Bit-level lattice implementation of enhanced adaptive predictors

The objective of this thesis study is to investigate the performance characteristics of a lattice structure implementation of the cascade forward adaptive predictor and the cascade forward-backward adaptive predictor in the presence of bit quantization distortion. Previous research has demonstrated that the mentioned predictors have proven effective in combating the misadjustment and eigenvalue spread effects. This thesis proves that the particular lattice implementation of the cascade forward adaptive predictor and the cascade forward-backward adaptive predictor is less affected by quantization distortion. This result translates into better performance in the speed of convergence and the lessening of the misadjustment at reduced bit quantization levels. In applications where a trade off between signal-to-noise ratio and bit rate is of importance, immunity to bit quantization leads to an increased signal-to-noise ratio at a required bit rate, or a decreased bit rate for a required signal-to-noise ratio. The chosen application of adaptive differential pulse code modulation demonstrates this is the case, where the lattice implementation demonstrates a better performance in the gain prediction of speech signals.