Improved Pseudo-Bayesian Access Class Barring for Bursty M2M Communications

Random access procedure of long-term evolution (LTE) cellular systems can cause excessive communication delay for bursty machine-type communications. In this paper, we propose an improved pseudo-Bayesian access class barring (ACB) algorithm by introducing a novel tuning for burst factor to support bursty traffic arrivals. As a result, the proposed algorithm estimates the number of active machine-type devices (MTDs) more accurately compared to the conventional ACB algorithms and significantly reduces access delay for bursty MTD activations. Simulation results show that the proposed algorithm shows closed performance to the ideal ACB algorithm which is assumed to know the actual number of active MTDs.