Adaptive prediction for DPCM coding of TV signals

Abstract

An adaptive minimum mean-squared error prediction algorithm has been derived for DPCM coding of video signals. Its parallel structure guarantees the possibility of high-speed implementation. Simulations show that there is only little gain in terms of signal-to-noise ratio as compared to optimum fixed prediction. However, if the autocorrelation function of the input is not known a priori and there is a mismatch in the prediction coefficients, the gain due to a fixed predictor experiences a severe dropoff, and adaptive prediction proves superior.