Low power signal processing architectures for network microsensors

Abstract

Low power signal processing systems are required for distributed network microsensor technology. Network microsensors now provide a new monitoring and control capability for civil and military applications in transportation, manufacturing, biomedical technology, environmental management, and safety and security systems. Signal processing methods for event detection have been developed with low power, parallel architectures that optimize performance for unique sensor system requirements. Implementation of parallel datapaths with shared arithmetic elements enables high throughput at low clock rate. This method has been used to implement a microsensor spectrum analyzer for a 200 sample/s measurement system. This 0.8 /spl mu/ CMOS device operates with a 1 /spl mu/A drain current at a 3 V supply bias.