Congestion Control in Optical Burst Switching Networks Using Differential Evolution Optimization

Abstract

Optical burst switching (OBS) is a trade-off between optical packet switching (OPS) and optical circuit switching (OCS). OBS is different than these two concepts which allow transmission of control packet separately over reserved optical channel. The most challenging task in OBS is congestion which varies as per the traffic load. In this paper, differential evolution (DE) optimization algorithm is presented for congestion control in OBS network. The performance of DE is evaluated primarily for the total number of congested links against total load and second for the total number of hops against load in the network. This was evaluated on the basis of reduction in the number of congested links. These simulations were carried out for different sparse and dense network like ARPANET, NSFNET, RANDOM-12, and TORUS-9 topology. The results of the same were compared with conventional Dijkstra’s shortest path (SP) algorithm, which is widely used as a routing technique and DE exhibits better result.