X-FAB Extends Statistical Modeling Support for Analog/Mixed-Signal Processes

Monte Carlo simulation and circuit resizing with statistical and worst-case corners enables robust, high-yielding designs

X-FAB Silicon Foundries, the world's leading foundry group for analog/mixed-signal semiconductor applications, today completed the rollout of its statistical modeling suite for Monte Carlo simulation on its 0.6 and 0.35 micrometer analog/ mixed-signal technology platforms. Using it, designers gain full flexibility to accommodate all variations associated with designing and manufacturing analog/mixed-signal integrated circuits (ICs).  The statistical models enable their designs to be more robust and meet performance specifications across environmental corners.

Monte Carlo simulation covers matching and process variations. The simulation randomly generates values, according to predefined distributions for dedicated design parameters, to find solutions for complex design problems that cannot easily be solved. Sizing runs that include statistical and worst-case corners lead to optimized analog-mixed/signal designs.

X-FAB’s combination of safe operating area check, parasitic and, now, full statistical modeling support, enables design teams to perform design centering and optimization. This approach results in reliable, high-yielding designs and faster time to market.

According to Thomas Ramsch, director of design support at X-FAB, "Designing analog/ mixed-signal ICs is quite complicated and requires a depth of experience. We are pleased to offer full statistical modeling support because it helps take the mystery out of this process, leading to a robust solution and faster time to market for our customers."