A unified framework for energy efficient physical activity tracking

Abstract

A unified framework of joint state tracking and control design is proposed for energy-efficient physical activity tracking in heterogeneous Wireless Body Area Networks (WBANs). The objective is to devise sensor selection strategies for the WBAN's fusion center to optimize the trade-off between tracking performance and energy consumption. Our recently proposed Kalman-like estimator is employed for state tracking. The associated mean-squared error and an appropriate energy consumption metric are used in a partially observable Markov decision process formulation to derive the optimal selection strategy. A low-complexity suboptimal strategy is also proposed. Numerical results are provided using real WBAN experimental data.

Markov Decision Processes With Applications in Wireless Sensor Networks: A Survey

“…The action space includes the command for activating or switching off sensors. Similarly, Zois and Mitra [121] introduced a model for activity detection in wireless body area networks (WBANs). The network is composed of heterogeneous nodes, and the node selection is optimized for maximum energy saving and max-Wall A node with sectorized antenna A node with sectorized antenna Sensor's coverage area Uncovered area due to wall effects Fig.…”