Multi-unmanned aerial vehicle identification method based on burst signal detection

Abstract

The invention belongs to the technical field of communication signal detection and identification, and particularly relates to a multi-unmanned aerial vehicle identification method based on burst signal detection, which comprises the following specific steps: S1, constructing an unmanned aerial vehicle signal time-frequency domain characteristic parameter library: based on a communication system of typical unmanned aerial vehicle signals, analyzing characteristic parameters, and establishing an unmanned aerial vehicle signal time-frequency domain characteristic parameter library; parameters of a working frequency band, a signal bandwidth, a burst transmission time slot length and a burst transmission time interval of an unmanned aerial vehicle signal are obtained and divided into time domain parameters and frequency domain parameters. According to the method, through statistical analysis of time domain and frequency domain characteristic parameters of unmanned aerial vehicle burst signals, unmanned aerial vehicle signal sorting of known signal characteristics can be rapidly and efficiently realized without complex processing, and meanwhile, signal characteristic database establishment of unknown unmanned aerial vehicles can be realized based on time domain and frequency domain parameter statistical analysis, so that the unmanned aerial vehicle signal sorting and identification problem can be well realized.

Description

Technical Field

[0001] The present invention relates to the technical field of communication signal detection and recognition, and specifically provides a multi-UAV recognition method based on burst signal detection. Background Art

[0002] With the popularization of UAV technology, the collaborative operation of multiple UAVs (such as logistics distribution, agricultural plant protection, emergency rescue, etc.) has become a trend. However, it also brings potential safety hazards. The recognition of multi-UAV signals is one of the core technologies to meet the control requirements of cluster UAVs, and can be used to distinguish legal and illegal UAVs, track the positions of UAVs, crack communication protocols, etc. The recognition of multi-UAV signals refers to the technical process of detecting, distinguishing, analyzing, and understanding the signals emitted by multiple UAVs. Its core goal is to accurately identify the signal characteristics of different UAVs from a complex electromagnetic environment, so as to realize the monitoring, control, and situation awareness of UAVs.

[0003] However, currently, when recognizing multi-UAV signals, the sorting of UAV signals with known signals is usually carried out by using methods such as machine learning, neural networks, and high-order statistics. As a result, there are problems such as complex calculations and low sorting efficiency. Therefore, a multi-UAV recognition method based on burst signal detection is invented. Summary of the Invention

[0004] To solve the above technical problems, according to one aspect of the present invention, the following technical solutions are provided:

[0005] A multi-UAV recognition method based on burst signal detection, including the following specific steps:

[0006] S1, construction of the time-frequency domain characteristic parameter library of UAV signals: Based on the analysis of the communication system and characteristic parameters of typical UAV signals, obtain the parameters of the working frequency band, signal bandwidth, burst transmission time slot length, and burst transmission time interval of UAV signals, and divide them into time domain parameters and frequency domain parameters;

[0007] S2, detection of UAV burst signals: First, receive broadband signals, then perform transformation, conduct burst signal detection, and then form a list of original burst signal detection results arranged in chronological order according to the start time of the burst signals ;

[0008] n is an integer;

[0009] S3, frequency domain sorting of burst signals: For the list of original burst signal detection results Sort the signals according to the signal center frequency and signal bandwidth;

[0010] S4, Time-domain analysis of burst signals: For the UAV burst signal list after frequency-domain screening, further sorting is performed according to time-domain characteristic parameters;

[0011] S5, Comprehensive decision of UAV signals: Compare the results of the burst signal frequency-domain analysis in S3 and the time-domain analysis in S4 with the UAV burst signal parameters.

[0012] The specific steps of S1 are as follows:

[0013] S11: The frequency-domain characteristic parameters of the UAV signal are the operating frequency band and the signal bandwidth BW;

[0014] S12: The time-domain characteristic parameters of the UAV signal are the burst transmission time slot length and the burst transmission time interval , where and The subscript K in represents different values of the two characteristic parameters. The burst transmission time interval is defined as the time interval between two adjacent time slot transmissions;

[0015] S13: Based on the above time and frequency domain parameter classification, construct the time-domain and frequency characteristic parameter libraries of different UAV signals. Different UAVs usually have different time-frequency domain characteristic parameters.

[0016] The specific steps of S2 are as follows:

[0017] S21: Use a broadband detection receiver to detect the broadband signal in the UAV operating frequency band, and generate a baseband signal after analog mixing and digital down-conversion;

[0018] S22: Perform a short-time Fourier transform STFT on the received broadband baseband signal to carry out burst signal detection, and obtain the parameters of the center frequency, signal bandwidth, start time of the burst signal, duration of the burst signal, and average power of the burst signal;

[0019] S23: Arrange the detected multiple burst signals in the order of the start time of the burst signal to form a list of original burst signal detection results arranged in chronological order:

[0020] ;

[0021] Each burst signal detection result respectively corresponds to the start time of the burst signal , the duration of the burst signal , the center frequency of the received signal , the bandwidth of the received signal , and the power of the received signal parameters.

[0022] The specific steps of S3 are as follows:

[0023] S31: According to the signal bandwidth of the burst signal, from the burst signal detection result list screen out the burst signal detection results that meet specific conditions , and the screening condition is the absolute value of the difference from the typical bandwidth value of the UAV signal does not exceed the bandwidth threshold ; After screening, form the burst detection results after preliminary screening , where m represents an integer;

[0024] S32: For the burst signal detection results after screening , perform two-dimensional histogram segmentation statistics according to the center frequency and signal bandwidth value of the burst signal , and count the occurrence times of burst signals with specific center frequencies and signal bandwidths , to form a frequency domain histogram ;

[0025] S33: Perform threshold processing on the original frequency domain histogram to filter out the detection results with the occurrence times lower than the histogram signal occurrence threshold T2; the threshold value typical value can be set to R times the highest occurrence times of the burst signal ; Through , obtain the frequency domain histogram after preliminary screening , where is the center frequency after preliminary screening, is the signal bandwidth value after preliminary screening, is the occurrence times of burst signals with specific center frequencies and signal bandwidths after preliminary screening;

[0026] S34: According to the frequency domain histogram after preliminary screening in which the signals that meet , conditions, perform further sorting on the burst signal detection list:

[0027] to obtain , as the input of the burst signal time domain analysis step.

[0028] The specific steps of S4 are as follows:

[0029] S41: Sort the burst signals after frequency domain screening above by time, calculate the starting time difference between adjacent time burst signals , and expand the burst signal detection results to ;

[0030] S42: According to the duration of the burst signal , further screen the burst signal detection list, and the screening condition is the absolute value of the difference of the UAV signal characteristic parameter does not exceed the threshold , and the burst signal transmission interval and the UAV signal characteristic parameter the difference of does not exceed the threshold ;

[0031] ;

[0032] When there are multiple values, satisfy and any one When the value satisfies the above conditions, it meets the above conditions;

[0033] When there are multiple values, satisfy and any one When the value satisfies the above conditions, it meets the above conditions;

[0034] , The typical value of is , that is ;

[0035] The burst signal list after screening according to the above steps is:

[0036] ;

[0037] .

[0038] In S5 for the time-domain signal, compare the burst signal duration list , the burst signal transmission interval time list with the UAV signal characteristic parameter library, and calculate the correlation coefficient; based on the correlation threshold judgment, determine the type of the UAV signal, and output the UAV signal alarm result; for the frequency-domain and time-frame structure analysis results that cannot be matched with the feature database parameters, classify them into the unknown UAV signal library, and gradually construct the unknown UAV signal feature library according to the method in S1.

[0039] The specific steps of S5 are as follows:

[0040] S51: For ; ;

[0041] According to the center frequency , further count the number of times the burst transmission time appears ;

[0042] S52: If the ratio of the number of occurrences of a burst signal at a certain operating frequency to the average number of occurrences of burst signals exceeds a specific threshold , and the absolute number of occurrences meets a specific threshold , then it is determined that there is a drone signal at that operating frequency;

[0043] S53: Report the drone signals that appear at the above-mentioned operating frequency, as well as their corresponding time and frequency domain parameters, and display them on the computer interface;

[0044] S54: For the frequency domain and time frame structure analysis results that cannot be matched with the parameters in the feature database, classify them as unknown drone signal parameters; if the number of relevant parameters exceeds a certain amount, construct an unknown drone signal feature library according to the parameters in S1.

[0045] Compared with the prior art:

[0046] Through the statistical analysis of the time domain and frequency domain characteristic parameters of drone burst signals, the present invention can quickly and efficiently realize the sorting of drone signals with known signal characteristics without complex processing. At the same time, based on the statistical analysis of time and frequency domain parameters, it can realize the construction of a signal feature library for unknown drones, and preferably solve the problem of drone signal sorting and recognition. Brief Description of the Drawings

[0047] Figure 1 It is a schematic diagram of the overall framework of the present invention. Detailed Embodiment

[0048] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will further describe the embodiments of the present invention in detail with reference to the accompanying drawings.

[0049] The present invention provides a multi-drone recognition method based on burst signal detection. Please refer to Figure 1 , and the specific steps are as follows:

[0050] S1, construction of the time-frequency domain characteristic parameter library of drone signals: Based on the analysis of the communication system and characteristic parameters of typical drone signals, obtain the parameters of the operating frequency band, signal bandwidth, burst transmission time slot length, and burst transmission time interval of drone signals, and classify them into time domain parameters and frequency domain parameters;

[0051] S2, detection of drone burst signals: First, receive broadband signals, then perform transformation, conduct burst signal detection, and then form a list of original burst signal detection results arranged in chronological order according to the start time of the burst signals ;

[0052] n represents an integer;

[0053] S3, Burst signal frequency-domain sorting: For the list of detection results of the original burst signals Sort the signals according to the signal center frequency and signal bandwidth;

[0054] S4, Burst signal time-domain analysis: For the list of UAV burst signals after frequency-domain screening, further sort according to the time-domain characteristic parameters;

[0055] S5, Comprehensive judgment of UAV signals: Compare the results of the burst signal frequency-domain analysis in S3 and the time-domain analysis in S4 with the UAV burst signal parameters.

[0056] The specific steps of S1 are as follows:

[0057] S11: The frequency-domain characteristic parameters of the UAV signal are the working frequency band and the signal bandwidth BW;

[0058] S12: The time-domain characteristic parameters of the UAV signal are the burst transmission time slot length and the burst transmission time interval , where and The subscript K in represents that the two characteristic parameters have different values, and the burst transmission time interval is defined as the time interval between two adjacent slot transmissions;

[0059] S13: Based on the above time and frequency domain parameter classification, construct the time-domain and frequency characteristic parameter libraries of different UAV signals. Different UAVs usually have different time-frequency domain characteristic parameters.

[0060] The specific steps of S2 are as follows:

[0061] S21: Use a broadband detection receiver to detect the broadband signals in the UAV working frequency band, and generate a baseband signal after analog mixing and digital down-conversion;

[0062] S22: Perform a short-time Fourier transform STFT on the received broadband baseband signal, carry out burst signal detection, and obtain the parameters of the burst signal center frequency, signal bandwidth, burst signal start time, burst signal duration, and burst signal average power;

[0063] S23: Arrange the detected multiple burst signals in the order of the burst signal start time to form a list of original burst signal detection results arranged in chronological order:

[0064] ;

[0065] Each burst signal detection result , respectively corresponding to the burst signal start time , Duration of the burst signal , Center frequency of the received signal , Bandwidth of the received signal , Power of the received signal Parameter

[0066] The specific steps of S3 are as follows:

[0067] S31: According to the signal bandwidth of the burst signal, select the burst signal detection results that meet specific conditions from the burst signal detection result list . The selection condition is that the absolute value of the difference from the typical bandwidth value of the UAV signal does not exceed the bandwidth threshold ; After selection, form the burst detection results after preliminary screening , where m represents an integer;

[0068] S32: For the burst signal detection results after selection , perform two-dimensional histogram segmentation statistics according to the center frequency and signal bandwidth value of the burst signal , and count the number of occurrences of burst signals with specific center frequencies and signal bandwidths to form a frequency-domain histogram ;

[0069] S33: Perform threshold processing on the original frequency-domain histogram to filter out the detection results with the number of occurrences lower than the histogram signal occurrence threshold T2; The threshold value typical value can be set to R times the highest number of occurrences of burst signals ; Through , obtain the frequency-domain histogram after preliminary screening , where is the center frequency after preliminary screening, is the signal bandwidth value after preliminary screening, is the number of occurrences of burst signals with specific center frequencies and signal bandwidths after preliminary screening;

[0070] S34: According to the signal detection results that meet the and conditions in the frequency-domain histogram after preliminary screening , further sort the burst signal detection list:

[0071] to obtain , which is used as the input for the burst signal time-domain analysis step

[0072] The specific steps of S4 are as follows:

[0073] ​S41: Sort the burst signals after frequency-domain screening above by time, and calculate the difference in start times between adjacent burst signals , and expand the burst signal detection result to ;

[0074] S42: Further screen the burst signal detection list according to the duration of the burst signal , , and the screening condition is that the absolute value of the difference of the UAV signal characteristic parameters does not exceed the threshold , and the difference between the burst signal transmission interval and the UAV signal characteristic parameter does not exceed the threshold ;

[0075] ;

[0076] When there are multiple values, it satisfies and any value satisfies the above conditions, then it meets the above conditions;

[0077] When there are multiple values, it satisfies and any value satisfies the above conditions, then it meets the above conditions;

[0078] , The typical values of are ;

[0079] The burst signal list screened according to the above steps is:

[0080] ;

[0081] .

[0082] In S5 above, for the time-domain signal, compare the burst signal duration list , the burst signal transmission interval time list with the UAV signal characteristic parameter library, and calculate the correlation coefficient; based on the correlation threshold decision, determine the type of the UAV signal and output the UAV signal alarm result; for the frequency-domain and time-frame structure analysis results that cannot be matched with the feature database parameters, classify them into the unknown UAV signal library, and gradually construct the unknown UAV signal feature library according to the method in S1.

[0083] The specific steps of S5 are as follows:

[0084] S51: For ; ;

[0085] According to the center frequency , further count the number of occurrences of the burst transmission time ;

[0086] S52: If the ratio of the number of occurrences of the burst signal at a certain operating frequency to the average number of occurrences of the burst signal exceeds a specific threshold , and the absolute number of occurrences meets a specific threshold , then it is determined that there is a drone signal at this operating frequency;

[0087] S53: Report the drone signal that appears at the above-mentioned operating frequency, and its corresponding time and frequency domain parameters, and display them on the computer interface;

[0088] S54: For the frequency domain and time frame structure analysis results that cannot be matched with the parameters in the feature database, classify them as unknown drone signal parameters; if the relevant parameters exceed a certain number, then construct an unknown drone signal feature library according to the parameters in S1.

[0089] Although the present invention has been described above with reference to the embodiments, various improvements can be made to it and its components can be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed in the present invention can be combined with each other in any way, and the exhaustive description of these combinations is not given in this specification only for the sake of saving space and resources. Therefore, the present invention is not limited to the specific embodiments disclosed in the text, but includes all technical solutions falling within the scope of the claims.

Claims

1. A multi-UAV recognition method based on burst signal detection, characterized in that, The specific steps are as follows: S1. Construction of the time-frequency domain characteristic parameter library for UAV signals: Based on the analysis of the communication systems and characteristic parameters of typical UAV signals, obtain the parameters of the working frequency band, signal bandwidth, burst transmission time slot length, and burst transmission time interval of UAV signals, and divide them into time-domain parameters and frequency-domain parameters; S2, Drone Burst Signal Detection: First, receive the broadband signal, then perform transformation, conduct burst signal detection, and then form a list of original burst signal detection results arranged in chronological order according to the start time of the burst signal ; n is an integer; S3, Burst signal frequency domain sorting: For the list of original burst signal detection results Sort the signals according to the signal center frequency and signal bandwidth; S4. Time-domain analysis of burst signals: For the list of UAV burst signals after frequency-domain screening, further sort them according to time-domain characteristic parameters; S5. Comprehensive judgment of UAV signals: Compare the results of the burst signal frequency-domain analysis in S3 and the time-domain analysis in S4 with the parameters of UAV burst signals.

2. The multi-UAV recognition method based on burst signal detection according to claim 1, wherein The specific steps of S1 are as follows: S11: The frequency-domain characteristic parameters of the UAV signal are the operating frequency band and the signal bandwidth BW; S12: The time-domain characteristic parameters of the UAV signal are the burst transmission slot length , the burst transmission time interval , where , the subscript K in indicates that the two characteristic parameters have different values, and the burst transmission time interval is defined as the time interval between two adjacent slot transmissions; S13: Based on the above time- and frequency-domain parameter classifications, construct the time-domain and frequency characteristic parameter libraries for different UAV signals. Different UAVs usually have different time-frequency domain characteristic parameters.

3. A multi-UAV recognition method based on burst signal detection according to claim 1, characterized in that, The specific steps of S2 are as follows: S21: Use a broadband detection receiver to detect the broadband signals in the working frequency band of UAVs, and generate baseband signals after analog mixing and digital down-conversion; S22: Perform short-time Fourier transform (STFT) on the received broadband baseband signals to conduct burst signal detection, and obtain the parameters of the center frequency, signal bandwidth, start time of the burst signal, duration of the burst signal, and average power of the burst signal; S23: Arrange the detected multiple burst signals in the order of the start time of the burst signal to form a list of the original burst signal detection results arranged in chronological order: ； Each burst signal detection result , corresponding to the start time of the burst signal , the duration of the burst signal , the center frequency of the received signal , the bandwidth of the received signal , the power of the received signal parameters.

4. A multi-UAV recognition method based on burst signal detection according to claim 1, characterized in that The specific steps of S3 are as follows: S31: According to the signal bandwidth of the burst signal, from the list of burst signal detection results screen the burst signal detection results that meet specific conditions , and the screening condition is the absolute value of the difference from the typical bandwidth value of the UAV signal does not exceed the bandwidth threshold ; After screening, form the burst detection results after preliminary screening , where m is an integer; S32: For the detection results of the filtered burst signals , according to the center frequency of the burst signal , and the signal bandwidth value, perform two-dimensional histogram segmentation statistics to count the occurrence times of burst signals with specific center frequencies and signal bandwidths , and form a frequency-domain histogram ; S33: Threshold process the original frequency domain histogram to filter out the detection results with the number of occurrences lower than the histogram signal appearance threshold T2; the threshold value The typical value can be set to R times the highest number of occurrences of the burst signal ; Through , obtain the frequency domain histogram after primary screening , where is the center frequency after primary screening, is the signal bandwidth value after primary screening, is the number of occurrences of the specific center frequency and signal bandwidth burst signal after primary screening; S34: According to the frequency domain histogram after primary screening that satisfies , the signal detection results of the conditions, for the burst signal detection list: Further sorting is performed to obtain , which is used as the input for the time-domain analysis step of the burst signal.

5. A multi-UAV recognition method based on burst signal detection according to claim 1, characterized in that, The specific steps of S4 are as follows: S41: Sort the burst signals after frequency domain screening as described above in chronological order, and calculate the difference in start times between adjacent burst signals , and expand the burst signal detection result to ; S42: According to the duration of the burst signal , further screen the burst signal detection list, and the screening condition is that the absolute value of the difference of the UAV signal feature parameters does not exceed the threshold , and the difference between the burst signal transmission interval and the UAV signal feature parameters does not exceed the threshold ; ； When there are multiple values, satisfying with any when the value satisfies the above conditions, it meets the above conditions; When there are multiple values, it satisfies with any one When the value satisfies the above conditions, it all conforms to the above conditions; , The typical value of is ; The list of burst signals screened according to the above steps is: ； 。 6. The multi-UAV recognition method based on burst signal detection according to claim 1, wherein In S5, for the time-domain signal, the list of burst signal duration , the list of burst signal transmission interval time is compared with the UAV signal feature parameter library to calculate the correlation coefficient; Based on relevant threshold judgments, determine the types of UAV signals and output the UAV signal alarm results; for the frequency-domain and time-frame structure analysis results that cannot be matched with the parameters in the feature database, classify them into the unknown UAV signal library, and gradually construct the unknown UAV signal feature library according to the method in S1.

7. A multi-UAV recognition method based on burst signal detection according to claim 6, characterized in that, The specific steps of S5 are as follows: S51: For ; ; According to the center frequency , further count the number of occurrences of the burst transmission time ; S52: If the ratio of the number of occurrences of a burst signal at a certain operating frequency , to the average number of occurrences of burst signals exceeds a specific threshold , and the absolute number of occurrences meets a specific threshold , then it is determined that there is a drone signal at that operating frequency; S53: Report the UAV signals that appear at the above working frequencies and their corresponding time- and frequency-domain parameters, and display them on the computer interface; S54: For the frequency-domain and time-frame structure analysis results that cannot be matched with the parameters in the feature database, classify them into unknown UAV signal parameters; if the relevant parameters exceed a certain number, construct the unknown UAV signal feature library according to the parameters in S1.

Images