As semiconductor manufacturing advances into ultra-scaled technology nodes, Static Random Access Memory (SRAM) verification faces critical challenges stemming from increasingly stringent geometries, heightened process variability, and the sheer density of contemporary designs. Existing methodologies, notably Design Rule Checking and conventional pattern matching, often fall short in capturing the intricate multilayer interactions and subtle geometric deviations characteristic of advanced SRAM layouts. These limitations can lead to missed layout anomalies and potential functional failures, resulting in diminished yield. This paper introduces a specialized pattern-matching-based algorithm designed to address these challenges in SRAM verification. The algorithm integrates novel localization techniques, efficient spatial indexing, and an overlap detection procedure that together achieve exact verification while delivering substantial improvements in runtime performance. Experimental results confirm that this approach not only maintains 100% detection accuracy on complex SRAM benchmarks but also achieves speedups ranging from threefold to over twentyfold compared to state-of-the-art methods, and from twofold to several thousandfold on classic single-layer pattern matching datasets.

As Integrated Circuit (IC) design complexity increases exponentially, Layout Pattern Matching (LPM) has become a critical technique for physical verification, yet faces significant challenges in efficiency and accuracy. This paper presents a high-performance two-stage LPM framework centered around the innovative GeoFA (Geometric Finite Automaton) engine. The first stage successfully applies Deterministic Finite Automaton (DFA) theory for the first time to 2D layout pattern matching with geometric tolerances, enabling rapid candidate region screening with near-linear time complexity. The second stage then performs precise geometric verification, leveraging optimized spatial indexing, robust Boolean operations, and specific optimizations for cell arrays, to guarantee 100% sign-off accuracy. Experimental results demonstrate that the framework achieves 100% accuracy across diverse layouts. On standard benchmarks, it provides a speedup of approximately 1.7x over state-of-the-art academic work. Furthermore, compared to the industry-standard tool Calibre, it delivers average speedups of up to 201.7x and 22.9x on complex single-layer and cell-level matching tasks, respectively, demonstrating exceptional performance and scalability.