A Hedge Algebras Based Fuzzy Inference System for clustering in Multi-hop WSNs

To prolong the lifetime of wireless sensor networks, the clustering technique is a key point to create optimized sensor node groups for effective transmission processes. There are many factors affected the efficiency of clustering algorithms that come from uncertain characteristics of both inside and outside network conditions. Therefore, the fuzzy logic-based clustering technique can be seen as a promising technique since it allows combining and evaluating diverse parameters in an efficient sense. Moreover, the Fuzzy Inference System (FIS) is an efficient modeling tool to utilize the best input data features and expert knowledge for supporting appropriate decisions. However, using the human reasoning process to assign linguistic terms for setting fuzzy rules may lead to ineffective results because of intuitively heavily dependent problem. Otherwise, thanks to a good property of hedge algebras that provides a mathematical formalism for designing the order-based semantic structure of term domains of linguistic variables as a quantitative model. Hence, this paper proposes a novel fuzzy inference system based hedge algebras which combine for forming clusters in multi-hop sensor networks. The numerical results are shown in this paper to validate the efficiency of the proposed model.