Energy-efficient predictive congestion control for wireless sensor networks

In recent years, wireless sensor networks have become an emerging technology in industries, environment monitoring and health care monitoring systems and so on. However, sensor node is a resource-constrained device in terms of memory, bandwidth and energy. These constraints impose congestion in the network, leading to large number of packet drops, low throughput and significant wastage of energy because of retransmission. This study presents a new approach for predicting congestion using probabilistic method, and controlling congestion using new rate control methods. The probabilistic approach used for the prediction of occurrence of congestion in a node is developed using data traffic and buffer occupancy. The rate control method uses rate allocation schemes, namely, rate reduction (RR), rate regulation (RRG) and split protocol (SP) to improve throughput and to reduce packet drops. In addition, an energy-efficient routing which finds the best forwarding node for data transmission is also proposed. Simulation results are compared with decentralised predictive congestion control (DPCC). The results show that the proposed method indeed reduces congestion and energy consumption, and improves the performance.