Maximizing Efficiency Of multiple–Path Source Routing in Presence of Jammer

Routing traffic through multiple paths is known as multipath routing. There was considerable research in the area of multiple-path source routing to overcome the problem of jamming. Many jamming- aware approaches existed. This paper explores the jamming aware concept of (1) with empirical studies using NS2. Moreover it also explores the possibility of relaxing the assumption of (1) with respect to in-network inference of correlations among related variables. In this implementation the source node allocates traffic to the available paths based on the awareness of jamming details at nodes. The practical implementation makes use of portfolio selection as explored in (1) and tries to explore the in-network inference of correlations among estimated random variables. The algorithm used to achieve this ensures that all available paths are optimally utilized without congestion while maximizing throughput. The simulation results revealed that the network is capable of performing jamming aware allocation of traffic. Index Terms-Multiple path source routing, traffic allocation, jamming, optimization, and in-network inference