Fuzzy multi-hop clustering protocol: Selection fuzzy input parameters and rule tuning for WSNs

Nowadays, the most important aspects of wireless sensor networks (WSNs) are to make optimal use of and direct the limited energy of sensor nodes towards the desired application and prolong the network lifetime for that application. Although a few studies exist that have addressed these special goals, they have been mostly focused on the process of selecting cluster heads (CHs) and forwarders. No studies have been conducted so far on the selection of fuzzy input parameters in clustering and routing processes as well as the application-based parameter selection process. Generally, a fixed number of parameters have always been selected by designers. Hence, the shuffled frog leaping algorithm (SFLA) was employed in this paper to propose a technique for selecting fuzzy input parameters in a fuzzy multi-hop clustering protocol named the PS-SFLA. This technique includes three main phases, introduced in three versions for the sake of stepwise evaluation. Based on the literature review, the most frequent and diverse parameters were extracted and formulated in the first version. The proposed technique used the SFLA in the second version to select the appropriate parameters fitting the application specifics and scenario and determine the coefficients of parameters simultaneously so that they could be used as the inputs of the fuzzy inference system. It was also utilized in the final version for the automated, accurate, application-based tuning of fuzzy rules before the network was set up. By design, different versions of the PS-SFLA act as the starting points of the next version in addition to the fact that they can be evaluated separately. The PS-SFLA was compared with the LEACH, ASLPR, SIF, ERA, and FSFLA in different scenarios and two applications from the perspective of the alive nodes, the number of packets received by the BS, lifetime, and other factors. According to the simulation results, the PS-SFLA outperformed all of the other methods greatly in all scenarios and applications and PS-SFLA increases the lifetime due to the appropriate selection of fuzzy input parameters based on application and purpose. (C) 2020 Elsevier B.V. All rights reserved.