CN116015771A - Detection method for malicious nodes of Internet of things - Google Patents
Detection method for malicious nodes of Internet of things Download PDFInfo
- Publication number
- CN116015771A CN116015771A CN202211589796.XA CN202211589796A CN116015771A CN 116015771 A CN116015771 A CN 116015771A CN 202211589796 A CN202211589796 A CN 202211589796A CN 116015771 A CN116015771 A CN 116015771A
- Authority
- CN
- China
- Prior art keywords
- node
- generation
- network
- malicious
- vector
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title abstract description 7
- 239000013598 vector Substances 0.000 claims abstract description 73
- 230000006399 behavior Effects 0.000 claims abstract description 23
- 238000013145 classification model Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000008485 antagonism Effects 0.000 claims abstract description 14
- 238000003064 k means clustering Methods 0.000 claims abstract description 11
- 230000005540 biological transmission Effects 0.000 claims description 13
- 238000000034 method Methods 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 8
- 230000001934 delay Effects 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 230000003111 delayed effect Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 230000003042 antagnostic effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention belongs to the technical field of information security detection, and relates to a detection method of malicious nodes of the Internet of things, which comprises the following steps: acquiring behavior data of malicious nodes in an Internet of things network; embedding behavior data of the malicious node into a vector representation to obtain a trust vector of the malicious node; creating a node classification model; training the node classification model by taking the trust vector of the malicious node as a training sample; the node classification model includes: the system comprises a first generation antagonism network, a second generation antagonism network, a third generation antagonism network and a K-MEANS clustering module; acquiring behavior data of a target node in an Internet of things network, generating a trust vector of the target node, inputting the trust vector of the target node into a node classification model, and outputting the type of the target node; managing the target node according to the type of the target node, wherein the managing the target node comprises the following steps: and removing the target node from the internet of things network, repairing the target node or replacing the target node.
Description
Technical Field
The invention belongs to the technical field of information security detection, and particularly relates to a detection method of malicious nodes of the Internet of things.
Background
With the continuous development and application of new technologies such as artificial intelligence and big data, the internet of things is widely applied to the sky, the air, the land and the sea, but also faces a plurality of problems and challenges, wherein the security problem is the first to go. The internet of things can collect and transmit data in real time, but due to the bad working environment and lack of special maintenance, the problems of fragile link, limited node energy, low transmission efficiency and the like exist, so that hardware nodes in the internet of things are captured by external nodes and converted into internal malicious nodes to damage the internet of things besides being threatened by external attacks such as replay attacks and denial of service, for example, when the nodes transmit data, the malicious nodes deliberately lose data packets or delay forwarding the data packets. Most attacks that cause the Internet of things to crash are initiated by internal malicious nodes. The malicious nodes hidden in the network consume the whole energy of the network, reduce the throughput and the service life of the Internet of things network, and cause immeasurable loss. Therefore, how to detect and classify these malicious nodes in the internet of things is an urgent problem to be solved in the field of internet of things security.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides a detection method of malicious nodes of the Internet of things, which comprises the following steps:
s1: acquiring behavior data of malicious nodes in an Internet of things network;
s2: embedding behavior data of the malicious node into a vector representation to obtain a trust vector of the malicious node;
s3: creating a node classification model; training the node classification model by taking the trust vector of the malicious node as a training sample; the node classification model includes: the system comprises a first generation antagonism network, a second generation antagonism network, a third generation antagonism network and a K-MEANS clustering module;
s4: acquiring behavior data of a target node in an Internet of things network, generating a trust vector of the target node, inputting the trust vector of the target node into a node classification model, and outputting the type of the target node;
s5: managing the target node according to the type of the target node, wherein the managing the target node comprises the following steps: and removing the target node from the internet of things network, repairing the target node or replacing the target node.
The invention has at least the following beneficial effects
According to the method, behavior data of part of malicious nodes in the Internet of things network are obtained, a node classification model is trained based on the behavior data of the malicious nodes, the first generation countermeasures network, the second generation countermeasures network and the third generation countermeasures network are adopted to generate generation vectors of the malicious nodes, and the malicious nodes are classified by using a K-MENAS algorithm according to the generation vectors of the malicious nodes; acquiring behavior data of a target node, inputting the behavior data of the target node into a node classification model, and rejecting, repairing or replacing the node by adopting different management modes according to the category of the node; according to the method and the device for detecting the malicious nodes, the malicious nodes can be timely found and managed correspondingly, the safety of the Internet of things network is improved, the energy consumption of the Internet of things network is reduced, and the transmission speed of the data packet is improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed description of the preferred embodiments
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the illustrations provided in the following embodiments merely illustrate the basic idea of the present invention by way of illustration, and the following embodiments and features in the embodiments may be combined with each other without conflict.
Referring to fig. 1, the invention provides a method for detecting malicious nodes of the internet of things, which comprises the following steps: ,
s1: acquiring behavior data of malicious nodes in an Internet of things network; wherein, malicious node in the thing networking network includes: hardware nodes in the Internet of things network have the probability of losing data packets and delaying forwarding the data packets in the data transmission process of more than 30 percent; the behavior data of the malicious node includes: the malicious node respectively loses the data packet and delays forwarding the data packet in N times of data transmission; if a malicious node loses a data packet in the ith data transmission process, R i =1 otherwise 0; if the malicious node delays forwarding the data packet in the ith data transmission process, X i =1 otherwise 0; for example, when N is equal to 10, then the behavior data of the malicious node includes: RX= { R 1 X 1 ,…,R i X i ,…R 10 X 10 I.e. r= {0,1,1,0,1,1,0,0,0,1} and x= {0,1,0,1,1,0,0,1,0,1}, then rx= {0,1,0,0,1,0,0,0,0,1}, where the value of N can be set randomly according to the person skilled in the art.
The hardware nodes in the internet of things network comprise, but are not limited to, intelligent devices such as computers, mobile phones, bluetooth headsets, unmanned aerial vehicles, unmanned ships, wiFi, intelligent voice, sound boxes, intelligent pens and the like.
The main work of the nodes in the Internet of things network is to collect data and forward the data, so that the behavior data selects two indexes of whether the nodes lose data packets or not and whether the nodes delay forwarding the data packets or not.
The data behavior data statistics of the malicious node can be performed by using the monitoring node, namely, when the malicious node performs data transmission with other nodes, the nodes in the communication range of the other nodes can be used as the monitoring node, when the monitoring node receives the data transmission request of the other nodes, the other nodes are considered to send data packets to the malicious node, if the malicious node does not have the corresponding data packets finally, the malicious node performs packet loss behavior, and whether the malicious node delays forwarding the data packets can be calculated similarly.
S2: embedding behavior data of the malicious node into a vector representation to obtain a trust vector of the malicious node;
generating an antagonizing network:
generating the antagonism network includes: a generator and a arbiter; randomly inputting a noise z to a generator G, and generating a generated data G (z) according to the random noise z by the generator; the discriminator discriminates the generated data G (z) according to errors of the real data and the generated data G (z) and updates parameters of the discriminator; the generator updates the parameters of the generator according to the discrimination result of the discriminator on the generated data G (z).
For the first generation countermeasure network and the second generation countermeasure network, the real data is white noise conforming to Gaussian distribution, and when the specific setting is carried out, the real data is obtained by training to gradually approach the real space distribution through regular Gaussian white noise because the space distribution is not known at first, which is the basic knowledge of the generation countermeasure network. Not further described in the present invention, for the third generation countermeasure network, the true data is the trust vector that was originally input to the malicious node.
If the real data X is true, the expected discriminator can score the real data as approaching or reaching 1; if data G (Z) is generated, it is desirable for the generator to score the discriminator as close to or as high as 1, while it is desirable for the discriminator to score the generated data as close to or as high as 0 (all of the scoring intervals are 0-1). The optimization formula of GAN is as follows:
wherein,,indicating that the expected x is taken from the Pdata distribution; x represents real data, pdata represents distribution of real data,/->Indicating that the desired z is taken from the P (z) distribution; z represents the generated data, and Pz (z) represents the distribution of the generated data.
S3: creating a node classification model; training the node classification model by taking the trust vector of the malicious node as a training sample; the node classification model includes: the system comprises a first generation antagonism network, a second generation antagonism network, a third generation antagonism network and a K-MEANS clustering module;
preferably, the training the node classification model using the trust vector of the malicious node as the training sample includes:
s31: the trust vector of the malicious node is respectively input into a first generation vector and a second generation vector of the malicious node generated by the first generation countercheck network and the second generation countercheck network, and parameters of a generator of the first generation countercheck network are updated through a back propagation mechanism according to a discrimination result of the first generation vector by a discriminator of the first generation countercheck network; updating parameters of a generator of the second generation countermeasure network through a back propagation mechanism according to a discrimination result of the second generation vector by a discriminator of the second generation countermeasure network;
the first generation reactance network maps trust vectors to intervals of (-1, 1) through an activation function; the second generation countermeasure network maps trust vectors to intervals of (0, 10) by activating functions;
the generator of the first generation reactance network comprises an input layer, a convolution layer 1 (1 x 10), a maximum pooling layer 1 (parameter 20 x 10), a convolution layer 2 (10 x 10), a pooling layer 2 (parameter 10 x 5), a full connection layer 1 (parameter 250), a full connection layer 2 (parameter 25), an activation function (tanh activation function,
the generator of the second generation countermeasure network comprises an input layer, a convolution layer 1 (1 x 10) and a maximum pooling layer 1 (parameter 20 x 10) which are connected in sequence, convolutional layer 2 (10×10), pooling layer 2 (parameter 10×5×5), fully connected layer 1 (parameter 250), fully connected layer 2 (parameter 25), activation function (ReLu activation function, y=max (0, x)).
The generation of two different output characteristic dimensions is used for acquiring different spatial characteristics of trust vectors of malicious nodes and generating two different characteristic dimension generation vectors, when the generated trust vectors are subsequently input into a third generation countermeasure network for decoding, the third generation countermeasure network can obtain a richer characteristic space, errors of the generation vectors of the third generation countermeasure network and the trust vectors of the malicious nodes are reduced, and the behavior data characteristics of the malicious nodes are fully mastered by the third generation countermeasure network.
S32: the first generation vector and the second generation vector of the malicious node are subjected to logical OR operation and then input into a third generation vector for generating the malicious node against the network; updating parameters of generators of the first generation countermeasure network, the second generation countermeasure network and the third generation countermeasure network in a gradient descending mode according to a discrimination result of the third generation countermeasure network by a discriminator of the third generation countermeasure network, and repeating steps S31-S32 to complete training of the first generation countermeasure network, the second generation countermeasure network and the third generation countermeasure network;
the third generation countermeasure network generator includes an input layer, a full connection layer 1 (parameter 25), a full connection layer 2 (parameter 250), a deconvolution layer 1 (parameter 10 x 5), an up-sampling layer 1 (parameter 10 x 10), a deconvolution layer 2 (parameter 20 x 10), an up-sampling layer 2 (parameter 1 x 10), and an activation function (Sigmoid activation function,) The method comprises the steps of carrying out a first treatment on the surface of the According to the invention, a large number of simulation samples can be generated under limited real samples through the first generation countermeasure network, the second generation countermeasure network and the third generation countermeasure network, and the generated simulation samples can be restored to the vector with the same dimension as the trust vector through the third generation countermeasure network, so that the classification result is more accurate. />
S33: inputting a third generated vector of the malicious node into a K-MEANS clustering module, classifying the malicious node by using a K-MEANS algorithm, and completing training of the K-MEANS clustering module;
s331: randomly selecting third generated vectors of k malicious nodes as a clustering center;
s332: distributing the third generated vector of each malicious node to the nearest cluster center by using a Euclidean distance formula to obtain k clusters;
dist(c i ,x)
wherein x represents a third generated vector of the malicious node, c i Represents the ith cluster center, dist (c) i X) represents the euclidean distance of the third generated vector of the malicious node from the i-th cluster center.
S333: calculating the center points of K clusters by using the error square sum SSE (sumofthe Squared Error) as the cluster centers of the next iteration, and repeatedly executing the steps S331-S333 until the cluster centers are unchanged, wherein each cluster of the K-MEANS cluster module after training represents one class of malicious nodes (similar to a classifier);
wherein x represents a third generated vector of the malicious node, c i Represents the ith cluster center, dist (c) i X) represents the Euclidean distance between the third generation vector of the malicious node and the ith clustering center, R i Represents the ith cluster, m i Is the value obtained by the derivative root-finding formula, c i ' is the cluster center of the next iteration.
S4: acquiring behavior data of a target node in an Internet of things network, generating a trust vector of the target node, inputting the trust vector of the target node into a node classification model, and outputting the type of the target node;
preferably, outputting the trust vector of the target node to the node classification model comprises:
respectively inputting trust vectors of the target nodes into a first generation reactance network to calculate to obtain first generation vectors and second generation vectors of the target nodes; the first generated vector and the second generated vector of the target node are subjected to logical OR operation and then input into a third generated vector of the target node generated by the antagonism network; calculating an error between a third generated vector and a trust vector of the target node; when the error is larger than the set threshold, the target node is a normal node, and if the error is smaller than the set threshold, the target node is a malicious node, a third generated vector of the target node is input into a K-MEANS clustering module for classification, and the type of the target node is output;
s5: managing the target node according to the type of the target node, wherein the managing the target node comprises the following steps: and removing the target node from the internet of things network, repairing the target node or replacing the target node.
The target nodes classified by the K-MEANS clustering module are in K categories, wherein each category represents the severity of different lost data packets or delayed converted data packets of the malicious node, and management operations such as rejecting, repairing or replacing the data packets according to the severity of the lost data packets or delayed converted data packets of the malicious node reduce the resource consumption of the Internet of things network and improve the data transmission efficiency of the Internet of things network.
The above preferred embodiments are only for illustrating the technical solution of the present invention and not for limiting it, and it should be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the present invention as defined by the appended claims.
Claims (6)
1. The method for detecting the malicious node of the Internet of things is characterized by comprising the following steps:
s1: acquiring behavior data of malicious nodes in an Internet of things network;
s2: embedding behavior data of the malicious node into a vector representation to obtain a trust vector of the malicious node;
s3: creating a node classification model; training the node classification model by taking the trust vector of the malicious node as a training sample; the node classification model includes: the system comprises a first generation antagonism network, a second generation antagonism network, a third generation antagonism network and a K-MEANS clustering module;
s4: acquiring behavior data of a target node in an Internet of things network, generating a trust vector of the target node, inputting the trust vector of the target node into a node classification model, and outputting the type of the target node;
s5: managing the target node according to the type of the target node, wherein the managing the target node comprises the following steps: and removing the target node from the internet of things network, repairing the target node or replacing the target node.
2. The method for detecting a malicious node of the internet of things according to claim 1, wherein the malicious node in the internet of things comprises:
hardware nodes in the Internet of things network have the probability of losing data packets and delaying forwarding the data packets in the data transmission process of more than 30 percent; the behavior data of the malicious node includes: the malicious node respectively loses the data packet and delays forwarding the data packet in N times of data transmission; if a malicious node loses a data packet in the ith data transmission process, R i =1 otherwise 0; if the malicious node delays forwarding the data packet in the ith data transmission process, X i And vice versa, is 0.
3. The method for detecting the malicious node of the internet of things according to claim 1, wherein the training the trust vector of the malicious node as the training sample input node classification model comprises:
s31: the trust vector of the malicious node is respectively input into a first generation vector and a second generation vector of the malicious node generated by the first generation countercheck network and the second generation countercheck network, and parameters of a generator of the first generation countercheck network are updated through a back propagation mechanism according to a discrimination result of the first generation vector by a discriminator of the first generation countercheck network; updating parameters of a generator of the second generation countermeasure network through a back propagation mechanism according to a discrimination result of the second generation vector by a discriminator of the second generation countermeasure network;
s32: the first generation vector and the second generation vector of the malicious node are subjected to logical OR operation and then input into a third generation vector for generating the malicious node against the network; updating parameters of generators of the first generation countermeasure network, the second generation countermeasure network and the third generation countermeasure network in a gradient descending mode according to a discrimination result of the third generation countermeasure network by a discriminator of the third generation countermeasure network, and repeating steps S31-S32 to complete training of the first generation countermeasure network, the second generation countermeasure network and the third generation countermeasure network;
s33: inputting a third generated vector of the malicious node into a K-MEANS clustering module, classifying the malicious node by using a K-MEANS algorithm, and completing training of the K-MEANS clustering module.
4. The method for detecting the malicious node of the internet of things according to claim 1, wherein the generator of the first generation reactive network comprises an input layer, a convolution layer 1, a maximum pooling layer 1, a convolution layer 2, a pooling layer 2, a full connection layer 1, a full connection layer 2 and a tanh activation function which are sequentially connected;
the second generation countermeasure network generator comprises an input layer, a convolution layer 1, a maximum pooling layer 1, a convolution layer 2, a pooling layer 2, a full connection layer 1, a full connection layer 2 and a ReLu activation function which are connected in sequence;
the third generation countermeasure network generator comprises an input layer, a full connection layer 1, a full connection layer 2, a deconvolution layer 1, an up-sampling layer 1, a deconvolution layer 2, an up-sampling layer 2 and a Sigmoid activation function which are connected in sequence.
5. The method according to claim 3, wherein inputting the third generated vector of the malicious node into the K-MEANS clustering module to classify the malicious node by using the K-MEANS algorithm comprises:
s331: randomly selecting third generated vectors of k malicious nodes as a clustering center;
s332: distributing the third generated vector of each malicious node to the nearest cluster center by using a Euclidean distance formula to obtain k clusters;
s333: calculating the center points of k clusters by using the square error and SSE as the cluster center of the next iteration and repeatedly executing steps S331-S333 until the cluster center is unchanged.
6. The method for detecting a malicious node of the internet of things according to claim 4, wherein calculating the center points of k clusters using the square error and the SSE as the cluster center of the next iteration comprises:
wherein x represents a third generated vector of the malicious node, c i Represents the ith cluster center, dist (c) i X) represents the Euclidean distance between the third generation vector of the malicious node and the ith clustering center, R i Represents the ith cluster, m i Is the value obtained by the derivative root-finding formula, c i ' is the cluster center of the next iteration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211589796.XA CN116015771B (en) | 2022-12-12 | 2022-12-12 | Detection method for malicious nodes of Internet of things |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211589796.XA CN116015771B (en) | 2022-12-12 | 2022-12-12 | Detection method for malicious nodes of Internet of things |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116015771A true CN116015771A (en) | 2023-04-25 |
CN116015771B CN116015771B (en) | 2024-07-16 |
Family
ID=86025729
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211589796.XA Active CN116015771B (en) | 2022-12-12 | 2022-12-12 | Detection method for malicious nodes of Internet of things |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116015771B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180285740A1 (en) * | 2017-04-03 | 2018-10-04 | Royal Bank Of Canada | Systems and methods for malicious code detection |
CN111210002A (en) * | 2019-12-30 | 2020-05-29 | 北京航空航天大学 | Multi-layer academic network community discovery method and system based on generation of confrontation network model |
CN113269274A (en) * | 2021-06-18 | 2021-08-17 | 南昌航空大学 | Zero sample identification method and system based on cycle consistency |
US20220014554A1 (en) * | 2020-07-10 | 2022-01-13 | International Business Machines Corporation | Deep learning network intrusion detection |
-
2022
- 2022-12-12 CN CN202211589796.XA patent/CN116015771B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180285740A1 (en) * | 2017-04-03 | 2018-10-04 | Royal Bank Of Canada | Systems and methods for malicious code detection |
CN111210002A (en) * | 2019-12-30 | 2020-05-29 | 北京航空航天大学 | Multi-layer academic network community discovery method and system based on generation of confrontation network model |
US20220014554A1 (en) * | 2020-07-10 | 2022-01-13 | International Business Machines Corporation | Deep learning network intrusion detection |
CN113269274A (en) * | 2021-06-18 | 2021-08-17 | 南昌航空大学 | Zero sample identification method and system based on cycle consistency |
Also Published As
Publication number | Publication date |
---|---|
CN116015771B (en) | 2024-07-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108390869B (en) | Vehicle-mounted intelligent gateway device integrating deep learning and command sequence detection method thereof | |
Lyu et al. | AoI-aware co-design of cooperative transmission and state estimation for marine IoT systems | |
CN111178504B (en) | Information processing method and system of robust compression model based on deep neural network | |
Liang et al. | GaDQN-IDS: A novel self-adaptive IDS for VANETs based on Bayesian game theory and deep reinforcement learning | |
CN115099133A (en) | TLMPA-BP-based cluster system reliability evaluation method | |
CN115296759A (en) | Interference identification method based on deep learning | |
CN113037778B (en) | Attack detection method for continuous variable quantum key distribution system | |
CN116015771B (en) | Detection method for malicious nodes of Internet of things | |
Murugan et al. | Optimizing Mobile Ad Hoc Network cluster based routing: Energy prediction via improved deep learning technique | |
CN116916317A (en) | Invasion detection method based on white shark and random forest | |
CN115640845A (en) | Method for generating few-category samples of neural network of graph based on generation of confrontation network | |
Zhang et al. | Energy-Aware Positioning Service Provisioning for Cloud-Edge-Vehicle Collaborative Network Based on DRL and Service Function Chain | |
CN115996135A (en) | Industrial Internet malicious behavior real-time detection method based on feature combination optimization | |
Hua et al. | A lightweight trust management mechanism based on conflict adjudication in underwater acoustic sensor networks | |
Wang et al. | Deep learning method for generalized modulation classification under varying noise condition | |
Zhu et al. | Algorithm for solving traveling salesman problem based on self-organizing mapping network | |
CN117376084A (en) | Fault detection method, electronic equipment and medium thereof | |
CN114265954A (en) | Graph representation learning method based on position and structure information | |
CN113810385A (en) | Network malicious flow detection and defense method for self-adaptive interference | |
Ye et al. | A New Frequency Hopping Strategy Based on Federated Reinforcement Learning for FANET | |
Kou et al. | Intelligent selection: A neural network-based mac protocol-selection mechanism for wireless ad hoc networks | |
CN111314157A (en) | Efficient deployment method for cluster communication network nodes of water quality monitoring robot | |
CN118233317B (en) | Topology confusion defense method based on time-based network inference | |
Xiong et al. | Automatic digital modulation classification for ors satellite relay communication | |
CN117998364B (en) | XGBoost WSN intrusion detection system based on mixed feature selection |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20240618 Address after: Wangda Mansion A1236, No. 1777 Binhai Avenue, Huangdao District, Qingdao City, Shandong Province, 266000 Applicant after: Zhongyan Southern Financial Technology (Qingdao) Co.,Ltd. Country or region after: China Address before: 400065 Chongwen Road, Nanshan Street, Nanan District, Chongqing Applicant before: CHONGQING University OF POSTS AND TELECOMMUNICATIONS Country or region before: China |
|
GR01 | Patent grant | ||
GR01 | Patent grant |