MCFL: an energy efficient multi-clustering algorithm using fuzzy logic in wireless sensor network

摘要

In this study, a multi-clustering algorithm based on fuzzy logic (MCFL) with an entirely different approach is presented to carry out node clustering in wsn. This approach minimizes energy dissipation and, consequently, prolongs network lifetime. In the past, numerous algorithms were tasked with clustering nodes in wireless sensors networks. The common denominator of all these approaches is the constancy of the algorithm in all the rounds of network lifetime that causes the selection of cluster heads in each round. Selecting cluster heads in each round indicates that throughout the process the most eligible nodes are not selected. By comparing the chance of each node to be selected as a cluster head using a random number, the majority of these clustering approaches, both fuzzy and non-fuzzy, destroy the chance of selecting the most eligible node as cluster head. As a result, all these approaches require the selection of cluster heads in each round. Performing selections in each round increases the rate of sent and received messages. By increasing the number of messages, the total number of sent messages in the network increases too. Therefore, in a network with a high number of nodes, any increase in the number of packets will augment network traffic and increase the collision probability. On the other hand, since nodes lose a certain amount of energy for each sent message, by increasing the number of messages, nodes' energy will correspondingly decrease which results in their premature death. However, by selecting the most eligible nodes as cluster heads and trusting them for at least a few rounds, the amount of sent and received messages is reduced. In this article, In addition to clustering nodes in different rounds using different clustering algorithms, MCFL avoids selecting new cluster heads by trusting previous cluster heads leading to a reduction in the number of messages and saving energy. MCFL is compared with other approaches in three different scenarios using indices such as total remaining energy, the number of dead nodes, first node dies, half of nodes die, and last node dies. Results reveal that MCFL has as advantage over other approaches.