Discrete teaching–learning-based optimization algorithm for clustering in wireless sensor networks

Clustering is an appealing paradigm exploited to improve the lifetime and scalability of wireless sensor networks (WSNs). Considering the NP-completeness of the clustering problem, numerous meta-heuristic algorithms are provided in the literature for the clustering of WSNs. Teaching–learning-based optimization (TLBO) is an optimization algorithm employed to tackle continuous optimization problems. In this paper, a novel discrete version of the TLBO algorithm is being presented that employs the swap and mutation operators to deal with discrete solutions. Subsequently, the new-fangled algorithm was utilized to design a hierarchical energy-aware clustering scheme for the WSNs to minimize the energy usage of the sensor nodes. In addition, an energy-aware local search algorithm was provided to enhance the network lifetime by taking factors such as energy and distance into account. Extensive simulations are conducted to indicate the effectiveness of this scheme in reducing the power usage of the sensor nodes and improving the WSN lifetime.