LAUSR

Poor signal-to-noise ratios (SNRs) and low spatial resolutions have impeded low-cost pyroelectric infrared (PIR) sensors from many intelligent applications for thermal target detection/recognition. This article presents a cognitive signal conditioning and modulation learning framework for PIR sensing with the following two innovations to solve these problems: 1) a reconfigurable signal conditioning circuit design to achieve high SNRs and 2) an optimal sensor mask design to achieve high recognition performance. By using a programmable system on chip, the PIR signal amplifier gain and filter bandwidth can be adjusted automatically according to working conditions. Based on the modeling between PIR physics and thermal images, sensor masks can be optimized through training convolution neural networks with large thermal image datasets for feature extraction of specific thermal targets. The experimental results verify the improved performance of PIR sensors in various working conditions and applications by using the developed reconfigurable circuit and application-specific masks.