Signaling and sleep mode enhancement strategies for machine to machine over cellular networks -

Abstract

In the future, the majority of everyday devices will connect and interact among each other using Machine-to-Machine (M2M) communications under the umbrella of the Internet of Things (IoT), initiating a new area of opportunities with innovative services and useful information. This anticipated growth in demand is expected to challenge the operation of wireless cellular networks due to the diverse characteristics and stringent requirements of M2M communications compared to traditional human-type communications. In this PhD dissertation, we present novel solutions for optimized M2M communications over cellular networks with focus on two main problems, namely the signaling overload problem due to a massive number of M2M devices transmitting small amounts of data and the energy efficiency problem due to the limited battery resources of M2M devices. The first part of this dissertation presents an extensive experimental measurement campaign for several M2M use cases over a live cellular network in order to derive empirical joint data-signaling traffic models. The proposed models take into account protocols from cellular technologies specifications in addition to the traffic behavior of different types of M2M services; we utilize them to develop and evaluate two effective signaling overload reduction solutions. The second part of the dissertation addresses the energy efficiency challenge by proposing a new deep sleep state for M2M devices in order to notably reduce their energy consumption and, thus, prolong battery lifetime. The novelty of the proposed solution is in the use of radio frequency (RF) signals to wake-up M2M devices on-demand, instead of periodic wake up to check paging channels. Stochastic geometry tools are utilized to model, analyze, and design M2M cellular networks with the proposed RF wake-up solution. We assess performance gains and analyze performance tradeoffs using simulations and numerical results. Furthermore, this dissertation includes a hardware direction that complements the analytical and s