Our interest is in analog signal processing as it might be applied to NDE, carrying out some sophisticated signal processing using an inexpensive, real-time analog system based on the surface acoustic- wave technology, or perhaps the CCD technology. To date we have considered an inverse filter for correcting for the frequency response of an NDT transducer: we have shown the feasibility of .making such a device and will show some experimental results here. We are making and testing some filters that have been designed to go with a commercial NDT transducer that we have purchased. We are also looking at the possibility of realizing the same sort of filter with a charge-coupled device. In the next year we should like to look at a more complicated analog signal processor - the convolver or correlator - to see how it can be realized and applied to defect signature recognition. At the end of my talk I'd like just to mention some work being done by another faculty member at Berkeley on an interesting receiving transducer which I don't think has come to the attention of this particular audience.