An energy efficient grid-based clustering algorithm using type-3 fuzzy system in wireless sensor networks

The efficient management of energy in wireless sensor networks (WSNs) is a primary concern among researchers. Clustering algorithms serve as a crucial technique to address this issue. However, the initial uncertainties in both measuring the WSN’s values and in node localization due to GPS lead to secondary uncertainties like residual energy of nodes, cluster centrality, and distance from the cluster to the base station in the higher layers of WSNs. In this study, we have incorporated five improvements to our previous algorithm “FSCVG: A Fuzzy Semi‑Distributed Clustering Using Virtual Grids in WSN”. Firstly, we have discussed and classified uncertainties into two categories: primary uncertainties and secondary uncertainties. Secondly, we have applied a Type-3 fuzzy system to handle secondary uncertainties. Thirdly, we have used an adaptive imaginary grid to generate uneven clusters and balance the load according to the base station location. Fourthly, both decentralized and centralized clustering have applied based on new adaptive imaginary grid updates. Finally, we have determined the threshold level of each cluster proportionally, based on the energy of nodes within the same cluster. The findings of these improvements indicate an increased lifetime of the network concerning comparable methods.