Design of product polar quantizers for A/D conversion of measurement signals with Gaussian distribution

Abstract

In this paper, a design of product polar quantizers for A/D (analog-to-digital) conversion of measurement signals and an analysis of their performances in the wide range of variances are considered, in a general manner, i.e. for any compression function. Special attention is devoted to the design of robust product polar quantizers, applying the developed theory on μ-law logarithmic compression function, which is very important since a lot of measurement signals are nonstationary. The proposed quantizers achieve much better performances than scalar quantizers (they have higher the maximal SQNR (signal-to-quantization noise ratio) for about 2.5dB and they are more robust), without increasing of complexity. Theory is proved by simulation. These quantizers can be applied for many different types of measurement signals and can be used in many measurement systems (telemetry, telemedicine, wireless sensor networks, etc.).