Font Size: a A A

Research On Security Scheduling Method Based On Automotive Time-sensitive Network System

Posted on:2023-08-06Degree:DoctorType:Dissertation
Country:ChinaCandidate:B W WangFull Text:PDF
GTID:1522307316451784Subject:Vehicle Engineering
Abstract/Summary:PDF Full Text Request
With the development of Intelligent and Connected Vehicles(ICVs),automotive networks need to support multiple communication requirements,such as high reliability,high real-time performance,low jitter,and so on.Automotive Ethernet makes these communication requirements possible.Time Sensitive Network(TSN)is an important solution for automotive Ethernet.At present,the research on automotive TSN is still in the initial stage,and there is a lack of complete and systematic research on both the simulation analysis method and the secure scheduling design.To this end,this paper studies the schedulability and secure scheduling design of the automotive TSN network of ICVs.A model framework of the automotive time-sensitive network communication system is established.Combining the development process of model-based and component-based automotive embedded systems,the time-sensitive network modeling is abstracted into an end-to-end communication model of embedded systems,and decomposed into a clock synchronization model,a switch model,and a traffic shaping scheduling model.The time characteristics of the traffic type are required as design goals and forming the design framework for automotive time-sensitive network systems.The scheduling mechanism of automotive time-sensitive network is studied,and the architecture and implementation method of TSN software simulation and hardware simulation platform are proposed.The time-aware scheduling,frame preemption,cyclic queuing and forwarding,asynchronous traffic shaping,seamless redundancy and other mechanisms suitable for automotive communication are analyzed in detail;an automotive TSN software simulation platform based on the OMNeT++simulation environment is constructed from the level of the vehicle network communication system.It can realize the complete performance and function simulation of the automotive TSN scheduling mechanism through parameter configuration,and reduce the simulation error to less than 0.6%in combination with the physical system.An arbitrary variable topology control method is proposed,and the automotive TSN physical simulation platform is developed.It realizes the automatic construction of arbitrary TSN topology and its communication performance analysis technology.A maximum end-to-end delay model for automotive time-sensitive network timesharing scheduling based on network calculus theory is established,and the schedulability analysis is carried out.The advantages of the time-aware scheduling mechanism when used alone are analyzed.The time-aware scheduling mechanism can ensure the purpose of real-time data transmission of vehicle traffic,but the guard band limits and reduces the resources of time slots.Through the comparative analysis of the frame preemption mechanism and the time-aware scheduling mechanism,the frame preemption mechanism can greatly reduce the time slot length of the guard band and reduce the negative impact of the guard band.At the same time,a resource optimization scheduling algorithm based on cyclic queuing and forwarding mechanism is proposed,which can greatly increase the number of schedulable flows in the network.The security scheduling method of automotive time-sensitive network is studied,and the per-stream filtering and policing mechanisms are analyzed in detail.The STRIDE threat model is established for the centralized automotive architecture based on TSN,and the security threat status under the TSN network architecture is analyzed.An anomaly detection system for time-sensitive network systems is proposed for detecting TSN Dos attacks,sensor anomalies and network bandwidth anomalies,etc.,and the function and performance of the anomaly detection system are verified based on the automotive TSN software simulation platform.The anomaly detection system can detect and discard four different types of abnormal data without affecting the performance of the original TSN network,which effectively improves the security of automotive TSN scheduling.
Keywords/Search Tags:Intelligent and Connected Vehicles, automotive Time-Sensitive Network, automotive network scheduling, automotive cyber security, automotive network simulation
PDF Full Text Request
Related items