Testing On-Die Process Variation in Nanometer VLSI

Abstract

Ring oscillators are not new, but the authors of this article use them in a novel, unconventional way to monitor process variation at different regions of a die in the frequency domain. Measuring the variation of each design or fabrication parameter is infeasible from a circuit designer's perspective. Therefore, we propose a methodology that approaches PV from a test perspective. This methodology advocates testing dies for process variation by monitoring parameter variations across a die and analyzing the data that the monitoring devices provide. We use ring oscillators (ROs) to map parameter variations into the frequency domain. Our use of ROs is far more rigorous than in standard practices. To keep complexity and overhead low, we neither employ analog channels nor use zero-crossing counters. Instead, we use a frequency domain analysis because it allows compacting RO signals using digital adders (thereby also reducing the number of wires), and decoupling frequencies to identify high PVs and problematic regions. Our PV test methodology includes defining the PV fault model; deciding on types, numbers, and positions of a small distributed network of frequency-sensitive sensors (RO) and designing an efficient, fully digital communication channel with sufficient bandwidth to transfer sensor information to an analysis point. With this methodology, users can trade off cost and accuracy by choosing the number or frequency of sensors and regions on the die to monitor