Plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform
Through the plaintext detection alarm system of multi-dimensional feature fusion and Fourier transform, the machine learning algorithm is used to determine that the data packet is plaintext or ciphertext, which solves the problem that the transmission of plaintext data cannot be effectively detected in the prior art, improves the detection accuracy and reduces the The phenomenon of plain text transmission has been improved and network security has been enhanced.
Patent Information
- Application Number
- CN202410993991.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-07-24
AI Technical Summary
The prior art cannot effectively and accurately detect and alert the transmission of plaintext data in the internal network, resulting in an increased risk of malicious attackers using sniffer tools to steal data.
Using a plaintext detection and alarm system based on multidimensional feature fusion and Fourier transform, the classification judgment model is trained using machine learning algorithms, and multidimensional features and frequency domain features are fused to perform The plain text or cipher text judgment of the data packet reduces the phenomenon of plain text transmission.
It improves the accuracy of detection of plaintext data during data packet transmission, reduces plaintext transmission phenomenon in internal networks, and enhances network security.
Smart Images

Figure CN118900196A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of network security technology, and in particular to a plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform. Background Art
[0002] In the current network security environment, enterprises and organizations have higher and higher requirements for data security. However, due to software implementation problems or improper configuration, there is a phenomenon of plaintext data transmission in the internal network, which provides opportunities for malicious attackers to use sniffer tools to steal data. There are a large number of security threats and attacks in the field of network security. In order to protect and prevent network security threats, it is necessary to introduce multi-dimensional feature fusion and Fourier transform technology for analysis and detection, so as to identify and respond accordingly in time.
[0003] Multidimensional feature fusion technology can fuse a variety of different feature information together to form a comprehensive feature vector for data analysis and classification. Fourier transform technology can convert signals (such as network traffic) into frequency domain. By analyzing and extracting the frequency, some useful information in the signal can be found. These two technologies are widely used in the field of network security and can identify attack traffic, abnormal traffic, etc.
[0004] Patent No. CN2019108454664 discloses a voice tampering detection method based on the fusion of multiple features, which detects whether the voice file is spliced, including the following steps: Step S1, divide the voice data to be detected into multiple groups of voice data frames by framing; Step S2, extract multidimensional features for each group of voice data frames; Step S3, build a model based on Attention-RNN as a classifier; Step S4, input the multidimensional features extracted in step S2 into the trained classifier to determine whether the current frame of voice has been tampered with. The method of the above invention can effectively mine the difference between the previous and next features in the voice signal by extracting frame-level features, combining multiple features, enriching the voice feature mining, and using the attention mechanism to give different importance to the parts of the same sample, and automatically learn the characteristics of the time series signal.
[0005] Patent No. CN2017107739554 discloses a harmonic detection method based on discrete Fourier transform, which includes a harmonic detection method based on discrete Fourier transform, characterized in that it includes collecting three-phase current on the power grid side within a set period, and discretely processing the collected three-phase current; constructing a calculation model for the fundamental active current amplitude and the fundamental reactive current amplitude of the three-phase current; performing a discrete Fourier transform forward transformation on the three-phase current collected on the power grid side through a sliding window iteration method to obtain the fundamental active current amplitude and the fundamental reactive current amplitude of the three-phase current respectively; using coordinate transformation to calculate the fundamental active current positive sequence component of the three-phase current; and calculating the detected three-phase harmonic current based on the collected three-phase current and the fundamental active current positive sequence component.
[0006] Although the above patents all use multi-dimensional feature fusion or Fourier transform to detect information data, a single processing method cannot effectively and accurately detect and alarm the transmission of plaintext data, and the phenomenon of plaintext transmission in the internal network cannot be accurately avoided. Summary of the invention
[0007] The purpose of the present invention is to provide a plaintext detection alarm system based on multidimensional feature fusion and Fourier transform, which can fuse multidimensional feature fusion and Fourier transform, and then use machine learning algorithm to train a classification judgment model to judge whether a data packet is plaintext or ciphertext; so as to improve the detection accuracy of plaintext data in the process of data packet transmission; and reduce the plaintext transmission phenomenon in the internal network.
[0008] The present invention utilizes the following technical solutions:
[0009] A plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform includes a collection unit, a processing unit, a measurement unit, a judgment unit and a verification unit; wherein,
[0010] A collection unit, used for collecting data packets during data transmission;
[0011] A processing unit, used for preprocessing the collected data packets to obtain multi-dimensional features of the common data payload with special fields of the data payload removed;
[0012] A calculation unit, used for performing multi-dimensional feature fusion and Fourier transformation on the pre-processed data packets to obtain a comprehensive feature vector;
[0013] A judgment unit, used to classify the comprehensive feature vector using a preset classification judgment model, and optimize the data packet using a messy information database;
[0014] The verification unit is used to re-verify the data packets that have completed the classification judgment.
[0015] Preferably, the collection unit continuously samples the data packet set during the data transmission process; the data transmission process includes but is not limited to a gateway and / or a router; the data packet set includes a plurality of data packets, each of which includes a data payload and a protocol header.
[0016] Preferably, the process of preprocessing the data packet by the processing unit is as follows:
[0017] S1: extract the data payload and protocol header of each data packet;
[0018] P i =D i ⊙H i (1)
[0019] Where i represents the packet number, P i represents the ith data packet, D i represents the data load of the ith data packet, ⊙ represents the Hadamard outer product, H i Represents the protocol header of the i-th data packet;
[0020] S2: Use the protocol type to identify and remove the special fields of the data payload to obtain the normal data payload;
[0021]
[0022] Among them, D i ′ represents the normal data payload of the ith data packet, special_fields() represents the special field removal function, T i Indicates the protocol type of the i-th data packet, represents the Hadamard inner product;
[0023] S3: Calculate and obtain multi-dimensional features of common data load; multi-dimensional features include but are not limited to entropy value H(D i ′ ), byte frequency distribution F(D i ′ )、Data length L(D i ′ ) and the time series characteristics T(D i ′ ).
[0024] Preferably, the process of the calculation unit calculating the preprocessed data packets is as follows:
[0025] First, a fast Fourier transform is performed on the common data payload to obtain the frequency domain characteristics of the common data payload;
[0026] F f (D i ′)=FFT(D i ′ ) (3)
[0027] Among them, F f (D′ i ) represents the frequency domain characteristics of ordinary data payload, and FFT() represents the fast Fourier function;
[0028] Then, the multi-dimensional features and frequency domain features are fused into a comprehensive feature vector V i ;
[0029] V i =[H(D′ i ), F(D′ i ), L(D′ i ), T(D′ i ), F f (D′ i )] (4).
[0030] Preferably, the process of the judgment unit classifying and judging the comprehensive feature vector is as follows:
[0031] A: Use the data set constructed from historical comprehensive feature vectors and machine learning algorithms to iteratively train the classification judgment model;
[0032] M=TrainModel(V1, V2,...,V n ) (5)
[0033] Among them, M represents the classification judgment model, TrainModel() represents the machine learning algorithm, and V n represents the nth historical comprehensive feature vector;
[0034] B: Input the comprehensive feature vector of each data packet into the classification judgment model and compare it with the entropy threshold H t Compare to obtain the classification result; the classification result is plain text or cipher text;
[0035] Result(P()=M(Vi) (6)
[0036] Among them, ResuIt() represents the classification judgment result;
[0037] C: Optimize the data load of the data packet using the messy information database to reduce the false positives of the classification judgment results; the messy information database includes the protocol header and special fields;
[0038]
[0039] Among them, D iRepresents the optimized data load, and known_fields() represents a messy information database.
[0040] Preferably, the classification judgment model includes a feature separation layer, an iterative training layer and a recognition classification layer; the feature separation layer first performs nonlinear processing on the historical comprehensive feature vector to obtain a historical feature matrix;
[0041]
[0042] Among them, MP n represents the historical feature matrix, SBox[] represents the substitution box function, ABox[] represents the permutation box function, Represents vector and, x (k,n) represents the chaotic sequence of the nth item and the kth dimension;
[0043] Then the historical feature matrix is input into the iterative training layer, and the multi-scale attention mechanism is used to perform several iterative trainings to obtain the feature weight matrix; the iterative training layer includes 4 3X3 convolutional layers, 2 3X3 maximum pooling layers, 6 residual blocks of Inception V3 blocks in parallel, 3 global pooling layers, 2 dimension expansion layers, 3 Sigmoid activation functions, 3 normalization layers and 3 5X5 convolutional layers; the 4 3X3 convolutional layers are connected in parallel with the 3 5X5 convolutional layers and 3 3X3 maximum pooling layers in series, and are connected in series with 2 normalization layers;
[0044] Finally, the recognition and classification layer is used to optimize and update the feature weight matrix to obtain the classification judgment result; the recognition and classification layer includes 4 Inception V1 blocks, 3 Inception V4 blocks, 2 5X5 upsampling layers, 2 3X3 average pooling layers and 2 MISH activation functions.
[0045] Preferably, the verification unit uses an advanced statistical algorithm to perform a secondary verification on the data packet whose classification result is plain text: if KS_statistic(D i ″ )≤F t , then the data packet is plain text; if KS_statistic(D i ″ )>F t , then the data packet is ciphertext; KS_statistic() represents the Kolmogorov-Smirnov test algorithm, F t represents the feature threshold; advanced statistical algorithms include but are not limited to the Kolmogorov-Smirnov test algorithm.
[0046] The present invention fuses multi-dimensional features with Fourier transform, and then uses a machine learning algorithm to train a classification judgment model to judge whether a data packet is plain text or cipher text; thereby improving the detection accuracy of plain text data during data packet transmission; and reducing the plain text transmission phenomenon in the internal network. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0048] Figure 1 This is the principle block diagram of the plain text detection alarm system. DETAILED DESCRIPTION
[0049] The present invention is described in detail below with reference to the accompanying drawings and embodiments:
[0050] like Figure 1 As shown, the plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform described in the present invention includes a collection unit, a processing unit, a measurement unit, a judgment unit and a verification unit; wherein,
[0051] A collection unit, used for collecting data packets during data transmission;
[0052] In the present invention, the collection unit continuously samples the data packet set during the data transmission process; the data transmission process includes but is not limited to a gateway and / or a router; the data packet set includes a plurality of data packets, each of which includes a data payload and a protocol header;
[0053] A processing unit, used for preprocessing the collected data packets to obtain multi-dimensional features of the common data payload with special fields of the data payload removed;
[0054] In the present invention, the process of preprocessing the data packet by the processing unit is as follows:
[0055] S1: extract the data payload and protocol header of each data packet;
[0056] P i =D i ⊙H i (1)
[0057] Where i represents the packet number, P i represents the ith data packet, D i represents the data load of the ith data packet, ⊙ represents the Hadamard outer product, H iRepresents the protocol header of the i-th data packet;
[0058] S2: Use the protocol type to identify and remove the special fields of the data payload to obtain the normal data payload;
[0059]
[0060] Among them, D i v represents the normal data payload of the ith data packet, special_fields() represents the special field removal function, T i Indicates the protocol type of the i-th data packet, represents the Hadamard inner product;
[0061] S3: Calculate and obtain multi-dimensional features of common data load; multi-dimensional features include but are not limited to entropy value H(D i ′ ), byte frequency distribution F(D i ′ )、Data length L(D i ′ ) and the time series characteristics T(D i ′ ).
[0062] In this embodiment, the entropy value represents a measure of uncertainty or information in a system and is often used to evaluate the randomness and complexity of data;
[0063] Frequency distribution is used to describe the frequency or proportion of different values in data, and is often used to analyze the central tendency and dispersion of data;
[0064] Time series features are used to describe the patterns and regularities of data changes over time and are often used in time series analysis and prediction;
[0065] Hadamard inner product is a way of expressing matrix operations. The matrix is obtained by multiplying the elements one by one. It is often used in signal processing and data compression.
[0066] The protocol header represents the control information part of the data packet, which is used to identify the protocol type and transmission information of the data packet, and help network devices to correctly parse and process the data packet;
[0067] Protocol type identification refers to determining the type of communication protocol used by a data packet by analyzing its protocol header. It is often used in network traffic analysis and protocol parsing.
[0068] A calculation unit, used for performing multi-dimensional feature fusion and Fourier transformation on the pre-processed data packets to obtain a comprehensive feature vector;
[0069] In the present invention, the process of calculating the pre-processed data packets by the measuring unit is as follows:
[0070] First, a fast Fourier transform is performed on the common data payload to obtain the frequency domain characteristics of the common data payload;
[0071] F f (D′ i ) = FFT(D′ i ) (3)
[0072] Among them, F f (D′ i ) represents the frequency domain characteristics of ordinary data payload, and FFT() represents the fast Fourier function;
[0073] Then, the multi-dimensional features and frequency domain features are fused into a comprehensive feature vector V i ;
[0074] V i =[H(D′ i ), F(D′ i ), L(D′ i ), T(D′ i ), F f (D′ i )] (4);
[0075] In this embodiment, the fast Fourier transform refers to a mathematical transformation that converts a signal from the time domain to the frequency domain, and reveals the periodicity and frequency characteristics of the signal by analyzing the frequency components of the signal;
[0076] A judgment unit, used to classify the comprehensive feature vector using a preset classification judgment model, and optimize the data packet using a messy information database;
[0077] In this embodiment, the messy information database is used to store the database of protocol headers and special fields, and is also used to optimize and reduce false positives in data classification;
[0078] In the present invention, the process of the judgment unit classifying and judging the comprehensive feature vector is as follows:
[0079] A: Use the data set constructed from historical comprehensive feature vectors and machine learning algorithms to iteratively train the classification judgment model;
[0080] M=TrainModel(V1, V2,...,V n ) (5)
[0081] Among them, M represents the classification judgment model, Train / Model() represents the machine learning algorithm, and V n represents the nth historical comprehensive feature vector; the machine learning algorithm can adopt a neural network;
[0082] B: Input the comprehensive feature vector of each data packet into the trained classification judgment model and compare it with the entropy threshold H t Compare to obtain the classification result; the classification result is plain text or cipher text;
[0083] Result(P i )=M(V i ) (6)
[0084] Among them, Result() represents the classification judgment result;
[0085] In the present invention, the classification judgment model includes a feature separation layer, an iterative training layer and a recognition classification layer; the feature separation layer first performs nonlinear processing on the historical comprehensive feature vector to obtain a historical feature matrix;
[0086]
[0087] Among them, MP n represents the historical feature matrix, SBox[] represents the substitution box function, ABox[] represents the permutation box function, Represents vector and, x (k,n) represents the chaotic sequence of the nth item and the kth dimension;
[0088] Then the historical feature matrix is input into the iterative training layer, and the multi-scale attention mechanism is used to perform several iterative trainings to obtain the feature weight matrix; the iterative training layer includes 4 3X3 convolutional layers, 2 3X3 maximum pooling layers, 6 residual blocks of Inception V3 blocks in parallel, 3 global pooling layers, 2 dimension expansion layers, 3 Sigmoid activation functions, 3 normalization layers and 3 5X5 convolutional layers; the 4 3X3 convolutional layers are connected in parallel with the 3 5X5 convolutional layers and 3 3X3 maximum pooling layers in series, and are connected in series with 2 normalization layers;
[0089] Finally, the recognition and classification layer is used to optimize and update the feature weight matrix to obtain the classification judgment result; the recognition and classification layer includes 4 Inception V1 blocks, 3 Inception V4 blocks, 2 5X5 upsampling layers, 2 3X3 average pooling layers and 2 MISH activation functions;
[0090] In this embodiment, the feature separation layer represents a part of the machine learning model and is also used to extract and separate different features of the input data for subsequent feature processing and classification;
[0091] The multi-scale attention mechanism is a technique in machine learning that improves the model's feature extraction and classification capabilities by focusing on different features of data at different scales;
[0092] The Inception block is a structure in a convolutional neural network that extracts features through convolution kernels of different scales to improve the model's feature extraction capability and classification effect.
[0093] The residual block is a structure in deep learning that alleviates the gradient vanishing problem by adding skip connections and improves the training effect of deep networks.
[0094] The Sigmoid activation function is a common activation function that maps the input to a value between 0 and 1 and is often used in binary classification problems.
[0095] The ReLU activation function is a common activation function that maps input values less than 0 to 0 and values greater than 0 remain unchanged. It is often used in deep learning models.
[0096] The MISH activation function is a new type of activation function that maps the input in a smooth manner to improve the classification accuracy and convergence speed of the model;
[0097] The dimension expansion layer is a layer in the deep learning model that increases the dimension of features to enhance the model's expressiveness and classification effect.
[0098] The working principle of the classification judgment model is:
[0099] The classification judgment model can be divided into three parts: feature separation layer, iterative training layer and recognition classification layer.
[0100] First, the feature separation layer performs nonlinear processing on the historical comprehensive feature vector to obtain the historical feature matrix. This process is mainly used for feature preprocessing and extraction to obtain high-dimensional feature expressions of the data. Among them, nonlinear processing can be performed using activation functions, such as the common ReLU function.
[0101] Then, the historical feature matrix is input into the iterative training layer, and several iterative trainings are performed through the multi-scale attention mechanism to obtain the feature weight matrix. The iterative training layer is mainly composed of multiple convolutional layers, pooling layers, residual blocks, global pooling layers, activation functions, normalization layers, and expansion layers. These layers perform multi-layer processing and information extraction on the input data through different parameter adjustments and connection methods to achieve feature optimization and abstraction. The multi-scale attention mechanism of the iterative training layer means that convolution kernels at different levels can focus on features of different dimensions to achieve more accurate representation.
[0102] Finally, the feature weight matrix is optimized and updated, and input to the recognition and classification layer for classification judgment. The recognition and classification layer is mainly composed of convolutional layers, Inception blocks, upsampling layers, average pooling layers, and activation functions. Through different levels and modules, the features are abstracted more complexly and comprehensively, and the classification results are finally output. Among them, the Inception block is a special convolutional neural network block with better feature extraction and processing capabilities. In addition, the upsampling layer and average pooling layer can make the feature resolution higher, which helps to better extract detail information. The MISH activation function can improve the classification accuracy of the model.
[0103] In summary, the classification judgment model adopts multiple levels and technologies, and combines algorithms such as nonlinear processing, attention mechanism, residual and Inception network to achieve good classification effect and generalization performance.
[0104] C: Optimize the data load of the data packet using the messy information database to reduce the false positives of the classification judgment results; the messy information database includes the protocol header and special fields;
[0105]
[0106] Among them, D i ″ represents the optimized data load, and known_fields() represents the messy information database;
[0107] A verification unit, used to re-verify the data packets that have completed the classification judgment;
[0108] In the present invention, the verification unit uses an advanced statistical algorithm to perform secondary verification on the data packets whose classification judgment results are plain text: if KS_statistic(D i ″ )≤F t , then the data packet is plain text; if KS_statistic(D i " )>F t , then the data packet is ciphertext; KS_statistic() represents the Kolmogorov-Smirnov test algorithm, F t Indicates feature threshold; advanced statistical algorithms include but are not limited to Kolmogorov-Smirnov test algorithm;
[0109] In this embodiment, the Kolmogorov-Smirnov test is a statistical test method used to compare whether there is a significant difference in the distribution of two samples, and is often used in data analysis and verification.
[0110] Example:
[0111] When the plaintext detection alarm system is used for plaintext data transmission monitoring alarm in internal and external communications of an enterprise: during the data transmission process, first, the acquisition unit continuously samples each data packet set passing through the gateway and / or router; the data packet set includes several data packets, each of which contains a data payload and a protocol header; the processing unit extracts the data payload and protocol header of each data packet, and then uses the protocol type to identify and remove the special fields of the data payload to obtain the common data payload, and calculates and obtains the multi-dimensional features of the common data payload: entropy value, byte frequency distribution, data length and timing features;
[0112] Then, the measurement unit performs fast Fourier transform on the common data load to obtain the frequency domain characteristics of the common data load; and fuses the multi-dimensional characteristics and frequency domain characteristics into a comprehensive feature vector; then, the judgment unit uses the data set constructed by the historical comprehensive feature vector and the machine learning algorithm to iteratively train the classification judgment model; then, the comprehensive feature vector of each data packet is input into the classification judgment model and compared with the entropy threshold H t Compare to obtain the classification judgment result: plain text or ciphertext; at the same time, optimize the data load of the data packet by using the messy information database to reduce the false positive of the classification judgment result; the messy information database includes the protocol header and special fields;
[0113] Finally, the verification unit uses an advanced statistical algorithm to perform a second verification on the data packets whose classification results are plain text: i " )≤F t , then the data packet is plain text; if KS_statistic(D i ″ )>F t , then the data packet is ciphertext; KS_statistic() represents the Kolmogorov-Smirnov test algorithm, F t Represents the feature threshold.
Claims
1. A plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform, characterized by: It includes a collection unit, a processing unit, a measurement unit, a judgment unit and a testing unit; wherein, A collection unit, used for collecting data packets during data transmission; A processing unit, used for preprocessing the collected data packets to obtain multi-dimensional features of the common data payload with special fields of the data payload removed; A calculation unit, used for performing multi-dimensional feature fusion and Fourier transformation on the pre-processed data packets to obtain a comprehensive feature vector; A judgment unit, used to classify and judge the comprehensive feature vector using a preset classification judgment model, and to optimize the data packet using a messy information database; The verification unit is used to re-verify the data packets that have completed the classification judgment.
2. The plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform according to claim 1 is characterized in that: The acquisition unit continuously samples the data packet set during the data transmission process; The data transmission process includes but is not limited to a gateway and / or a router; the data packet set includes a plurality of data packets, each of which includes a data payload and a protocol header.
3. The plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform according to claim 1 is characterized in that: The process of the processing unit preprocessing the data packet is as follows: S1: extract the data payload and protocol header of each data packet; P i =D i ⊙H i (1) Where i represents the packet number, P i represents the ith data packet, D i represents the data load of the ith data packet, ⊙ represents the Hadamard outer product, H i Represents the protocol header of the i-th data packet; S2: Use the protocol type to identify and remove the special fields of the data payload to obtain the normal data payload; Among them, D′ i represents the normal data payload of the ith data packet, special_fields() represents the special field removal function, T i Indicates the protocol type of the i-th data packet, represents the Hadamard inner product; S3: Calculate and obtain multi-dimensional features of common data load; multi-dimensional features include but are not limited to entropy value H(D i ′ ), byte frequency distribution F(D i ′ )、Data length L(D i ′ ) and the time series characteristics T(D i ′ ).
4. The plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform according to claim 1 is characterized in that: The process of the measurement unit calculating the pre-processed data packets is as follows: First, a fast Fourier transform is performed on the common data payload to obtain the frequency domain characteristics of the common data payload; F f (D′ i )=FFT(D′ i ) (3) Among them, F f (D′ i ) represents the frequency domain characteristics of ordinary data payload, and FFT() represents the fast Fourier function; Then, the multi-dimensional features and frequency domain features are fused into a comprehensive feature vector; V i =[H(D′ i ),F(D′ i ),L(D′ i ),T(D′ i ),F f (D′ i )] (4) Among them, V ′ represents the comprehensive feature vector.
5. The plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform according to claim 1 is characterized in that: The process of the judgment unit classifying and judging the comprehensive feature vector is as follows: A: Use the data set constructed from historical comprehensive feature vectors and machine learning algorithms to iteratively train the classification judgment model; M=TrainModel(V1,V2,…,V n ) (5) Among them, M represents the classification judgment model, TrainModel() represents the machine learning algorithm, and V n represents the nth historical comprehensive feature vector; B: Input the comprehensive feature vector of each data packet into the classification judgment model and compare it with the entropy threshold H t Compare to obtain the classification result; the classification result is plain text or cipher text; Result(P i )=M(V i ) (6) Among them, Result() represents the classification judgment result; C: Optimize the data load of the data packet using the messy information database to reduce the false positives of the classification judgment results; the messy information database includes the protocol header and special fields; Among them, D i ″ Represents the optimized data load, and known_fields() represents a messy information database.
6. The plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform according to claim 5 is characterized in that: The classification judgment model includes a feature separation layer, an iterative training layer and a recognition classification layer; the feature separation layer first performs nonlinear processing on the historical comprehensive feature vector to obtain a historical feature matrix; Among them, MP n represents the historical feature matrix, SBox[] represents the substitution box function, ABox[] represents the permutation box function, Represents vector and, x (k,n) represents the chaotic sequence of the nth item and the kth dimension; Then the historical feature matrix is input into the iterative training layer, and the multi-scale attention mechanism is used to perform several iterative trainings to obtain the feature weight matrix; the iterative training layer includes 4 3X3 convolutional layers, 2 3X3 maximum pooling layers, 6 residual blocks of Inception V3 blocks in parallel, 3 global pooling layers, 2 dimension expansion layers, 3 Sigmoid activation functions, 3 normalization layers and 3 5X5 convolutional layers; the 4 3X3 convolutional layers are connected in parallel with the 3 5X5 convolutional layers and 3 3X3 maximum pooling layers in series, and are connected in series with 2 normalization layers; Finally, the recognition and classification layer is used to optimize and update the feature weight matrix to obtain the classification judgment result; the recognition and classification layer includes 4 Inception V1 blocks, 3 Inception V4 blocks, 2 5X5 upsampling layers, 2 3X3 average pooling layers and 2 MISH activation functions.
7. The plaintext detection alarm system based on multi-dimensional feature fusion and Fourier transform according to claim 1 is characterized in that: The verification unit uses an advanced statistical algorithm to perform a secondary verification on the data packet whose classification result is plain text: if KS_statistic(D″ i )≤F t , then the data packet is plaintext; if KS_statistic(D″ i )>F t , then the data packet is ciphertext; KS_statistic() represents the Kolmogorov-Smirnov test algorithm, F t represents the feature threshold; advanced statistical algorithms include but are not limited to the Kolmogorov-Smirnov test algorithm.
Citation Information
Patent Citations
Encrypted traffic network threat key node automatic extraction method based on GCN-DL
CN114172688A
Content type identification method and device of traffic data, equipment and storage medium
CN115225583A
Malicious encrypted traffic detection method and system based on multi-level attention mechanism
CN115987599A
Network traffic classification model training method, classification method and training device
CN117076923A
Unsupervised time sequence anomaly detection method based on multi-dimensional feature fusion
CN117556311A
