Mobility-aware pro-active low energy (MAPLE) clustering in ad hoc mobile wireless networks

We propose a framework for mobility-aware pro-active low energy (MAPLE) clustering in ad hoc mobile wireless networks. Most of the clustering approaches proposed in the literature primarily focus on the algorithmic aspects of clustering without considering the practical implementation issues, and these are often reactive in nature. The proposed approach addresses the problem of clustering in a medium-access control framework and enables pro-active and energy-efficient clustering exploiting the node mobility information. In particular, for pro-active clustering, we introduce a method to exploit the link level information to estimate the mobility pattern of the wireless nodes and the cost of using a wireless link in terms of required transmission power. A channel reservation technique is used to reduce the number of contentions among the nodes while accessing the channel during cluster formation. Simulation results show that the proposed framework results in superior clustering performance in terms of control overhead, average number of link failures and load distribution compared to other clustering approaches proposed in the literature, such as the LCC-LID (least cluster change lowest ID)-based clustering.