Reliable and Load Balanced Multi-path Routing for Multiple Sinks in Wireless Sensor Networks

Abstract

Wireless Sensor Networks, the existing techniques for sink deployment consider load balancing and multipath routing. But the fault detection and recovery of sinks are rarely considered. In this paper, we propose to develop a Fault detection and recovery mechanism for multiple sinks. The nodes are grouped into clusters based on the data generation rate and multiple sinks are deployed in the clusters in which the data generation rate is more. An energy efficient multi path tree is constructed towards the sink and the aggregated data from the sensors are transmitted to their nearest sinks through this multi path tree. The rate adjustment value and the inter arrival delay of each path is calculated and sent back as a feedback to the source to by the respective sinks. Then the path with the minimum feedback value is chosen as the best path for transmission. When a failure occurs in a sink, then it can be detected based on delay and remaining energy. Once the fault sink is detected, then the data is transferred to another sink or the sinks are deployed again. From our simulation results, we show that an efficient load balanced multi path routing for multiple sinks is obtained and the fault detection and recovery can be made effectively.