Adaptive sensing for energy-efficient manufacturing system and process monitoring

Continued advancement in telecommunication and computing power has accelerated the use of wireless sensors in monitoring of a wide range of manufacturing systems and processes. In most scenarios wireless sensors sample and transmit measured data continuously at a fixed sampling rate. This is a suboptimal method of operation from the perspectives of data and power management. Continuous sampling and transmission limits the service life of the wireless sensors and sensor networks because of the limited energy storage capacity of the power source. Furthermore, high rate sampling of process related signals that only exhibit high frequency characteristics for short durations interspersed with long durations of low frequency content leads to data redundancy and processing overheads. To address these limitations this paper presents an adaptive sensing technique consisting of a novel adaptive sampling algorithm that dynamically adjusts the data sampling rate to reduce data redundancy and improve energy efficiency. Experimental evaluation of this technique on data microprocessor based wireless sensor node confirms the validity of this adaptive and reconfigurable sensing method.