In-Sensor Analytics and Energy-Aware Self-Optimization in a Wireless Sensor Node
With the proliferation of distributed sensors and Internet of Thing end-nodes, aggregate data transfer to the backend servers in the cloud is expected to become prohibitively large which not only results in network congestion, but also high energy expenditure a the sensor nodes. This motivates in-sensor data analytics that can perform context-aware acquisition and processing of data; and transmit data only when required. This paper presents a camera based wireless sensor node with in sensor computation, wireless communication and end-to-end system optimization. Depending on the amount of information content and the wireless channel quality, the system chooses the minimum energy operating-point by dynamically adjusting the processing depth (PD) and power amplifier (PA) gain. We demonstrate a complete end-to-end system and measure 3:7× reduction in energy consumption compared to a baseline design where only rudimentary image compression is performed.