MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202611087134 A) filed by Noida Institute Of Engineering And Technology Niet on July 16, 2026, for Adaptive Multi-Modal Data Ingestion And Synthetic-To-Real Training System For Wireless Channel Prediction.
Inventors include Ranjana Singh; and Rahul Agarwal.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: A system and method for adaptive multi-modal data ingestion and synthetic-to-real domain transfer training for wireless channel prediction in sixth-generation communication networks is disclosed. The system comprises a multi-modal data ingestion pipeline that synchronizes and aligns sensor data from cameras, LiDAR, radar, and GNSS receivers with measured channel impulse responses. A synthetic channel data generation module employs environment-aware ray-tracing with four- category stochastic augmentation. A domain adaptation bridge uses adversarial feature alignment and confidence-weighted sample selection to bridge the synthetic-to- real distribution gap. A multi-modal channel prediction engine fuses visual, geometric, radar, and positional features through cross-modal transformer attention. An active learning feedback module estimates prediction uncertainty using Monte Carlo dropout and schedules targeted measurements for continuous accuracy improvement.
Disclaimer: Curated by HT Syndication.
