Random Access of Machine Type Communications: Modeling and Optimization

With the growing popularity of the Internet of Things and rapid proliferation of embedded devices, Machine Type Communications (MTC) has garnered significant attention in recent years. It is predicted that 50 billion devices will be connected to mobile networks by 2020, and MTC will become the dominant communications paradigm for a wide range of services in smart grid, transportation, health care, manufacturing and monitoring. For such a huge number of Machine Type Devices (MTDs), how to efficiently facilitate their access is becoming a significant challenge that attracts enormous attention from both academia and industry.To enable pervasive access for MTC, the Long Term Evolution (LTE) cellular network has become the most widely accepted solution owing to its ubiquitous coverage. Its random access process is, however, originally designed for Human-to-Human (H2H) communications witha relatively small number of users, which could easily be overwhelmed by the massive access requests of MTDs. There have been extensive studies on improving the access performance of MTC in LTE networks. Various algorithms were proposed to adaptively tune key system parameters according to estimated or measured network status. Due to the lack of scalable yet accurate models, nevertheless, how to optimize the random access process of MTC has received relatively less attention, and is unfortunately little understood.To address the above challenges, in this project, we will develop a theoretical framework to characterize the fundamental limits of random access performance of MTC and investigate the optimal access design for MTC. Our goal is three-fold: (1) We will establish a unified and scalable analytical framework for random access of MTC where key features of MTC, including a massive number of MTDs, small packet payload, bursty data arrivals and diverse Quality-ofService (QoS) requirements, will be incorporated. (2) We will characterize the maximum sum rates with various delay and energy constraints, which provide the fundamental limits of access performance of MTDs with diverse applications. (3) We will develop optimal tuning principles of access parameters and criteria for connection initialization and sensing, both of which offer direct guidance on the optimal access design of MTC.Our preliminary results have shown that the proposed approach has high potential to address the key issues outlined above. Built upon the PI’s expertise on random access theory and ongoing studies on MTC in LTE networks, the proposed research will not only contribute to the fundamental understanding of multiple access theory, but also shed important light on the successful development of MTC in next-generation communication networks.