CN-116305121-B - GAN data enhancement-based vehicle networking malicious software propagation control method

CN116305121BCN 116305121 BCN116305121 BCN 116305121BCN-116305121-B

Abstract

The invention discloses a vehicle networking malicious software propagation control method based on GAN data enhancement, which is characterized in that a vehicle networking malicious software propagation optimal control solution is obtained by modeling the propagation rule of malicious software in the vehicle networking and applying Pang Deli < Jin Jizhi > principle, and then the scale of a data set is controlled by expanding an optimal state through GAN so as to be used for a training data set for controlling the malicious software propagation in the vehicle networking by a neural network, so that the propagation of the malicious software in the vehicle networking can be greatly prevented, and the adverse effect of the malicious software on vehicles and users is reduced.

Inventors

LIU GUIYUN
TAN ZHIHAO
LIANG ZHONGWEI
ZHONG XIAOJING
CHENG LEFENG
LIU XIAOCHU

Assignees

广州大学

Dates

Publication Date: 20260508
Application Date: 20230224

Claims (4)

1. The vehicle networking malicious software propagation control method based on GAN data enhancement is characterized by comprising the following steps of: S1, establishing a traffic flow model to obtain an expression of a vehicle distance in an IDM balanced flow state; s2, establishing a vehicle-to-vehicle communication channel model to obtain the probability of successfully receiving the wireless signal; S3, establishing a vehicle networking malicious software propagation model to obtain a vehicle networking malicious software propagation fractional differential equation; S4, calculating an optimal control solution for a vehicle networking malicious software propagation fractional differential equation based on Pang Deli < Jin Jizhi > principle; S5, expanding the original data set based on the optimal state control of the GAN to obtain a new data set; in S3, the infection rate probability is expanded by combining a traffic flow model and a vehicle-to-vehicle communication channel model: S ij represents the clear distance between the vehicle j and the adjacent front vehicle i, and then a vehicle networking malicious software transmission fractional differential equation is obtained based on infectious disease dynamics and fractional theory; in the step S4, combining with an immunization and treatment control strategy, obtaining a new internet of vehicle malicious software propagation fractional differential equation, introducing Pang Deli to Jin Jizhi principle, solving the optimal control problem of the new internet of vehicle malicious software propagation fractional differential equation, and finally obtaining the minimum control quantity of an objective function J (u 1 (t),u 2 (t),u 3 (t)) of the internet of vehicle malicious software propagation fractional differential equation; u 1 (t) represents the treatment rate of the roadside equipment which is successfully attacked by the malicious software at the moment t, u 2 (t) represents the immunity rate of the roadside equipment which is easily attacked by the malicious software at the moment t, and u 3 (t) represents the treatment rate of the mobile equipment which is successfully attacked by the malicious software at the moment t; in S2, the probability density function of the received wireless signal power obtained by the distance between the vehicles in S1 is as follows: And is also provided with Where p t denotes the wireless signal transmission power of the equipped vehicle, λ denotes the wavelength, G t denotes the gain of the transmitting antenna, G r denotes the gain of the receiving antenna, f (p r (s ij )) denotes the probability density function of the received wireless signal power p r (s ij ); By giving the critical receiving power p su , that is, the signal power received by the vehicle is greater than p su , the wireless signal can be successfully transmitted, and the probability that the vehicle successfully receives the wireless signal is obtained as follows:
2. The GAN data-based enhanced internet of vehicles malware propagation control method according to claim 1, wherein in S1, the expression of the vehicle distance in the IDM balanced flow state is as follows: where T represents reaction time, s 0 represents minimum bumper spacing at full stop traffic, v represents speed in the state of balanced flow of the vehicle, v 0 represents desired speed in free flow of traffic, δ represents acceleration index, and s e represents vehicle clearance.
3. The method for controlling the spread of internet of vehicles malicious software based on GAN data enhancement according to claim 1, wherein in S5, a GAN model is obtained through an objective function, and the training process is regarded as a min-max game process, and new optimal state control and simulation data are synthesized from the learned distribution, and the expression of the objective function is as follows: Wherein, the Representing the expectation of the distribution of the real data x, z represents random noise, The expectation of gaussian noise distribution, D (x), represents the probability that the data generated by the generator is noisy z, G (z), and D (G (z)) represents the probability that the data generated is true data.
4. The method for controlling the spread of the internet of vehicle malicious software based on the GAN data enhancement according to claim 1, wherein in S5, the GAN is introduced to expand the data set of the optimal control solution, so as to increase the size of the data set, and make the trained neural network more robust, so as to enhance the robustness of controlling the spread of the internet of vehicle malicious software through the neural network.

Description

GAN data enhancement-based vehicle networking malicious software propagation control method Technical Field The invention relates to the technical field of information security of the Internet of vehicles, in particular to an Internet of vehicles malicious software propagation control method based on GAN data enhancement. Background The Internet of vehicles (Internet of Vehicles, IOV) is an application of an information physical fusion system (Cyber-PHYSICAL SYSTEMS, CPS) in an intelligent transportation system (INTELLIGENT TRANSPORTATION SYSTEMS, ITS), and advanced technologies such as a highly integrated mobile communication technology, an information processing technology, a sensor technology, an automatic control technology and the like can organically connect traffic participation factors such as people, vehicles and roads together, so that information among vehicles, people and road infrastructure can be shared and exchanged quickly, and a traveling vehicle is converted into a large-range wireless mobile network. Meanwhile, the Internet of vehicles is used as a high-efficiency information exchange network, and can play a great role in reducing traffic accident rate, relieving traffic congestion and providing personalized services, so that the Internet of vehicles has very wide application prospect and economic benefit. However, the vehicle nodes in the internet of vehicles highly depend on the wireless communication technology to transmit mutual information, and the information is presented to the user through the upper layer application, so that convenience is brought to the user, and meanwhile, the internet of vehicles has the characteristics of openness, instantaneity, dynamic change of topological structure and the like, so that the risk of the internet of vehicles suffering from invasion and propagation of malicious software is obviously increased. As with the common network, the internet of vehicles is also faced with the network security threat of malicious attacks such as viruses, worms, trojans and the like. The spreading of the malicious software in the network may reveal the privacy data of the vehicle and the user and also affect the receiving and responding of the vehicle to the related traffic information, so that traffic accidents occur, and the life and property security of the user is seriously threatened. Disclosure of Invention The invention provides a vehicle networking malicious software propagation control method based on GAN data enhancement based on the information security problem of the vehicle networking. The method is realized by the following technical scheme: a vehicle networking malicious software propagation control method based on GAN data enhancement comprises the following steps: s1, establishing a traffic flow model to obtain an expression of a vehicle distance in an IDM (intelligent driver mode) balanced flow state; s2, establishing a vehicle-to-vehicle communication channel model to obtain the probability of successfully receiving the wireless signal; S3, establishing a vehicle networking malicious software propagation model to obtain a vehicle networking malicious software propagation fractional differential equation; S4, calculating an optimal control solution for a vehicle networking malicious software propagation fractional differential equation based on Pang Deli < Jin Jizhi > principle; And S5, expanding the original data set based on the optimal state control of the GAN to obtain a new data set. Preferably, in S1, the expression of the vehicle distance in the IDM equilibrium flow state is as follows: Where s ij＝xi-xj represents the net distance of vehicle j from its neighboring front vehicle i, T represents the reaction time, s 0 represents the minimum bumper spacing when traffic is completely stopped, v represents the speed in the state of equilibrium flow of the vehicle, v 0 represents the desired speed in the free flow of traffic, δ represents the acceleration index, and s e represents the vehicle clearance. More preferably, in S2, the probability density function of the received wireless signal power obtained by the distance between the vehicles in S1 is as follows: And is also provided with Where p t denotes the wireless signal transmission power of the equipped vehicle, λ denotes the wavelength, G t denotes the gain of the transmitting antenna, G r denotes the gain of the receiving antenna, and f (p r(sij)) denotes the probability density function of the received wireless signal power p r(sij). More preferably, the wireless signal can be successfully transmitted by giving the critical receiving power p su, that is, the signal power received by the vehicle is greater than p su, and the probability that the vehicle successfully receives the wireless signal is obtained as follows: more preferably, in S3, the infection rate probability is extended by combining the traffic flow model and the vehicle-to-vehicle communication model as follows: And combining