Publication
Approximate computing is an emerging paradigm for designing error-resilient applications. It reduces circuit area, power, and delay at the cost of allowingintroducing errors. This paper introducesproposes a powerful technique, termed Approximate Resubstitution (AppResub), to approximately simplify the circuit. AppResub re-expressesreplaces a node's function with a simpler approximate function onusing existing nodes in the circuit, by replacing it with a simpler approximate function to reduce the hardware cost. Leveraging AppResub, an efficient flow for approximate logic synthesis (ALS) is developed by iteratively applying a set of promising AppResubs for circuit simplification. To evaluate errors caused by a set of AppResubs, a novel error model capable of efficiently computing an error upper bound is used to smartly apply AppResubs in the ALS flow. The experimental results demonstrate that, compared to a state-of-the-art method, the proposed flow further reduces 20.9% area and 21.7% delay under the mean error distance constraint, making itwhile being 400× faster. The code of our flow is open-source.