Activity management in battery-powered embedded systems: A case study of ZigBee® WSN

Portable battery-powered embedded systems necessitate sustainable energy-aware computing. For energy-efficient realization of such systems, static and possibly dynamic optimizations need to be applied to both the hardware and software design abstraction layers. In this paper, models for estimating the energy consumption based on application “activity” are proposed. By “activity” we mean the rate at which the computing platform performs a set of predefined application functions; and managing them aims at extending battery life. The proposed models can be used for both static and/or dynamic (on the fly) design/exploration of such systems. In order to validate the models, a generic low duty-cycle ZigBee^® Wireless Sensor Network (WSN) application has been considered as a case study. Experimental results confirm fair accuracy for the proposed models (3% error on average); based on comparisons of estimated values with those obtained from experimentations.