Modeling contention-based M2M transmissions over 3GPP LTE cellular networks

In this paper, we consider massive deployments of unattended wireless devices communicating over 3GPP Long Term Evolution (LTE) cellular networks. By contrast to conventional human-to-human connectivity which typically involves streaming of data, machine-to-machine (M2M) communication is often characterized by small-size and infrequent data patterns. With LTE technology, M2M devices may employ random-access based channels to connect to the network and update it on their current traffic demand to promptly deliver their data. However, existing literature has paid very limited attention to analytical characterization of such contention-based behavior when a large number of M2M devices transmit their dynamic (unsaturated) traffic. Therefore, in this work we propose a novel mathematical model that essentially captures contention in multi-channel environment and further apply it to several LTE-based M2M scenarios. As confirmed by extensive protocol-level simulations, our model is a powerful and precise tool that is able to provide important numerical insights into the performance of realistic M2M deployments.