Dithering Loopback-Based Prediction Technique for Mixed-Signal Embedded System Specifications

Lower cost test solutions are required to overcome high test cost issue of conventional characterization for the mixed-signal circuits, which is primarily caused by test equipment cost. This brief proposes a cost-effective self-characterization technique that accurately predicts the harmonics of individual mixed-signal circuits by dithering a loopback path with Gaussian-distributed noise. A set of scale factors is applied to the dithered root-mean-square noise. The different scale factors yield a corresponding change in the harmonic magnitudes of the loopback responses. Based on this, we derive a precise nonlinear loopback behavior model to quantitatively identify the harmonics of a device under test. The results show that the proposed method can be used for practical characterization.

I. Introduction

Conventional testing of mixed-signal circuits in a system on a chip (SoC) requires costly test equipment to accurately measure the characteristics of high-frequency small-amplitude signals [1]– [3].