User Behavior Driven MAC Scheduling for Body Sensor Networks: A Cross-Layer Approach

In deploying body sensor networks (BSNs), sampling rates might be dynamically tuned to fit application requirements (e.g., monitoring patients' different activities), which helps conserving energy for battery-powered sensors. However, this results in variable data rates among sensors, which further requires an efficient resource allocation to maintain reliable transmission accommodating all traffic loads. We thereby address this joint problem of transmission reliability and energy efficiency by proposing a BSN system that autonomously detects user behaviors, which, in turn, trigger dynamic sampling and resource scheduling via an adaptive MAC scheduling scheme. This cross-layer scheme uses time-slotted channel hopping (TSCH) in IEEE 802.15.4, which is a reliable low-power MAC protocol. Specifically, the proposed solution determines the best TSCH slotframe length for specific application requirements and the number of timeslots to be added/removed according to dynamic sampling rates, and then allocates timeslots via an equally spaced timeslot allocation algorithm. We implement our proposed approach on a BSN testbed, which features both state-of-the-art hardware and software architectures. The experimental results are conducted to evaluate our proposed solution in terms of throughput, packet delivery ratio, and energy per bit, which demonstrates that our cross-layer solution ensures reliable data transmission and energy efficiency compared to existing techniques.