MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641051551 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on April 22, for 'system and method for continuous intent-based access control in zero trust networks using sequential learning.'
Inventor(s) include Rampriya R S; Vedansh Gupta; Rounak Agarwal; and Jerart Julus L.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a system (100) and method for continuous intent-based access control in a network environment. The system comprises a data engineering layer (102) configured to process enterprise telemetry into chronological sequences using a temporal sliding window. A deep sequential learning module (104) processes the sequences using an autoencoder architecture to compute reconstruction error. An adaptive enforcement layer (106) applies extreme value modelling to determine a dynamic threshold based on the reconstruction error. A sequential error localization module within the adaptive enforcement layer (106) computes localized reconstruction errors at individual timesteps. An external enterprise network orchestration module (108) receives anomaly-related data and interfaces with network control components. The invention enables continuous sequential analysis and dynamic threshold-based evaluation of telemetry data within a unified architecture."
Disclaimer: Curated by HT Syndication.
