DRL-Optimized UAV-Relayed 6G Multicast/Broadcast for Over-the-Air IoT Firmware Updates

The evolution toward 6G networks introduces unprecedented opportunities for efficient multicast and broadcast services, particularly for Over-the-Air (OTA) firmware updates in massive IoT deployments. This paper proposes a DRL-optimized UAV-relayed 6G multicast/broadcast architecture that delivers large-scale OTA IoT updates while maintaining high spectral efficiency and stringent delay targets. We address the critical challenge of delivering OTA firmware updates to millions of IoT devices across diverse geographical regions by leveraging 6G broadcast/multicast capabilities through a hybrid satellite-UAV architecture. The proposed DRL-based solution employs a Deep Q-Network (DQN) with prioritized experience replay to dynamically optimize broadcast coverage areas, multicast group assignments, and transmission scheduling for OTA updates. Our framework incorporates action dimension reduction techniques to handle the massive scale of IoT devices while ensuring efficient spectrum utilization and minimizing update completion time. Experimental results show that compared with the unicast method, the DRL-enabled broadcast system can significantly shorten the OTA update completion time and maintain higher update reliability in challenging environmental conditions. The proposed architecture supports scalable firmware distribution to over 100,000 IoT devices simultaneously, making it suitable for smart city and industrial IoT applications requiring frequent security patches and feature updates.