Charge-coupled device (CCD) adaptive discrete analog signal processing

Abstract

A CCD adaptive signal processor is described which uses a so-called `clipped-data' least mean square (LMS) error algorithm to optimize the selection of tap weights in a CCD filter. A detailed description of a 16-tap monolithic silicon CCD analog adaptive filter is also presented. The filter is comprised of a basic linear combiner formed with a nondestructively tapped CCD analog delay line and electrically reprogrammable MOS analog conductances as the tap weights. To demonstrate the feasibility of adaptive analog signal processing, a 2-tap weight CCD adaptive filter is described and experimental results presented.