Reliable routing protocol for mobile adhoc network

Abstract

quotMANET help us in setting up a network of mobile nodes like laptop, smart phones, tablet etc. without the need of any infrastructure setup. The topology of the network may change frequently due to node movement. Nodes are resource constrained and uses wireless media for communication. Each node has to act as a router to forward data of other nodes. Due to these characteristics, the routing is vulnerable to many security attacks like data loss, delay etc. Apart from these attacks, the MANET is vulnerable to low quality of service, wrong advertisement and malicious activities due to its open nature which are difficult to control and detect. To detect attacker nodes researchers have come up with cryptography based solutions. In a MANET, behaviour of a node changes due to malicious behaviour of the nodes/hardware failure/mobility of the node. The cryptographic approaches cannot detect such continuously changing behaviour of nodes. Researchers come up with a trust based routing solutions which use communication parameters for calculating trust value of a node and aggregate them to calculate trust value. They create a route based on trust of nodes and give one trustworthy route with most trusted nodes. This may add extra burden to trustworthy nodes and free other nodes from routing. To address this we can think of searching multiple trusted route between same source to destination during route formation. This allows us to use other available routes, if any route fails and the load of routing is divided among multiple nodes. In our routing scheme TMA-AODV (Trust based Mobility Aware AODV), we have used a number of packets dropped, the number of packets delayed and number of time node move out of the active route due to its movement for trust computation. The first two parameters detect the packet drop and packet delay attack and the latter parameter provides a stable route. For the implementation, we used the OPNET simulator. We have implemented drop attack and delay attacks to study its effect on route discovery time