MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641043099 A) filed by St. Peter'S Engineering College; A Narmada; Dr K Viswanath Allamraju; Arikatla Sireesha; Dr. Venkateswarulu Naik. B; Y B Avinash; Mr. K. Shyama Satya Naga Teja; Bhargavi Marri; Raman Kumar M; and Malla Reddy Mr Deemed To Be University on April 04, 2026, for Ai-Optimized Dynamic Spectrum Allocation System For Congestion-Aware Wireless Networks.
Inventors include Guna Hari Keerthana; A Narmada; Dr K Viswanath Allamraju; Arikatla Sireesha; Dr. Venkateswarulu Naik. B; Y B Avinash; Mr. K. Shyama Satya Naga Teja; Bhargavi Marri; Raman Kumar M; and Dr G Prasanna Kumar.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: ABSTRACT The present invention discloses a technical framework for an AI-optimized dynamic spectrum allocation system specifically engineered to mitigate spectral congestion and enhance bandwidth utilization in next-generation wireless networks. Conventional frequency management protocols rely on static assignment or simple opportunistic sensing, which often result in "spectrum holes" and significant interference during peak traffic periods. The invention overcomes these limitations by providing an intelligent analytical infrastructure that utilizes deep reinforcement learning to predict localized congestion patterns and reconfigure frequency assignments in real-time. The system architecture utilizes a combination of multi-agent neural networks and spatial-temporal spectral sensing to monitor the electromagnetic environment across diverse frequency bands. Traditional cognitive radio systems frequently suffer from high latency and decision-making overhead when operating in high-density urban environments, whereas the proposed model leverages predictive analytics to anticipate user demand surges before they result in packet loss. This dual-pathway approach ensures that the network maintains high throughput and low latency across heterogeneous cells, facilitating seamless connectivity for ultra-reliable communication services. The invention further incorporates an automated interference-mitigation module that dynamically adjusts transmission parameters, such as power levels and modulation schemes, based on the predicted spectral density. By utilizing refined feature extraction from raw RF signals, the framework provides a robust solution for spectrum sharing between licensed primary users and unlicensed secondary users. The result is a proactive network management tool that significantly reduces spectral waste, ensuring that wireless resources are allocated with maximum efficiency in alignment with real-time congestion metrics. Furthermore, the system integrates a decentralized coordination layer that allows for the simultaneous synchronization of spectrum policies across geographically dispersed base stations. This capability ensures that the allocation process remains resilient to localized network failures and adapts to varying regulatory constraints in different jurisdictions. By maintaining a dynamic repository of spectral fingerprints, the invention provides an adaptive electromagnetic solution that supports the development of future-proof telecommunications infrastructures and automated frequency planning protocols. Finally, the invention provides a cloud- native scalability interface that allows for the integrated management of licensed and unlicensed bands through a unified administrative dashboard. This architecture enables network operators to identify underserved zones and implement data-driven resource slicing for diverse IoT and mobile applications. By utilizing automated data ingestion and AI-based pattern synthesis, the invention minimizes the manual effort required for RF optimization, ensuring that high-priority communication links are maintained with zero-compromise reliability.
Disclaimer: Curated by HT Syndication.
