Heterogeneous sensor networks traffic control algorithm based on CARIMA prediction

Aiming at the problems that they are slowly responded, working unstably etc. in sensor networks traffic control algorithms, we propose a traffic control algorithm based on CARIMA prediction. According to the traffic distribution, the parameter in CARIMA model are recognized online. The CARIMA predictive control sequences are obtained by objective function. The generation rates of source nodes are updated based on control sequences. Simulation results show that the proposed algorithm can effectively improve network throughput and maintain the intermediate node queue length, which achieves congestion control and improves the network QoS performance.