Truly wearable physiological sensors, monitoring for example breathing or the electroencephalogram (EEG), require accurate and reliable algorithms for the automated analysis of the collected signal. This facilitates real-time signal interpretation and reduces the burden on human interpreters. It is well known that to reduce the total device power in many physiological sensors the automated analysis is best carried out using dedicated circuits in the sensor device itself, rather than transmitting all of the raw data and using an external system for the processing.

To allow the physiological sensor to operate from the physically smallest batteries and energy harvesters new algorithms optimized for low power operation are thus required. This results in designers being presented with new trade-offs between the algorithm performance (for example the number of correct detections of an event and the number of false detections) and the power consumption of the circuit implementation. This presentation explores the state-of-the-art algorithms and circuits for use in these situations, drawing on particular examples from algorithms and circuits for use in breathing monitoring and EEG analysis.