Bayesian Compressed Sensing for IoT: Application to EEG Recording

Abstract

Internet of Things (IoT) is a hot research topic since several years. IoT has gained a large interest in many application fields such as digital health, smart agriculture or industry. The main focus of the IoT community remains the design of appropriate applications and performant connected objects. In this paper, we address this topic from a signal processing viewpoint. We propose a model to perfom compressed sensing with connected objects where energy and communication constraints araise. The proposed model is formulated in a Bayesian framework and promising results demonstrate its potential in application to EEG signal recording from a connected MindWave device.