CN115225584A - Encrypted traffic classification method and system based on graph neural network - Google Patents
Encrypted traffic classification method and system based on graph neural network Download PDFInfo
- Publication number
- CN115225584A CN115225584A CN202210878574.3A CN202210878574A CN115225584A CN 115225584 A CN115225584 A CN 115225584A CN 202210878574 A CN202210878574 A CN 202210878574A CN 115225584 A CN115225584 A CN 115225584A
- Authority
- CN
- China
- Prior art keywords
- graph
- encrypted
- traffic
- flow
- classification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an encrypted traffic classification method and system based on a graph neural network, and relates to the technical field of encrypted traffic classification, wherein firstly, an encrypted traffic data packet sequence to be classified is captured and converted into a traffic interaction graph; and constructing a graph classification model based on a graph neural network, performing operations such as pooling on the topological structure of the flow interaction graph according to the importance parameter, and the like, and finally classifying the flow interaction graph to obtain a flow interaction graph classification result, namely an encrypted flow classification result. The invention converts the encrypted flow classification problem into the graph classification problem, fully considers the characteristics of the nodes and the topological structure of the graph, reserves the appearance and the characteristics of the original data stream, and has high accuracy, high speed and certain generalization function.
Description
Technical Field
The invention relates to the technical field of encrypted traffic classification, in particular to an encrypted traffic classification method and system based on a graph neural network.
Background
In recent years, while computer network technology is developing, the awareness of security and privacy of network users is also increasing, and the awareness of data protection is becoming stronger. In order to protect the privacy of users and improve the network security, network traffic needs to be encrypted and transmitted in the communication process, and browser manufacturers such as google start to force websites to use security transmission protocols such as TLS, so that the privacy security of people is protected to a certain extent.
With the widespread use of Secure Sockets Layer (SSL), secure shell protocol (SSH), virtual Private Network (VPN), and anonymous communication (Tor), encryption traffic is growing rapidly. Network traffic encryption techniques also present unprecedented challenges while protecting network data security. The encryption technology can be utilized by a network attacker to carry out hidden malicious activities, the traditional classifier is difficult to identify the malicious TLS or SSL encrypted traffic data, and even a firewall cannot intercept the malicious TLS or SSL encrypted traffic data, so that great network safety hidden dangers can be brought. Therefore, for reasons of improving network management, improving network service quality and privacy, etc., it is important to classify encrypted network traffic without decrypting the encrypted traffic.
Network traffic identification is to classify network traffic into different sets by observing features, such as encryption protocol type, application, service type, and abnormal encryption traffic, etc., for a specific target. With the rapid development of the internet, network protocols become more and more complex, so that the network traffic based on the protocols is increased sharply, and how to realize the effective identification of the network protocols becomes an important problem for the network development. The network flow identification is the research focus in the fields of network behavior analysis, network flow planning, network anomaly detection and electronic marketing, and the high-accuracy identification and detection of encrypted flow have important practical significance for ensuring the network information safety and maintaining the normal, stable and reliable operation of the network.
At present, the network traffic classification technology mainly includes a port-based method, a Deep Packet Inspection (DPI) -based method, a conventional machine learning method, and a deep learning method. Among these, early port-based methods are less accurate; because the traffic characteristics change after encryption, the encrypted traffic cannot be identified based on a deep packet inspection method, and the calculation complexity is high; in recent years, most of researches are methods based on machine learning, a group of flow characteristic sets are designed firstly, then modeling and training are carried out on the group of flow characteristic sets, and a trained model can distinguish and classify new flow, but the traditional machine learning method needs to manually select characteristics to fit the model from historical data for classification; the deep learning method can automatically learn high-level features from input original data, overcomes the difficulty of feature design of traditional machine learning, is limited by an inherent neural network architecture, and can only input data with a fixed shape.
In summary, a need exists in the art for a method for classifying encrypted traffic that supports the input of any number of data packets, retains original stream data information, and ensures a high classification accuracy.
Disclosure of Invention
In view of this, the present invention provides a method and a system for classifying encrypted traffic based on a graph neural network, which construct a traffic interaction graph and classify the encrypted traffic to be classified according to a protocol by using the graph neural network.
In order to achieve the above purpose, the invention provides the following technical scheme:
a method for classifying encrypted traffic based on a graph neural network comprises the following steps:
step 1, capturing a flow data packet sequence to be classified and encrypted;
step 2, converting the encrypted flow data packet sequence to be classified into a flow interaction diagram;
and 3, classifying the traffic interaction graph by using a graph classification model based on a graph neural network to obtain an encrypted traffic classification result.
Optionally, in step 2, the traffic interaction graph is represented by a binary group G = (V, E), where V represents a vertex set, and each vertex V ∈ V represents the size of a data packet; and E represents an edge set, each burst edge E belongs to E and represents the relationship between data packets, and the burst edges have two types, namely burst inner edges connecting two continuous vertexes in a burst and burst inter edges connecting the starting or ending vertexes of the two continuous bursts.
Optionally, in step 2, the method for converting the encrypted traffic data packet sequence to be classified into a traffic interaction graph includes:
step 2.1, initializing a vertex set and an edge set of the flow interaction graph;
step 2.2, adding a vertex to a vertex set according to the encrypted flow data packet sequence to be classified;
step 2.3, iteratively adding burst inner edges and burst inter edges between vertexes according to the encrypted flow data packet sequence to be classified;
and 2.4, outputting to obtain a flow interaction diagram.
Optionally, in step 3, the graph classification model based on the graph neural network includes three graph convolution layers, a splicing layer, a graph pooling layer, a readout layer, a full connection layer, and a classification layer, which are connected in sequence.
Optionally, in step 3, the method for classifying the flow interaction graph by using the graph classification model includes:
step 3.1, assigning an importance parameter to each vertex in the flow interaction graph by using a graph convolution layer in a graph convolution mode;
3.2, splicing the output results of the graph convolution layers by using the splicing layers;
3.3, pooling the topological structure of the flow interaction diagram by using the diagram pooling layer according to the importance parameter;
step 3.4, using the reading layer to perform aggregation operation on the whole situation, and aggregating the vertex features to reduce the dimension of the flow interaction diagram to a preset uniform dimension;
step 3.5, classifying according to the vertex characteristics of the flow interaction diagram by using a full connection layer;
and 3.6, outputting the classification result of the traffic interaction diagram by using a classification layer, namely the classification result of the encrypted traffic.
Optionally, in step 3.1, the calculation method of the importance parameter includes:
where, σ denotes the activation function,
indicating that the adjacency matrix of the self-join is added,
is composed of
Degree matrix of (c), X represents the characteristic of the node, Θ att ∈R N×1 Denotes a weight parameter, R N×1 And representing an N multiplied by 1 dimensional weight set, wherein N represents the number of data packets in the encrypted flow data packet sequence to be classified.
A graph neural network-based encrypted traffic classification system, comprising:
the data acquisition module is used for capturing the encrypted flow data packet sequence to be classified;
the conversion module is used for converting the encrypted flow data packet sequence to be classified into a flow interaction diagram;
and the classification module is used for classifying the traffic interaction graph by using a graph classification model based on a graph neural network to obtain an encrypted traffic classification result.
According to the technical scheme, the invention provides the encrypted traffic classification method and system based on the graph neural network, and compared with the prior art, the encrypted traffic classification method and system based on the graph neural network have the following beneficial effects:
(1) The invention converts the captured encrypted flow data packet sequence to be classified into a flow interaction graph and converts the encrypted flow classification problem into a graph classification problem.
(2) The graph classification model is constructed based on the self-attention mechanism, the characteristics of the nodes and the topological structure of the graph are fully considered, and the encrypted flow classification result is output through the graph classification model.
(3) The flow interaction graph constructed by the invention keeps the same amount of information of original flow data, including the size, the direction, the sequence and the like of a data packet, and keeps the appearance and the characteristics of the original flow data.
(4) The invention uses end-to-end deep learning without feature selection.
(5) The scheme of the invention has high accuracy and high speed, does not need to detect the content of the data packet, is also effective to unknown protocols and has certain generalization function.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method for classifying encrypted traffic according to the present invention;
FIG. 2 is a schematic diagram illustrating interaction between two ends of a sequence of encrypted traffic packets to be classified according to an embodiment;
FIG. 3 is a flow chart of a method for constructing a traffic interaction graph according to the present invention;
fig. 4 is a flow interaction diagram corresponding to a sequence of encrypted flow packets to be classified in an embodiment;
FIG. 5 is a diagram classification model schematic diagram based on a diagram neural network.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses an encryption traffic classification method based on a graph neural network, which is shown in figure 1 and comprises the following steps:
step 1, capturing an encrypted flow data packet sequence to be classified.
And 2, converting the encrypted traffic data packet sequence to be classified into a traffic interaction diagram.
The traffic interaction graph is represented by a binary group G = (V, E), wherein V represents a vertex set, and each vertex V belongs to V and represents the size of a data packet; and E represents an edge set, each burst edge E belongs to E and represents the relationship between data packets, and the burst edges have two types, namely burst inner edges connecting two continuous vertexes in a burst and burst inter edges connecting the starting or ending vertexes of the two continuous bursts.
Referring to fig. 3, the method for converting the encrypted traffic data packet sequence to be classified into a traffic interaction graph specifically includes the following steps:
step 2.1, initializing a vertex set and an edge set of the flow interaction graph to be null;
step 2.2, adding a vertex to a vertex set according to the encrypted flow data packet sequence to be classified;
step 2.3, according to the encrypted flow data packet sequence to be classified, iteratively adding burst inner edges and burst inter edges between vertexes;
and 2.4, outputting to obtain a flow interaction diagram.
Referring to fig. 2, a schematic diagram of interaction between a client and a server of an encrypted traffic data packet sequence to be classified according to a specific embodiment is shown, and fig. 4 is a traffic interaction diagram constructed according to the above method.
And 3, classifying the traffic interaction graph by using a graph classification model based on a graph neural network to obtain an encrypted traffic classification result.
And constructing a graph classification model based on a graph neural network, and referring to fig. 5, wherein the graph classification model comprises three graph convolution layers, a splicing layer, a graph pooling layer, a reading layer, a full-connection layer and a classification layer which are sequentially connected.
The method for classifying the flow interaction graph by using the graph classification model comprises the following steps:
step 3.1, assigning an importance parameter to each vertex in the flow interaction graph by using the graph convolution layer in a graph convolution mode, wherein the calculation method of the importance parameter comprises the following steps:
where, σ denotes the activation function,
indicating that the adjacency matrix of the self-join is added,
is composed of
Degree matrix of (c), X represents the characteristic of the node, Θ att ∈R N×1 Denotes a weight parameter, R N×1 Representing a weight set with dimension of Nx 1, wherein N represents the number of data packets in the encrypted flow data packet sequence to be classified;
3.2, splicing the output results of the three graph volume layers by using a splicing layer;
3.3, pooling the topological structure of the flow interaction graph by using a graph pooling layer according to the importance parameter based on a self-attention mechanism, discarding part of unimportant nodes, and updating the adjacent matrix and the node characteristics to obtain a pooling result;
step 3.4, using the reading layer to perform aggregation operation on the whole situation, and aggregating the vertex features to reduce the dimension of the flow interaction diagram to a preset uniform dimension;
step 3.5, classifying by using a full connection layer according to the vertex characteristics of the flow interaction graph;
and 3.6, outputting the classification result of the traffic interaction diagram by using a classification layer, namely the classification result of the encrypted traffic.
In another embodiment, there is provided a graph neural network-based encrypted traffic classification system, including:
the data acquisition module is used for capturing the encrypted flow data packet sequence to be classified;
the conversion module is used for converting the encrypted flow data packet sequence to be classified into a flow interaction diagram;
and the classification module is used for classifying the traffic interaction graph by using a graph classification model based on a graph neural network to obtain an encrypted traffic classification result.
For the system module disclosed by the embodiment, the description is relatively simple because the system module corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
Experiments prove that the method provided by the invention can accurately classify the encrypted flow of SSL, SSH and QUIC protocols, the average accuracy rate is 99.516%, the average accuracy rate is 99.521%, the average recall rate is 99.516%, and the average F1 value is 99.516%. Compared with other methods in the prior art, the method has the advantages of high accuracy and high speed, does not need to detect the content of the data packet, is effective to unknown protocols, and has a certain generalization effect.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. An encryption traffic classification method based on a graph neural network is characterized by comprising the following steps:
step 1, capturing an encrypted flow data packet sequence to be classified;
step 2, converting the encrypted flow data packet sequence to be classified into a flow interaction diagram;
and 3, classifying the traffic interaction graph by using a graph classification model based on a graph neural network to obtain an encrypted traffic classification result.
2. The method for classifying the encrypted traffic based on the graph neural network as claimed in claim 1, wherein in the step 2, the traffic interaction graph is represented by using a binary group G = (V, E), wherein V represents a set of vertices, and each vertex V ∈ V represents the size of a data packet; e represents an edge set, each burst edge E belongs to E and represents the relation between data packets, and the burst edges have two types, namely burst inner edges connecting two continuous vertexes in a burst and burst inter edges connecting the starting or ending vertexes of the two continuous bursts.
3. The method for classifying encrypted traffic based on the graph neural network according to claim 2, wherein in the step 2, the method for converting the sequence of encrypted traffic data packets to be classified into the traffic interaction graph comprises:
step 2.1, initializing a vertex set and an edge set of the flow interaction graph;
step 2.2, adding a vertex to a vertex set according to the encrypted flow data packet sequence to be classified;
step 2.3, iteratively adding burst inner edges and burst inter edges between vertexes according to the encrypted flow data packet sequence to be classified;
and 2.4, outputting to obtain a flow interaction diagram.
4. The method according to claim 1, wherein in the step 3, the graph classification model based on the graph neural network comprises three graph convolution layers, a splicing layer, a graph pooling layer, a reading layer, a full connection layer and a classification layer which are connected in sequence.
5. The method for classifying encrypted traffic based on the graph neural network according to claim 4, wherein in the step 3, the method for classifying the traffic interaction graph by using the graph classification model is as follows:
step 3.1, endowing each vertex in the flow interaction graph with an importance parameter by using a graph convolution layer in a graph convolution mode;
3.2, splicing the output results of the graph convolution layers by using the splicing layers;
3.3, pooling the topological structure of the flow interaction graph by using a graph pooling layer according to the importance parameter;
step 3.4, using the reading layer to perform aggregation operation on the whole situation, and aggregating the vertex features to reduce the dimension of the flow interaction diagram to a preset uniform dimension;
step 3.5, classifying according to the vertex characteristics of the flow interaction diagram by using a full connection layer;
and 3.6, outputting a traffic interaction diagram classification result by using a classification layer, namely an encrypted traffic classification result.
6. The method for classifying encrypted traffic based on the neural network of the graph as claimed in claim 5, wherein in the step 3.1, the calculation method of the importance parameter is as follows:
where, σ denotes the activation function,
indicating that the adjacency matrix of the self-join is added,
is composed of
Degree matrix of (a), X represents the characteristic of the node, theta att ∈R N×1 Denotes a weight parameter, R N×1 And representing a weight set with dimension of Nx 1, wherein N represents the number of data packets in the encrypted flow data packet sequence to be classified.
7. An encrypted traffic classification system based on a graph neural network, comprising:
the data acquisition module is used for capturing the encrypted flow data packet sequence to be classified;
the conversion module is used for converting the encrypted flow data packet sequence to be classified into a flow interaction diagram;
and the classification module is used for classifying the traffic interaction graph by using a graph classification model based on a graph neural network to obtain an encrypted traffic classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210878574.3A CN115225584A (en) | 2022-07-25 | 2022-07-25 | Encrypted traffic classification method and system based on graph neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210878574.3A CN115225584A (en) | 2022-07-25 | 2022-07-25 | Encrypted traffic classification method and system based on graph neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115225584A true CN115225584A (en) | 2022-10-21 |
Family
ID=83614170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210878574.3A Pending CN115225584A (en) | 2022-07-25 | 2022-07-25 | Encrypted traffic classification method and system based on graph neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115225584A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112217834A (en) * | 2020-10-21 | 2021-01-12 | 北京理工大学 | Internet encryption flow interactive feature extraction method based on graph structure |
| CN112468324A (en) * | 2020-11-11 | 2021-03-09 | 国网冀北电力有限公司信息通信分公司 | Graph convolution neural network-based encrypted traffic classification method and device |
| CN112862015A (en) * | 2021-04-01 | 2021-05-28 | 北京理工大学 | Paper classification method and system based on hypergraph neural network |
| CN114389966A (en) * | 2022-03-24 | 2022-04-22 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Network traffic identification method and system based on graph neural network and stream space-time correlation |
-
2022
- 2022-07-25 CN CN202210878574.3A patent/CN115225584A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112217834A (en) * | 2020-10-21 | 2021-01-12 | 北京理工大学 | Internet encryption flow interactive feature extraction method based on graph structure |
| CN112468324A (en) * | 2020-11-11 | 2021-03-09 | 国网冀北电力有限公司信息通信分公司 | Graph convolution neural network-based encrypted traffic classification method and device |
| CN112862015A (en) * | 2021-04-01 | 2021-05-28 | 北京理工大学 | Paper classification method and system based on hypergraph neural network |
| CN114389966A (en) * | 2022-03-24 | 2022-04-22 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Network traffic identification method and system based on graph neural network and stream space-time correlation |
Non-Patent Citations (3)
| Title |
|---|
| BOYU SUN 等: "An Encrypted Traffic Classification Method Combining Graph Convolutional Network and Autoencoder", 2020 IEEE 39TH INTERNATIONAL PERFORMANCE COMPUTING AND COMMUNICATIONS CONFERENCE (IPCCC), pages 3 - 4 * |
| 冯翔 等: "基于嵌网融合结构的卷积神经网络手势图像识别方法", 生物医学工程研究 * |
| 王勤凡 等: "一种基于图卷积神经网络的加密流量分类方法", 电子测量技术, vol. 45, no. 14 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Towards automatic fingerprinting of IoT devices in the cyberspace | |
| CN111865815B (en) | Flow classification method and system based on federal learning | |
| Shen et al. | Machine learning-powered encrypted network traffic analysis: A comprehensive survey | |
| CN109450721B (en) | Network abnormal behavior identification method based on deep neural network | |
| CN113989583A (en) | Method and system for detecting malicious traffic of internet | |
| He et al. | Deep‐Feature‐Based Autoencoder Network for Few‐Shot Malicious Traffic Detection | |
| CN114422211B (en) | HTTP malicious traffic detection method and device based on graph attention network | |
| CN114172688B (en) | Method for automatically extracting key nodes of network threat of encrypted traffic based on GCN-DL (generalized traffic channel-DL) | |
| CN113821793B (en) | Multi-stage attack scene construction method and system based on graph convolution neural network | |
| CN112333706A (en) | Internet of things equipment anomaly detection method and device, computing equipment and storage medium | |
| Amru et al. | Network intrusion detection system by applying ensemble model for smart home. | |
| CN112491894A (en) | Internet of things network attack flow monitoring system based on space-time feature learning | |
| Yusof et al. | Visualizing realistic benchmarked IDS dataset: CIRA-CIC-DoHBrw-2020 | |
| Hameed et al. | IOTA‐Based Mobile Crowd Sensing: Detection of Fake Sensing Using Logit‐Boosted Machine Learning Algorithms | |
| CN114362988B (en) | Network traffic identification method and device | |
| Pham et al. | Lightweight Convolutional Neural Network Based Intrusion Detection System. | |
| Yin et al. | Identifying iot devices based on spatial and temporal features from network traffic | |
| Ma et al. | GraphNEI: A GNN-based network entity identification method for IP geolocation | |
| Yang et al. | Fingerprinting Industrial IoT devices based on multi-branch neural network | |
| Zhou et al. | Malicious encrypted traffic features extraction model based on unsupervised feature adaptive learning | |
| CN115225584A (en) | Encrypted traffic classification method and system based on graph neural network | |
| Xu et al. | Trafficgcn: Mobile application encrypted traffic classification based on gcn | |
| Liu et al. | Spatial‐Temporal Feature with Dual‐Attention Mechanism for Encrypted Malicious Traffic Detection | |
| CN116760569A (en) | Internet of things intrusion detection method based on graph neural network model | |
| Dener et al. | RFSE-GRU: Data balanced classification model for mobile encrypted traffic in big data environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |