Improved data-acquisition system for microprocessor-based applications using a least significant word algorithm

A novel technique to compress and expand a digitized signal is presented. The proposed methodology allows the storage memory to be cut in half and the sample frequency of the acquisition system to be doubled. It is useful in those cases in which the number of bits of the analog-to-digital converter is greater than the databus length of the microprocessor and the signal slew-rate is known. The proposed method also permits the clipping of unwanted spikes present in the signal. An application using a 12-bit analog-to-digital converter and an 8-bit microprocessor is given.<
>