Short-wave frequency band frequency hopping signal blind detection method and system

Abstract

The invention provides a short-wave frequency band frequency hopping signal blind detection method and system, and the method comprises the steps: collecting spectrum data of short-wave full-band monitoring, removing a pulse interference spectrum, and then carrying out the statistics of a self-adaptive threshold; signal sorting is carried out based on the maximum value frequency spectrum and the adaptive threshold, and statistics is carried out on the time domain occurrence rule and the corresponding complete occurrence and disappearance time of each signal; performing burst signal detection on the sorted basic signals according to the duration time, and marking the occurrence state and the duration time of the burst signals to obtain a burst signal list; deleting overtime burst signals in the burst signal list, and carrying out standard jump speed and duty ratio matching on the remaining burst signals; and carrying out network station sorting on the matched frequency hopping signals to obtain a final frequency hopping detection result. According to the method, the signal change of adjacent time is counted in real time based on broadband monitoring data to judge whether pulse interference exists or not, the frequency spectrum of the pulse interference does not participate in statistics and detection, and the frequency hopping detection effectiveness of the maximum frequency spectrum is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic reconnaissance, and in particular to a shortwave frequency band frequency hopping signal blind detection method and system. Background Art

[0002] The shortwave frequency band has the characteristic of long transmission distance. It can achieve signal transmission of thousands of kilometers by using ionospheric reflection. Therefore, military and civilian communications in various countries are very common in the shortwave frequency band. Frequency hopping communication has significant advantages such as strong anti-interference and low interception rate. It is widely used in military communications. Various radio stations using frequency hopping technology are widely used in the military field, which greatly improves the anti-interception and anti-interference capabilities of military equipment. The hopping rate of the frequency hopping signal in the shortwave frequency band usually ranges from 5 hops per second to 200 hops per second.

[0003] Signal detection is the basic function of signal reconnaissance. Currently, shortwave signal blind reconnaissance mostly uses direct sampling receivers, which can realize real-time acquisition of the full-band 28.5M bandwidth (1.5MHz~30MHz), and output the full-band 28.5MHz broadband real-time spectrum after time-frequency changes. Shortwave signal detection is based on broadband real-time spectrum, realizing signal detection within the specified frequency band bandwidth, and realizing frequency hopping signal detection and network station sorting based on conventional signal and burst signal detection. Frequency hopping signal can obtain frequency hopping parameters such as hopping speed, frequency hopping frequency set, and signal duty cycle.

[0004] For the sorting of multiple frequency hopping signals in the shortwave band, most algorithms estimate the parameters of frequency hopping signals and sort the stations, such as hopping speed, frequency hopping frequency set, etc., by processing the time-frequency diagram. However, the error of parameter estimation of traditional algorithms is large and the complexity is high. At present, the commonly used blind detection methods of frequency hopping signals are mainly time-frequency analysis technology, including linear time-frequency analysis technology represented by short-time Fourier transform and nonlinear time-frequency analysis technology represented by Wigner transform. These algorithms are all special detection algorithms for frequency hopping signals. The algorithms perform time-frequency transformation based on the original IQ data to obtain the time-frequency diagram, and perform frequency hopping signal detection and parameter estimation based on the time-frequency diagram.

[0005] Patent document CN116455424A discloses a method and system for blind estimation of multi-frequency hopping parameters based on a quadratic difference algorithm. The reconnaissance platform collects IQ data, obtains the time-frequency diagram of the received electromagnetic signal through short-time Fourier transform, realizes signal enhancement of the time-frequency diagram through the frequency domain enhancement process, obtains a clear time-frequency distribution diagram, and realizes blind estimation of multi-frequency hopping signal parameters through a quadratic difference algorithm. Patent document CN116979992A discloses a method for estimating hopping period based on FFT transformation. The reconnaissance platform collects IQ data, obtains the time-frequency diagram after processing, obtains the time-frequency peak value of the hopping pulse through the time-frequency peak detection algorithm, creates the time center sequence of the hopping pulse according to the time-frequency peak value, performs FFT transformation on the time center sequence, and reads the peak coordinate information related to the hopping period in the transformed FFT spectrum. The value of the hopping period can be calculated based on the information coordinates.

[0006] However, the frequency hopping period in the above patent document is estimated by averaging the frequency hopping period of multi-hop signals or calculating by using the FFT transformation of the time center sequence. These methods are not accurate in the actual detection environment of the shortwave frequency band due to the influence of noise transformation. In view of this, a network station frequency hopping signal parameter estimation method based on time-frequency diagram correction is proposed, which can effectively estimate the number of frequency hopping network stations, hopping period, hopping time and carrier frequency of the received signal. The spectrum in the time-frequency diagram is a short-time spectrum. When the original IQ for time-frequency calculation only contains a part of a certain hopping signal in time, which is equivalent to a pulse signal, the spectrum will leak, which will be reflected in the time-frequency diagram as a bright line, thereby affecting the accuracy of subsequent frequency hopping detection.

[0007] The hopping rate of the frequency hopping signal in the shortwave band is relatively low, usually between 5 hops per second and 200 hops per second. In actual reconnaissance, due to the large number of shortwave band signals and the significant impact of ionospheric changes when transmitted through sky waves, the hopping rate parameter errors estimated by conventional frequency hopping signal detection algorithms are relatively large. Summary of the invention

[0008] In view of the defects in the prior art, the object of the present invention is to provide a method and system for blind detection of frequency hopping signals in a shortwave frequency band.

[0009] A shortwave frequency band frequency hopping signal blind detection method provided by the present invention comprises:

[0010] Step S1: monitoring antenna sensing, collecting spectrum data of shortwave full-band monitoring, and removing pulse interference spectrum;

[0011] Step S2: Calculating an adaptive threshold from the spectrum data without the pulse interference spectrum;

[0012] Step S3: performing signal sorting based on the maximum value spectrum and the adaptive threshold, and counting the time domain appearance rule of each signal and the corresponding complete appearance and disappearance time;

[0013] Step S4: performing burst signal detection on the selected basic signals according to the duration, marking the occurrence state and duration of the burst signals, and obtaining a burst signal list;

[0014] Step S5: deleting the timed-out burst signals in the burst signal list, and matching the standard hopping rates and duty cycles of the remaining burst signals;

[0015] Step S6: performing network station sorting on the matched frequency hopping signal to obtain the final frequency hopping detection result.

[0016] Preferably, the step S1 comprises:

[0017] Step S1.1: Continuously receive spectrum data of shortwave full-band monitoring;

[0018] Step S1.2: Set the data statistics period to T cal , the refresh cycle is T ref ;

[0019] Step S1.3: Accumulate multiple frames of full-band spectrum data over time, and when the accumulation period is met and pulse interference is detected, remove the spectrum with pulse interference.

[0020] Preferably, the step S1.3 comprises: performing a differential operation on three consecutive frames of full-band spectrum data, and the calculation formula is as follows:

[0021]

[0022] Among them, i represents the i-th frame spectrum, j represents the j-th frequency point, and N is the number of points in a frame spectrum. Then, it is determined whether there is pulse interference in the three consecutive frames of spectrum involved in the operation. If so, the frame data is deleted. If ΔSpec0 and ΔSpec1 are both greater than the judgment threshold, the middle frame data among the three frames of data is pulse interference.

[0023] Preferably, step S2 comprises:

[0024] Step S2.1: Calculate the mean spectrum meanSpec, the maximum spectrum maxSpec and the minimum spectrum minSpec. The formula is as follows:

[0025]

[0026] maxSpec i =MAX(spec j,i )

[0027] minSpec i =MIN(spec j,i )

[0028] Among them, meanSpec i Indicates the mean spectrum value of the i-th frequency point, i indicates the i-th frequency point, spec j,i Indicates the value of the i-th frequency point of the j-th instantaneous spectrum, N is the cumulative number of frames, maxSpec i Indicates the maximum spectrum value of the i-th frequency point, MAX indicates a comparison operation, taking the maximum value, minSpec i Indicates the minimum spectrum value of the i-th frequency point, MIN indicates comparison operation, taking the minimum value;

[0029] Step S2.2: Perform windowed median filtering on the mean spectrum meanSpec to obtain a basic threshold BasicTh, and the window function width is set to be slightly larger than 2 times the target signal width;

[0030] Step S2.3: Calculate the floating threshold addTh, the formula is as follows:

[0031]

[0032] Wherein, M represents the number of frequency points in a full-band spectrum frame;

[0033] Step S2.4: Calculate the adaptive threshold FinalTh, the formula is as follows:

[0034] FinalTh=BasicTh+addTh.

[0035] Preferably, step S3 includes: comparing the maximum spectrum maxSpec of each frequency point and the adaptive threshold FinalTh, if the maximum spectrum is higher than the adaptive threshold, it is marked as a signal frequency point, otherwise it is marked as a noise frequency point; if it is continuously marked as a signal frequency point, it is counted as a basic signal BasicSignal, and the starting frequency point BS1 and the cutoff frequency point BS2 of the basic signal are recorded.

[0036] Preferably, the step S4 comprises:

[0037] Step S4.1: Take the basic signal BasicSignal in the statistical period T cal All spectra within, get the signal spectrum Sig_Spec i , where i∈[0T cal ), the frequency range of Sig_Spec is from BS1 to BS2;

[0038] Step S4.2: Sig_Spec i The signal status statistics are performed for each frame in the program. If the current spectrum is higher than the corresponding adaptive threshold, it is a signal at the current moment, otherwise it is noise.

[0039] Step S4.3: Set all T cal The signal status is counted. If the frequency appears continuously for a certain period of time, it is a signal. Otherwise, it is recorded as a new signal. If the signal has a complete start and stop time, it is marked as a burst signal. The record status is disappeared, and the signal appearance time, disappearance time and duration KeepLen are recorded. If the head and tail are incomplete, the signal needs to be merged with the previous result and the next result, and then marked according to the signal integrity.

[0040] Step S4.4: Determine the duration of the marked burst signal, where the duration is the last appearance time minus the first appearance time. When the signal duration is between 0.1ms and 400ms, it is marked as a burst signal and added to the burst signal list S_Puls.

[0041] Preferably, the standard jump speed and duty cycle matching includes the following sub-steps:

[0042] Step S5.1: Match according to the hop rate of 5 to 200 hops. The current hop rate is StandHopRate, and the time difference between two hops StandTime is:

[0043] StandTime = 1 / StandHopRate;

[0044] Step S5.2: Match the duty cycle of the frequency hopping signal. The duty cycle matching range is 50% to 100%, the step is 5%, and the current duty cycle is OccRate. Then the theoretical duration StandOccTime of each frequency hopping signal is:

[0045] StandOccTime=StandTime*OccRate

[0046] The idle time of each hop StandIdleTIme is:

[0047] StandIdleTIme=StandTime-StandOccTime;

[0048] Step S5.3: Perform duration matching on each signal in the burst signal list in turn. When the deviation ErrTime between the signal duration and the theoretical duration is less than the duration threshold, the signal is added to the duration matching queue MeetHop, and the center time HopCenterTime of each hop point, the index HopIndex of the hop signal in the frequency hopping frequency set, and the error HopErrTime of the center time of the signal are calculated. The calculation formulas are as follows:

[0049] ErrTime=|KeepLen-StandOccTime|

[0050] HopCenterTime=OffsetTime+0.5*KeepLen

[0051] HopIndex=Int(HopCenterTime / StandTime)

[0052] HopErrTime=HopCenterTime-HopIndex*StandTime

[0053] Among them, KeepLen represents the duration, Int represents rounding, and OffsetTime represents the time interval between each hop signal and the latest time signal;

[0054] Step S5.4: If the number of frequency points in the duration matching queue MeetHop is less than 4, return to step S5.2 and increase the duty cycle in steps; if the number of frequency points is greater than or equal to 4, perform regularity screening of the frequency hopping sequence and execute step S5.5;

[0055] Step S5.5: Cluster the frequencies in the duration matching queue MeetHop according to HopErrTime. If the frequency points after clustering are greater than or equal to 4, they are added to the same time matching queue MeetHop_Sub i and calculate the average error MeanErrTimei; where i represents the i-th group of clustering;

[0056] Step S5.6: Match the MeetHop_Sub queues at the same time in sequence i The frequency points are compared according to HopIndex, and the frequency points with the same HopIndex select the HopErrTime that is closer to the mean error MeanErrTime i The frequency hopping point is added to the optimal hopping point matching queue PreHopList i middle;

[0057] Step S5.7: Record the current frequency hopping signal into the frequency hopping signal queue HopRes.

[0058] Preferably, the frequency hopping signal parameters include a hopping rate StandHopRate, a duty cycle OccRate, a duration error mean MeanKeepLenErr, a root mean square of a center time error stdTimeErr, and a frequency hopping frequency parameter;

[0059] The frequency hopping parameters include frequency, bandwidth, duration, and occurrence time.

[0060] Preferably, the step S6 includes sorting the multiple frequency hopping signals in the frequency hopping signal queue HopRes according to the hopping speed and frequency set, refreshing the frequency set and parameters if it is the same frequency hopping network signal, merging the frequency sets if the frequency sets have the same frequency, the same appearance time, and a small duration error, and taking the hopping speed parameter as the measurement result with the smaller center time root mean square error stdTimeErr; otherwise, recording a new frequency hopping signal.

[0061] A shortwave frequency band frequency hopping signal blind detection system provided by the present invention comprises:

[0062] Module M1: monitoring antenna sensing, collecting spectrum data of shortwave full-band monitoring, and removing pulse interference spectrum;

[0063] Module M2: Calculate adaptive threshold from spectrum data without pulse interference spectrum;

[0064] Module M3: Signal sorting based on maximum spectrum and adaptive threshold, statistics of time domain appearance rules of each signal and corresponding complete appearance and disappearance time;

[0065] Module M4: Perform burst signal detection on the selected basic signals according to the duration, mark the occurrence state and duration of the burst signal, and obtain a burst signal list;

[0066] Module M5: deleting the timed-out burst signals in the burst signal list, and matching the standard hopping speed and duty cycle of the remaining burst signals;

[0067] Module M6: perform network station sorting on the matched frequency hopping signal to obtain the final frequency hopping detection result.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. The present invention is based on the characteristics of shortwave frequency hopping signals and the real-time statistics of signal changes in adjacent time based on broadband monitoring data to determine whether there is pulse interference. The spectrum of pulse interference is not involved in statistics and detection, thereby ensuring the effectiveness of frequency hopping detection of the maximum value spectrum.

[0070] 2. The present invention utilizes the hopping speed characteristics of the shortwave frequency hopping signal to perform hopping speed matching, and implements the optimal hopping speed matching by counting the duration error of the frequency hopping frequency set and the center time error of the frequency set, thereby improving the accuracy of the frequency hopping parameters of the shortwave frequency band frequency hopping signal.

[0071] 3. The present invention combines the sorting algorithms of shortwave conventional signals, burst signals and frequency hopping signals according to the actual electromagnetic environment, so as to realize blind detection of shortwave frequency band signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0073] Figure 1 This is a schematic diagram of the blind detection process of the shortwave frequency band frequency hopping signal of the present invention;

[0074] Figure 2 It is a schematic diagram of the process of matching the jump speed and duty cycle in the present invention;

[0075] Figure 3 A schematic diagram of burst signal detection in the present invention;

[0076] Figure 4 It is a frequency hopping detection-burst signal list in the present invention;

[0077] Figure 5 This is frequency hopping detection-duration matching and error classification in the present invention. DETAILED DESCRIPTION

[0078] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0079] The present invention realizes frequency hopping signal detection, parameter estimation and network station sorting during broadband detection of shortwave frequency band signals, uses continuous signal differential spectrum to detect pulse interference, and uses a frequency hopping parameter matching method to improve the accuracy of frequency hopping parameter estimation.

[0080] According to a shortwave frequency band frequency hopping signal blind detection method provided by the present invention, Figure 1 As shown, including:

[0081] Step S1: Monitoring antenna sensing, collecting spectrum data of shortwave full-band monitoring, and removing pulse interference spectrum. Step S1 includes:

[0082] Step S1.1: Continuously receive spectrum data of shortwave full-band monitoring; the spectrum resolution is 1250 Hz, and the interval time of each spectrum frame is 0.8 ms.

[0083] Step S1.2: Set the data statistics period to T cal , the refresh cycle is T ref , generally one third of the statistical period; the accumulated data meets T cal Then perform the calculation. After the calculation is completed, delete T ref The statistical data are re-accumulated.

[0084] Step S1.3: Accumulate multiple frames of full-band spectrum data over time. When the accumulation period is satisfied and pulse interference is detected, the spectrum with pulse interference does not participate in subsequent operations. Step S1.3 includes: performing differential operations on three consecutive frames of full-band spectrum data, and the calculation formula is as follows:

[0085]

[0086] Among them, i represents the i-th frame spectrum, j represents the j-th frequency point, and N is the number of points in a frame spectrum. Then, it is determined whether there is pulse interference in the three consecutive frames of spectrum involved in the operation. If so, the frame data is deleted. If ΔSpec0 and ΔSpec1 are both greater than the judgment threshold, the middle frame data among the three frames of data is pulse interference.

[0087] Step S2: Calculate the adaptive threshold from the spectrum data in the spectrum without the pulse interference. Step S2 includes:

[0088] Step S2.1: Calculate the mean spectrum meanSpec, the maximum spectrum maxSpec and the minimum spectrum minSpec. The formula is as follows:

[0089]

[0090] maxSpec i =MAX(spec j,i )

[0091] minSpec i =MIN(spec j,i )

[0092] Among them, meanSpec i Indicates the mean spectrum value of the i-th frequency point, i indicates the i-th frequency point, spec j,i Indicates the value of the i-th frequency point of the j-th instantaneous spectrum, N is the cumulative number of frames, maxSpec i Indicates the maximum spectrum value of the i-th frequency point, MAX indicates a comparison operation, taking the maximum value, minSpec i Indicates the minimum spectrum value of the i-th frequency point. MIN indicates comparison operation and takes the minimum value.

[0093] Step S2.2: Perform windowed median filtering on the mean spectrum meanSpec to obtain a basic threshold BasicTh, and the window function width is set to be slightly larger than 2 times the target signal width.

[0094] Step S2.3: Calculate the floating threshold addTh, the formula is as follows:

[0095]

[0096] Wherein, M represents the number of frequency points in a frame of full-band spectrum.

[0097] Step S2.4: Calculate the adaptive threshold FinalTh, the formula is as follows:

[0098] FinalTh=BasicTh+addTh.

[0099] Step S3: Signal sorting is performed based on the maximum spectrum and adaptive threshold, and the time domain appearance rule and corresponding complete appearance and disappearance time of each signal are counted. The step S3 includes: comparing the maximum spectrum maxSpec and the adaptive threshold FinalTh of each frequency point, and marking the maximum spectrum higher than the adaptive threshold as a signal frequency point, otherwise it is marked as a noise frequency point; the frequency points continuously marked as signal points are counted as a basic signal BasicSignal, and the starting frequency point BS1 and the cutoff frequency point BS2 of the basic signal are recorded.

[0100] Step S4: Perform burst signal detection on the selected basic signals according to the duration, mark the burst signal occurrence state and duration, and obtain a burst signal list. Step S4 includes:

[0101] Step S4.1: Take the basic signal BasicSignal in the statistical period T cal All spectra within, get the signal spectrum Sig_Spec i , where i∈[0T cal ), the frequency range of Sig_Spec is from BS1 to BS2;

[0102] Step S4.2: Sig_Spec i The signal status statistics are performed for each frame in the signal processing. If the current spectrum is higher than the corresponding adaptive threshold, it is a signal at the current moment, otherwise it is noise.

[0103] Step S4.3: Set all T cal The signal status is counted. If the frequency appears continuously for a certain period of time, it is a signal. Otherwise, it is recorded as a new signal. If the signal has a complete start and stop time, it is marked as a burst signal. The record status is disappeared, and the signal appearance time, disappearance time and duration KeepLen are recorded. If the head and tail are incomplete, the signal needs to be merged with the previous result and the next result. After merging, it is marked according to the signal integrity, such as Figure 3 shown.

[0104] Step S4.4: Determine the duration of the marked burst signal, where the duration is the last appearance time minus the first appearance time. When the signal duration is between 0.1ms and 400ms, it is marked as a burst signal and added to the burst signal list S_Puls.

[0105] Step S5: Delete the burst signals that have timed out in the burst signal list, and match the standard hopping rate and duty cycle of the remaining burst signals. Specifically, filter the burst signals that have timed out, and based on the latest data time, filter the burst signals that exceed the timeout time and remove them from S_Puls; calculate the time interval OffsetTime of each hopping signal relative to the latest time signal; perform frequency hopping matching on the signals in the burst signal list S_Puls. Figure 2 As shown, the standard jump speed and duty cycle matching includes the following sub-steps:

[0106] Step S5.1: Match according to the hop rate of 5 to 200 hops, the current hop rate is StandHopRate, and the time difference between two hops StandTime = 1 / StandHopRate;

[0107] Step S5.2: match the duty cycle of the frequency hopping signal, the duty cycle matching range is 50% to 100%, the step is 5%, the current duty cycle is OccRate, then the theoretical duration of each hopping frequency hopping signal is StandOccTime = StandTime * OccRate, and the idle time of each hop is StandIdleTIme = StandTime - StandOccTime;

[0108] Step S5.3: Perform duration matching on each signal in S_Puls in turn. When the deviation ErrTime between the signal duration and the theoretical duration is less than the duration threshold, add it to the duration matching queue MeetHop, and calculate the center time HopCenterTime of each hop point, the index HopIndex of the hop signal in the hop frequency set, and the error HopErrTime of the center time of the signal. Figure 2 and Figure 3 As shown, the calculation is as follows:

[0109] ErrTime=|KeepLen-StandOccTime|

[0110] HopCenterTime=OffsetTime+0.5*KeepLen

[0111] HopIndex=Int(HopCenterTime / StandTime)

[0112] HopErrTime=HopCenterTime-HopIndex*StandTime

[0113] Among them, Int means rounding.

[0114] Step S5.4: If the number of frequency points in the duration matching queue MeetHop is less than 4, return to step S5.2 and increase the duty cycle in steps; if the number of frequency points is greater than or equal to 4, perform regularity screening of the frequency hopping sequence and execute step S5.5.

[0115] Step S5.5: Cluster the frequency points in the duration matching queue MeetHop according to HopErrTime in turn. If the frequency points after clustering are greater than or equal to 4, they are added to the same time matching queue MeetHop_Subi (i represents the i-th group of the cluster), and the average error MeanErrTime is calculated. i .

[0116] Step S5.6: Match the MeetHop_Sub queues at the same time in sequence i The frequency points are compared according to HopIndex, and the frequency points with the same HopIndex select the HopErrTime that is closer to the mean error MeanErrTime i The frequency hopping point is added to the optimal hopping point matching queue PreHopList i middle.

[0117] Step S5.7: Record the current frequency hopping signal into the frequency hopping signal queue HopRes. The frequency hopping signal parameters include the hopping rate StandHopRate, the duty cycle OccRate, the mean value of the duration error MeanKeepLenErr, the root mean square of the center time error stdTimeErr, and the frequency hopping frequency parameters. The frequency hopping frequency parameters include frequency, bandwidth, duration, occurrence time, etc. Figure 4 and Figure 5 The specific calculation is as follows:

[0118]

[0119] Among them, meanKeepLen represents the mean of the duration, KeepLen represents the duration, MeanKeepLenErr represents the mean of the duration error, MeanErrTime represents the average error, HopErrTime represents the center time error, stdTimeErr represents the root mean square of the center time error, and N represents the number of spectrum points.

[0120] Step S6: sort the multiple frequency hopping signals in HopRes according to the hopping rate and frequency set, and refresh the frequency set and parameters if it is the same frequency hopping signal. The specific rules are as follows: if the frequency sets have the same frequency, the same appearance time, and the duration error is small, then merge the frequency sets, and take the measurement result with the smaller stdTimeErr for the hopping rate and other parameters; otherwise, record a new frequency hopping signal.

[0121] The present invention aims to solve the problem of recognition error caused by pulse interference during frequency hopping signal detection in shortwave reconnaissance, and to improve the estimation accuracy of frequency hopping parameters by using a frequency hopping parameter matching method.

[0122] Embodiment 2

[0123] The present invention also provides a shortwave frequency band frequency hopping signal blind detection system, which can be implemented by executing the process steps of the shortwave frequency band frequency hopping signal blind detection method, that is, those skilled in the art can understand the shortwave frequency band frequency hopping signal blind detection method as a preferred implementation of the shortwave frequency band frequency hopping signal blind detection system.

[0124] A shortwave frequency band frequency hopping signal blind detection system provided by the present invention comprises:

[0125] Module M1: Monitoring antenna sensing, collecting spectrum data of shortwave full-band monitoring, and removing pulse interference spectrum. Module M1 includes: Module M1.1: Continuously receiving spectrum data of shortwave full-band monitoring. Module M1.2: Setting the data statistics period to T cal , the refresh cycle is T ref Module M1.3: accumulate multiple frames of full-band spectrum data over time, and when the accumulation period is satisfied and pulse interference is detected, remove the spectrum with pulse interference. Module M1.3 includes: performing differential operation on three consecutive frames of full-band spectrum data, and the calculation formula is as follows:

[0126]

[0127] Among them, i represents the i-th frame spectrum, j represents the j-th frequency point, and N is the number of points in a frame spectrum. Then, it is determined whether there is pulse interference in the three consecutive frames of spectrum involved in the operation. If there is, the frame data is deleted. If ΔSpec0 and ΔSpec1 are both greater than the judgment threshold, the middle frame data among the three frames of data is pulse interference.

[0128] Module M2: Count the adaptive threshold from the spectrum data in the spectrum without pulse interference. Module M2 includes: Module M2.1: Count the mean spectrum meanSpec, the maximum spectrum maxSpec and the minimum spectrum minSpec, the formula is as follows:

[0129]

[0130] maxSpec i =MAX(spec j,i )

[0131] minSpec i =MIN(spec j,i )

[0132] Among them, meanSpec i Indicates the mean spectrum value of the i-th frequency point, i indicates the i-th frequency point, spec j,i Indicates the value of the i-th frequency point of the j-th instantaneous spectrum, N is the cumulative number of frames, maxSpec i Indicates the maximum spectrum value of the i-th frequency point, MAX indicates a comparison operation, taking the maximum value, minSpec i Indicates the minimum spectrum value of the i-th frequency point, MIN indicates comparison operation, and takes the minimum value. Module M2.2: Perform windowed median filtering on the mean spectrum meanSpec to obtain the basic threshold BasicTh, and the window function width is set to be slightly larger than 2 times the target signal width. Module M2.3: Statistical floating threshold addTh, the formula is as follows: Wherein, M represents the number of frequency points of a full-band spectrum of a frame. Module M2.4: Calculate the adaptive threshold FinalTh, the formula is as follows: FinlTh = BasicTh + addTh.

[0133] Module M3: Signal sorting based on maximum spectrum and adaptive threshold, statistics of time domain appearance rules and corresponding complete appearance and disappearance time of each signal. Module M3 includes: comparing the maximum spectrum maxSpec and adaptive threshold FinalTh of each frequency point, the maximum spectrum higher than the adaptive threshold is marked as a signal frequency point, otherwise it is marked as a noise frequency point. The frequency points continuously marked as signal points are counted as a basic signal BasicSignal, and the starting frequency point BS1 and the cutoff frequency point BS2 of the basic signal are recorded.

[0134] Module M4: Perform burst signal detection on the selected basic signals according to the duration, mark the occurrence state and duration of the burst signal, and obtain a burst signal list. Module M4 includes: Module M4.1: Take the basic signal BasicSignal in the statistical period T cal All spectra within, get the signal spectrum Sig_Spec i , where i∈[0T cal ), the frequency range of Sig_Spec is from BS1 to BS2. Module M4.2: For Sig_Spec i The signal status statistics are performed for each frame in the . If the current spectrum is higher than the corresponding adaptive threshold, then the current moment is a signal, otherwise it is noise. Module M4.3: All T calThe signal status is counted. If the frequency appears continuously for a certain period of time, it is a signal. Otherwise, it is recorded as a new signal. If the signal has a complete start and stop time, it is marked as a burst signal and the record status is disappeared. The signal appearance time, disappearance time and duration KeepLen are recorded. Incomplete signals at the beginning and end need to be merged with the previous result and the next result. After merging, they are marked according to the signal integrity. Module M4.4: Determine the duration of the marked burst signal. The duration is the last appearance time minus the first appearance time. When the signal duration is between 0.1ms and 400ms, it is marked as a burst signal and added to the burst signal list S_Puls.

[0135] Module M5: Delete the timed-out burst signals in the burst signal list, and match the standard hopping rate and duty cycle of the remaining burst signals. The standard hopping rate and duty cycle matching includes the following submodules: Module M5.1: Match according to the hopping rate of 5 to 200 hops, the current hopping rate is StandHopRate, and the time difference between two hops StandTime is: StandTime = 1 / StandHopRate. Module M5.2: Match the duty cycle of the frequency hopping signal, the duty cycle matching range is 50% to 100%, the step is 5%, and the current duty cycle is OccRate, then the theoretical duration StandOccTime of each hop frequency hopping signal is: StandOccTime = StandTime*OccRate, and the idle time StandIdleTIme of each hop is: StandIdleTIme = StandTime-StandOccTime. Module M5.3: Perform duration matching on each signal in the burst signal list in turn. When the deviation ErrTime between the signal duration and the theoretical duration is less than the duration threshold, it is added to the duration matching queue MeetHop, and the center time HopCenterTime of each hop point, the index HopIndex of the hop signal in the hopping frequency set, and the error HopErrTime of the center time of the signal are calculated. The calculation formulas are as follows:

[0136] ErrTime=|KeepLen-StandOccTime|

[0137] HopCenterTime=OffsetTime+0.5*KeepLen

[0138] HopIndex=Int(HopCenterTime / StandTime)

[0139] HopErrTime=HopCenterTime-HopIndex*StandTime

[0140] Among them, KeepLen represents the duration, Int represents rounding, and OffsetTime represents the time interval of each hop signal relative to the latest time signal. Module M5.4: If the number of frequency points in the duration matching queue MeetHop is less than 4, return to module M5.2 and increase the duty cycle in steps. If the number of frequency points is greater than or equal to 4, the regularity of the frequency hopping sequence is screened and module M5.5 is triggered. Module M5.5: Cluster the frequency points in the duration matching queue MeetHop according to HopErrTime in turn. If the clustered frequency points are greater than or equal to 4 frequency points, they are added to the same time matching queue MeetHop_Sub i , and calculate the average error MeanErrTimei. Where i represents the i-th group of clustering. Module M5.6: Sequentially match the continuous same time queue MeetHop_Sub i The frequency points are compared according to HopIndex, and the frequency points with the same HopIndex select the HopErrTime that is closer to the mean error MeanErrTime i The frequency hopping point is added to the optimal hopping point matching queue PreHopList i Module M5.7: Record the current frequency hopping signal into the frequency hopping signal queue HopRes. The frequency hopping signal parameters include the hopping rate StandHopRate, the duty cycle OccRate, the mean value of the duration error MeanKeepLenErr, the root mean square of the center time error stdTimeErr, and the frequency hopping frequency parameters. The frequency hopping frequency parameters include frequency, bandwidth, duration, and occurrence time.

[0141] Module M6: sort the matched frequency hopping signals to obtain the final frequency hopping detection result. Module M6 includes sorting the multiple frequency hopping signals in the frequency hopping signal queue HopRes according to the hopping speed and frequency set. If the same frequency hopping signal is determined, the frequency set and parameters are refreshed. If the frequency set has the same frequency, the same appearance time, and the duration error is small, the frequency set is merged, and the hopping speed parameter is the measurement result with the smaller center time error root mean square stdTimeErr. Otherwise, the new frequency hopping signal is recorded.

[0142] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0143] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A shortwave frequency band frequency hopping signal blind detection method, characterized in that: include: Step S1: monitoring antenna sensing, collecting spectrum data of shortwave full-band monitoring, and removing pulse interference spectrum; Step S2: Calculating an adaptive threshold from the spectrum data without the pulse interference spectrum; Step S3: performing signal sorting based on the maximum value spectrum and the adaptive threshold, and counting the time domain appearance rule of each signal and the corresponding complete appearance and disappearance time; Step S4: performing burst signal detection on the selected basic signals according to the duration, marking the occurrence state and duration of the burst signals, and obtaining a burst signal list; Step S5: deleting the timed-out burst signals in the burst signal list, and matching the standard hopping rates and duty cycles of the remaining burst signals; Step S6: performing network station sorting on the matched frequency hopping signal to obtain the final frequency hopping detection result.

2. The shortwave frequency band frequency hopping signal blind detection method according to claim 1, characterized in that: The step S1 comprises: Step S1.1: Continuously receive spectrum data of shortwave full-band monitoring; Step S1.2: Set the data statistics period to T cal , the refresh cycle is T ref ; Step S1.3: Accumulate multiple frames of full-band spectrum data over time, and when the accumulation period is met and pulse interference is detected, remove the spectrum with pulse interference.

3. The shortwave frequency band frequency hopping signal blind detection method according to claim 2, characterized in that: The step S1.3 includes: performing a differential operation on three consecutive frames of full-band spectrum data, and the calculation formula is as follows: Among them, i represents the i-th frame spectrum, j represents the j-th frequency point, and N is the number of points in a frame spectrum. Then, it is determined whether there is pulse interference in the three consecutive frames of spectrum involved in the operation. If so, the frame data is deleted. If ΔSpec0 and ΔSpec1 are both greater than the judgment threshold, the middle frame data among the three frames of data is pulse interference.

4. The method for blind detection of shortwave frequency hopping signals according to claim 1, characterized in that: The step S2 comprises: Step S2.1: Calculate the mean spectrum meanSpec, the maximum spectrum maxSpec and the minimum spectrum minSpec. The formula is as follows: maxSpec i =MAX(spec j,i ) mySpec i =MIN(spec j,i ) Among them, meanSpec i Indicates the mean spectrum value of the i-th frequency point, i indicates the i-th frequency point, spec j,i Indicates the value of the i-th frequency point of the j-th instantaneous spectrum, N is the cumulative number of frames, maxSpec i Indicates the maximum spectrum value of the i-th frequency point, MAX indicates a comparison operation, taking the maximum value, minSpec i Indicates the minimum spectrum value of the i-th frequency point, MIN indicates comparison operation, taking the minimum value; Step S2.2: Perform windowed median filtering on the mean spectrum meanSpec to obtain a basic threshold BasicTh, and the window function width is set to be slightly larger than 2 times the target signal width; Step S2.3: Calculate the floating threshold addTh, the formula is as follows: Wherein, M represents the number of frequency points in a full-band spectrum frame; Step S2.4: Calculate the adaptive threshold FinalTh, the formula is as follows: FinalTh=BasicTh+addTh.

5. The method for blind detection of shortwave frequency hopping signals according to claim 1, characterized in that: The step S3 includes: comparing the maximum spectrum maxSpec of each frequency point with the adaptive threshold FinalTh, and marking the maximum spectrum higher than the adaptive threshold as a signal frequency point, otherwise marking it as a noise frequency point; the frequency points continuously marked as signal frequencies are counted as a basic signal BasicSignal, and the starting frequency point BS1 and the cutoff frequency point BS2 of the basic signal are recorded.

6. The method for blind detection of shortwave frequency hopping signals according to claim 1, characterized in that: The step S4 comprises: Step S4.1: Take the basic signal BasicSignal in the statistical period T cal All spectra within, get the signal spectrum Sig_Spec i , where i∈[0T cal ), the frequency range of Sig_Spec is from BS1 to BS2; Step S4.2: Sig_Spec i The signal status statistics are performed for each frame in the signal processing. If the current spectrum is higher than the corresponding adaptive threshold, it is a signal at the current moment, otherwise it is noise. Step S4.3: Set all T cal The signal status is counted. If the frequency appears continuously for a certain period of time, it is a signal. Otherwise, it is recorded as a new signal. If the signal has a complete start and stop time, it is marked as a burst signal. The record status is disappeared, and the signal appearance time, disappearance time and duration KeepLen are recorded. If the head and tail are incomplete, the signal needs to be merged with the previous result and the next result, and then marked according to the signal integrity. Step S4.4: Determine the duration of the marked burst signal, where the duration is the last appearance time minus the first appearance time. When the signal duration is between 0.1ms and 400ms, it is marked as a burst signal and added to the burst signal list S_Puls.

7. The method for blind detection of shortwave frequency hopping signals according to claim 1, characterized in that: The standard hopping speed and duty cycle matching includes the following sub-steps: Step S5.1: Match according to the hop rate of 5 to 200 hops. The current hop rate is StandHopRate, and the time difference between two hops StandTime is: StandTime = 1 / StandHopRate; Step S5.2: Match the duty cycle of the frequency hopping signal. The duty cycle matching range is 50% to 100%, the step is 5%, and the current duty cycle is OccRate. Then the theoretical duration StandOccTime of each frequency hopping signal is: StandOccTime=StandTime*OccRate The idle time of each hop StandIdleTIme is: StandIdleTIme=StandTime-StandOccTime; Step S5.3: Perform duration matching on each signal in the burst signal list in turn. When the deviation ErrTime between the signal duration and the theoretical duration is less than the duration threshold, the signal is added to the duration matching queue MeetHop, and the center time HopCenterTime of each hop point, the index HopIndex of the hop signal in the frequency hopping frequency set, and the error HopErrTime of the center time of the signal are calculated. The calculation formulas are as follows: ErrTime=|KeepLen-StandOccTime| HopCenterTime=OffsetTime+0.5*KeepLen HopIndex=Int(HopCenterTime / StandTime) HopErrTime=HopCenterTime-HopIndex*StandTime Among them, KeepLen represents the duration, Int represents rounding, and OffsetTime represents the time interval between each hop signal and the latest time signal; Step S5.4: If the number of frequency points in the duration matching queue MeetHop is less than 4, return to step S5.2 and increase the duty cycle in steps; if the number of frequency points is greater than or equal to 4, perform regularity screening of the frequency hopping sequence and execute step S5.5; Step S5.5: Cluster the frequencies in the duration matching queue MeetHop according to HopErrTime. If the frequency points after clustering are greater than or equal to 4, they are added to the same time matching queue MeetHop_Sub i and calculate the average error MeanErrTimei; where i represents the i-th group of clustering; Step S5.6: Match the MeetHop_Sub queues at the same time in sequence i The frequency points are compared according to HopIndex, and the frequency points with the same HopIndex select the HopErrTime that is closer to the mean error MeanErrTime i The frequency hopping point is added to the optimal hopping point matching queue PreHopList i middle; Step S5.7: Record the current frequency hopping signal into the frequency hopping signal queue HopRes.

8. The method for blind detection of shortwave frequency hopping signals according to claim 7, characterized in that: The frequency hopping signal parameters include the hopping rate StandHopRate, the duty cycle OccRate, the mean value of the duration error MeanKeepLenErr, the root mean square of the center time error stdTimeErr and the frequency hopping frequency parameter; The frequency hopping parameters include frequency, bandwidth, duration, and occurrence time.

9. The method for blind detection of shortwave frequency hopping signals according to claim 1, characterized in that: The step S6 includes sorting the multiple frequency hopping signals in the frequency hopping signal queue HopRes according to the hopping speed and frequency set, refreshing the frequency set and parameters if it is the same frequency hopping network signal, merging the frequency sets if the frequency sets have the same frequency, the same appearance time, and a small duration error, and taking the hopping speed parameter as the measurement result with the smaller center time root mean square error stdTimeErr; otherwise, recording a new frequency hopping signal.

10. A shortwave frequency hopping signal blind detection system, characterized in that: include: Module M1: monitoring antenna sensing, collecting spectrum data of shortwave full-band monitoring, and removing pulse interference spectrum; Module M2: Calculate adaptive threshold from spectrum data without pulse interference spectrum; Module M3: Signal sorting based on maximum spectrum and adaptive threshold, statistics of time domain appearance rules of each signal and corresponding complete appearance and disappearance time; Module M4: Perform burst signal detection on the selected basic signals according to the duration, mark the occurrence state and duration of the burst signal, and obtain a burst signal list; Module M5: deleting the timed-out burst signals in the burst signal list, and matching the standard hopping speed and duty cycle of the remaining burst signals; Module M6: perform network station sorting on the matched frequency hopping signal to obtain the final frequency hopping detection result.

Images