A system-level approach to power/performance optimization in wearable computers

The paper describes a system-level design approach to the power and performance of Carnegie Mellon's wearable computers dedicated to speech processing-the Speech Translator Smart Modules. While processor speed and type affect power consumption and performance, memory size and type of secondary storage have a significant influence as well. A system-level approach to power/performance optimization is described that improved the metric of (performance/(weight*volume*power)) by over a factor of 300 through the four generations of wearable computers for speech processing.