A statistical analysis of adaptive quantization based on causalpast

摘要

A statistical estimation framework is proposed for adaptivequantization based on causal past. Different estimation methods aregiven for the marginal density based on the quantized sample. For astationary and ergodic source process, if its marginal density is in aparametric family with a dimension less than the quantization level,then “adaptation” can be achieved when the sample size islarge, i.e., the marginal density can be estimated consistently