Low-complexity unmanned aerial vehicle signal reconnaissance and interference system

Abstract

A low-complexity unmanned aerial vehicle signal reconnaissance and interference system comprises a software radio receiver, a software radio transmitter, an unmanned aerial vehicle signal reconnaissance module and an unmanned aerial vehicle signal interference module. The software radio receiver sweeps frequency in a specified frequency range and outputs frequency spectrum slices, and the unmanned aerial vehicle signal reconnaissance module stores reconnaissance results in structural body data after carrying out fixed-frequency signal detection and parameter estimation, frequency hopping signal detection and parameter estimation and broadband signal parameter estimation processing on the frequency spectrum slices step by step. And the software radio transmitter radiates interference signals and interferes the unmanned aerial vehicle signal with the largest threat. According to the method, a general software radio receiver and a transmitter are used as an algorithm deployment platform, unmanned aerial vehicle signal reconnaissance in a complex electromagnetic environment is realized by using a low-complexity signal detection and parameter estimation algorithm, and unmanned aerial vehicle signal interference is guided by rapidly utilizing reconnaissance information through a simple interference strategy; and detection and countering of the unmanned aerial vehicle are realized.

Description

Technical field:

[0001] The invention belongs to the technical field of radio monitoring, and in particular relates to a low-complexity unmanned aerial vehicle signal reconnaissance and jamming system. Background technology:

[0002] Drones have been widely used in various scenarios such as aerial photography, surveying and mapping, electricity, security, and firefighting. The emergence of a large number of consumer-grade drones has greatly lowered the threshold for drone use, but has also brought huge challenges to drone control. In particular, the widespread use of drones in the military in recent years has made people more fully aware of the huge value and threats brought by drones. Ground equipment without drone defense capabilities will be completely suppressed by drones, and key facilities without drone countermeasure capabilities will have no privacy and security in the face of drones. It is very urgent to carry out research on drone detection and countermeasures.

[0003] The detection methods of drones include radar detection, photoelectric detection, acoustic wave detection and radio detection. The radio detection method uses the radio signals used in the operation of drones for passive detection. By adopting general software radio equipment, the system cost can be greatly reduced. Moreover, compared with radar detection methods, under some complex terrain conditions, its blind spot is smaller and early warning can be issued more quickly. Interfering with the control signals, image transmission signals and navigation signals of drones can cut off the control of drone operators, interfere with the judgment of drone operators, and prevent drone operators from completing their tasks. It is an effective means to counter drones.

[0004] Directly using the receiver IQ data stream mode for spectrum sensing requires the receiver to have an instantaneous bandwidth covering the frequency band of interest, which will increase the cost of the receiver, and the huge real-time data stream will bring great pressure to the back-end data processing. Using the receiver's sweep mode, a large amount of spectrum calculations can be completed on the front end using FPGA, and a wider spectrum range can be covered at a lower cost by splicing multiple instantaneous bandwidths, reducing the back-end data processing pressure, reducing the complexity and cost of the system, and only through the back-end software development, the reconnaissance and interference of drone signals can be completed flexibly and quickly. Summary of the invention:

[0005] In order to overcome the above-mentioned deficiencies, the present invention provides a low-complexity UAV signal reconnaissance and jamming system.

[0006] The technical solution adopted by the present invention to solve its technical problem is:

[0007] A low-complexity UAV signal reconnaissance and jamming system, comprising:

[0008] The software radio receiver scans the electromagnetic environment within a specified frequency range through the receiving antenna to form a spectrum slice;

[0009] The UAV signal reconnaissance module performs signal detection and parameter estimation on the spectrum slice, and saves the reconnaissance results into three structures: fixed frequency signal parameters, frequency hopping signal parameters and broadband signal parameters;

[0010] The UAV signal jamming module matches the jamming strategy according to the reconnaissance results and dynamically adjusts the jamming resources;

[0011] Software radio transmitter, transmitting antenna radiation source interference signal, interferes with drone signal.

[0012] The steps for detecting and estimating the fixed frequency signal parameters are as follows:

[0013] 1) Slice the spectrum into spectrum blocks;

[0014] 2) Calculate the cumulative average spectrum of the spectrum block;

[0015] 3) Use the average of the first 90% of the amplitudes in the cumulative average spectrum as the noise floor estimate;

[0016] 4) Perform multi-signal parameter estimation;

[0017] 5) Get the fixed frequency signal parameters and save them as structure data.

[0018] The steps for detecting and estimating the frequency hopping signal parameters are as follows:

[0019] 1) Use spectrum cancellation method to eliminate fixed frequency signal;

[0020] 2) Use morphological opening operation to eliminate the noise in the spectrum block;

[0021] 3) Obtaining parameters of multiple signals existing in a spectrum block by a multi-signal parameter estimation method;

[0022] 4) retaining the region mask whose region width is less than the threshold, and using the four-connected region to detect the mask of the entire spectrum block;

[0023] 5) Use the DSCAN algorithm to cluster and sort the bandwidth, and construct an estimate to determine whether a frequency hopping signal exists. If a frequency hopping signal exists, output the estimated value of the frequency hopping signal parameter and save it as structure data.

[0024] The multi-signal parameter estimation step is:

[0025] 1) Set the detection threshold, and determine whether there is a signal at the frequency point corresponding to the spectrum slice based on the estimated noise floor of the current spectrum block, and use a mask to represent the area where the signal exists in a spectrum slice;

[0026] 2) Convert the mask representation into a region representation;

[0027] 3) Fusion of signal areas where the area interval is less than the minimum channel interval;

[0028] 4) removing the signal area whose interval width is less than the minimum signal bandwidth in the signal area to obtain a rough signal area;

[0029] 5) After each rough signal region is expanded to the left and backward, parameters are re-estimated in the expanded signal region to obtain the variance, skewness, kurtosis and symmetry parameter estimates of the precise signal region and the signal spectrum in the region.

[0030] The steps for parameter estimation of broadband signals are:

[0031] 1) Retain the area mask whose width is greater than the threshold, and use morphological closing operation to fill the concave area that may exist in the middle of the broadband signal;

[0032] 2) Using the four-connected region detection method to mark the mask area of ​​the broadband signal, the center frequency, center time, bandwidth, and duration parameters of each signal area are obtained;

[0033] 3) Use the DSCAN clustering method to sort the center frequencies, obtain broadband signal parameters, and save them as structured data.

[0034] There are four interference strategies:

[0035] Interference strategy 1: According to the number, center frequency, bandwidth and power amplitude of the fixed-frequency signals obtained by the reconnaissance module, at the center frequency positions of the first two fixed-frequency signals with the largest signals, the transmitter RF channel 1 generates a QPSK signal covering the corresponding fixed-frequency signal bandwidth; according to the frequency hopping frequency set obtained by the reconnaissance module, the transmitter RF channel 2 generates a multi-tone interference signal at the corresponding frequency point;

[0036] Interference strategy 2, based on the difference between the minimum frequency and the maximum frequency of the frequency hopping frequency set obtained by the reconnaissance module, estimates the frequency hopping sequence bandwidth of the frequency hopping signal, and generates a QPSK signal with the same bandwidth by transmitter RF channel 1; based on the frequency hopping frequency set, transmitter RF channel 2 generates a multi-tone interference signal of the corresponding frequency point;

[0037] Interference strategy 3: According to the number, center frequency, bandwidth and power amplitude of the broadband signals obtained by the reconnaissance module, at the center frequency positions of the first two broadband signals with the largest signal peak power, the transmitter RF channel 1 generates a QPSK signal with the same signal bandwidth; according to the number, center frequency, bandwidth and power amplitude of the fixed-frequency signals obtained by the reconnaissance module, at the center frequency positions of the first two fixed-frequency signals with the largest signal peak power, the transmitter RF channel 2 generates a QPSK signal with the same signal bandwidth;

[0038] Interference strategy 4, according to the number, center frequency, bandwidth and peak power of the broadband signals obtained by the reconnaissance module, at the center frequency positions of the first two broadband signals with the largest signal peak power, the transmitter RF channel 1 generates a QPSK signal with the same signal bandwidth, and the transmitter RF channel 2 generates a comb spectrum signal with the same bandwidth.

[0039] The steps of dynamically regulating the interference resources by matching the interference strategy according to the reconnaissance results are as follows: if there is a frequency hopping signal in the information output by the reconnaissance module, continue to determine whether there is a fixed frequency signal; if there is no fixed frequency signal, directly execute interference strategy 2; if there is a fixed frequency signal and the peak value of the fixed frequency signal exceeds the set threshold, directly use interference strategy 2; if there is a fixed frequency signal but the peak value of the fixed frequency signal does not exceed the set threshold, execute interference strategy 1; if there is no frequency hopping signal in the detection information output by the reconnaissance module, continue to determine whether there is a broadband signal; if there is no broadband signal, continue to conduct reconnaissance and monitor the radio signal spectrum; if there is a broadband signal but no fixed frequency signal, execute interference strategy 4; if there is a broadband signal but the peak value of the fixed frequency signal does not exceed the set threshold, execute interference strategy 4; if there is a broadband signal and the peak value of the fixed frequency signal exceeds the set threshold, execute interference strategy 3; when the execution time of any one of the four interference strategies exceeds times the spectrum length of the reconnaissance module, jump to the drone signal reconnaissance and start a new round of drone signal reconnaissance and interference.

[0040] Due to the adoption of the above-mentioned technical solution, the present invention has the following advantages:

[0041] The present invention provides a low-complexity UAV signal reconnaissance and jamming system, which is aimed at application scenarios in complex electromagnetic environments. It takes radio spectrum data as input, can use FPGA at the front end to complete a large number of spectrum calculations, and can also cover a wider spectrum range at a lower cost through a method of splicing multiple instantaneous bandwidths, reducing the back-end data processing pressure. It can quickly and effectively interfere with UAV signals through simple jamming strategies, and can be combined with a low-cost general software radio platform to flexibly and quickly complete the reconnaissance and jamming of UAV signals. Description of the drawings:

[0042] Figure 1 It is the overall flow chart of the present invention;

[0043] Figure 2 It is a schematic diagram of the multi-signal parameter estimation method;

[0044] Figure 3 Drone signal jamming strategy flow chart;

[0045] Figure 4 Spectrum blocks for 2400MHz-2483MHz radio signals;

[0046] Figure 5 The cumulative average spectrum of the fixed frequency signal;

[0047] Figure 6 Spectrum block mask for frequency hopping signals;

[0048] Figure 7 Spectral block mask for wideband signals. Specific implementation method:

[0049] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0050] Drones operate in the open ISM frequency band, which is shared by IEEE802.11 series signals, Bluetooth signals, ZigBee signals, etc. Some manufacturers also develop dedicated drone signals, which means that drone signal reconnaissance and interference need to be carried out in a complex, non-cooperative electromagnetic environment.

[0051] like Figure 1 As shown, the present invention provides a low-complexity UAV signal reconnaissance and jamming system, including: a software radio receiver, a software radio transmitter, a UAV signal reconnaissance module, a UAV signal jamming module, a receiving antenna and a transmitting antenna.

[0052] The software radio receiver scans the electromagnetic environment within the specified frequency range through the receiving antenna to form a spectrum slice; the drone signal reconnaissance module performs signal detection and parameter estimation on the spectrum slice, and saves the reconnaissance results into three structures: fixed-frequency signal parameters, frequency-hopping signal parameters, and broadband signal parameters; the drone signal interference module matches the interference strategy according to the reconnaissance results, generates targeted interference signals, and dynamically regulates the interference resources; the software radio transmitter transmits the antenna radiation source interference signal to interfere with the drone signal.

[0053] a) Fixed frequency signal detection and parameter estimation

[0054] Set the software radio receiver to sweep mode and set parameters such as center frequency, sweep bandwidth, reference power level, sweep time, etc. After the software radio receiver is working, it will continuously output radio spectrum slices, referred to as spectrum slices. Spectrum slices are 1×N vectors, denoted as SpecSlice1×N . Spectrum slices reflect the amplitude of different frequency components in the electromagnetic environment within a certain period of time. The time interval between adjacent spectrum slices is related to the sweep time SweepTime. The interval dF between adjacent frequency points in a spectrum slice is related to the sweep bandwidth SweepSpan and the total number of frequency points N, satisfying SweepSpan = dF*N. See the attached for the detection and parameter estimation process of fixed frequency signals. Figure 1 .

[0055] The detailed steps for detecting and estimating the fixed-frequency signal parameters are as follows:

[0056] 1. Forming spectrum blocks

[0057] Arrange the spectrum slices output by the software radio receiver from top to bottom in chronological order. When M spectrum slices are received cumulatively, an M×N spectrum block SpecBlock can be formed. M×N ,The spectrum block is the key object for subsequent signal processing.

[0058] 2. Cumulative average spectrum

[0059] Calculate the spectrum block SpecBlock M×N The average value of each column gives the cumulative average spectrum SpecBlockAVg 1×N , see equation (1). The cumulative average spectrum is the average value of each frequency component within the M spectrum slice time.

[0060]

[0061] Due to the significant differences in the time-frequency domain characteristics of fixed-frequency signals, frequency-hopping signals and burst signals, after cumulative averaging calculation, the amplitude of short-duration signals such as frequency-hopping signals and burst signals will be greatly reduced, while the amplitude of long-duration signals such as fixed-frequency signals will not change much. Cumulative averaging can suppress frequency-hopping signals, burst signals, etc. and extract the fixed-frequency signal in the spectrum.

[0062] 3. Noise Estimation

[0063] First, the cumulative average spectrum SpecBlockAvg 1×N Arrange them in ascending order by amplitude; then, take the average of the amplitudes of the first 90% of the sorting as the estimated value of the spectrum background noise (BGNoise).

[0064] 4. Multi-signal parameter estimation

[0065] Multi-signal parameter estimation estimates the number of signals, center frequency, bandwidth, peak value, area mask, and statistical characteristic parameters in a spectrum slice by setting appropriate minimum signal bandwidth BW_Min and minimum channel interval ChannMin parameters. The multi-signal parameter estimation method will be reused in subsequent frequency hopping signals. The specific processing flow is shown in the attached figure. Figure 2 shown.

[0066] First, set the detection threshold Thold and determine a spectrum slice SpecSlice based on the background noise BGNoise estimated by the current spectrum block. j Whether there is a signal in the N frequency components of j∈[1,N], if the amplitude of the jth frequency component is greater than BGNoise+Thold, it is judged that there is a signal at the jth frequency component. Construct Mask 1×N Describe the area where the signal exists. When the spectrum amplitude of the jth frequency component is greater than or equal to the threshold, the mask corresponding to the frequency component is 1, otherwise it is 0, see formula (2). Through this operation, the area where the signal exists in a spectrum slice can be represented in the form of a mask.

[0067]

[0068] Secondly, Mask 1×N The mask representation is converted to the region representation Roi. For example, a mask representation Mask 1×27 =[0 0 1 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 1 1 1 1 1 1 1 1], converted to the region representation Roi = [3 6; 11 13; 19 27].

[0069] Thirdly, the regional representation forms are fused, and the signal areas where the regional interval RoiSpan is smaller than the minimum channel interval ChannMin are fused to obtain the fused signal area, where the value of the minimum channel interval ChannMin is related to dF.

[0070] Then, the signal area in the fusion signal area with interval width RoiWidth less than the minimum signal bandwidth BW_Min is removed (the signal with smaller bandwidth is treated as noise) to obtain a rough signal area, where the value of the minimum signal bandwidth BW_Min is related to dF.

[0071] Finally, each rough signal area is expanded to the left and back by GuardCell guard cells, and the parameters are estimated again in the expanded signal area to obtain the precise signal area and the estimated values ​​of the parameters of the signal in the area. Assuming that L rough signal areas are obtained, for the kth rough signal area, k∈[1,L], its area range is [RoiRoughL k ,RoiRoughR k ], expand the area to the left and right by GuardCell protection units, GuardCell k =floor((RoiRoughR k -RoiRoughL k ) / 2). In the extended area [RoiRoughL k -GuardCell k ,RoiRoughR k +GuardCell k ]Search for the maximum value, RoiRoughL k -GuardCell k ≥1,RoiRoughR k +GuardCell k ≤N, take the maximum value in the area as the signal peak, and search left and right from the peak for the corresponding index when the signal amplitude drops by 3dB. k and IndexR k , get the left boundary of the precise signal area RoiAcuL k =RoiRoughL k -GuardCell k +IndexL k ,

[0072] The right border is RoiAcuR k =RoiRoughL k -GuardCell k +IndexRt k , then the signal bandwidth of this area is [IndexR k -IndexL k ], the signal center frequency is RoiRoughL k -GuardCell k +floor((IndexR k -IndexL k ) / 2). The variance, skewness, kurtosis and symmetry of the spectrum are calculated in the 3dB bandwidth region as the statistical characteristics of the signal in this region.

[0073] 5. Fixed frequency signal parameters

[0074] The cumulative average spectrum SpecBlockAVg is estimated by using the above multi-parameter estimation method. 1×N Processing is performed to estimate the number of fixed-frequency signals and various parameters. A structure data format FixedSignal.Block{Num}.{Parameters} is constructed to save the parameter estimation value of the fixed-frequency signal, where Num is used to count the processed spectrum blocks, and Parameters stores the center frequency CF, bandwidth BW, peak Peak, area mask Mask, and statistical feature StatisticFeature, where StatisticFeature stores the variance, kurtosis, skewness, and symmetry parameters. By constructing a structure about the estimated value of the fixed-frequency signal, the intermediate processing results can be saved to guide the interference of drone signals.

[0075] b) Frequency hopping signal detection and parameter estimation

[0076] The electromagnetic environment within the working frequency band of drones is complex, and it is difficult to obtain accurate frequency hopping signal parameter estimation results by directly estimating the frequency hopping signal. First, the spectrum cancellation method is used to eliminate the fixed-frequency signal; second, the morphological opening operation is used to eliminate the noise in the spectrum block; third, the parameters of multiple signals in a spectrum block are obtained through the multi-signal parameter estimation method; then, the area mask with a width less than the threshold is retained, and the four-connected area is used to detect the entire spectrum block mask; finally, the DSCAN (Density-Based Clustering of Applications with Noise) algorithm is used to cluster and sort the bandwidth, and an estimator is constructed to determine whether there is a frequency hopping signal. If there is a frequency hopping signal, the frequency hopping signal parameter estimation value is output. The specific process is shown in the attached Figure 1 , the detailed steps are as follows:

[0077] 1. Use spectrum cancellation method to eliminate fixed frequency signal.

[0078] Subtract the cumulative average spectrum SpecBlockAvg from each spectrum slice SpecSlice in the spectrum block to obtain the cancellation spectrum block SpecBlockEqu M×N , see formula (3), this operation can cancel the fixed frequency signal, and the algorithm is simple.

[0079] SpecBlockEqu M×N =SpecBlock M×N -repmat(SpecBlockAvg 1×N , M1)+BGNoise (3)

[0081] Among them, repmat means to copy M rows of the 1×N dimension row vector SpecBlockAvg into an M×N matrix, and BGNois e is the noise floor estimate for this spectral block.

[0082] 2. Use morphological opening operation to remove the noise present in the spectrum block.

[0083] Construct a two-dimensional rectangular structure element SeRec1=ones(p,q). The length q of the two-dimensional rectangular structure element is set according to dF and the minimum signal bandwidth BW1 Min. The reference value is q=round(BW1_Min / dF). If dF is 50KHz and BW1_Min is 500KHz, then q=10. The width p is set according to the spectrum scanning time SweepTime and the minimum signal duration DT_Min. Independent areas with very short duration and very small bandwidth are regarded as noise, p=round(DT_Min / SweepTime). If SweepTime=1ms and DT_Min=2ms, then p=2.

[0084] Spectrum cancellation module SpecBlockEqu using two-dimensional rectangular structure element SeRec1 M×N Perform an open operation to remove the influence of noise and obtain the spectrum block SpecBlockEquOpen after the open operation M×N .

[0085] 3. Obtain the parameters of multiple signals in a spectrum block through a multi-signal parameter estimation method;

[0086] Process spectrum blocks one by one using multi-signal parameter estimation method SpecBlockquOpen M×N The M spectrum slices in the spectrum block are used to obtain the area mask MaskWideNarrow and the corresponding signal bandwidth where the signal exists in the spectrum block. After spectrum cancellation and opening operations, the fixed-frequency signal and the noise area have been eliminated in the area mask MaskWideNarrow, but frequency-hopping signals and WiFi-type broadband signals may also exist at the same time. The bandwidth of WiFi-type signals is usually larger than the bandwidth of frequency-hopping signals. A signal area width threshold RoiMin is set to distinguish frequency-hopping signals from WiFi-type broadband signals for processing. If the width of the signal area in the area mask MaskWideNarrow is greater than RoiMin, the signal area mask is assigned 0 to obtain the narrowband signal area mask MaskNarrow.

[0087] 4. Connected area detection

[0088] Perform four-connected region detection on the narrowband signal area mask MaskNarrow, and mark the center frequency, center time, bandwidth and duration of each signal existence area.

[0089] 5. DSCAN clustering bandwidth sorting

[0090] Assuming that the bandwidth of the UAV frequency hopping signal remains unchanged during operation, the DSCAN algorithm is used to perform cluster analysis on the signal bandwidth characteristics, estimate the different bandwidth signals existing in the spectrum, and classify the signals with similar bandwidth into the same frequency hopping signal.

[0091] 6. Frequency Hopping Signal Detection and Parameter Estimation

[0092] First, construct an estimator η, which represents the number of signals with similar bandwidths in a spectrum block, and set the bandwidth detection threshold Thold Hop , if η ≥ Thold Hop , then it is judged that there is a frequency hopping signal with a frequency hopping bandwidth, and there are several frequency hopping signals exceeding Thold Hop Different bandwidths of the thresholds indicate that there are several frequency hopping signals. The center frequency and center time of the signal with the same frequency hopping bandwidth are recorded at the same time. Then, the same frequency hopping signal is arranged in the order of center time, and the difference is calculated to obtain the frequency hopping time of adjacent frequencies. The middle number of all frequency hopping times is calculated to obtain the estimated value of the hopping period of the frequency hopping signal. Finally, the estimated parameters of the frequency hopping signal, such as the frequency hopping bandwidth, frequency hopping set, and hopping period, can be obtained.

[0093] 7. Frequency hopping signal parameters

[0094] Using structure data format

[0095] HopSignal.Block{Num_a}.Signal{Num_b}.{Parameters}

[0096] Save the estimation of the frequency hopping signal, where Num_a is used to count the processed spectrum blocks, Num_b indicates the number of frequency hopping signals detected in a spectrum block, and Parameters stores the frequency set CF, bandwidth BW, and frequency hopping period Period.

[0097] c) Wideband signal parameter estimation

[0098] In the above steps, the signal area width is less than RoiMin, and there may be WiFi-type broadband signals in the signal area larger than RoiMin, which need further processing. First, use morphological closing operation to fill the concave area in the middle of the broadband signal, and then use the connected area detection method to mark the area of ​​the broadband signal to obtain the center frequency, center time, bandwidth, and duration parameters of each existing signal; finally, use DSCAN clustering center frequency sorting. See the attached for the specific process. Figure 1 , the detailed steps are as follows:

[0099] 1. Morphological closing operation

[0100] The following formula (4) is used to calculate the mask MaskWide of the area where the broadband signal exists.

[0101] MaskWide=MaskWideNarrow-MaskNarrow (4) After processing by formula (4), the signal area mask MaskWideNarrow in which the signal area width is less than RoiMin will be removed, and the signal area with a larger bandwidth, such as the WiFi-type drone signal, will be retained. WiFi-type signals use OFDM modulation, and there is an obvious depression in the middle of the spectrum shape. The energy threshold detection method used in the above processing may determine the middle area as an area where there is no signal, resulting in the same WiFi-type broadband signal being split. The morphological closing operation first performs an expansion operation and then an erosion operation, which can reconnect the disconnected adjacent areas.

[0102] Construct a two-dimensional rectangular structure element SeRec2=ones(r,s). The length s of the two-dimensional rectangular structure element is set according to dF and the minimum filling interval PaddingMin. The reference value is s=round(PaddingMin / dF). For example, if dF is 50KHz and PaddingMin is 2MHz, then s=40. The width p is set according to the spectrum scanning time SweepTime and the minimum signal duration DT_Min, r=round(DT_Min / SweepTime). For example, if SweepTime=1ms and DT_Min=2ms, then r=2.

[0103] Using the two-dimensional rectangular structure element SeRec2 to close the broadband signal area mask MaskWide, the continuity of the same broadband signal can be maintained, and the complete broadband signal area mask MaskWideClose is obtained to obtain the spectrum block MaskWideClose after the opening operation. M×N .

[0104] 2. Connected Area Detection

[0105] MaskWideClose for broadband signal area M×N Perform four-connected area detection, mark the areas where broadband signals exist, and obtain the center frequency, center time, bandwidth, and duration of each broadband signal area.

[0106] 3. DSCAN cluster center frequency sorting

[0107] The DSCAN algorithm is used to perform cluster analysis on the center frequencies of various broadband signals. It is assumed that different center frequencies correspond to different broadband signals, and signals with similar center frequencies are regarded as broadband signals using the same channel.

[0108] 4. Broadband signal parameters

[0109] Using structure data format

[0110] WideSignal.Block{Num_a}.Signal{Num_b}.{Parameters} stores the estimation of the frequency hopping signal, where Num_a is used to count the processed spectrum blocks, Num_b is used to distinguish different detected wideband signals, and Parameters stores the center frequency CF, bandwidth BW, and region mask Mask.

[0111] Effective interference with the drone control link signal and the image transmission signal can prevent the drone operator from taking further action. Since the invading drone is in a relatively open sky, it is closer to the defender's reconnaissance receiver and jammer, and farther from the drone operator. For the defender, it is easier to gain power advantages by interfering with the drone's uplink control link. According to the characteristics of various drone signals in the time-frequency spectrum, drone signals can be divided into three types: fixed-frequency signals, frequency-hopping signals, and broadband signals. The priority of interference is from high to low: frequency-hopping signals, broadband signals, and fixed-frequency signals. Fixed-frequency signals are usually image transmission signals emitted by drones to operators. The intensity of fixed-frequency signals reflects the straight-line distance between the drone and the defender, and is closely related to the threat level of the drone associated with the fixed-frequency signal. The method described in the present invention uses the results of the aforementioned drone signal reconnaissance module to guide the interference center frequency, coverage bandwidth, and interference signal style, and interferes with drone signals according to the interference strategy.

[0112] The control link of drones often uses WiFi-type signals and frequency-hopping signals. WiFi-type signals are broadband signals, which are usually used for low-cost drones and can also transmit videos. It is planned to use QPSK modulation signals or comb spectrum signals to interfere with them. The drone frequency-hopping signal is a series of narrowband signals with a short duration. The frequency set of frequency-hopping signals obtained by the reconnaissance module is used to guide the generation of targeted multi-tone signals for interference. The drone image transmission link is usually a fixed-frequency signal. The center frequency and bandwidth used by different models of drones are different. It is planned to use QPSK signals with the same bandwidth to interfere with each fixed-frequency signal. The present invention intends to construct a simple interference pattern set {QPSK signal, comb spectrum interference, multi-tone interference}. The center frequency, bandwidth, and interference frequency parameters of the three interference signals are configured according to the output parameters of the drone signal reconnaissance module.

[0113] There are four jamming strategies used by the drone signal jamming module of the present invention, namely:

[0114] Interference strategy 1: Based on the number, center frequency, bandwidth, and peak value of the fixed-frequency signals obtained by the reconnaissance module, at the center frequency positions of the first two fixed-frequency signals with the largest signals, the transmitter RF channel 1 generates a QPSK signal covering the corresponding fixed-frequency signal bandwidth. Based on the frequency hopping frequency set obtained by the reconnaissance module, the transmitter RF channel 2 generates a multi-tone interference signal at the corresponding frequency point.

[0115] Interference strategy 2: Based on the difference between the minimum frequency and the maximum frequency of the frequency hopping frequency set obtained by the reconnaissance module, the frequency hopping sequence bandwidth of the frequency hopping signal is estimated, and a QPSK signal with the same bandwidth is generated by transmitter RF channel 1. Based on the frequency hopping frequency set, a multi-tone interference signal with the corresponding frequency point is generated by transmitter RF channel 2.

[0116] Interference strategy 3: Based on the number, center frequency, bandwidth and peak value of the wideband signals obtained by the reconnaissance module, at the center frequency positions of the first two wideband signals with the largest signal peak power, the transmitter RF channel 1 generates a QPSK signal with the same signal bandwidth. Based on the number, center frequency, bandwidth and peak value of the fixed-frequency signals obtained by the reconnaissance module, at the center frequency positions of the first two fixed-frequency signals with the largest signal peak power, the transmitter RF channel 2 generates a QPSK signal with the same signal bandwidth.

[0117] Interference strategy 4: According to the number, center frequency, bandwidth and peak value of the broadband signals obtained by the reconnaissance module, at the center frequency positions of the first two broadband signals with the largest signal peak values, the transmitter RF channel 1 generates a QPSK signal with the same signal bandwidth, and the transmitter RF channel 2 generates a comb spectrum signal with the same bandwidth.

[0118] The jamming strategy is matched according to the reconnaissance results, and the jamming resources are dynamically adjusted.

[0119] Under the constraints of the three interference modes of QPSK signal, comb spectrum interference, and multi-tone interference and two RF transmission channels, the interference resources are automatically and dynamically adjusted according to the information obtained by the reconnaissance module. The process is shown in the attached Figure 3 , the steps are as follows:

[0120] 1. After the system is running, the reconnaissance module continuously scans the radio signal spectrum within the specified frequency band, and when a spectrum block SpecBlock is formed M×N Afterwards, the spectrum block is processed using the aforementioned drone signal reconnaissance method of the present invention, and reconnaissance information is output, including structure data named after three types of signals: HopSignal, WdieSignal, and FixedSignal;

[0121] 2. If there is a frequency hopping signal in the information output by the reconnaissance module, continue to determine whether there is a fixed frequency signal; if there is no fixed frequency signal, directly execute interference strategy 2; if there is a fixed frequency signal and the peak value of the fixed frequency signal exceeds the set threshold, directly use interference strategy 2; if there is a fixed frequency signal but the peak value of the fixed frequency signal does not exceed the set threshold, execute interference strategy 1;

[0122] 3. If there is no frequency hopping signal in the detection information output by the reconnaissance module, continue to determine whether there is a broadband signal; if there is no broadband signal, continue to conduct reconnaissance and monitor the radio signal spectrum; if there is a broadband signal but no fixed-frequency signal, execute interference strategy 4; if there is a broadband signal but the peak value of the fixed-frequency signal does not exceed the set threshold, execute interference strategy 4; if there is a broadband signal and the peak value of the fixed-frequency signal exceeds the threshold, execute interference strategy 3;

[0123] 4. When the execution time of any of the four jamming strategies exceeds 9*M*SweepTime, stop jamming and switch to reconnaissance mode, then repeat the above steps to conduct reconnaissance and jamming of drone signals.

[0124] Specifically, it is an embodiment:

[0125] A BB60C receiver is used as a software radio receiver, a USRP X310 is used as a software radio transmitter, and three rubber stick antennas are used as transceiver antennas. A drone and its remote controller are selected as research objects, and the drone signal reconnaissance module and drone signal jamming module described in the present invention are implemented through programming.

[0126] Turn on the software radio receiver, software radio transmitter and drone, connect the drone and drone remote control, confirm that the remote control can control the drone to move, and use another signal source to continuously transmit QPSK modulated fixed-frequency signals at 2430MHz, 2440MHz and 2460MHz to simulate three image transmission signals.

[0127] Set the receiver to sweep mode, sweep range from 2400MHz to 2483MHz, reference level to -20dBm, sweep time to 1ms, frequency interval dF = 39.022KHz, and set the spectrum block to consist of 100 spectrum slices. After running the program, when 100 spectrum slices are accumulated at a time, a spectrum block will be formed. The result is as shown in the attached figure. Figure 4 As shown in the figure, it can be seen that in the electromagnetic environment where the drone works, there are frequency hopping signals, fixed frequency signals, broadband burst signals, noise, etc.

[0128] Set the energy detection threshold Thold = 3dB, set the minimum signal bandwidth BW_Min = 500KHz and the minimum channel spacing ChannMin = 1MHz in the multi-signal parameter estimation of the fixed frequency signal, and set the minimum signal bandwidth BW_Min = 50KHz and the minimum channel spacing ChannMin = 200KHz in the multi-signal parameter estimation of the frequency hopping signal. The detection threshold Thold of the number of frequency hopping signal bandwidths Hop =6, the threshold for determining the width of the signal region RoiMin=51 (the corresponding signal bandwidth is 2MHz).

[0129] Using the algorithm described in the present invention, three fixed frequency signals can be detected. Figure 5 , with center frequencies of 2.43GHz, 2.44GHz and 2.4599GHz, bandwidths of 1.3658MHz, 1.2877MHz and 1.2877MHz, and peak powers of -63.48dBm, -51.81dBm and -46.39dBm respectively.

[0130] The frequency hopping signal that can be detected by the algorithm of the present invention is shown in the attached Figure 6, the detected frequency hopping bandwidth is 780KHz, and the frequency hopping signal set is HopSets = {2406.985MHz, 2409.989MHz, 2411.472MHz, 2414.477MHz, 2415.96MHz, 2418.964MHz, 2420.486MHz, 2423.452MHz, 2424.974MHz, 2427.978MHz, 2432.466MHz, 2433.949MHz, 24 36.954MHz, 2441.441MHz, 2442.924MHz, 2445.929MHz, 2447.412MHz, 2451.938MHz, 2454.943MHz, 2456.42 6MHz, 2460.913MHz, 2463.918MHz, 2465.401MHz, 2468.406MHz, 2470.201MHz, 2471.41MHz, 2472.893MHz}.

[0131] The broadband signal detected by the algorithm described in the present invention is shown in the attached Figure 7 The center frequencies are 2.4203GHz, 2.4254GHz, and 2.4562GHz, and the bandwidths are 9.2873MHz, 4.6436MHz, and 9.5604MHz, respectively.

[0132] When the drone signal reconnaissance module detects both frequency hopping signals and fixed frequency signals, and the peak value of the fixed frequency signal is greater than -70dBm, interference strategy 2 is used to interfere with the drone signal. The software radio transmitter RF channel 1 generates a QPSK signal with a center frequency of 2422MHz and a bandwidth of 65.9MHz. The software radio transmitter RF channel 2 will generate a multi-tone interference signal with the corresponding frequency point according to the frequency set HopSets.

[0133] The parts not described in detail in the above content are prior art, so they are not described in detail.

Claims

1. A low-complexity drone signal reconnaissance and jamming system, characterized in that: include: The software radio receiver scans the electromagnetic environment within a specified frequency range through the receiving antenna to form a spectrum slice; The UAV signal reconnaissance module performs signal detection and parameter estimation on the spectrum slice, and saves the reconnaissance results into three structures: fixed frequency signal parameters, frequency hopping signal parameters and broadband signal parameters; The UAV signal jamming module matches the jamming strategy according to the reconnaissance results and dynamically adjusts the jamming resources; Software radio transmitter, transmitting antenna radiation source interference signal, interferes with drone signal.

2. A low-complexity UAV signal reconnaissance and jamming system according to claim 1, characterized in that: The steps for detecting and estimating the fixed frequency signal parameters are as follows: 1) Slice the spectrum into spectrum blocks; 2) Calculate the cumulative average spectrum of the spectrum block; 3) Use the average amplitude of the first 90% of the sorted cumulative average spectrum as the noise floor estimate; 4) Perform multi-signal parameter estimation; 5) Get the fixed-frequency signal parameters and save them as structure data.

3. A low-complexity UAV signal reconnaissance and jamming system according to claim 1, characterized in that: The steps for detecting and estimating the frequency hopping signal parameters are as follows: 1) Use spectrum cancellation method to eliminate fixed frequency signal; 2) Use morphological opening operation to eliminate the noise in the spectrum block; 3) Obtain the parameters of multiple signals in a spectrum block through a multi-signal parameter estimation method; 4) Keep the region mask whose region width is less than the threshold, and use the four-connected region to detect the mask of the entire spectrum block; 5) Use the DSCAN algorithm to cluster and sort the bandwidth, and construct an estimator to determine whether a frequency hopping signal exists. If a frequency hopping signal exists, output the estimated value of the frequency hopping signal parameter and save it as structure data.

4. A low-complexity UAV signal reconnaissance and jamming system according to claim 2 or 3, characterized in that: The multi-signal parameter estimation step is: 1) Set the detection threshold and determine whether there is a signal at the frequency point corresponding to the spectrum slice based on the noise floor estimated by the current spectrum block, and use a mask to represent the area where the signal exists in a spectrum slice; 2) Convert the mask representation to the region representation; 3) Fusion of signal areas where the area interval is less than the minimum channel interval; 4) Remove the signal area whose interval width is less than the minimum signal bandwidth in the signal area to obtain a rough signal area; 5) After each rough signal region is expanded to the left and backward, the parameters are re-estimated in the expanded signal region to obtain the variance, skewness, kurtosis and symmetry parameter estimates of the precise signal region and the signal spectrum in the region.

5. A low-complexity UAV signal reconnaissance and jamming system according to claim 3, characterized in that: The steps for parameter estimation of broadband signals are: 1) Retain the area mask whose width is greater than the threshold, and use morphological closing operation to fill the concave area that may exist in the middle of the broadband signal; 2) Use the four-connected region detection method to mark the mask area of ​​the broadband signal and obtain the center frequency, center time, bandwidth, and duration parameters of each signal area; 3) Use the DSCAN clustering method to sort the center frequencies, obtain broadband signal parameters, and save them as structure data.

6. A low-complexity UAV signal reconnaissance and jamming system according to claim 1, characterized in that: There are four interference strategies: Interference strategy 1: According to the number, center frequency, bandwidth and power amplitude of the fixed-frequency signals obtained by the reconnaissance module, at the center frequency positions of the first two fixed-frequency signals with the largest signals, the transmitter RF channel 1 generates a QPSK signal covering the corresponding fixed-frequency signal bandwidth; according to the frequency hopping frequency set obtained by the reconnaissance module, the transmitter RF channel 2 generates a multi-tone interference signal at the corresponding frequency point; Interference strategy 2, based on the difference between the minimum frequency and the maximum frequency of the frequency hopping frequency set obtained by the reconnaissance module, estimates the frequency hopping sequence bandwidth of the frequency hopping signal, and generates a QPSK signal with the same bandwidth by the transmitter RF channel 1; According to the frequency hopping frequency set, the transmitter RF channel 2 generates a multi-tone interference signal of the corresponding frequency point; Interference strategy 3: According to the number, center frequency, bandwidth and power amplitude of the broadband signals obtained by the reconnaissance module, at the center frequency positions of the first two broadband signals with the largest signal peak power, the transmitter RF channel 1 generates a QPSK signal with the same signal bandwidth; according to the number, center frequency, bandwidth and power amplitude of the fixed-frequency signals obtained by the reconnaissance module, at the center frequency positions of the first two fixed-frequency signals with the largest signal peak power, the transmitter RF channel 2 generates a QPSK signal with the same signal bandwidth; Interference strategy 4, according to the number, center frequency, bandwidth and peak power of the broadband signals obtained by the reconnaissance module, at the center frequency positions of the first two broadband signals with the largest signal peak power, the transmitter RF channel 1 generates a QPSK signal with the same signal bandwidth, and the transmitter RF channel 2 generates a comb spectrum signal with the same bandwidth.

7. A low-complexity UAV signal reconnaissance and jamming system according to claim 6, characterized in that: The steps of dynamically regulating the interference resources by matching the interference strategy according to the reconnaissance results are as follows: if there is a frequency hopping signal in the information output by the reconnaissance module, continue to determine whether there is a fixed frequency signal; if there is no fixed frequency signal, directly execute interference strategy 2; if there is a fixed frequency signal and the peak value of the fixed frequency signal exceeds the set threshold, directly use interference strategy 2; if there is a fixed frequency signal but the peak value of the fixed frequency signal does not exceed the set threshold, execute interference strategy 1; if there is no frequency hopping signal in the detection information output by the reconnaissance module, continue to determine whether there is a broadband signal; if there is no broadband signal, continue to conduct reconnaissance and monitor the radio signal spectrum; if there is a broadband signal but no fixed frequency signal, execute interference strategy 4; if there is a broadband signal but the peak value of the fixed frequency signal does not exceed the set threshold, execute interference strategy 4; if there is a broadband signal and the peak value of the fixed frequency signal exceeds the set threshold, execute interference strategy 3; when the execution time of any one of the four interference strategies exceeds times the spectrum length of the reconnaissance module, jump to the drone signal reconnaissance and start a new round of drone signal reconnaissance and interference.

Images