Energy and delay aware massive access management in machine-to-machine communications

Abstract

Machine-to-machine (M2M) communication is a challenging topic in the Internet-of-Things era. The increasing growth of machines and high rate of packet production have result in massive access to the network. Therefore, one of the media access control (MAC) challenges in 5G cellular networks is the management of massive access to the wireless media regarding quality-of-service (QoS) requirements and battery restrictions of the machines. Delay is one of the QoS requirements that should be guaranteed in most applications. To the best of our knowledge, previous studies on scheduled MAC mechanisms have not addressed energy efficiency and delay requirements appropriately. In this paper, the problem of scheduled massive access management has been considered with the aim of simultaneously meeting delay requirements of machines and increasing energy efficiency. The proposed solution exploits gateways to reuse the spectrum dedicated to M2M communications and introduces a clustering algorithm to allocate radio resources to the machines appropriately. Moreover, an optimization problem is defined and solved using genetic algorithms (GA) to adjust the optimal transmission power of machines, aiming at maximizing the energy efficiency. To preserve delay constraints, the proposed method exploits an existing scheduling algorithm. Simulation results demonstrate more than 50% reduction in probability of packet delay violation for various classes of service and in mean packet delay compared to the baseline method. The results have also shown more than 20% improvement in energy efficiency of proposed method compared to the baseline. © 2019 John Wiley & Sons, Ltd.