MANET highly efficient clustering technique based on coverage k-means algorithm

Abstract

Mobile ad hoc networks (MANET) are employed as an alternative access for established infrastructure in areas lacking permanent connections. MANET are dynamic and flexible, all devices can communicate. This makes them particularly useful in situations like disaster recovery, military operations, or remote areas where traditional networks are unavailable. The routing techniques are more important to managing the communication and improving the network stability, reliability, and efficiency. In this paper, a Coverage K-Means cluster-based Routing Protocol (CKRP) is proposed. The CKRP has better link connectivity, a higher route lifetime, and is more reliable with minimum transmission delay. The MANET square area is divided into multiple zones based on the zone generation proposed model based on the ratio of the maximum and minimum boundary to the node coverage. The node density and zone will used to compute the number of clusters. The k-Means algorithm will used in the cluster formulation stage. In this paper, a new Cluster Head (CH) selection model was proposed as a final stage of the CKRP model to elect the optimal node as a CH. The experiment simulated the proposed model and compared it with Ad hoc On-Demand Distance Vector Routing (AODV) and Fuzzy Logic- AODV (FL-AODV). The CKRP routing reliability has 25% improvements compared with AODV for the number of nodes exceeding 90. The proposed model has a minimum delay compared with the other algorithms.