A transport protocol for congestion avoidance in Wireless Sensor Networks using cluster-based single-hop-tree topology

This paper proposes a transport protocol using cluster-based single hop tree topology for congestion avoidance in Wireless Sensor Networks (WSNs) under noisy environments. In this model, a WSN is divided into several clusters containing varied density of sensors according to the accuracy desired where a sensor as the head of one cluster is a child node of some other cluster so that the entire network is arranged in a tree topology with the root node as the sink. A sensor senses and measures predefined set of parameters from the noisy environment to transmit these to other sensors of the cluster. Each sensor continues to update its stored data integrating the parameter values received from other sensors till these converges to the estimated true value. The cluster heads transmit the received data to their corresponding parent nodes hop-by-hop until these reach the sink node. Simulation results show that a rapid convergence coupled with the cluster-tree topology, leads to congestion avoidance and energy minimization in the WSN. Analysis reveals that this protocol scheme is suitable for both continuous and event based monitoring in noisy environment and can be made adaptive to the changing requirements.