequency domain universal reference sequence; Performing a conjugate operation on the frequency domain universal reference sequence to obtain a frequency domain reference signal; Performing a convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; A constant false alarm rate test is performed on the frequency domain convolution result to obtain a drone detection result.

2. The method for detecting a drone based on pilot synchronization sequence matching according to claim 1, characterized in that: The drone detection result is used to determine whether there is a drone image transmission signal in the signal to be detected; The step of constructing a universal reference sequence comprises: According to the bandwidth characteristics of the UAV image transmission signal, a pilot synchronization sequence of the UAV image transmission signal is obtained, where the pilot synchronization sequence is composed of a ZC sequence; Construct an all-zero sequence whose pilot synchronization sequence points match the sampling rate of the drone image transmission signal, and embed the pilot synchronization sequences corresponding to different bandwidths into the specified positions of the all-zero sequence to obtain a frequency domain pilot synchronization sequence; wherein the preset number of points is an integer multiple n of the number of points of the pilot synchronization sequence, and n≥2; The frequency domain pilot synchronization sequence is converted into the time domain through inverse fast Fourier transform to obtain the universal reference sequence.

3. The method for detecting a drone based on pilot synchronization sequence matching according to claim 2, characterized in that: The steps of performing constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result include: Analyzing peak features in the frequency domain convolution result; Determining whether the peak characteristic meets a preset threshold condition; If so, it is determined that there is a drone image transmission signal in the signal to be detected, that is, the drone is detected; if not, it is determined that there is no drone image transmission signal in the signal to be detected, that is, the drone is not detected.

4. The method for detecting a drone based on pilot synchronization sequence matching according to claim 1, characterized in that: The drone detection result is used to determine whether there is a drone remote control signal in the signal to be detected; The step of constructing a universal reference sequence comprises: According to the bandwidth characteristics of the UAV remote control signal, a pilot synchronization sequence of the UAV remote control signal is obtained, where the pilot synchronization sequence is composed of a ZC sequence; Construct an all-zero sequence whose pilot synchronization sequence points match the sampling rate of the drone remote control signal, and embed the pilot synchronization sequences corresponding to different bandwidths into the specified positions of the all-zero sequence to obtain a frequency domain pilot synchronization sequence; wherein the preset number of points is an integer multiple n of the number of points of the pilot synchronization sequence, and n≥2; The frequency domain pilot synchronization sequence is converted into the time domain through inverse fast Fourier transform to obtain the universal reference sequence.

5. The method for detecting a drone based on pilot synchronization sequence matching according to claim 4, characterized in that: The steps of performing constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result include: Analyzing peak features in the frequency domain convolution result; Determining whether the peak characteristic meets a preset threshold condition; If so, it is determined that the drone remote control signal exists in the signal to be detected, that is, the drone is detected; if not, it is determined that the drone remote control signal does not exist in the signal to be detected, that is, the drone is not detected.

6. A drone detection method based on pilot synchronization sequence matching according to any one of claims 1 to 5, characterized in that: The constant false alarm rate detection includes a bilateral average constant false alarm rate strategy, a two-side unit average selection strategy or a two-side unit average selection strategy.

7. The method for detecting a drone based on pilot synchronization sequence matching according to claim 3 or 5, characterized in that: The preset threshold conditions include a preset number of protection units, a preset number of reference units and a preset detection threshold.

8. A drone detection system based on pilot synchronization sequence matching, characterized in that: The detection system comprises: Sequence building module, used to build universal reference sequences; A signal receiving module, used for receiving a signal to be detected; A signal segment interception module, used for intercepting a signal segment with a length of a preset number of points from the signal to be detected; A signal segment processing module, used for performing a fast Fourier transform on the signal segment to obtain a frequency domain representation of the signal to be detected; A frequency domain conversion module, used for performing a fast Fourier transform on the universal reference sequence according to the preset number of points to obtain a frequency domain universal reference sequence; A conjugate operation module, used for performing a conjugate operation on the frequency domain universal reference sequence to obtain a frequency domain reference signal; A convolution module, used for performing a convolution operation on the frequency domain representation of the signal to be detected and the frequency domain reference signal to obtain a frequency domain convolution result; The constant false alarm rate detection module is used to perform constant false alarm rate detection on the frequency domain convolution result to obtain the drone detection result.

9. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 7.

Images