A performance efficient joint clustering and routing approach for heterogeneous wireless sensor networks

Wireless sensor networks have proven to be a promising paradigm due to their wide applicability in the real-world scenarios. Because of the resource constraint, type-variant and battery-powered features of the sensor nodes, the key challenge is how to improve the energy usage and network life in the heterogeneous sensor networks. In this paper, the Mahalanobis distance based K-Means algorithm is integrated with a novel evolutionary approach for clustering called calf search optimization algorithm (K-CSOA) that creates energy efficient clusters. The routing in the network is performed using a power efficient ant colony optimization algorithm (ACO). The extensive and multiple simulations are performed to validate the effectiveness of the proposed method. It is observed that the running time of the proposed scheme is lowered due to the implementation of hybrid clustering. The results are validated for various performance metrics wherein the proposed method shows 41.6% and 4.3%-12% reduction in energy expenditure and end-to-end delay respectively and around 60% enhancement in the throughput of the network when compared with other similar state-of-the-art methods. In addition to this, statistical analysis is also carried out to identify the performance of the proposed algorithm.