An adaptive channel selection scheme for reliable TSCH-based communication

Abstract

Energy consumption is one of the main issues of Internet of things applications. The IEEE 802.15.4 standard defines the medium access control and physical layer for low data rate wireless personal area networks with low energy consumption. The limitations of IEEE 802.15.4 in reliability are overcome by the IEEE 802.15.4e standard amendment defining three more MAC modes. Time slotted channel hopping (TSCH) is the most interesting one, providing high reliability. However, the link quality degradation still exists in the TSCH protocol, especially with high amount of interference and deep fading. This comes from the fact that the channel list in TSCH is fixed. In this paper we propose an adaptive channel selection scheme based on the multi-arm bandit problem. The selection of each channel is formulated as an independent process with an associated variable called the Gittins index. For optimal channel selection, the algorithm is modified to adapt to a varying environment by using a greedy strategy. The trade off between exploring all the channels to gather information and exploiting the selected good channels to avoid interference or fading, is investigated to achieve the best performance. Network simulation is done to explore and verify the algorithm design with NS-3. The throughput of TSCH with our channel selection scheme can achieve 85% of the ideal case, while TSCH with default hopping list can only achieve 54% in our simulation scenario.