MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641116271 A) filed by Sunil Begumpur; Poornima Badiger; Shruti Hiremath; Kiran Kotin; Shreedevi Lali; Supriya Shidling; Uzma Mujawar; Sadiqa Kushtar Attar; Sujata Betageri; and Veena Kurudagi on September 28, 2026, for System And Method For Energy-Aware Source Code Optimization Using Artificial Intelligence And Execution-Based Energy Analysis.
Inventors include Sunil Begumpur; Poornima Badiger; Shruti Hiremath; Kiran Kotin; Shreedevi Lali; Supriya Shidling; Uzma Mujawar; Sadiqa Kushtar Attar; Sujata Betageri; and Veena Kurudagi.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: An energy-aware source-code optimization system and method are disclosed. Source code and a representative workload are converted into a dependency-aware intermediate representation having identifiable executable regions. Synchronized execution traces and energy observations are collected during baseline execution, and an energy-attribution engine estimates region-level energy consumption with uncertainty values. An artificial intelligence transformation engine generates dependency- constrained candidate modifications for identified high-energy regions. A verification engine compiles and executes each candidate and evaluates functional equivalence, including defined outputs and observable side effects. A feedback controller compares repeated baseline and candidate executions under matched operating conditions and accepts a candidate only when a specified energy-improvement confidence criterion and resource constraints are satisfied. Accepted modifications are provided with measurement provenance and verification evidence, while rejected modifications are reverted. Accompanied Drawing [FIGS. 1-2].
Disclaimer: Curated by HT Syndication.
