A study of Bayesian scheduling for M2M traffic in wireless LTE network

In the event of phenomenal growth in Internet of Things (IoT), Machine-to-Machine (M2M) devices are projected to reach figure of multiple billions in foreseeable future. Operators around the world are aggressively refarming their spectrum from older network and moving quickly to Long Term Evolution (LTE) for guarantee of future-proof services. With every new M2M application being invented or deployed at this pace, unprecedented factors are unceasingly induced to existing LTE protocols which caused undesirable performance degradation. In this paper, a realistic LTE network environment are modelled and simulated with tractable M2M traffic modules to observe such impacts on conventional scheduling schemes and to identify the major causes. A prominent conventional Bayesian approach is hence adopted for revision to adapt the new M2M traffics. The results obtained shown that the proposed M2M-enabled True Bayesian Estimate (TBE-M) algorithm is capable of outperforming the conventional TBE on a great scale.