MUMBAI, India, Sept. 7 -- Intellectual Property India has published a patent application (202641104348 A) filed by Rahul Tavva on August 31, 2026, for Ai-Driven Network Intelligence System For Predictive Fault Detection And Remediation.
Inventor includes Rahul Tavva.
The application for the patent was published on September 04, 2026, under issue no. 36/2026.
Abstract: AI-DRIVEN NETWORK INTELLIGENCE SYSTEM FOR PREDICTIVE FAULT DETECTION AND REMEDIATION ABSTRACT An AI-driven network intelligence system is disclosed for predicting, detecting, diagnosing, and remediating faults across heterogeneous communication networks. The system continuously collects telemetry, traffic statistics, event logs, configuration data, topology information, and performance indicators from distributed network elements. A data processing engine cleans, correlates, and transforms the collected information into temporal and contextual feature sets. Machine learning models analyze the feature sets to identify abnormal behavior, forecast probable failures, estimate fault severity, and determine affected devices, links, services, or users. A reasoning engine combines model outputs with topology relationships, operational policies, historical incidents, and dependency mappings to generate root-cause hypotheses and recommended corrective actions. Based on confidence thresholds and predefined authorization rules, an orchestration module automatically executes remediation procedures, including configuration adjustment, traffic rerouting, resource reallocation, service restart, or isolation of defective components. The system validates remediation outcomes, measures network recovery, and feeds results back to the models for continuous learning. A visual interface presents risk scores, predicted fault timelines, explanations, and audit records, thereby enabling proactive maintenance, reduced downtime, improved service reliability, and efficient network operations with minimal intervention.
Disclaimer: Curated by HT Syndication.
