Startup plans neural network circuit for low-power wireless sensors
The company is aiming at improving battery life in wireless sensor applications by reducing the communication and data storage requirements. The chip architecture allows detection and identification of randomly distributed events. The circuit will have digital configuration of a fully analog data path.
ACS was awarded an STTR Phase I grant in July 2014 for the development of a configurable analog circuit for use with myoelectric prosthetics. A portion of the effort will be done in collaboration with Indiana University Purdue University Indianapolis Biomedical Engineering Department.
The company has numerous examples of where such an analog network can help save power consumption from neural spike detection in prosthetic devices to heart arrhythmia detection to engine misfire detection and on to gesture recognition and multi-sensor signature detection.
Analog Computing Solutions: www.analogcomputingsolutions.com