MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202541010118 A) filed by Centre Of Excellence In Wireless Technology; and Indian Institute Of Technology Madras Iit Madras on February 06, 2025, for Methods And Apparatuses For Managing Ground Truth Label Acquisition For Artificial Intelligence/Machine Learning Based Positioning In A Wireless Network.
Inventors include Jishnu Ponnappan; Abhijeet Abhimanyu Masal; Anusuya Balasubramanian; Jeniston Deviraj Klutto Milleth; and Bhaskar Ramamurthi.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: ABSTRACT METHODS AND APPARATUSES FOR MANAGING GROUND TRUTH LABEL ACQUISITION FOR ARTIFICIAL INTELLIGENCE/MACHINE LEARNING BASED POSITIONING IN A WIRELESS NETWORK A method for subscription-based GTL acquisition and AI/ML model training at an NG-RAN node is disclosed. The method for collecting GTL of UE by next generation random access network (NG-RAN) node to train and monitor AI/ML model used for user equipment (UE) positioning, comprises sending, by the NG-RAN to an access and mobility management function (AMF), a subscription request to obtain a ground truth label (GTL) associated with at least one UE connected to the NG-RAN node, obtaining, by the NG-RAN, at least one radio signal measurement from a target UE during a positioning session associated with a location service (LCS) request, receiving, by the NG-RAN, from a location management function (LMF), the GTL corresponding to the target UE based on the AMF indication to the LMF to share the GTL with the NG-RAN and training, by the NG-RAN, the AI/ML model for UE positioning by using at least one of the GTL and at least one radio signal measurement. Fig. 4
Disclaimer: Curated by HT Syndication.
