Revisiting Probabilistic Schedule-Based Asynchronous Duty Cycling

Schedule-based asynchronous duty cycling is a simple duty cycling technique that requires no synchronization, control traffic or expensive computation. Thus, it is an interesting choice for WSNs. However, most of the literature on the design of asynchronous schedules has been limited by the exaggerated restriction that the schedules present a property named rotation closure. While more flexible probabilistic schedules have been proposed, they were considered inferior under the argument that they do not provide an upper bound on the communication latency. In this paper we revisit the probabilistic schedules with the goal of showing they may be the preferred method in several scenarios. We argue that under realistic assumptions no schedule can guarantee a latency upper bound. We show that probabilistic schedules better support asymmetric duty cycle operation, while also providing unlimited duty cycle granularity and range. Finally, by means of simulation, we compare a probabilistic schedule belonging to the family of Birthday Protocols, the Birthday–Listen–and–Transmit (BLT), with traditional deterministic schedules. Results show that, on top of its other qualitative advantages, BLT achieves competitive latencies.