Neural-network compensation methods for capacitive micromachined accelerometers for use in telecare medicine

Transducers represent a key component of medical instrumentation systems. In this paper, sensors that perform the task of measuring the physical quantity of acceleration are discussed. These sensors are of special significance since, by integrating their output signals, accelerometers can additionally provide measures of velocity and position. Applications for such measurements, and thus of accelerometers, range from early diagnosis procedures for tremor-related diseases (e.g. Parkinson's disease) to monitoring daily patterns of patient activity using telemetry systems. The system-level requirements in such applications are considered, and two novel neural-network transducer designs developed by the authors are presented, which aim to satisfy such requirements. Both designs are based on a micromachined sensing element with capacitive signal pickoff. The first is an open-loop design utilizing a direct-inverse control strategy, while the second is a closed-loop design, where electrostatic actuation is used as a form of feedback. Both transducers are nonlinearly compensated, capable of self-testing, and provide digital outputs.