A Fault Recovery Scheme for Wireless Sensor Networks

Abstract

In wireless sensor networks, sensor nodes may fail due to energy depletion or physical damage. To recover the data of a failed node, we propose a fault recovery scheme which enables the remaining alive sensor nodes to use the redundant information with regard to the failed node to fulfill the recovery. The idea is similar to a level 4 RAID (Redundant Array of Independent Disks) in the sense that each sensor serves as the dedicated parity disk for its neighbors. We compare the network lifetime with and without recovery being involved, where the network lifetime is defined as the time interval from the point that a network starts operation to the point that a node failure is observed and can not be recovered. We explore the impact of parameters such as node failure probability, communication range, and node density on the network lifetime.