Study and Performance Analysis of LTE MAC Schedulers for M2M

Abstract

Cellular systems are forecasted to play a fundamental role in the future Machine-to-Machine scenario. The 3GPP LTE networks appear to be the de facto standard for machine type communications. Beside opportunities given by M2M devices spreading, such as vehicular-to-vehicular communications and environmental monitoring, the operators will have to deal with a greater number of devices connected, which don t fulfil with the nowadays human centred traffic standard characteristics and will force to re-standardize the current uplink scheduling procedures. In this report an extensive study of the literature papers, both standards and M2M centred, is carried on a dense machine scenario. Problems such as Human-to-Human throughput drops are high lined with the raise of the number of M2M devices present. Moreover the channel utilization, both in uplink and in downlink, drops drastically among the studied scheduler schemes. A new M2M Aware Scheduler, with the goal of maximise the medium utilization, is designed and implemented. Its results over the simulations shows that M2M and Human-to-Human devices could live over the same LTE Network but new schedulers, such as the one here presented, have to be further studied and analysed. This thesis work was carried on during an exchange period at the Norwegian University of Science and Technology in Trondheim with the collaboration and supervision of Telenor Norge AS.