MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202641091836 A) filed by Sr University on July 29, 2026, for An Intelligent System And Method For Early-Stage Iot Botnet Detection Using Spatiotemporal Network Traffic Features.
Inventors include Ms. B. Koteswari; and V. Thirupathi.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: AN INTELLIGENT SYSTEM AND METHOD FOR EARLY-STAGE IOT BOTNET DETECTION USING SPATIOTEMPORAL NETWORK TRAFFIC FEATURES ABSTRACT The invention discloses an intelligent system and method for detecting Internet of Things botnets at an early stage by analyzing spatiotemporal characteristics of network traffic. The system continuously captures communication flows generated by heterogeneous IoT devices and extracts spatial features representing interactions among devices, endpoints, ports, protocols, and network segments, together with temporal features describing packet rates, flow duration, burst patterns, request intervals, and behavioral changes over time. A preprocessing module filters noise, normalizes traffic records, and constructs feature sequences for real-time evaluation. An intelligent detection engine applies machine learning or deep learning models to identify deviations from legitimate device behavior and recognize coordinated activities associated with botnet formation, command propagation, scanning, and initial compromise. The system may dynamically update behavioral profiles to accommodate evolving devices and network conditions while reducing false alerts. Upon detecting suspicious activity, a response module generates risk scores, alerts administrators, and initiates isolation or traffic restriction measures. The proposed approach enables scalable, adaptive, and low-latency detection before large-scale malicious operations occur, thereby improving IoT network resilience, security monitoring, and proactive threat containment across connected environments.
Disclaimer: Curated by HT Syndication.
