Data-centric clustering for data gathering in machine-to-machine wireless networks

Abstract

While clustered communication has been considered as one key technology for supporting machine-to-machine (M2M) wireless networks, existing clustering techniques have predominantly been designed with the objectives of maximizing the service quality for individual machines. Many M2M applications, however, are characterized by the large amount of correlated data to transport, and hence existing “machine-centric” clustering techniques fail to effectively address the “big data” problem introduced by these M2M applications. In this paper, we propose the concept of “data-centric” clustering to exploit the correlation of data to be gathered by a large number of machines. We first formulate an optimization problem for the target problem that involves cluster formation and power control. We then propose an anytime algorithm for solving the optimization problem iteratively in two phases. Compared with other approaches for cluster formation, we show through evaluation that data-centric clustering can achieve noticeable performance gain for dense M2M communications with big data.