Learning Automaton based Adaptive Access Barring for M2M Communications in LTE Network

Massive access of machine-to-machine (M2M) devices over long term evolution (LTE) network as infrastructure is a challenging problem due to the possible overload in the radio access network (RAN). The access class barring (ACB) scheme is an efficient and simple scheme which is proposed in the 3GPP documents to relieve the massive access by barring some M2M devices in each contention cycle. In this paper, we propose a learning automaton (LA) based algorithm to adaptively and dynamically adjust the ACB factor at evolved Node B (eNB) taking into account the number of collided preambles in the previous contention cycle. We show that by using an appropriate LA at the eNB, the system performance asymptotically converges to the optimal performance in which the eNB knows the number of access-attempting devices a priori. Simulation results are provided to show the performance of the proposed approach in adjusting the barring factor properly.