A Study on Real-Time Scheduling for Low-Power Sensor Node Platforms

Abstract

This paper presents a technique to integrate energyaware real-time scheduling into sensor node platforms from the perspective of the combined periodic (or time-bounded) and aperiodic (or time-unbounded) scheduling policy. In the scheduling policy, the first objective is to meet the timing constraints of periodic tasks as well as to improve the response time of aperiodic tasks. The second objective is to minimize the energy consumption in a battery-powered sensor node without missing the timing constraints of periodic tasks. Experiments showed that the proposed scheduling technique achieved efficient performance in terms of the minimal deadline miss ratio of periodic tasks, a fast average response time of aperiodic tasks, and low power consumption.