Abstarct: Vehicle ad hoc network (VANET) has been proposed as a new plane to provide road safety, traffic management and convenience applications for drivers and passengers on the road. In this network, communications are in two ways: vehicle to vehicle and vehicle to infrastructure. Different messages containing events warning about road are exchanges in these communications. In these networks, wireless communications leads to many challenges such as: injecting false information, modifying and replaying the disseminated messages in the network, impersonation and distributed denial of service (DDOS). One of the security threats in VANETs is the DDOS attack which aims to busy and unavailable the Road Side Unit for legitimate vehicles. On the other hand, a number of vehicles on the roads have computational resources and can be used as an unexploited resource. Therefor. The concept of cloud computing for vehicles was introduced. To use cloud technology in VANET, we need to use an infrastructure as a service to centralize network management and secure attacks. In this thesis, the software-defined network (SDN) have been used to design an intelligent secure schema for vehicles. One of the security challenges of SDN baxsed VANETs is availability of the controller. In this thesis, DDOS attack has been investigated to address the security needs of the controller's availability. By launching legitimate and attack traffic to SDN baxsed VANETs baxsed on the entropy criterion and traffic characteristics, the network behavior pattern is investigated. Then by adaptive an improved algorithm with network behavior by traffic characteristics, a dynamic schema was obtained to detect DDOS attack, which the attack detection rate achieved 65/61%.