MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641106360 A) filed by Presidency University on September 03, 2026, for Machine Learning-Assisted Adaptive Hybrid Counter System For Metastability-Resilient Clock Domain Crossing.
Inventor includes Aruna M.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: ABSTRACT Machine Learning-Assisted Adaptive Hybrid Counter System for Metastability-Resilient Clock Domain Crossing The machine learning-assisted adaptive hybrid counter system (100) for metastability-resilient clock domain crossing is disclosed. The system comprises the following: an offline training unit (102), and a machine learning inference engine (104). An adaptive clock controller (106) provides clock gating and dynamic voltage and frequency scaling to a multi-mode hybrid counter (108) selectively employing binary, Gray-code, or toggle counting. A transition activity monitor (110) extracts Hamming distance and toggle activity, while a CDC interface wrapper (112) transfers counter data between clock domains using handshake and FIFO mechanisms. A multi-stage synchronizer (114) provides metastability settling, and a metastability monitor (116) detects unstable synchronization conditions. An error correction unit (118) corrects erroneous counter data. An output register bank (120) and final output bus (122) provide corrected data and feedback for adaptive control, thereby reducing switching activity and dynamic power consumption while improving data-transfer reliability.
Disclaimer: Curated by HT Syndication.
