Fault Recovery in Self Training Wireless Sensor Networks

摘要

The topic of Wireless Sensor Networks (WSNs) has gained considerable attention in the research community due to the variety of applications and interesting challenges in developing and deploying such networks. The typical WSN is significantly energy constrained and often deployed in harsh or even hostile environments, resulting in sensor nodes that are prone to failure. Failing nodes alter the topology of the network resulting in segmented routing paths and lost messages, utlimately reducing network efficiency. These issues spur the desire to develop energy-efficient, Fault-Tolerant (FT) algorithms that enable the network to persist in spite of the failed nodes. A FT recovery mechanism is proposed to mitigate an observed failure mechanism and increase the aggregate network efficiency. The recovery mechanism is evaluated through simulation and the results are presented in a characterization of the effect of the proposal on the network's performance.