MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641105926 A) filed by Aditya Institute Of Technology And Management on September 03, 2026, for Ai-Enabled Adaptive 6g Wireless Communication System For Dynamic Spectrum Allocation And Interference Mitigation.
Inventors include Dr. M. V. H. Bhaskara Murthy; Dr. G. Sateeshkumar; and Dr. G. S. S. S. S. V. Krishna Mohan.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: ABSTRACT OF THE INVENTION The present invention relates to an AI-Enabled Adaptive 6G Wireless Communication System for Dynamic Spectrum Allocation and Interference Mitigation. Sixth-generation wireless networks demand ultra-high data rates, ultra-low latency, massive connectivity, and efficient utilization of scarce spectrum resources in highly dynamic environments spanning millimetre-wave, sub-THz, and THz bands as well as integrated terrestrial and non-terrestrial deployments. Conventional static or reactive spectrum management techniques are inadequate for these conditions. The invented system employs a hierarchical multi- agent architecture comprising Edge Spectrum Agents located at base stations or edge nodes and an optional Global Coordination Agent. Each Edge Spectrum Agent continuously observes a rich state that includes hybrid spectrum sensing results (energy detection, cyclostationary features, and deep-learning occupancy prediction), channel quality indicators, traffic demands, residual energy, and predicted trajectories of potential interferers. A deep reinforcement learning engine (preferably PPO or MADDPG under centralized training with decentralized execution) selects actions that jointly determine sub-band allocation, bandwidth portions, transmit power, beamforming parameters, and reconfigurable intelligent surface configurations. A multi-objective reward function balances spectral efficiency, latency of priority traffic, energy efficiency, interference temperature, and fairness. Federated learning enables collaborative policy improvement across agents while preserving privacy of local radio data. A Predictive Interference Mitigation Module and an accompanying Digital Twin allow the system to anticipate interference events and perform safe offline exploration. Simulation case studies in dense urban scenarios demonstrate substantial gains—typically more than 35–40 % higher aggregate throughput, significant latency reduction, improved energy efficiency, and markedly lower interference levels—compared with enhanced dynamic spectrum access baselines, while maintaining decision latencies suitable for real-time operation. The system is compatible with open RAN and 3GPP service-based architectures and is extensible to integrated sensing and communication as well as multi-tier heterogeneous networks. By closing the loop between sensing, prediction, multi-agent learning, and actuation, the invention provides a practical, scalable pathway toward AI-native spectrum management for future 6G wireless systems.
Disclaimer: Curated by HT Syndication.
