Degree Type

Dissertation

Date of Award

2013

Degree Name

Doctor of Philosophy

Department

Electrical and Computer Engineering

First Advisor

Degang Chen Chen

Abstract

For several decades, technology scaling has brought many orders of magnitude improvements in digital CMOS performance and similar economic benefits to consumers. Feature size is quickly approaching nanometer scale, and the associated large variability imposes grand challenges in achieving reliable and robust operation. This is especially so for high-precision analog and mixed-signal circuits since they have always relied on accurate device matching which will not be available in nanometer CMOS or emerging technologies. This dissertation is aiming to develop design methodologies for overcoming such grand challenges without the conventional matching requirements. The underlining hypothesis is that, from a population of devices with significant variability, correct interconnection and sequencing can produce an effective system level matching that is several orders of magnitude better than the original devices. The optimal solution is non-deterministic polynomial-time hard but a simple ordered element matching strategy based on ordered statistics produces dramatically improved matching. Practical implementation of the new matching strategy is demonstrated on a 15-bit binary-weighted current-steering digital-to-analog converter design in a 130nm CMOS technology. The core area of the chip is less than 0.42mm2, among which the MSB current source area is well within 0.021mm2. Measurement results have shown that the differential nonlinearity and integral nonlinearity can be reduced from 9.85LSB and 17.41LSB to 0.34LSB and 0.77LSB, respectively.

Copyright Owner

Tao Zeng

Language

en

File Format

application/pdf

File Size

127 pages

Share

COinS