Inverse Gaussian distribution based timing analysis of Sub-threshold CMOS circuits

• An accurate and flexible delay model for sub-threshold CMOS circuits is proposed.
• Both combinational and sequential circuits are considered.
• Fan-out effect is also taken into account.
• High accuracy of our approach is achieved against SPICE Monte Carlo simulation.

Aggressive technology scaling and ultra low power constraints have resulted in less predictable device behavior complicating timing analysis/estimation. The traditional delay models fail to accurately capture the circuit behavior under such conditions. This paper proposes a novel highly accurate Inverse Gaussian Distribution (IGD) based delay model applicable to both combinational and sequential elements for sub-powered circuits. The IGD based delay estimation accuracy is demonstrated by evaluating multiple circuits, i.e., D Flip Flops (DFFs) + 8-bit Ripple Carry Adder, and 8-bit De-multiplexer (DEMUX) and Multiplexer (MUX). Our experiments indicate that the IGD based approach provides a high matching against HSPICE Monte Carlo simulation results, with an average error less than 1.9% and 1.2% for the two circuits, respectively, while sparing orders of magnitude simulation time. Moreover, the IGD model outperforms the traditional Gaussian Distribution (GD) model by providing 6 × better average accuracy with no extra simulation time overhead.