Satellite navigation jamming signal type identification method and system

Abstract

The application provides a satellite navigation interference signal type identification method and system, and belongs to the field of satellite navigation.The method comprises the following steps: collecting complete navigation signals from a satellite receiver, calculating a carrier-to-noise ratio and pseudorange residual information, and judging whether a navigation interference signal exists in the complete navigation signals; if the navigation interference signal is detected, collecting intermediate frequency data in the complete navigation signals, performing spectrum analysis, adding a sliding window to the intermediate frequency data, extracting frequency domain features of the complete navigation signals based on the sliding window and a sliding window weight, combining a gated recurrent unit, extracting time domain features of the complete navigation signals by performing time domain analysis on the intermediate frequency data in the complete navigation signals; and combining the frequency domain features of the complete navigation signals and the time domain features of the complete navigation signals to calculate a navigation interference signal type.The application improves the adaptability of satellite navigation interference signal type identification and the accuracy of the result.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of satellite navigation, and particularly relates to a satellite navigation jamming signal type identification method and system. BACKGROUND

[0002] Satellite navigation refers to a technology for performing navigation positioning on users on the ground, sea, air and space by using navigation satellites. Common satellite navigation systems include a GPS navigation system, a Beidou navigation system and the like. With the advantages of high precision, all-weather, global coverage and the like, the satellite navigation system has become an indispensable navigation means in many fields. Anti-jamming is an important index for evaluating navigation capability. Jamming signals in satellite navigation include suppression jamming, deception jamming and disturbance jamming, which can seriously affect the reception quality of satellite navigation signals and cause positioning precision to decrease or even completely fail. Satellite navigation jamming signal type identification refers to analyzing and processing received satellite navigation signals to accurately determine whether there is jamming in the signals and further identify the type of the jamming signals, which is of great significance for improving the anti-jamming capability of the satellite navigation system and ensuring the stability of navigation signals.

[0003] In the prior art, in terms of satellite navigation jamming signal type identification, traditional methods mainly include methods based on signal processing, statistical models and machine learning. Among them, the method based on signal processing usually relies on filtering, spectrum analysis and the like, extracts features in the signal, and determines the existence and type of the jamming signal by using a preset rule or threshold. However, this kind of method is difficult to cope with complex and changeable jamming environments, and the extraction and identification of signal features depend on artificial experience and preset rules, which lack flexibility and adaptability. The method based on statistical models uses statistical parameters to identify jamming signals by establishing a statistical model of the signal. Although the accuracy and robustness of identification are improved to some extent, the establishment and optimization process of the model is complex, and it is difficult to adapt to rapidly changing jamming environments. The method based on machine learning usually adopts the strategy of providing a special identification algorithm for each sub-type when facing numerous jamming types. Although the identification accuracy of a specific jamming is improved, the algorithm complexity and maintenance cost are increased, a large amount of resources is consumed, and false positives are caused due to conflicts between algorithms, which affects the overall identification efficiency. SUMMARY

[0004] In view of the above defects or deficiencies in the prior art, the present application aims to provide a satellite navigation jamming signal type identification method and system to solve the problems of insufficient adaptability of traditional satellite navigation jamming signal type identification methods and false positives caused by algorithm conflicts.

[0005] To achieve the above-mentioned purpose, the embodiments of the present application adopt the following technical solutions:

[0006] In a first aspect, the embodiments of the present application provide a satellite navigation interference signal type identification method, the method comprising:

[0007] Step S1, collecting a complete navigation signal from a satellite receiver;

[0008] Step S2, calculating the carrier-to-noise ratio (CNR) and pseudo-range residual information of the complete navigation signal, and determining whether a navigation interference signal exists in the complete navigation signal;

[0009] Step S3, if the navigation interference signal is detected, extracting intermediate frequency (IF) data in the complete navigation signal and performing preprocessing, and if the navigation interference signal is not detected, returning to step S1;

[0010] Step S4, performing spectrum analysis on the IF data, and extracting frequency domain features of the complete navigation signal;

[0011] Step S5, performing time domain analysis on the IF data in the complete navigation signal, and extracting time domain features of the complete navigation signal;

[0012] Step S6, constructing a navigation interference signal identification network, combining the frequency domain features of the complete navigation signal and the time domain features of the complete navigation signal, and calculating the type of the navigation interference signal.

[0013] As a preferred embodiment of the present application, step S1 further comprises: receiving navigation signals from no less than four satellites using the satellite receiver, and demodulating the received navigation signals to convert them from radio frequency (RF) signals to baseband signals as the complete navigation signal.

[0014] As a preferred embodiment of the present application, step S2 further comprises:

[0015] Step S21, decoding the complete navigation signal to extract satellite orbit parameters, signal propagation time, receiver position and other information, recording signal power through the RF front-end circuit of the receiver, and recording noise power through the noise measurement circuit of the receiver;

[0016] Step S22, calculating the CNR using the signal power and the noise power, and the calculation formula (1) is as follows:

[0017]

[0018] In formula (1), CNR is the carrier-to-noise ratio, lg() is a logarithm operation with base 10, P1 is the signal power, and P2 is the noise power.

[0019] Step S23: Multiply the signal propagation time measured by the receiver by the speed of light to calculate the actual measured pseudorange; use the satellite orbit parameters and receiver position to calculate the theoretical pseudorange using the least squares method; subtract the theoretical pseudorange from the actual measured pseudorange to obtain the pseudorange residual ρ.

[0020] Step S24: Calculate the probability of the existence of an interference signal, and determine whether an interference signal exists based on the probability of the existence of an interference signal.

[0021] As a preferred embodiment of the present invention, the probability formula (2) for calculating the existence of interference signals in step S24 is as follows:

[0022] pro=sigmoid(MLP(ρ,CNR)) (2)

[0023]

[0024] In equation (2), pro is the probability of the presence of interference signal, sigmoid() is the sigmoid function, and MLP() is the multilayer perceptron operation;

[0025] When determining whether there is a navigation interference signal, if pro = 1, then the navigation interference signal is considered to exist; if pro = 0, then the navigation interference signal is considered not to exist.

[0026] In a preferred embodiment of the present invention, step S4 further includes:

[0027] Step S41: Use the Fast Fourier Transform algorithm to convert the intermediate frequency data in the complete navigation signal into a frequency domain signal;

[0028] Step S42: Add a sliding window to the frequency domain signal, and extract the frequency domain features of the complete navigation signal based on the frequency domain signal in the sliding window. The extraction formulas (3)-(7) are as follows:

[0029] F frequency =

[0030] MLP(Concat(F weighted window ,F global ),F global ,F weighted window (3)

[0031] In equation (3), F weighted window For weighted sliding window features; F global It is a global feature; and,

[0032] F weighted window =∑ i w i ·F i,window (4)

[0033]

[0034] w i =Attention(F i,window ) (6)

[0035]

[0036] In formula (4)-(7), w i is an attention weight, F i,window is a feature of the i-th sliding window, i is a sliding window index, and n_window is a sliding window number; MLP() is a multi-layer perception operation, Concat() is a concatenation operation, W is a window size, X i is a frequency domain signal of the i-th sliding window, CNN() is a convolutional neural network operation, Attention() is an attention mechanism operation, and GRU() is a gated recurrent unit operation; and X is a frequency domain signal of intermediate frequency data in a complete navigation signal.

[0037] As a preferred embodiment of the present application, step S5 further comprises:

[0038] Step S51, a gated recurrent unit network is used to extract preliminary time domain features based on intermediate frequency data in a complete navigation signal, and a calculation formula (8) is as follows:

[0039] h T =GRU(T) (8)

[0040] In formula (8), h T is a preliminary time domain feature, and T is intermediate frequency data in a complete navigation signal.

[0041] Step S52, an attention mechanism is used to calculate weights of each time step and perform normalization, and a calculation formula is as follows:

[0042] w′ t =Attention(h t ) (9)

[0043]

[0044] In formula (9) and (10), t is a time step index, w′ t is a weight of each time step, and a t is a normalized weight of each time step.

[0045] Step S53, the normalized attention weights a t are sorted in a positive order, and the first n time steps are selected, n is a number of time steps that satisfy that a cumulative attention weight exceeds a threshold value p, and a calculation formula (11) is as follows:

[0046]

[0047] In formula (11), TimeSteps is a time step set, and j is a count index.

[0048] In step S54, the gated recurrent unit network is modified, and the specific operation is as follows: for the selected n time steps, a neural network connection is added between adjacent two time steps, so that information can be directly transmitted on the n time steps; the modified gated recurrent unit network is represented as GRU'(T).

[0049] In step S55, the time domain feature of the complete navigation signal is calculated, and the calculation formula (12) is as follows:

[0050]

[0051] h′ T = GRU'(T)

[0052] In formula (12), F time is the time domain feature of the complete navigation signal, h1, h2, …, hn respectively represent the hidden states corresponding to the first, second, …, and n-th time steps in the time step set TimeSteps, and h′ T is the modified preliminary time domain feature.

[0053] As a preferred embodiment of the present application, step S6 further comprises:

[0054] In step S61, based on the frequency domain feature and the time domain feature, the navigation interference signal type feature is calculated, and the calculation formula is as follows:

[0055] f1 = MLP(Concat(F time ,F frequency )) (13)

[0056]

[0057] g1 = tanh(β1·f2+b1) (15)

[0058] g2 = tanh(β2·f1+b2) (16)

[0059] f′1 = f1·g1 (17)

[0060] f′2 = f2·g2 (18)

[0061] f = MLP(Concat(f′1,f′2)) (19)

[0062] In formula (13)-(19), f1 is a longitudinal fusion feature, f2 is a transverse fusion feature, CNN (1) is a longitudinal one-dimensional convolution operation, g1 is a longitudinal gate, tanh() is a hyperbolic tangent function, β1 is a longitudinal parameter, b1 is a longitudinal bias term, g2 is a transverse gate, β2 is a transverse parameter, b2 is a transverse bias term, and f is a navigation interference signal type feature.

[0063] In step S62, the probabilities of each type of navigation interference signal are calculated according to the navigation interference signal type features, and the type of navigation interference signal is identified according to the calculated probabilities of each type of navigation interference signal.

[0064] As a preferred embodiment of the present application, the formula (20) for calculating the probabilities of each type of navigation interference signal is as follows:

[0065] p=softmax(f) (20)

[0066] In formula (20), p is the probability of each type of navigation interference signal, and softmax() is a softmax function.

[0067] As a preferred embodiment of the present application, the method further comprises:

[0068] In step S7, a random gradient descent optimization algorithm is used to optimize the navigation interference signal identification network in combination with an L2 regularization technique.

[0069] And step S7 specifically operates as follows:

[0070] In step S71, the weight and bias parameters of the navigation interference signal identification network are randomly initialized, the learning rate is set to 0.001, and the regularization coefficient is 0.01.

[0071] In step S72, a cross-entropy loss function is used to measure the difference between the network prediction result and the true label.

[0072] In step S73, the network parameters are updated using a random gradient descent algorithm until the navigation interference signal identification network converges.

[0073] In a second aspect, the embodiments of the present application also provide a satellite navigation interference signal type identification system, which comprises a signal acquisition module, a presence probability calculation module, a preliminary judgment module, an intermediate frequency data acquisition module, a frequency domain feature extraction module, a time domain feature extraction module, and an interference type identification module, wherein

[0074] The signal acquisition module is used to acquire complete navigation signals from a satellite receiver.

[0075] The presence probability calculation module is configured to calculate the carrier-to-noise ratio and pseudo-range residual information of the complete navigation signal, determine whether a navigation interference signal exists in the complete navigation signal, and send the determination result to the preliminary determination module.

[0076] The preliminary determination module is configured to start the intermediate frequency data acquisition module when receiving the result that a navigation interference signal exists, and start the signal acquisition module when receiving the result that a navigation interference signal does not exist.

[0077] The intermediate frequency data acquisition module is configured to extract the intermediate frequency data in the complete navigation signal and perform preprocessing when detecting a navigation interference signal.

[0078] The frequency domain feature extraction module is configured to perform spectrum analysis on the intermediate frequency data and extract the frequency domain features of the complete navigation signal.

[0079] The time domain feature extraction module is configured to perform time domain analysis on the intermediate frequency data in the complete navigation signal and extract the time domain features of the complete navigation signal.

[0080] The interference type identification module is configured to combine the frequency domain features of the complete navigation signal and the time domain features of the complete navigation signal to calculate the type of the navigation interference signal.

[0081] The technical solution provided by the embodiment of the application has the following beneficial effects:

[0082] (1) The deep learning idea is introduced, the presence of interference in the satellite navigation signal is preliminarily determined according to the carrier-to-noise ratio and pseudo-range residual information, the frequency domain and time domain features of the navigation signal are extracted respectively, and finally the navigation interference signal recognition network is constructed to fuse the frequency domain and time domain features, thereby improving the adaptability of the satellite navigation interference signal type recognition method, avoiding the problem that each type of interference needs a specific algorithm for targeted recognition, and avoiding the situation that various signal interference recognition algorithms alarm at the same time;

[0083] (2) The frequency domain feature extraction method is innovated, the sliding window calculation idea is introduced for local feature extraction, the attention mechanism is used to assign weights to each sliding window feature, the importance of the local feature is considered, GRU is applied to the global frequency domain signal to extract features, the weighted sliding window features are spliced, and finally the frequency domain features of the complete navigation signal are obtained through the multilayer perception, thereby improving the effect of frequency domain feature extraction and laying a data foundation for subsequent navigation interference signal classification;

[0084] (3) Innovated the time domain feature extraction mode, considering the problem of long-term information loss of the traditional gated recurrent unit network, which is not conducive to analyzing the dynamic characteristics of signal interference, based on the time step importance, an information transmission channel is added in the gated recurrent unit network, so that the information can be transmitted by jumping, reducing the information attenuation and improving the overall time domain feature extraction capability;

[0085] (4) Innovated the feature fusion mode of frequency domain and time domain, proposed a gating mechanism based on multilayer perceptron, considering the characteristics of the two dimensions of vertical and horizontal information, highlighting important features and suppressing irrelevant features; compared with the feature fusion based on attention mechanism, the gating mechanism proposed in the embodiment of the application is based on simple mathematical operation, the calculation process is more concise and efficient; and the strength of each gating signal directly reflects the importance of the corresponding feature, which is helpful for understanding and debugging the model.

[0086] Of course, implementing any product or method of the present application does not necessarily require all the advantages described above. BRIEF DESCRIPTION OF DRAWINGS

[0087] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0088] Figure 1 is a flow chart of the satellite navigation jamming signal type identification method described in the embodiments of the present application. DETAILED DESCRIPTION

[0089] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0090] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In the description of the present application, the terms "first", "second", "third", "fourth" and the like are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0091] In view of the problems existing in the identification of the navigation interference signal type, the embodiments of the present application provide a satellite navigation interference signal type identification method and system, aiming at solving the problem of poor adaptability of the traditional satellite navigation interference signal type identification method and the false alarm problem caused by algorithm conflict.

[0092] Referring to Figure 1 , the satellite navigation interference signal type identification method comprises the following steps:

[0093] Step S1, collecting complete navigation signals from a satellite receiver.

[0094] In this step, the satellite receiver is used to receive navigation signals from no less than 4 satellites, and the received navigation signals are demodulated to convert them from radio frequency signals to baseband signals as complete navigation signals.

[0095] Step S2, calculating the carrier-to-noise ratio (CNR) and pseudo-range residual information of the complete navigation signals to determine whether there is a navigation interference signal in the complete navigation signals.

[0096] Specifically, this step further comprises:

[0097] Step S21, decoding the complete navigation signals to extract satellite orbit parameters, signal propagation time, receiver position and other information, and recording the signal power through the radio frequency front-end circuit of the receiver and recording the noise power through the noise measurement circuit of the receiver;

[0098] Step S22, calculating the CNR using the signal power and the noise power, and the calculation formula (1) is as follows:

[0099]

[0100] In formula (1), CNR is the carrier-to-noise ratio, lg() is the logarithm operation with base 10, P1 is the signal power, and P2 is the noise power.

[0101] Step S23, multiplying the signal propagation time measured by the receiver by the speed of light to calculate the actual measured pseudo-range, using the satellite orbit parameters and the receiver position to calculate the theoretical pseudo-range by the least square method, and subtracting the theoretical pseudo-range from the actual measured pseudo-range to obtain the pseudo-range residual ρ;

[0102] Step S24, calculating the probability of the existence of the interference signal, and the calculation formula (2) is as follows:

[0103] pro=sigmoid(MLP(ρ,CNR)) (2)

[0104]

[0105] In formula (2), pro is a probability of existence of an interference signal, sigmoid() is a sigmoid function, and MLP() is a multi-layer perception operation.

[0106] According to the probability of existence of the interference signal, it is determined whether the interference signal exists. The specific determination process is as follows: if pro=1, it is considered that the navigation interference signal exists; and if pro=0, it is considered that the navigation interference signal does not exist.

[0107] In step S3, if the navigation interference signal is detected, the intermediate frequency data in the complete navigation signal is extracted and preprocessed; and if the navigation interference signal is not detected, step S1 is returned.

[0108] In this step, the pre-processing of the intermediate frequency data includes: performing direct current component removal and downsampling processing on the intermediate frequency data; the direct current component removal is to eliminate the direct current offset in the signal and avoid affecting the subsequent analysis; and the downsampling processing reduces the calculation complexity and removes redundant data.

[0109] In step S4, the intermediate frequency data is subjected to spectrum analysis, and the frequency domain features of the complete navigation signal are extracted.

[0110] Specifically, this step further includes:

[0111] In step S41, a fast Fourier transform algorithm is used to convert the intermediate frequency data in the complete navigation signal into a frequency domain signal;

[0112] In step S42, a sliding window is added to the frequency domain signal, and the frequency domain features of the complete navigation signal are extracted based on the frequency domain signal in the sliding window. The extraction formulas (3)-(7) are as follows: Σ

[0113] F frequency =

[0114] MLP(Concat(F weighted window ,F global ),F global ,F weighted window ) (3)

[0115] In formula (3), F weightedwindow is a weighted sliding window feature; F global is a global feature; and,

[0116] F weighted window = i w i ·F i,window (4)

[0117]

[0118] w i = Attention(Fi,window ) (6)

[0119]

[0120] in formula (4)-(7), w i is an attention weight, F i,window is a feature of the i-th sliding window, i is a sliding window index, n_window is a sliding window number; MLP() is a multi-layer perception operation, Concta() is a concatenation operation, W is a window size, X i is a frequency domain signal of the i-th sliding window, CNN() is a convolutional neural network operation; Attention() is an attention mechanism operation; GRU() is a gated recurrent unit operation, and X is a frequency domain signal of the intermediate frequency data in the complete navigation signal.

[0121] In this step, when performing frequency domain feature extraction, a sliding window calculation idea is introduced, and each sliding window can perform local feature extraction, and the convolutional neural network and the mean feature are integrated when performing local feature extraction. Then, the local features are concatenated and further processed by MLP to obtain sliding window features. Then, considering that not all sliding window information should be treated equally, but the key information should be extracted and the irrelevant information should be ignored, the attention mechanism is used to assign weights to each sliding window feature, and the weighted sum is obtained. Weighted sliding window features; At the same time, the global frequency domain signal is applied to GRU to extract features, and the weighted sliding window features are concatenated, and finally the multi-layer perception is obtained to obtain the frequency domain features of the complete navigation signal.

[0122] The traditional frequency domain feature extraction method relies on frequency spectrum analysis or single feature extraction, and cannot fully capture the complex interference characteristics in the navigation signal. Due to the limited feature extraction capability, in the subsequent identification of the type of interference signal, different algorithm models can only be compared one by one. When a certain interference mode meets the discrimination conditions of several algorithms at the same time, the traditional method will fail. In this step, by introducing a sliding window and a weight, a new network feature extraction structure is constructed, and the extracted frequency domain features provide rich input information for the subsequent navigation interference signal recognition network, so in the subsequent type identification, it is only necessary to regard it as a multi-classification problem and solve it.

[0123] Step S5, performing time domain analysis on the intermediate frequency data in the complete navigation signal to extract time domain features of the complete navigation signal.

[0124] Specifically, this step further includes:

[0125] Step S51, using a gated recurrent unit network to extract preliminary time domain features based on the intermediate frequency data in the complete navigation signal, and the calculation formula (8) is as follows:

[0126] h T= GRU(T) (8)

[0127] In formula (8), h T is a preliminary time domain feature, and T is intermediate frequency data in a complete navigation signal.

[0128] In step S52, an attention mechanism is used to calculate the weight of each time step and normalize it, and the calculation formula is as follows:

[0129] w′ t = Attention(h t ) (9)

[0130]

[0131] In formulas (9) and (10), t is a time step index, w′ t is the weight of each time step, and a t is the normalized weight of each time step.

[0132] In step S53, the normalized attention weight a t is sorted in a positive order, and the first n time steps are selected, where n is the number of time steps that satisfy that the cumulative attention weight exceeds a threshold value p, and the calculation formula (11) is as follows:

[0133]

[0134] In formula (11), TimeSteps is a time step set, and j is a counting index.

[0135] In step S54, the gated recurrent unit network is modified, and the specific operation is as follows: for the selected n time steps, a neural network connection is added between adjacent two time steps, so that information can be directly transmitted among the n time steps; and the modified gated recurrent unit network is represented as GRU′().

[0136] In step S55, the time domain feature of the complete navigation signal is calculated, and the calculation formula (12) is as follows:

[0137]

[0138] h′ T = CRU′(T)

[0139] In formula (12), F time is a time domain feature of a complete navigation signal, are hidden states corresponding to the first, second, …, and n time steps in the time step set TimeSteps, respectively, and h′ T is a modified preliminary time domain feature.

[0140] In this step, the time domain information is the key to understanding the dynamic behavior of the navigation signal. The traditional method only relies on time domain analysis (time domain waveform analysis, probability density analysis, etc.) or gated recurrent unit network. The traditional time domain analysis method cannot fully capture the complex time domain characteristics in the navigation signal, and the gated recurrent unit network has the problem of long-term information loss. In contrast, in this step, the gated recurrent unit network is first used to extract the time domain features of the navigation signal, capturing the time dependence of the signal. Then, the attention mechanism is used to calculate the weight of each time step, highlighting the information of important time steps. Finally, through the multi-layer perception and feature fusion operations, the features of these key time steps are integrated, and the global features updated by the gated recurrent unit network are considered to obtain the time domain features of the complete navigation signal.

[0141] The traditional attention mechanism combined with the gated recurrent unit directly takes the last hidden state as the network output. However, in this step, the first n key time steps are selected according to the attention weight, and neural network connections are added between these time steps, so that information can be transmitted through the newly added neural network connections, solving the problem of long-term information loss of the gated recurrent unit.

[0142] Step S6, construct a navigation interference signal recognition network, combine the frequency domain features of the complete navigation signal and the time domain features of the complete navigation signal to calculate the type of navigation interference signal.

[0143] Specifically, this step further includes:

[0144] Step S61, based on the frequency domain features and the time domain features, calculate the navigation interference signal type features, the calculation formula is as follows:

[0145] f1=MLP(Concat(F time ,F frequency )) (13)

[0146]

[0147] g1=tanh(β1·f2+b1) (15)

[0148] g2=tanh(β2·f1+b2) (16)

[0149] f′1=f1·g1 (17)

[0150] f′2=f2·g2 (18)

[0151] f=MLP(Concat(f′1,f′2)) (19)

[0152] In equations (13)-(19), f1 is the longitudinal fusion feature, f2 is the transverse fusion feature, CNN(1) For the longitudinal one-dimensional convolution operation, g1 is a longitudinal gate, tanh() is a hyperbolic tangent function, β1 is a longitudinal parameter, b1 is a longitudinal bias term, g2 is a transverse gate, β2 is a transverse parameter, b2 is a transverse bias term, and f is a navigation jamming signal type feature.

[0153] In this step, the longitudinal fusion feature and the transverse fusion feature are calculated respectively. The longitudinal fusion feature aims to directly fuse the frequency domain and time domain features. The transverse fusion feature aims to analyze the information of the time domain and the frequency domain step by step through one-dimensional convolution, capture the local correlation between features, and highlight important features and suppress irrelevant features. In addition, a gating mechanism based on a multilayer perception is proposed. Considering the characteristics of the longitudinal and transverse dimensions, the fusion features are weighted, important features are highlighted, and irrelevant features are suppressed. Compared with the traditional attention mechanism, the gating mechanism is based on simple mathematical operations, and the calculation process is more concise and efficient. In addition, the strength of each gating signal directly reflects the importance of the corresponding feature, which is helpful for understanding and debugging the model. Although the attention mechanism can also achieve feature weighting, its internal operation process is relatively complex and not easy to intuitively explain.

[0154] In step S62, the probability of each type of navigation jamming signal is calculated according to the navigation jamming signal type feature. The calculation formula (20) is as follows:

[0155] p = softmax(f) (20)

[0156] In formula (20), p is the probability of each type of navigation jamming signal, and softmax() is a softmax function.

[0157] According to the calculated probability of each type of navigation jamming signal, the type of navigation jamming signal is identified.

[0158] For example, the probability p of each type of navigation jamming signal is calculated as: [0.2, 0.45, 0.35], which represents the probability of the signal being a suppression jamming is 0.2, the probability of being a deception jamming is 0.45, and the probability of being a disturbance jamming is 0.35. Therefore, the type of the navigation jamming signal is finally judged to be a deception jamming.

[0159] Here, the situation of preliminary judgment error in S1 step can also be solved by adding the option of non-interference signal, that is, the signal is preliminarily judged as existing interference, but actually no interference exists. For example, when the pro calculated in S1 step of a certain signal is 0.501, the signal is preliminarily judged as existing interference according to the judgment rule; but after deep judgment of time domain and frequency domain feature extraction and fusion, it is found that the signal is a normal signal in both time domain and frequency domain; in order to deal with this situation, the probability of each type of navigation interference signal needs to be added with a type, so that it changes from a 3-dimensional vector to a 4-dimensional vector, such as [0.2, 0.25, 0.2, 0.35], which represents the maximum probability of the signal being "non-interference", and finally judges the type of the navigation interference signal as non-interference, and corrects the preliminary judgment result.

[0160] The embodiment can further include:

[0161] Step S7, using a stochastic gradient descent optimization algorithm, combining an L2 regularization technique, optimizing the navigation interference signal recognition network.

[0162] Specifically, the step further includes:

[0163] Step S71, randomly initializing the weight and bias parameters of the navigation interference signal recognition network, setting the learning rate to 0.001 and the regularization coefficient to 0.01;

[0164] Step S72, using a cross-entropy loss function to measure the difference between the network prediction result and the true label;

[0165] Step S73, using a stochastic gradient descent algorithm to update the network parameters until the navigation interference signal recognition network converges.

[0166] Based on the same idea, the embodiment of the application also provides a satellite navigation interference signal type recognition system. The system includes:

[0167] The system includes a signal acquisition module, an existence probability calculation module, a preliminary judgment module, an intermediate frequency data acquisition module, a frequency domain feature extraction module, a time domain feature extraction module, and an interference type recognition module; preferably, the system can also include an algorithm optimization module; wherein,

[0168] The signal acquisition module is used to acquire complete navigation signals from a satellite receiver;

[0169] The existence probability calculation module is used to calculate the carrier-to-noise ratio and pseudo-range residual information of the complete navigation signal, judge whether a navigation interference signal exists in the complete navigation signal, and send the judgment result to the preliminary judgment module;

[0170] The preliminary judgment module is configured to start the intermediate frequency data acquisition module when receiving the result of the existing navigation interference signal, and start the signal acquisition module when receiving the result of the non-existing navigation interference signal.

[0171] The intermediate frequency data acquisition module is configured to extract intermediate frequency data in the complete navigation signal and perform preprocessing when detecting the navigation interference signal.

[0172] The frequency domain feature extraction module is configured to perform spectrum analysis on the intermediate frequency data and extract the frequency domain feature of the complete navigation signal.

[0173] The time domain feature extraction module is configured to perform time domain analysis on the intermediate frequency data in the complete navigation signal and extract the time domain feature of the complete navigation signal.

[0174] The interference type identification module is configured to calculate the type of the navigation interference signal in combination with the frequency domain feature of the complete navigation signal and the time domain feature of the complete navigation signal.

[0175] The algorithm optimization module is configured to use a stochastic gradient descent optimization algorithm and combine an L2 regularization technique to optimize the navigation interference signal identification network.

[0176] The modules in the embodiment are implemented by a processor, and a memory is appropriately added when storage is needed. The processor can be, but is not limited to, a microprocessor (MPU), a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), other programmable logic devices, discrete gates, transistor logic devices, discrete hardware components, and the like. The memory can include a random access memory (RAM) and can also include a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory can also be at least one storage device located away from the aforementioned processor.

[0177] In the above embodiments, all or part can be realized by software, hardware, firmware or any combination thereof. When realized by software, all or part can be realized in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, the computer instructions produce the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server or data center to another via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode.

[0178] In addition, it should be noted that the satellite navigation jamming signal type identification system and the satellite navigation jamming signal type identification method described in the embodiments are corresponding. The description and limitation of the method also apply to the system, which will not be described here.

[0179] From the above technical solutions, it can be seen that the satellite navigation jamming signal type identification method and system provided by the embodiments of the present application collect complete navigation signals from a satellite receiver, calculate the carrier-to-noise ratio and pseudo-range residual information, and determine whether there is a navigation jamming signal in the complete navigation signal. If the navigation jamming signal is detected, the intermediate frequency data in the complete navigation signal is collected and spectrum analysis is performed. A sliding window is added to the intermediate frequency data, and based on the sliding window and the sliding window weight, the frequency domain features of the complete navigation signal are extracted by combining the gated recurrent unit. The time domain features of the complete navigation signal are extracted by performing time domain analysis on the intermediate frequency data in the complete navigation signal. The type of navigation jamming signal is calculated by combining the frequency domain features of the complete navigation signal and the time domain features of the complete navigation signal. The present application has the following advantages:

[0180] (1) The deep learning idea is introduced. First, according to the carrier-to-noise ratio and pseudo-range residual information, it is preliminarily judged whether there is interference in the satellite navigation signal. Then, the features of the frequency domain and time domain of the navigation signal are extracted respectively. Finally, the navigation jamming signal recognition network is constructed to fuse the frequency domain and time domain features, which improves the adaptability of the satellite navigation jamming signal type identification method and avoids the problem that each type of interference needs a specific algorithm for targeted recognition. Since the present application adopts the fusion recognition technical idea, there is no step of judging each type of interference, so the situation of simultaneous alarm of various signal interference recognition algorithms is avoided.

[0181] (2) Innovated the frequency domain feature extraction method, introduced the calculation idea of sliding window, carried out local feature extraction, and used attention mechanism to assign weight to each sliding window feature, considering the importance of local features; at the same time, the global frequency domain signal was applied to GRU to extract features, and the weighted sliding window features were spliced, and finally the frequency domain features of the complete navigation signal were obtained through multilayer perception, which improved the effect of frequency domain feature extraction and laid data foundation for subsequent navigation interference signal classification.

[0182] (3) Innovated the time domain feature extraction method, considering the problem of long-term information loss of traditional gated recurrent unit network, which is not conducive to analyzing the dynamic characteristics of signal interference, the application is based on time step importance, adds information transmission channel in the gated recurrent unit network, so that information can be transmitted by jumping, reduces information attenuation, and improves the overall time domain feature extraction capability.

[0183] (4) Innovated the feature fusion method of frequency domain and time domain, proposed a gating mechanism based on multilayer perception, considering the characteristics of longitudinal and transverse two-dimensional information, highlighting important features and suppressing irrelevant features; compared with the feature fusion based on attention mechanism, the gating mechanism proposed in the application is based on simple mathematical operation, and the calculation process is more concise and efficient; and the strength of each gating signal directly reflects the importance of the corresponding feature, which is helpful for understanding and debugging the model.

[0184] The above description is only the preferred embodiment of the application and the explanation of the applied technical principles, and is not intended to limit the scope of the claimed application, but only represents the preferred embodiment of the application. Those skilled in the art should understand that the scope of the application involved in the application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the application.

Claims

1. A method for identifying satellite navigation interference signal types, characterized in that, The method includes: Step S1: Acquire complete navigation signals from the satellite receiver; Step S2: Calculate the carrier-to-noise ratio and pseudorange residual information of the complete navigation signal to determine whether there is a navigation interference signal in the complete navigation signal; Step S3: If a navigation interference signal is detected, extract the intermediate frequency data from the complete navigation signal and perform preprocessing; if no navigation interference signal is detected, return to step S1. Step S4, perform spectrum analysis on the intermediate frequency data to extract the frequency domain features of the complete navigation signal; including: Step S41: Use the Fast Fourier Transform algorithm to convert the intermediate frequency data in the complete navigation signal into a frequency domain signal; Step S42: Add a sliding window to the frequency domain signal, and extract the frequency domain features of the complete navigation signal based on the frequency domain signal in the sliding window. The extraction formulas (3)-(7) are as follows: F frequency =MLP(Concat(F weighted window ,F global ),F global ,F weighted window ) (3) In equation (3), F weighted window For weighted sliding window features; F global It is a global feature; and, F weighted window =∑ i w i ·F i,window (4) w i =Attention(F i,window ) (6) In equations (4)-(7), w i For attention weights, F i,window Let be the feature of the i-th sliding window, where i is the sliding window index and n_window is the number of sliding windows; MLP() is the multilayer perceptron operation, Concat() is the concatenation operation, W is the window size, and X is the window size. i Let X be the frequency domain signal of the i-th sliding window; CNN() is the operation of a convolutional neural network; Attention() is the operation of an attention mechanism; GRU() is the operation of a gated recurrent unit; and X is the frequency domain signal of the intermediate frequency data in the complete navigation signal. Step S5: Perform time-domain analysis on the intermediate frequency data in the complete navigation signal to extract the time-domain features of the complete navigation signal; Step S6: Construct a navigation interference signal identification network, and calculate the navigation interference signal type by combining the frequency domain characteristics and time domain characteristics of the complete navigation signal.

2. The satellite navigation interference signal type identification method according to claim 1, characterized in that, Step S1 further includes: using a satellite receiver to receive navigation signals from no less than 4 satellites, and demodulating the received navigation signals to convert them from radio frequency signals to baseband signals as complete navigation signals.

3. The satellite navigation interference signal type identification method according to claim 1, characterized in that, Step S2 further includes: Step S21: Decode the complete navigation signal to extract satellite orbit parameters, signal propagation time and receiver position information. At the same time, record the signal power through the receiver's radio frequency front-end circuit and the noise power through the receiver's noise measurement circuit. Step S22: Calculate the carrier-to-noise ratio using the signal power and noise power. The calculation formula (1) is as follows: In equation (1), CNR is the carrier-to-noise ratio, lg() is the logarithmic operation with base 10, P1 is the signal power, and P2 is the noise power. Step S23: Multiply the signal propagation time measured by the receiver by the speed of light to calculate the actual measured pseudorange; use the satellite orbit parameters and receiver position to calculate the theoretical pseudorange using the least squares method; subtract the theoretical pseudorange from the actual measured pseudorange to obtain the pseudorange residual ρ. Step S24: Calculate the probability of the existence of an interference signal, and determine whether an interference signal exists based on the probability of the existence of an interference signal.

4. The satellite navigation interference signal type identification method according to claim 3, characterized in that, Step S24 calculates the probability formula (2) for the presence of interference signals as follows: pro=sigmoid(MLP(ρ,CNR)) (2) In equation (2), pro is the probability of the presence of interference signal, sigmoid() is the sigmoid function, and MLP() is the multilayer perceptron operation; When determining whether there is a navigation interference signal, if pro = 1, then the navigation interference signal is considered to exist; if pro = 0, then the navigation interference signal is considered not to exist.

5. The satellite navigation interference signal type identification method according to claim 1, characterized in that, Step S5 further includes: Step S51: Using a gated cyclic cell network, preliminary time-domain features are extracted based on the intermediate frequency data in the complete navigation signal. The calculation formula (8) is as follows: h T =GRU(T) (8) In equation (8), h T This represents the preliminary time-domain characteristics, where T is the intermediate frequency data in the complete navigation signal; Step S52: Using an attention mechanism, calculate and normalize the weights for each time step. The calculation formula is as follows: w′ t =Attention(h t ) (9) In equations (9) and (10), t is the time step index, and h t For the hidden state corresponding to time step index t, w′ t As the weight of each time step, a t These are the weights for each time step after normalization; Step S53, adjust the normalized attention weights a t Sort the data in ascending order and select the first n time steps, where n is the number of time steps where the cumulative attention weight exceeds the threshold Q. The calculation formula (11) is as follows: In equation (11), TimeSteps is the set of time steps, and j is the counting index; Step S54, modify the gated recurrent unit network, specifically as follows: for the selected n time steps, add a neural network connection between two adjacent time steps so that information can be directly transmitted in these n time steps; the modified gated recurrent unit network is represented as GRU′(); Step S55, calculate the time-domain characteristics of the complete navigation signal, and the calculation formula (12) is as follows: h′ T =GRU′(T) In equation (12), F time For the time-domain characteristics of the complete navigation signal, Let h′ represent the hidden states corresponding to the 1st, 2nd, ..., nth time steps in the time step set TimeSteps. T This represents the modified preliminary time-domain features.

6. The satellite navigation interference signal type identification method according to claim 5, characterized in that, Step S6 further includes: Step S61: Based on frequency domain and time domain characteristics, calculate the navigation interference signal type characteristics. The calculation formula is as follows: f1=MLP(Concat(F time ,F frequency )) (13) g1=tanh(β1·f2+b1) (15) g2=tanh(β2·f1+b2) (16) f′1=f1·g1 (17) f′2=f2·g2 (18) f=MLP(Concat(f′1,f′2)) (19) In equations (13)-(19), f1 represents the vertical fusion feature, f2 represents the horizontal fusion feature, and CNN... (1) For vertical one-dimensional convolution operation, g1 is the vertical gate, tanh() is the hyperbolic tangent function, β1 is the vertical parameter, b1 is the vertical bias term, g2 is the horizontal gate, β2 is the horizontal parameter, b2 is the horizontal bias term, and f is the navigation interference signal type feature. Step S62: Calculate the probability of each type of navigation interference signal based on the characteristics of the navigation interference signal type; identify the type of navigation interference signal based on the calculated probabilities of each type of navigation interference signal.

7. The satellite navigation interference signal type identification method according to claim 6, characterized in that, The probability formula (20) for calculating each type of navigation interference signal is as follows: p = softmax(f) (20) In equation (20), p represents the probability of each type of navigation interference signal, and softmax() is the softmax function.

8. The satellite navigation interference signal type identification method according to claim 1, characterized in that, The method further includes: Step S7: The navigation interference signal identification network is optimized by using the stochastic gradient descent optimization algorithm combined with L2 regularization. The specific operation of step S7 is as follows: Step S71: Randomly initialize the weights and bias parameters of the navigation interference signal identification network, set the learning rate to 0.001, and the regularization coefficient to 0.01; Step S72: Use the cross-entropy loss function to measure the difference between the network prediction results and the true labels; Step S73: Update the network parameters using the stochastic gradient descent algorithm until the navigation interference signal identification network converges.

9. A satellite navigation interference signal type identification system, characterized in that, The system includes: a signal acquisition module, an existence probability calculation module, a preliminary judgment module, an intermediate frequency data acquisition module, a frequency domain feature extraction module, a time domain feature extraction module, and an interference type identification module; wherein... The signal acquisition module is used to acquire complete navigation signals from the satellite receiver; The existence probability calculation module is used to calculate the carrier-to-noise ratio and pseudorange residual information of the complete navigation signal, determine whether there is a navigation interference signal in the complete navigation signal, and send the judgment result to the preliminary judgment module. The preliminary judgment module is used to activate the intermediate frequency data acquisition module when a result indicating the presence of navigation interference signals is received; and to activate the signal acquisition module when a result indicating the absence of navigation interference signals is received. The intermediate frequency data acquisition module is used to extract intermediate frequency data from the complete navigation signal and perform preprocessing when navigation interference signals are detected; and is further used to perform the following steps: Step S41: Use the Fast Fourier Transform algorithm to convert the intermediate frequency data in the complete navigation signal into a frequency domain signal; Step S42: Add a sliding window to the frequency domain signal, and extract the frequency domain features of the complete navigation signal based on the frequency domain signal in the sliding window. The extraction formulas (3)-(7) are as follows: F frequency =MLP(Concat(F weighted window ,F global ),F global ,F weighted window ) (3) In equation (3), F weighted window For weighted sliding window features; F global It is a global feature; and, F weighted window =∑ i w i ·F i,window (4) w i =Attention(F i,window ) (6) In equations (4)-(7), w i For attention weights, F i,window Let be the feature of the i-th sliding window, where i is the sliding window index and n_window is the number of sliding windows; MLP() is the multilayer perceptron operation, Concat() is the concatenation operation, W is the window size, and X is the window size. i Let X be the frequency domain signal of the i-th sliding window; CNN() is the operation of a convolutional neural network; Attention() is the operation of an attention mechanism; GRU() is the operation of a gated recurrent unit; and X is the frequency domain signal of the intermediate frequency data in the complete navigation signal. The frequency domain feature extraction module is used to perform spectrum analysis on the intermediate frequency data and extract the frequency domain features of the complete navigation signal; The time-domain feature extraction module is used to perform time-domain analysis on the intermediate frequency data in the complete navigation signal and extract the time-domain features of the complete navigation signal; The interference type identification module is used to calculate the type of navigation interference signal by combining the frequency domain characteristics and time domain characteristics of the complete navigation signal.

Images