MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621096800 A) filed by Lakshmi Narain College Of Technology Excellence on August 11, 2026, for System And Method For Variability-Aware Optimization Of Static Random-Access Memory.
Inventors include Dr. Deepika Sharma; and Satyarth Tiwari.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: ABSTRACT SYSTEM AND METHOD FOR VARIABILITY-AWARE OPTIMIZATION OF STATIC RANDOM-ACCESS MEMORY The present invention describes a system and method for variability-aware design optimization for static random-access memory. Information about design, variability, and operating information as well as responses are acquired at varying levels of fidelity for memory cell configurations that can be potential candidates. The correlation preserving variability state preserves the statistical correlation between the variations and operating conditions. The multi-fidelity performance prediction model produces probability distributions and uncertainty for the performance. A particular memory cell configuration and evaluation fidelity level are chosen based on uncertainty and expected uncertainty reduction at a particular operating constraint. The evaluation resource evaluates the configuration and the cross-fidelity residual updates the performance prediction model. Memory cell parameters and semiconductor process or interconnect parameters are changed in an iterative process, whereas confidence bounded constraints screen out inappropriate memory cell configurations. A configuration not screened out undergoes high-fidelity verification. Information about implementation that defines the optimized memory cell configuration and compatible manufacturing window is revealed only when the verified response falls in the prediction interval and meets the operating constraints. FIG. 3
Disclaimer: Curated by HT Syndication.
