signal encoder, decoder and methods using predictive models

摘要

a signal encoder divides the signal into segments and uses prediction models to approximate samples of each segment each local prediction model, each applicable to a segment, is applied to its own system of translated axes within the segment and the offset is given by the last predicted value for the previous segment. when the signal is reasonably continuous, it alleviates the need to parameterize the displacement for each local predictor model, as each local predictor model can be based on the last predicted sample value of the previous segment. the encoder, as a consequence, does not accumulate errors even though the offset is not transmitted, but the last predicted value from the last sample of the previous segment is used. Prediction errors are obtained for the approximate samples transmitted to the decoder, along with the predictor model parameters and seed value to allow accurate signal reconstruction by the decoder.