MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202621065306 A) filed by Vinay Keswani; Dr. Pushpinder Singh Patheja; Prof. Gajanan Nanaji Tikhe; Dr. Vikas Rameshrao Palekar; Dr. Kavita Rawat; Chandu Dajiba Vaidya; and Mr. Mukesh Madhukar Tarone on May 23, 2026, for A Self-Learning Iot Framework For Dynamic Resource Allocation In Smart Environments.
Inventors include Vinay Keswani; Dr. Pushpinder Singh Patheja; Prof. Gajanan Nanaji Tikhe; Dr. Vikas Rameshrao Palekar; Dr. Kavita Rawat; Chandu Dajiba Vaidya; and Mr. Mukesh Madhukar Tarone.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: The present invention discloses a self-learning IoT framework for dynamic resource allocation in smart environments configured to continuously monitor, analyze, and optimize the distribution of computational, network, and energy resources across heterogeneous IoT deployments in real time. The framework comprises a machine learning-based resource management engine, a distributed sensing module for continuous acquisition of device and environmental data, an adaptive workload classification and prediction module, and a programmable task scheduling engine configured to dynamically allocate resources across IoT edge nodes and cloud infrastructure. The proposed framework incorporates self-learning algorithms capable of detecting workload imbalances, compute burst events, network congestion, and energy-critical conditions with high accuracy, and automatically adjusts resource allocation parameters including CPU time slices, bandwidth assignments, task priorities, and energy budgets based on environment-specific operational conditions. The framework further includes a wireless communication interface for remote monitoring, administrator configuration, and over-the-air updates, enabling secure bidirectional data exchange with an external management platform. An integrated energy management unit with adaptive duty cycling and battery health monitoring extends operational longevity of resource-constrained IoT devices. The system also incorporates safety verification logic and fail-safe fallback mechanisms to prevent inappropriate resource starvation or system unavailability. The invention provides enhanced personalization of resource scheduling, improved energy efficiency, reduced manual configuration overhead, and increased system reliability compared to conventional static IoT resource management approaches. The adaptive self-learning framework enables responsive resource management under dynamically varying environmental and operational conditions, thereby improving smart environment performance, sustainability, and long-term operational stability
Disclaimer: Curated by HT Syndication.
