Vehicular ad-hoc networks have been a research hotspot in wireless communication field. Security providing for data transmission in between nodes, attacks that are occurring in VANETs has become a big issue in VANETs.Usage of V2V (vehicle to vehicle) ad hoc networks and its internet applications has becoming an important task in present days, because of significant changes in network attacks. Intrusion detection system is a defense measure that reduces different tasks of computer networks and generates the attack sequences to the organizer of the network. So privacy and security is the most and effective measure for any type of network organization. So intrusion detection is an important research topic in network communication. AODV (Ad hoc On-demand Distance Vector) and Enhanced AODV's are the two approaches were used to support intrusion detection in static V2V (vehicle to vehicle ad hoc networks. To provide effective intrusion detection for dynamic ad hoc networks, in this paper, we propose and introduce a novel semi supervised approach i.e. Extended AODV Design. This approach is introduced to support two main issues, first one is select most relevant feature from network communication based on information gain, and second one is to split the value is chosen in such a way that makes the classifier impartial towards most regular values. Our experimental results will perform based on different attributes and also maintain equivalence simulation time in dynamic V2V (vehicle to vehicle) transmission. Proposed algorithm will use for signature based intrusion detection in V2V (vehicle to vehicle ad hoc networks. Keywords: V2V (vehicle to vehicle) Ad hoc networks, AODV (Ad hoc On-demand Distance Vector) , Classification, Feature Comparison.
I. INTRODUCTIONNow a days VANET technology V2V(vehicle to vehicle communication) V2I(vehicle to infrastructure communication) growing enormously to maintain the driving system more and more secure and trust worthy. Each vehicle is assuming as node. The main motive of this technology is to make the vehicle intelligent in transportation, fast decision taking in the risky situations, quick reaction to environmental situations during rain, fog and identifying the fake messages to not be deceived etc. It is also training in transforming the messages between the vehicles about road conditions like traffic, accidents, etc., to reduce the transportation time. For this purpose each vehicle has to exchange data each other about the conditions it had faced. By this data the neighbor vehicle can take decisions, in changing the direction due to heavy traffic and accidents. This type of actions reduces the congestion on the road makes transportation very fast. In this case every action is taken after exchanging the information. It is very essential to secure the data that is exchanging between the nodes. Sometimes the selfish node transmits a fake message for its own desire, showing traffic condition to other nodes to make itself clean road. Many types of attacks are injecting in to ...