MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202621078462 A) filed by Dr. Dhanshyam Anokhilal Waghmare; Dr. Shubhangi Subhash Borekar; Dr. Sheetal Babanrao Vidhate; Dr. Kavita Mukund Meshkar; Ankita Sudhakar Wankhade; and Shrikant Gangadhar Nikam on June 25, 2026, for Artificial Intelligence-Driven Personalized Learning Systems: Enhancing Student Engagement And Academic Performance.
Inventors include Dr. Dhanshyam Anokhilal Waghmare; Dr. Shubhangi Subhash Borekar; Dr. Sheetal Babanrao Vidhate; Dr. Kavita Mukund Meshkar; Ankita Sudhakar Wankhade; and Shrikant Gangadhar Nikam.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: A hardware-secured adaptive learning system (100) integrates an FPGA-based inference accelerator (106) with tamper-proof audit trail generation for personalized educational content delivery. A multimodal sensor array (102) captures visual, acoustic, and physiological learner data processed by a sensor fusion controller (104) into composite feature vectors transmitted through an interconnect firewall (132) to the FPGA-based inference accelerator (106). The accelerator (106) executes an engagement prediction pipeline (108) with feature extraction (138), classification (140), content selection (142), and commit (144) stages governed by a finite state machine controller (146). An IOMMU controller (112) with memory protection unit registers (114) write-locked by a hardware latch (128) enforces isolation of a learner profile memory region (110). An attestation engine (124) rooted in a physically unclonable function module (120) signs each classification output, committed sequentially through a monotonic counter (122) to an immutable audit store (126), ensuring tamper-proof accountability for all adaptive learning decisions.
Disclaimer: Curated by HT Syndication.
