Interference real-time sensing device and method suitable for spread spectrum communication system

Abstract

The invention provides an interference real-time sensing device and method suitable for a spread spectrum communication system, and belongs to the technical field of signal sensing of a communication system.A key frame extraction module extracts cepstrum features and time domain cross-correlation features of link signals in an offline state; the off-line training module optimizes weight parameters based on cepstrum features and time domain cross-correlation features obtained in an off-line state in combination with a decision method of a decision tree; the cepstrum processing module extracts cepstrum characteristics of link signals on line in real time, and the cepstrum characteristics serve as first detection statistics; the correlation processing module extracts time domain cross-correlation characteristics of the link signals online in real time as second detection statistics; the online judgment module is used for comparing the first detection statistical magnitude and the second detection statistical magnitude with a threshold to complete real-time interference sensing of the spread spectrum communication system; and the parameter setting module adjusts the threshold. According to the invention, real-time sensing of different large dynamic multi-style interference signals can be realized.

Description

Technical Field

[0001] The present application belongs to the technical field of signal perception of communication systems, and more specifically, to a real-time interference perception device and method suitable for spread spectrum communication systems. Background Art

[0002] Spread spectrum communication technology is one of the information processing technologies for high-speed transmission in the information age. It was proposed in the 1940s and has the advantages of strong anti-interference ability, good concealment, and the ability to implement code division multiple access. Currently, broadband spread spectrum communication technology is widely used to improve the anti-interference performance of communication systems. However, spread spectrum communication systems have a certain interference tolerance. When encountering high-power interference signals that exceed the interference tolerance that the system itself can withstand, the communication of the system will suffer serious damage, and the anti-inductive performance of the spread spectrum communication system will decrease. At this time, corresponding measures must be taken to improve the anti-interference ability of the system, and interference detection technology is an effective solution.

[0003] Interference detection provides necessary information for the communication system to suppress interference by obtaining information such as whether the interference signal exists, the energy of the interference signal, the frequency band of the interference, and the type of interference signal, and finally eliminates the impact of the interference signal on the system communication. At present, a lot of research has been carried out on interference detection technology. Among them, the energy detection method does not require prior knowledge of the interference signal and is suitable for all types of interference signal detection. However, the energy detection method is easily affected by background noise, and the detection performance is greatly reduced at low signal-to-noise ratios. The matched filter method has a short detection time, but requires accurate prior knowledge of the signal, and is not very practical. Cyclostationary detection uses the cyclostationary characteristics of the signal for detection. It is highly robust in low signal-to-noise ratio scenarios, but has high computational complexity. Summary of the invention

[0004] In view of the defects of the prior art, the purpose of the present application is to provide a real-time interference perception device and method suitable for spread spectrum communication systems, aiming to solve the problem of poor robustness of interference detection in complex environments.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides a real-time interference perception device suitable for a spread spectrum communication system, comprising: a key frame extraction module, an offline training module, a cepstrum processing module, a correlation processing module, an online decision module and a parameter setting module;

[0006] The output end of the key frame extraction module is connected to the input end of the offline training module; the output end of the offline training module is connected to the input end of the parameter setting module and the online decision module; the output end of the cepstrum processing module and the output end of the correlation processing module are connected to the input end of the online decision module;

[0007] The key frame extraction module is used to extract the cepstrum features and time domain cross-correlation features of the link signal in an offline state;

[0008] The offline training module is used to optimize the weight parameters by grid search method based on the cepstrum features and time domain cross-correlation features obtained offline and combined with the decision tree judgment method;

[0009] The cepstrum processing module is used to extract the cepstrum features of the link signal online and in real time, and use it as the first detection statistic;

[0010] The correlation processing module is used to extract the time domain cross-correlation characteristics of the link signal online and in real time, and use it as the second detection statistic;

[0011] The online decision module is used to compare the first detection statistic and the second detection statistic with the threshold to complete the real-time perception of interference in the spread spectrum communication system;

[0012] The parameter setting module is used to adjust the threshold according to the anti-interference tolerance and optimal weight parameter of the spread spectrum system.

[0013] Further preferably, the real-time interference perception device suitable for a spread spectrum communication system also includes a data preprocessing module, whose input end is connected to the receiving link, the first output end of the data preprocessing module is connected to the key frame extraction module, and the second output end is connected to the input end of the cepstrum processing module and the related processing module; the data preprocessing module is used to down-convert, filter and down-sample the collected link signal in sequence to filter out redundant information and interference information in the link signal and reduce the amount of calculation.

[0014] Further preferably, the data preprocessing module comprises a mixing unit, a digital filtering unit, a downsampling unit and a normalization unit connected in sequence;

[0015] The mixing unit is used to mix the collected link signal and move the link signal from the radio frequency to the baseband;

[0016] The digital filter unit is used to perform digital filtering on the mixed signal to filter out the local oscillator and out-of-band interference;

[0017] The downsampling unit is used to downsample the filtered signal according to the signal characteristics of the spread spectrum communication system, reduce the sampling rate of the link signal to the same as the communication signal bandwidth, and remove redundant information in the link signal;

[0018] The normalization unit is used to normalize the power of the downsampled signal.

[0019] Further preferably, the power normalization formula is:

[0020]

[0021] Wherein, x(n) is the sampled link signal; N is the number of sampling points.

[0022] Further preferably, the real-time interference perception device suitable for the spread spectrum communication system also includes an audio-visual alarm module, an input end of which is connected to the output end of the online decision module, for outputting the interference perception result in a graphical manner.

[0023] In a second aspect, the present application provides a real-time interference perception method applicable to a spread spectrum communication system, comprising the following steps:

[0024] Step 1: Extract the cepstrum features and time domain cross-correlation features of the link signal in an offline state;

[0025] Step 2: construct the objective function based on the cepstrum feature parameters and time domain cross-correlation features obtained in step 1, and optimize the weight parameters through the grid search method in combination with the decision tree-based judgment method;

[0026] Step 3: Collect the cepstrum characteristics and time domain cross-correlation characteristics of the link signal in real time online as the first detection statistic and the second detection statistic respectively;

[0027] Step 4: Combine the optimal weight parameters, select the threshold according to the working mode of the communication system, and call the interference detection method under the corresponding working mode;

[0028] Step 5: Based on the interference detection method in step 4, the first detection statistic and the second detection statistic are compared with the threshold to determine whether there is interference in the real-time link signal, and output a judgment result.

[0029] Further preferably, the real-time interference perception method applicable to a spread spectrum communication system further includes the following steps before step 1:

[0030] Perform mixing processing on the collected link signal and move the link signal from RF to baseband;

[0031] The mixed signal is digitally filtered to remove the local oscillator and out-of-band interference;

[0032] According to the signal characteristics of the spread spectrum communication system, the filtered signal is downsampled to reduce the sampling rate of the link signal to the same as the communication signal bandwidth, and the redundant information in the link signal is removed;

[0033] The downsampled signal is power normalized.

[0034] Further preferably, step one specifically includes the following steps:

[0035] Introducing white noise with different signal-to-noise ratios into the link signal to obtain communication signals with different signal-to-noise ratios;

[0036] The communication signals with different signal-to-noise ratios are segmented, and the cepstrum and time-domain cross-correlation values ​​of each segment are calculated respectively, and then the average values ​​are calculated respectively, and the average values ​​are used as the cepstrum characteristics and time-domain cross-correlation characteristic parameters of the communication signals.

[0037] Further preferably, the calculation formula of the cepstrum feature is:

[0038] C(t)=|FFT{lg|FFT[r(t)]| 2}| 2

[0039] Wherein, r(t) is the link signal; FFT is Fourier transform;

[0040] The calculation formula of the time domain cross-correlation feature is:

[0041] R rz (τ) = R sz (τ)+R jz (τ)+R n (τ)

[0042] Among them, R rz (τ) is the cross-correlation parameter between the link signal and the reference signal; R sz (τ) is the correlation function between the direct spread signal and the reference signal in the link signal; R jz (τ) is the correlation function between the interference signal and the reference signal; R n (τ) is the autocorrelation function of the noise; the reference signal is the local communication signal, which has the same parameters as the direct spread signal in the link signal.

[0043] Further preferably, the objective function in step 2 is:

[0044] F = 0.5 (P {J c >γor J a ≤λ|H0}+P{J c ≤γ&J a >λ|H1})

[0045] Among them, J c is the cepstrum statistic, J a is the time domain cross-correlation statistic; λ is the set time domain cross-correlation threshold; γ is the set cepstrum threshold; P{*} represents the probability that * is true; H0 represents the absence of a communication signal; H1 represents the presence of a communication signal.

[0046] Further preferably, the decision tree-based decision method comprises the following steps:

[0047] Step A: Calculate a first detection statistic and a second detection statistic according to the link signal collected in real time;

[0048] Step B: Compare the first detection statistic with the cepstrum threshold. If the first detection statistic is greater than the cepstrum threshold, it is determined that there is an interference signal of 0 to 4000 kHz in the current link signal; if the first detection statistic is less than or equal to the cepstrum threshold, it is determined that there is an interference signal greater than 4000 kHz or a direct spread signal in the current link signal, and go to step C;

[0049] Step C: Compare the second detection statistic with the time domain cross-correlation threshold. If the second detection statistic is less than or equal to the time domain cross-correlation threshold, it is determined that there is an interference signal greater than 4000kHz in the current link signal; if the second detection statistic is greater than the time domain cross-correlation threshold, it is determined that the current link signal is a direct-spread signal.

[0050] Further preferably, the real-time interference perception method further includes step six, specifically:

[0051] The judgment result data stream output in step five is encapsulated, and the judgment result is visualized based on PyQT design interface by function calling method, and the real-time data stream perception result is given in a graphical way.

[0052] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art:

[0053] The present application provides a real-time interference perception device suitable for spread spectrum communication systems, wherein a cepstrum processing module and a correlation processing module are used to extract cepstrum features and time domain cross-correlation features in link signals online and in real time, and a decision tree is combined for detection and judgment, that is, real-time data frame online calculation and judgment, which can greatly improve the speed of interference perception. On the other hand, the present application fully considers the characteristics of spread spectrum communication systems in adversarial environments, and can select thresholds and interference detection methods under working modes according to the system working mode. By extracting cepstrum features and time domain cross-correlation features, and combining the machine learning "offline training + online judgment" idea, compared with traditional interference detection algorithms, the present application can realize real-time perception of different large dynamic and multi-style interference signals.

[0054] The present application provides a real-time interference perception device suitable for a spread spectrum communication system, wherein a receiving link adopts the idea of ​​software radio, a mixing unit in a data preprocessing module is used to directly convert a radio frequency signal into a baseband signal for processing, and a downsampling unit is used to reduce the amount of data storage and improve the interference detection bandwidth. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of a real-time interference perception device applicable to a spread spectrum communication system provided in an embodiment of the present application;

[0056] Figure 2It is a principle block diagram of data preprocessing in a real-time interference perception device applicable to a spread spectrum communication system provided in an embodiment of the present application;

[0057] Figure 3 It is a detection and decision flow chart of a real-time interference perception method applicable to a spread spectrum communication system provided in an embodiment of the present application;

[0058] Figure 4 It is the cepstrum feature of the direct spread signal under different interference-to-signal ratios in the real-time interference perception method applicable to the spread spectrum communication system provided in the embodiment of the present application;

[0059] Figure 5 It is the time domain cross-correlation characteristics of direct spread signals under different interference-to-signal ratios in the real-time interference perception method applicable to a spread spectrum communication system provided in an embodiment of the present application;

[0060] Figure 6 This is a diagram of perception results when the time domain cross-correlation threshold provided in the embodiment of the present application is 0.15 and the cepstrum threshold takes different values. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and specific implementation of this application clearer, the specific implementation steps of this application are further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation cases described here are only used to explain this application and are not used to limit this application.

[0062] This application provides a new real-time interference perception device and method, which collects spatial signals through antennas, preprocesses received signals, obtains key frame data, obtains the optimal weight of the interference perception model through offline training, completes cepstrum processing and correlation processing online, completes real-time interference perception through interference judgment algorithm, and outputs the results on the display control interface based on PyQT. The characteristics of interference signals with different bandwidths and different styles in spread spectrum communication systems are fully explored, and online interference perception is realized on a CPU-based operating system. The proposed structure can meet the requirements of large dynamic and multi-style interference signals and the real-time needs of interference perception.

[0063] like Figure 1 The figure shows an interference detection device based on deep support vector data description provided by an embodiment of the present application. For the convenience of description, only the part related to the embodiment of the present application is shown, which is described in detail as follows:

[0064] In the first aspect, the present application provides a real-time interference perception device suitable for a spread spectrum communication system, comprising: a data preprocessing module 00, a key frame extraction module 01, an offline training module 02, a cepstrum processing module 03, a correlation processing module 04, an online judgment module 05, a parameter setting module 06 and an audio-visual alarm module 07;

[0065] The input end of the data preprocessing module 00 is connected to the receiving link, the first output end of the data preprocessing module 00 is connected to the key frame extraction module 01, and the second output end of the data preprocessing module 00 is connected to the input end of the cepstrum processing module 03 and the correlation processing module 04. The data preprocessing module is used to sequentially down-convert, filter and down-sample the collected link signal to filter out redundant information and interference information in the signal and reduce the amount of calculation;

[0066] The input end of the receiving link is connected to the output end of the receiving antenna, and is used to complete analog signal processing such as frequency conversion, filtering and amplification of the received signal;

[0067] The input end of the key frame extraction module 01 is connected to the first output end of the data preprocessing module, and the output of the key frame extraction module 01 is connected to the input end of the offline training module 02. The key frame extraction module is used to extract the cepstrum and cross-correlation features of the link signal in an offline state;

[0068] The input end of the offline training module 02 is connected to the key frame extraction module 01, and the output end thereof is connected to the online judgment module; the offline training module is used to optimize the weight parameters of the offline training network based on the cepstrum and cross-correlation features of the link signal extracted in the offline state to improve the detection performance;

[0069] The input end of the cepstrum processing module 03 is connected to the second output end of the data preprocessing module 00, and the output end of the cepstrum processing module 03 is connected to the second input end of the online decision module 05; the cepstrum processing module 03 is used to extract the cepstrum features of the link signal online in real time and use it as a detection statistic;

[0070] The input end of the correlation processing module 04 is connected to the third output end of the data preprocessing module 00, and the output end of the correlation processing module 04 is connected to the third input end of the online decision module 05. The correlation processing module 04 is used to extract the time domain cross-correlation characteristics of the link signal online in real time and use it as a detection statistic;

[0071] The first input end of the online decision module 05 is connected to the output end of the offline training module 02, the second input end of the online decision module 05 is connected to the output end of the cepstrum processing module 03, and the third input end of the online decision module 05 is connected to the output end of the correlation processing module 04. The online decision module 05 is used to make a decision on the detection statistic and output a decision result, that is, to compare the detection statistic with the threshold, so as to complete the real-time perception of interference in the spread spectrum communication system;

[0072] The output end of the parameter setting module 06 is connected to the fourth input end of the online decision module 05, and the parameter setting module 06 is used to adjust the threshold according to the anti-interference tolerance of the spread spectrum communication system and the optimal weight parameter obtained by offline training;

[0073] The input end of the sound and light alarm module 07 is connected to the output end of the online decision module 05, and the sound and light alarm module 07 is used to present the perception result in a graphical manner.

[0074] like Figure 2 As shown, the data preprocessing module 00 includes a mixing unit 001, a digital filtering unit 002, a downsampling unit 003 and a normalization unit 004; the output end of the mixing unit 001 is connected to the input end of the digital filtering unit 002, and the output end of the digital filter 002 is connected to the input end of the downsampling unit 003; the output end of the downsampling unit 003 is connected to the normalization unit 004; the output end of the normalization unit 004 is connected to the input end of the key frame extraction module 01.

[0075] In a second aspect, the present application provides a real-time interference perception method applicable to a spread spectrum communication system, comprising the following steps:

[0076] Step S1: pre-processing the collected link signal to remove redundant information in the link signal to avoid unnecessary interference of signals of different strengths on the detection result;

[0077] Step S2: extracting cepstrum characteristic parameters and time domain cross-correlation characteristic parameters of the link signal offline;

[0078] Step S3: constructing an objective function according to the cepstrum characteristic parameters and time domain cross-correlation characteristic parameters obtained offline in step S2, and optimizing weight parameters by a grid search method based on a decision tree-based decision process;

[0079] Step S4: combining the optimal weight parameter, selecting a threshold according to the working mode of the communication system, and calling the interference detection method in the corresponding working mode to determine whether there is interference in the real-time frame data;

[0080] Step S5: Design a visualization interface based on PyQT for the judgment result, and present the perception result in a graphical manner.

[0081] Further preferably, step S1 specifically includes the following steps:

[0082] Step S11: performing mixing processing on the collected link signal, and moving the link signal from the radio frequency to the baseband 5MHz;

[0083] Step S12: performing digital filtering on the mixed signal to filter out the local oscillator and out-of-band interference;

[0084] Step S13: according to the signal characteristics of the spread spectrum communication system, down-sampling the filtered signal, reducing the sampling rate of the link signal to be consistent with the communication signal bandwidth, and removing redundant information in the link signal;

[0085] Step S14: normalize the power of the downsampled signal to avoid unnecessary interference of signals of different strengths on the detection result.

[0086] Further preferably, the power normalization formula is:

[0087]

[0088] Where x(n) is the sampled signal; N is the number of sampling points;

[0089] Furthermore, step S2 specifically includes the following steps:

[0090] Step S21: introducing white noise (interference signal) with a signal-to-noise ratio within a certain range into the preprocessed link signal, obtaining communication signals with different signal-to-noise ratios, and improving the robustness of the algorithm to white noise;

[0091] Step S23: Divide the communication signals with different signal-to-noise ratios into segments and calculate the cepstrum and cross-correlation value of each segment respectively, then calculate the average value, and use the average value as the cepstrum feature and cross-correlation feature parameters of the communication signal respectively.

[0092] More preferably, the calculation formula of the cepstrum is:

[0093] C(t)=|FFT{lg|FFT[r(t)]| 2}| 2

[0094] Where r(t) is the link signal;

[0095] The calculation formula for time domain cross-correlation is:

[0096] R rz (τ) = R sz (τ)+R jz (τ)+R n (τ)

[0097] Among them, R rz (τ) is the cross-correlation parameter between the link signal and the reference signal; R sz (τ) is the correlation function between the direct spread signal and the reference signal in the link signal; R jz (τ) is the correlation function between the interference signal and the reference signal; R n (τ) is the autocorrelation function of the noise; the reference signal is the local communication signal, which has the same parameters as the direct spread signal in the link signal; in order to reduce the error, the cross-correlation value is normalized, and the normalized time domain cross-correlation calculation formula is:

[0098]

[0099] Among them, R rr(0) represents the autocorrelation function of the received signal when the time delay is 0, R zz (0) represents the autocorrelation function of the reference signal when the time delay is 0.

[0100] Further preferably, step S3 specifically includes the following steps:

[0101] Step S31: determining the target function according to the cepstrum and cross-correlation characteristic parameters of the mixed signal;

[0102] Step S33: Combined with the minimization of the objective function and the decision process based on the decision tree, the optimal weight parameter is searched through the network search method.

[0103] Furthermore, the objective function expression is:

[0104] F = 0.5 (P {J c >γor J a ≤λ|H0}+P{J c ≤γ&J a >λ|H1})

[0105] Among them, J c is the cepstrum statistic, J a is the time domain cross-correlation statistic; λ is the set cross-correlation threshold; γ is the set cepstrum threshold; P{*} represents the probability that * is true; H0 represents the absence of a communication signal; H1 represents the presence of a communication signal.

[0106] Furthermore, step S4 specifically includes the following steps:

[0107] Step S41: adjusting the threshold according to the anti-interference tolerance of the spread spectrum communication system and the optimal weight parameter obtained by offline training;

[0108] Step S42: performing cepstrum calculation and time domain cross-correlation calculation on the real-time data frame;

[0109] Step S43: The cepstrum and time domain cross-correlation features of the real-time data frame are used as detection statistics, and the detection statistics are compared with the threshold, and finally the decision result is output through the interference decision algorithm.

[0110] Furthermore, step S5 specifically includes the following steps:

[0111] Step S52: encapsulate the judgment result data stream obtained in step S43, and transmit it to the sound and light alarm module by function calling method;

[0112] Step S53: Design a visualization interface based on PyQT to present the real-time data stream perception results in a graphical manner.

[0113] Example

[0114] Figure 3 The decision flow chart based on the decision tree first uses the extracted cepstrum and time domain cross-correlation features as detection statistics, and then compares the detection statistics with the set threshold. When the cepstrum statistic of the received signal is less than the threshold and the time domain cross-correlation statistic is greater than the threshold, it is determined to be a direct spread signal, otherwise it is an interference signal.

[0115] Figure 4 and Figure 5 They are the cepstrum characteristics and time domain cross-correlation characteristics of the DSSS signal under different interference-to-signal ratios, where the interference-to-signal ratio (ISR) is the ratio of the interference signal to the DSSS signal energy; when ISR = -30 dB, the eigenvalue is close to the theoretical value of the DSSS signal, and when ISR = 30 dB, the eigenvalue is close to the theoretical value of the interference signal; from Figure 4 and Figure 5 It can be seen that there are differences in the cepstrum characteristics and time domain cross-correlation characteristics of the DSSS signal and the interference signal. The interference signal and the DSSS signal can be distinguished by setting a threshold;

[0116] Figure 6 This is the perception result diagram of this application when the time domain cross-correlation threshold is 0.15 and the cepstrum threshold takes different values. The detection performance indicators are the accuracy rate, false alarm rate and missed detection rate, which are expressed as follows:

[0117]

[0118]

[0119]

[0120] Among them, P a is the accuracy, P f is the false alarm rate, P r is the missed detection rate, N a N is the total number of samples that detect the communication signal as a communication signal. t N is the total number of samples that detect the interference signal as interference signal. f N is the total number of samples that detect the communication signal as an interference signal. r N is the total number of samples that detect the interference signal as a useful signal. p is the total number of samples; Figure 6 The results are obtained by testing in an experimental environment, and it can be seen that the application has good detection performance. Under different cepstrum thresholds, the accuracy rate is always greater than 87%, the false alarm rate is 12%, and the missed detection rate is less than 4%. When the time domain cross-correlation threshold is 0.15 and the cepstrum threshold is 1.9, the accuracy rate is the largest, reaching 92.44%, and the missed detection rate is the smallest, reaching 0.73%.

[0121] In summary, compared with the prior art, the present application has the following advantages:

[0122] Realize real-time perception of large dynamic and diverse interference signals: This application uses the inverse spectrum and time domain cross-correlation method to extract signal features, and combines the decision tree for detection and judgment. Compared with traditional interference detection algorithms, this application can realize real-time perception of different large dynamic and diverse interference signals.

[0123] Interference perception is more real-time: This application adopts a method of online calculation and judgment of real-time data frames, which can greatly improve the speed of interference perception.

[0124] Effectively improve the interference detection bandwidth: The receiving link of this application adopts the idea of ​​software radio, directly converts the RF signal into a baseband signal for processing, reduces the amount of data storage, and improves the interference detection bandwidth.

[0125] It should be understood that expressions such as "including" and "may include" that may be used in the present application indicate the existence of the disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements. In the present application, terms such as "including" and / or "having" may be interpreted as indicating specific characteristics, numbers, operations, constituent elements, components, or combinations thereof, but may not be interpreted as excluding the existence or possibility of adding one or more other characteristics, numbers, operations, constituent elements, components, or combinations thereof.

[0126] In the description of the embodiments of the present application, it should be noted that, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense.

[0127] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A real-time interference perception device suitable for spread spectrum communication systems, It is characterized in that include: Key frame extraction module, offline training module, cepstrum processing module, correlation processing module, online judgment module and parameter setting module; The output end of the key frame extraction module is connected to the input end of the offline training module; the output end of the offline training module is connected to the input end of the parameter setting module and the online decision module; the output end of the cepstrum processing module and the output end of the correlation processing module are connected to the input end of the online decision module; The key frame extraction module is used to extract the cepstrum features and time domain cross-correlation features of the link signal in an offline state; the offline training module is used to optimize the weight parameters through a grid search method based on the cepstrum features and time domain cross-correlation features obtained in an offline state and combined with the decision tree judgment method; The cepstrum processing module is used to extract the cepstrum features of the link signal online and in real time, and use it as the first detection statistic; The correlation processing module is used to extract the time domain cross-correlation characteristics of the link signal online and in real time, and use it as the second detection statistic; The online decision module is used to compare the first detection statistic and the second detection statistic with the threshold to complete the real-time perception of interference in the spread spectrum communication system; The parameter setting module is used to adjust the threshold according to the anti-interference tolerance and optimal weight parameter of the spread spectrum system.

2. The real-time interference perception device according to claim 1, It is characterized in that It also includes a data preprocessing module, whose input end is connected to the receiving link, whose first output end is connected to the key frame extraction module, and whose second output end is connected to the input end of the cepstrum processing module and the related processing module; The data preprocessing module is used to down-convert, filter and down-sample the collected link signals in sequence to filter out redundant information and interference information in the link signals and reduce the amount of calculation.

3. The real-time interference perception device according to claim 2, It is characterized in that The data preprocessing module includes a mixing unit, a digital filtering unit, a downsampling unit and a normalization unit connected in sequence; The mixing unit is used to mix the collected link signal and move the link signal from the radio frequency to the baseband; The digital filter unit is used to perform digital filtering on the mixed signal to filter out the local oscillator and out-of-band interference; The downsampling unit is used to downsample the filtered signal according to the signal characteristics of the spread spectrum communication system, reduce the sampling rate of the link signal to the same as the communication signal bandwidth, and remove redundant information in the link signal; The normalization unit is used to normalize the power of the downsampled signal.

4. The real-time interference perception device according to claim 3, It is characterized in that The power normalization formula is: Wherein, x(n) is the sampled link signal; N is the number of sampling points.

5. The real-time interference sensing device according to any one of claims 1 to 4, It is characterized in that It also includes an audible and visual alarm module, whose input end is connected to the output end of the online judgment module and is used to output the interference perception result in a graphical manner.

6. A real-time interference perception method for spread spectrum communication systems, It is characterized in that The following steps are involved: Step 1: Extract the cepstrum features and time domain cross-correlation features of the link signal in an offline state; Step 2: construct the objective function based on the cepstrum feature parameters and time domain cross-correlation features obtained in step 1, and optimize the weight parameters through the grid search method in combination with the decision tree-based judgment method; Step 3: Collect the cepstrum characteristics and time domain cross-correlation characteristics of the link signal in real time online as the first detection statistic and the second detection statistic respectively; Step 4: Combine the optimal weight parameters, select the threshold according to the working mode of the communication system, and call the interference detection method under the corresponding working mode; Step 5: Based on the interference detection method in step 4, the first detection statistic and the second detection statistic are compared with the threshold to determine whether there is interference in the real-time link signal, and output a judgment result.

7. The real-time interference perception method according to claim 6, It is characterized in that Before step 1, the following steps are also included: Perform mixing processing on the collected link signal and move the link signal from RF to baseband; The mixed signal is digitally filtered to remove the local oscillator and out-of-band interference; According to the signal characteristics of the spread spectrum communication system, the filtered signal is downsampled to reduce the sampling rate of the link signal to the same as the communication signal bandwidth, and the redundant information in the link signal is removed; The downsampled signal is power normalized.

8. The real-time interference perception method according to claim 6 or 7, It is characterized in that Step 1 specifically includes the following steps: The link signal is introduced into white noise with different signal-to-noise ratios to obtain communication signals with different signal-to-noise ratios; The communication signals with different signal-to-noise ratios are segmented, and the cepstrum and time-domain cross-correlation values ​​of each segment are calculated respectively, and then the average values ​​are calculated respectively, and the average values ​​are used as the cepstrum characteristics and time-domain cross-correlation characteristic parameters of the communication signals.

9. The real-time interference perception method according to claim 8, It is characterized in that The calculation formula of the cepstrum feature is: C(t)=FFT{lgFFT[r(t)] 2 } 2 Wherein, r(t) is the link signal; FFT is Fourier transform; The calculation formula of the time domain cross-correlation feature is: R rz (τ)=R sz (t)+R jz (t)+R n (t) Among them, R rz (τ) is the cross-correlation parameter between the link signal and the reference signal; R sz (τ) is the correlation function between the direct spread signal and the reference signal in the link signal; R jz (τ) is the correlation function between the interference signal and the reference signal; R n (τ) is the autocorrelation function of the noise; the reference signal is the local communication signal, which has the same parameters as the direct spread signal in the link signal.

10. The real-time interference perception method according to claim 6, It is characterized in that The objective function in step 2 is: F=0.5(P{J c >γor J a ≤λ|H 0 }+P{J c ≤γ&J a >λ|H 1 }) Among them, J c is the cepstrum statistic, J a is the time domain cross-correlation statistic; λ is the set time domain cross-correlation threshold; γ is the set cepstrum threshold; P{*} represents the probability that * is true; H 0 Indicates that there is no communication signal; H1 indicates that there is a communication signal.

11. The real-time interference perception method according to claim 6 or 10, It is characterized in that The decision tree-based decision method includes the following steps: Step A: Calculate a first detection statistic and a second detection statistic according to the link signal collected in real time; Step B: Compare the first detection statistic with the cepstrum threshold. If the first detection statistic is greater than the cepstrum threshold, it is determined that there is an interference signal of 0 to 4000 kHz in the current link signal; if the first detection statistic is less than or equal to the cepstrum threshold, it is determined that there is an interference signal greater than 4000 kHz or a direct spread signal in the current link signal, and go to step C; Step C: Compare the second detection statistic with the time domain cross-correlation threshold. If the second detection statistic is less than or equal to the time domain cross-correlation threshold, it is determined that there is an interference signal greater than 4000kHz in the current link signal; if the second detection statistic is greater than the time domain cross-correlation threshold, it is determined that the current link signal is a direct-spread signal.

12. The real-time interference perception method according to claim 6, It is characterized in that Step 5 also includes step 6, which is specifically: The judgment result data stream output in step five is encapsulated, and the judgment result is visualized based on PyQT design interface by function calling method, and the real-time data stream perception result is given in a graphical way.

Images