Prediction based topology reconfiguration using clustering for power efficient operation in mobile adhoc networks

The topology of a Mobile Ad hoc Network (MANET) has substantial influence on its performance in that the transmission range and battery power of the nodes are not uniform in nature. The topology is susceptible to node and link failures which lead to network partitioning. Topology reconfiguration aims to recover from the failure and maintain the connectivity. Existing topology reconfiguration algorithms does topology updates at regular intervals which increases the communication overhead and leads to energy depletion. We propose a distributed load balanced Cluster-based Topology Reconfiguration algorithm that achieves higher data delivery and connectivity. In this work we estimate the failure of the nodes and links and reconfigured the topology by selecting optimal transmission power to avoid partitioning. Topology updates are done only at failing conditions. The simulation results show that the proposed algorithm increases the node life time and reduces the power consumption.