Analog Coding for Bandwidth Expansion

This paper presents practical analog coding schemes for bandwidth expansion, specifically for 1:2 and 1:3 systems, based on successive quantization of the input samples and the corresponding residual errors. The proposed encoder is very simple, and can be optimized by properly choosing the required quantizers and by performing intelligent energy allocation. We study different decoding strategies, ranging from an involved optimal scheme to very simple straightforward direct decoding, and assess the trade-off between performance and decoding complexity. Insights into optimal energy allocation and quantization strategies are provided. By properly optimizing the system parameters, it is possible to obtain very reasonable performance, even for the straightforward decoding technique, which achieves a performance that, contrary to many other analog coding/decoding schemes, does not experience significant degradation relative to the theoretical limit when the signal to noise ratio increases.