Efficient Deep Post-Decision State Learning for Privacy-Conscious Offloading in MEC-Enabled 6G Networks

Abstract

With the growing demand for data-intensive applications, offloading tasks to multi-access edge computing (MEC) servers has become essential. However, privacy concerns arise as sensitive data is transmitted to remote servers. To address this, the proposed method uses deep learning techniques to intelligently analyze and classify data packets, ensuring that only non-sensitive information is offloaded. This paper proposes a novel approach called a deep post-decision state (PDS) learning privacy-concerned offloading in MEC-enabled 6G Environments. A Post-Decision State (PDS) approach that selectively selects data packets while maintaining privacy is used to train the deep learning model. The suggested method successfully balances the trade-off between increasing system performance and safeguarding user privacy to achieve privacy-aware offloading in MEC-enabled 6G networks, as shown by the experimental findings. Also, it performs noticeably better than the traditional DQN, according to simulation data.