The case for using traffic forecasting in schedule-based channel access

In this paper, we explore the idea of using traffic forecasting to improve the delay performance of a schedule-based medium access control protocol. Schedule-based channel access has been shown to utilize network and energy resources efficiently but is often hindered by the extra delay that scheduling introduces. We explore the use of traffic forecasting to anticipate transmission schedules instead of establishing them reactively, thereby reducing scheduling delays. We show the potential performance benefits traffic forecasting can bring to schedule-based medium access in the context of an existing MAC protocol called DYNAMMA. Preliminary results using a machine-learning based traffic forecasting technique are also presented.