Probabilistic Algorithms for Cost-based Integrated MAC and Routing in Wireless Sensor Networks

In this paper we propose a new integrated MAC/routing algorithm for CSMA channel access based wireless sensor networks (WSNs). In modern joint MAC/routing solutions for WSNs, forwarding decisions are often made based on metrics related to the advancements toward a specific destination (sink) and this is usually achieved through geographical routing and/or local indicators (virtual topologies) of the direction to be followed to get to the sink. In these settings, an important problem to be solved consists of providing extremely efficient next hop selection procedures, where relay nodes should be selected in a short time, in an energy efficient manner and promoting the nodes leading to positive geographical advancements toward the sink. In this work, we generalize the concept of node energy by referring to generic node costs, that we introduce here to represent the suitability of a given sensor node to be selected for data forwarding. Costs may therefore be associated with node energies, queue states (network congestion) as well as link qualities. Further, we propose to jointly exploit local routing rules (coordinates or virtual topologies) and node costs right into the channel access phase. In our framework, potential relay nodes contend for the channel based on a cost-based probabilistic approach where all these aspects are jointly considered and the final goal is to promote the selection of cost-efficient paths.