Abstract: Vehicle networking intrusion detection solutions (IDS) can be used to confirm the authenticity of the events described in the notice of traffic incidents. The current Vehicle networking IDS frequently use detection scheme based on the consistency of redundant data, to reduce dependence on redundant data, an intrusion detection scheme based on neural network is presented. The program can be described as a lot of traffic event types , and the integrated use of the back-propagation (BP) and support vector machine (SVM) two learning algorithms. The two algorithms respectively applicable to personal safety driving fast and efficient transportation system with high detection applications. Simulation results and performance analysis show that our scheme has a faster speed intrusion detection, and has a high detection rate and low false alarm rate.