Energy-efficient signal acquisition in wireless sensor networks: a compressive sensing framework

The sampling rate of the sensors in wireless sensor networks (WSNs) determines the rate of its energy consumption since most of the energy is used in sampling and transmission. To save the energy in WSNs and thus prolong the network lifetime, we present a novel approach based on the compressive sensing (CS) framework to monitor 1-D environmental information in WSNs. The proposed technique is based on CS theory to minimize the number of samples taken by sensor nodes. An innovative feature of our approach is a new random sampling scheme that considers the causality of sampling, hardware limitations and the trade-off between the randomization scheme and computational complexity. In addition, a sampling rate indicator (SRI) feedback scheme is proposed to enable the sensor to adjust its sampling rate to maintain an acceptable reconstruction performance while minimizing the number of samples. A significant reduction in the number of samples required to achieve acceptable reconstruction error is demonstrated using real data gathered by a WSN located in the Hessle Anchorage of the Humber Bridge.