Intellectual Property India Publishes Patent Application for Context-Aware Application Sandboxing Framework Using Behavioral Pattern Recognition For Malware Detection

MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641096254 A) filed by Thiyagarajan T; Dr. Kalpesh. H. Vandra; Dr. Jayasri Indiran; Dr. M. Sandeep Kumar; Dr. Gyana Ranjana Panigrahi; Dr. M. Sathis Kumar; Mr. Sabarivasan A; Dr. Adalarasu B; Dr. Mamatha B; Dr. Suvendu Kumar Nayak; and Dr. Lalmohan Pattnaik on August 10, 2026, for Context-Aware Application Sandboxing Framework Using Behavioral Pattern Recognition For Malware Detection.

Inventors include Thiyagarajan T; Dr. Kalpesh. H. Vandra; Dr. Jayasri Indiran; Dr. M. Sandeep Kumar; Dr. Gyana Ranjana Panigrahi; Dr. M. Sathis Kumar; Mr. Sabarivasan A; Dr. Adalarasu B; Dr. Mamatha B; Dr. Suvendu Kumar Nayak; and Dr. Lalmohan Pattnaik.

The application for the patent was published on August 14, 2026, under issue no. 33/2026.

Abstract: A context-aware application sandboxing system and method for malware detection are disclosed. A context analysis module (110) extracts contextual attributes of a target application (190) prior to execution, including application source, requested permissions, and deployment environment. A dynamic sandbox configuration module (120) instantiates an isolated execution environment (122) configured according to the extracted context, while an evasion countermeasure module (160) injects simulated user interactions and decoy artifacts to resist sandbox-detection techniques employed by malware. A behavioral monitoring module (130) captures multi-layer telemetry during execution, which a pattern recognition engine (140) converts into a behavioral pattern score using trained classification models. A context-behavior correlation engine (150), referencing a context cluster baseline repository (152), computes a context-relative anomaly score, enabling a decision and response module (170) to apply graduated responses. A feedback module (180) continuously updates baselines and retrains models, improving detection accuracy while reducing false positives and evasion susceptibility.

Disclaimer: Curated by HT Syndication.