SemanticBlock: Blockchain-Assisted Secure and Energy-Aware UAV-BS Deployment Leveraging Semantic Learning

초록

Unmanned aerial vehicles (UAVs) are emerging as flexible, rapid-deployment platforms for wireless communication, particularly in remote or disaster-affected areas. However, effective UAV-based base station (UAV-BS) deployment faces two major challenges: limited flight energy and communication security. This paper presents SemanticBlock, a blockchain-assisted framework that leverages semantic learning to enable secure, energy-aware UAV-BS deployment. The proposed system integrates a context-aware deep learning model, a long short-term memory (LSTM) network that learns semantic relationships among environmental and operational parameters, such as temperature, wind speed, battery state, and payload, to accurately predict UAV energy consumption for flight duration. These predictions guide UAV-BS positioning and migration decisions to maintain service continuity and energy efficiency. To address security vulnerabilities, a blockchain network is introduced to authenticate UAVs, ensure trustworthy communication, and prevent malicious access without centralized control. The joint use of semantic learning and blockchain provides a robust mechanism for both intelligent energy management and decentralized security. Simulation results demonstrate that SemanticBlock achieves 93.7% prediction accuracy, surpassing state-of-the-art models while reducing communication latency, confirming its potential for reliable UAV-BS deployment in dynamic, disaster-prone environments.