K-Means Clustering Based Localization Techniques for Multi-Hop Wireless Sensor Networks

Ok-means clustering is a well-known and influential clustering method in the data processing sphere. It has been widely applied in several domains, including earnings statistics mining and characteristic recognition. Recently, an increasing interest has been seen in wireless sensor networks (WSNs) localization techniques. Multi-hop WSNs have gained widespread attention due to their ability to offer broader coverage and increased connectivity over single-hop networks [45]. Nevertheless, as the types of nodes in a multi-hop community grow, traditional localization schemes become cumbersome and inefficient. By dividing the network into clusters and estimating the location of each cluster with anchor nodes, K-means clustering provides a simple yet efficient localization solution for multi-hop WSNs. These locations of anchor nodes are then used to help estimate the locations of the remaining unknown nodes belonging to that cluster using distance estimation methods such as trilateration or multilateration.