On design and deployment of fuzzy-based metric for routing in low-power and lossy networks

Minimizing the energy consumption and hence extending the network lifetime is a key requirement when designing an efficient sensor network protocol. QoS-aware routing in Wireless Sensor Network (WSN), aims to take into account other networks performance aspects as minimizing end-to-end delay (as well as jitter), reducing packet loss rate and minimizing the energy consumption of the network during data transmission. These objectives are sometimes conflicting therefore tradeoffs must be made between energy conservation and QoS considerations. The general problem can be reformulated as a Multi-Constrained Optimal Path problem (MCOP) which is known as NP-complete. The latter raises a real challenge as sensor nodes are very limited in resources capabilities. We propose to use fuzzy inference mechanism to seek a good tradeoff between all given metrics and constraints. This paper discusses the implementation of combining several routing metric using fuzzy logic to design a RPL objective function, the routing standard for the Internet of Things. The proposal is integrated on Contiki operating system, the deployment was performed on a real world indoor WSN. Obtained results show improvements compared to the common implementation of the RPL protocol and demonstrate relevance of our contribution.