Key Technologies Of Edge Computing Security Protection For Internet Of Vehicles

The vehicle edge computing network deploys the computing capacity to an edge position closer to the vehicle,making it capable of processing computing-intensive and latency-sensitive IoV services with low-latency,high real-time and large-bandwidth requirements,thus providing vehicle users with higher quality service.As one of the most important issues,the security problem should be considered carefully during the further development and evolution of the vehicle edge computing network,which mainly face the following challenges:(1)The method of verifying the legality of data sources and data is rigid and inefficient,and it is difficult to achieve all-round trusted protection;(2)The security policies of "patch type" and "plug-in type" are difficult to resist large-scale distributed network attacks initiated by hackers;(3)The task offloading and processing methods of the vehicle edge computing network are unreasonable,resulting in task processing efficiency is low and edge computing nodes with heavy load is easy to become potential attack targets.Therefore,how to design a diversified and effective security protection mechanism for the above challenges is the edge computing network environment of the Internet of Vehicles is a key issue that needs to be solved.To cope with the above challenges,this thesis takes the vehicle edge computing network as the background and deeply studies the security protection technology in the Internet of Vehicles.Innovative solutions such as trusted authentication and reputation sharing mechanisms,detection and defense methods for low-rate DDoS attacks in the Internet of Vehicles,and safe and efficient processing technology for collaborative vehicle tasks have been proposed.And through theoretical analysis and simulation experiments,the performance of functionally feasible and effective for the proposed schemes have been proved and verified.Specifically,the main research works and innovations of this thesis can be divided into the following three aspects:(1)In response to the data security requirements of authorization authentication and trusted consensus in the Internet of Vehicles,a blockchain-based trusted authentication and reputation consensus mechanism is proposed to manage the trust of vehicles and road-side units(RSUs),which ensures the security of the authentication consensus of the Internet of Vehicles.Using decentralized identification(DID)and practical byzantine fault tolerance(PBFT)to design authentication and consensus methods,based on tamper-proof ability,enhances the ability of system decentralization,security,and credibility.The identification and processing capabilities of malicious RSU are improved.The deep reinforcement learning algorithm is introduced to optimize the consensus process,which improves the system consensus efficiency while ensuring the security and credibility of RSU.Theoretical analysis and simulation experiments show that the method proposed in this study not only ensures the security of the system but also optimizes the performance such as delay and throughput.(2)Aiming at the device security problem that distributed deployed RSUs are vulnerable to low-rate DDoS attacks and are illegally controlled,a detection and defense algorithm for low-rate DDoS attacks is proposed.The detection algorithm introduces the generalized entropy metric and the information distance metric into the vehicle edge computing network,uses the RSU to collect the characteristic value of the received traffic,calculates the information metric in real-time and compares it with the threshold value,and detects the low-rate DDoS traffic at the edge of the network,reducing network computing.consumption of resources.The defense algorithm innovatively proposes an edge node cooperative defense and conflict detection strategy.Multiple nodes cooperate to share security data within a certain range,and then dynamically track and exclude attackers who use the identity of normal vehicles under the unified scheduling of node controllers.Ensuring the service link of the normal vehicle whose identity has been fraudulently used.Through theoretical analysis and simulation experiments,the proposed detection algorithm and defense algorithm have high sensitivity and accuracy in the identification of low-rate DDoS attack traffic in the vehicle edge computing network.(3)Aiming at the problem that the heavy-loaded RSU is easy to become a potential attack target and cannot process the Internet of Vehicles services safely and efficiently,the safe and efficient processing technology for vehicle cooperative tasks based on a greedy algorithm is proposed.Under the premise that RSUs cannot provide task offloading and processing capability due to physical/network environment problems,consider the dynamic factors such as the resource distribution and time-varying state of the vehicle alliance,and take flexible and safe task allocation and efficient and low-cost task processing as the optimization goal,to establish a vehicle alliance combined auction model.Through the vehicle cooperative task processing greedy algorithm,the optimal vehicle alliance task unloading and processing scheme is obtained to ensure the safe and efficient processing of system tasks.The theoretical analysis and simulation experiments prove that the proposed scheme can guarantee lower processing costs such as latency and resource consumption during task processing.