Analysis and Design of Adaptive MAC Layer for Machine to Machine Communication over Next Generation Wireless Networks

Abstract

Narrowband Internet of Things (NB-IoT) proves superiority to the other IoT networks from many different aspects. Where, it became as a complement part of the fifth-generation (5G) networks specifications. However, the increasing future demand for IoT applications requires that resource utilization be optimized to serve the maximum possible number of NB-IoT (User Equipment) UEs. Additionally, the relation between the NB-IoT spectral efficiency and the type of IoT traffic is not studied enough. Therefore, the achievement of traffic management over NB-IoT for efficient spectrum utilization is currently a new area for IoT 5G networks. In this dissertation, we introduced a holistic vision that achieved two main goals. First, we improved the NB-IoT spectral efficiency to serve the increasing number of NB-IoT UEs. Second, we alleviated the burden of two issues; the signaling during each transmission request for periodic and aperiodic traffic such as Smart-Meters (SM) and emergency alarms and the queuing burden for periodic traffic. Additionally, this dissertation addresses the modeling issue of the NB-IoT from modeling ease, model scalability, and the impact of other upper layers of its protocol architecture. Accordingly, this dissertation is divided into the three following phases. First, we model the protocol stack of the NB-IoT access network by using the state-machine modeling methodology by using the State-flow toolbox of the Simulink environment. Our model's architecture is open-source and comprises all (Long Term Evolution) LTE and NB-IoT protocol layers. Therefore, our model is scalable and can be used for any future works. In the second phase, we evaluate the performance of the NB-IoT uplink scheduler in terms of the number of periodic NB-IoT versus four main Key Performance Indicators (KPIs): resources utilization, throughput, accessibility, and retain-ability. We apply our analysis to the periodic IoT traffic due to its domination in the IoT field. We assume that the access network model provides four different data rates for NB-IoT traffics. To ensure fairness, we divide the cell spectrum equally on the four data rates. Also, we assume ideal propagation without any transmission impairments and use round-robin scheduling. The simulation results are verified according to the 3GPP standard and are validated with another trusted modeling technique. According to the above assumptions, the evaluation results depict that the NB-IoT single-tone data rate achieved the highest resources utilization and, at the same time, the highest channel holding time. However, the single-tone data rate's low throughput value motivates us to improve the NB-IoT spectral efficiency. In the final phase, we introduce our holistic improvement approach.