A prediction-based traffic aggregation scheme for green networking

Reduction of energy consumption is becoming a major concern in communication networks, because of its potential economical benefits and environmental impact. In this paper, we propose a prediction-based traffic aggregation scheme to realize energy saving for green networking. By predicting the traffic loads, the scheme aggregates and reroutes the traffic in some nodes to make them idle and then sleep in the next control time interval when the traffic is predicted light. Simulation results demonstrate that the proposed scheme can yield the accuracy of traffic prediction with a small normalized one-step prediction error and significant energy saving corresponding to the traffic loads in a simple network. Meanwhile, the throughput and delay of the network are almost not affected by the scheme.