Probabilistic-Based Energy Aware Routing for Network Lifetime and Source Level Fairness in Wireless Sensor Networks

Wireless sensor networks are often used to sample the state of physical phenomena in space. Sensor nodes, which usually have limited energy, collect data locally and forward them to a sink node. Thus there are two operating domains pertinent to the performance of such networks: (1) in time, which is the network longevity or lifetime, and (2) in space, which is the network coverage or source level fairness, i.e., the effective rate at which the state of the phenomena observed is reported at the sink. In this design paper we propose a simple, localized, probabilistic-based protocol that addresses performance issues in both domains. Our protocol addresses the time domain by exploring multiple paths for data delivery so that energy consumption can be uniformly distributed throughout the network, and addresses the spatial domain by dropping packets selectively in congested nodes, so that data flows from all sources will have their share of bandwidth to reach the sink.