DRIFT COUNTERACTION IN ODOR RECOGNITION APPLICATIONS - LIFELONG CALIBRATION METHOD

Sensor drift is addressed as one of the most serious impairments affli cting chemical and biochemical sensors. One possible solution to this problem is to view sensor arrays as time-varying dynamic systems, whos e variations have to be tracked by adaptive estimation algorithms. A t heory of hidden variable dynamics for the rejection of common mode dri fting of sensors has previously been developed and is here coupled wit h a recursive least squares algorithm. In Section 7 a smart solution i s provided for a difficult vapour recognition problem vexed by drift t hat have failed with traditional pattern recognition techniques. Among the many advantages we distinguish that model adaptation to changes i n the sensor array makes lifelong calibration possible without interru pting the operation of the array. (C) 1997 Elsevier Science S.A.