Prioritised Random Access Channel Protocols for Delay Critical M2M Communication over Cellular Networks

With the ever-increasing technological evolution, the current and future generation communication systems are geared towards accommodating Machine to Machine (M2M) communication as a necessary prerequisite for Internet of Things (IoT). Machine Type Communication (MTC) can sustain many promising applications through connecting a huge number of devices into one network. As current studies indicate, the number of devices is escalating at a high rate. Consequently, the network becomes congested because of its lower capacity, when the massive number of devices attempts simultaneous connection through the Random Access Channel (RACH). This results in RACH resource shortage, which can lead to high collision probability and massive access delay. Hence, it is critical to upgrade conventional Random Access (RA) techniques to support a massive number of Machine Type Communication (MTC) devices including Delay-Critical (DC) MTC. This thesis approaches to tackle this problem by modeling and optimising the access throughput and access delay performance of massive random access of M2M communications in Long-Term Evolution (LTE) networks. This thesis investigates the performance of different random access schemes in different scenarios. The study begins with the design and inspection of a group based 2-step Slotted-Aloha RACH (SA-RACH) scheme considering the coexistence of Human-to-Human (H2H) and M2M communication, the latter of which is categorised as: Delay-Critical user equipments (DC-UEs) and Non-Delay-Critical user equipments (NDC-UEs). Next, a novel RACH scheme termed the Priority-based Dynamic RACH (PD-RACH) model is proposed which utilises a coded preamble based collision probability model. Finally, being a key enabler of IoT, Machine Learning, i.e. a Q-learning based approach has been adopted, and a learning assisted Prioritised RACH scheme has been developed and investigated to prioritise a specific user group. In this work, the performance analysis of these novel RACH schemes show promising results compared to that of conventional RACH