Forwarding Algorithm Using Prediction and Optimation Theory Over Multipath Network

In the Internet today, multipath routing is proved a popular technique that improves reliability and robustness of data delivery; however, current forwarding algorithm over multipath network have problems such as lacking of preparing for sudden spikes. This paper proposes a new forwarding algorithm, which is based on time series prediction and PSO (particle swarm optimization). We describe how the algorithm dynamically predicts loss levels over all available network paths and adjust the allocation proportion of packets for next prediction period by the adaptive PSO. We present our simulation work using NS-2 to show the superiority of our algorithm over equality splitting algorithm.