An adaptive and fault-tolerant scheme for gateway assignment in sensor networks

Gateways are important elements in a sensor network since they provide the ability to establish long range reachback communication for the purpose of retrieving critical data from remote locations. However, gateways are prone to failures just like regular sensor, and they also consume significantly more energy since they transmit over longer distance compared to sensor-to-sensor links. In this paper an adaptive and fault-tolerant method for gateway assignment in sensor networks is introduced. The proposed approach is fully distributed and it achieves the following objectives: 1. It allows surviving gateways to recover for other gateways that have failed. 2. It distributes energy usage and traffic load incurred by gateway nodes within the sensor network among several regional gateway nodes, while each gateway adaptively controls its own region of influence based on local conditions such as energy and traffic load. This methodology was evaluated via simulation using a realistic network model containing 400 sensor with virtual targets. The simulation results indicate that the scheme recovers quite well when gateways fail. Moreover, during periods when no gateway failures are present, the scheme also balances the traffic load among gateways.