Dynamic rough-based clustering for vehicular ad-hoc networks

Abstract

Due to the spatio-temporal aspects of vehicles within vehicular ad-hoc networks, traditional clustering techniques are not effective as they rely on static configuration. In this paper, we proposed a dynamic clustering technique that is based on rough theory of grouping data. The contributions of this research are to propose: A self-organising clustering technique as an extension to dynamic rough clustering and a framework that manages the integration among different algorithmic components, which are required to develop such soft computing systems. We performed extensive experiments in order to evaluate the effectiveness of the proposed technique in terms of: communication load, inter and intra connectivity, threshold analysis, and relationship among data clusters. Furthermore, a performance comparison with relevant techniques has been reported. The results indicated that the proposed technique is robust and promising in comparison with existing techniques in the domain of wireless sensor networks.