Application of Q-Learning for RACH Access to Support M2M Traffic over a Cellular Network

This paper considers the coexistence of Machine-to-Machine (M2M) and Human-to-Human (H2H) based traffic sharing the Random Access Channel (RACH) of an existing cellular network. A novel combined RACH access scheme is proposed to control M2M traffic in order to reduce its impact on a cellular network. A Q-learning RACH access scheme (QL-RACH) which allows interaction of M2M with H2H via a Slotted Aloha RACH access (SA-RACH) scheme is presented. The QL-RACH access scheme uses an intelligent slot assignment strategy in order to avoid collisions amongst the M2M users and is generic and compatible with all cellular network standards. The learning is applied so that no central entity is involved in the slot selection process, to avoid tampering with the existing network standards. Simulation results show that this approach overcomes the negative impact of M2M traffic on existing H2H traffic in the RACH access contest and improves the total RACH-throughput to 55%.