Secrecy Outage Performance and Deep Learning Evaluation of Multihop Energy Harvesting IoT Networks over Nakagami-m Fading Channels

In this paper, we study the secrecy outage performance of multi-hop energy harvesting Internet-of-Things (IoT) networks, where all IoT devices harvest energy from a power beacon for conveying a confidential message to multiple legitimate users in the presence of an eavesdropper. To enhance the secrecy outage probability (SOP), we propose and analyze the best relay selection (BRE) and best-path selection (BPA) schemes under Nakagami-m fading environments. Based on the analysis results, we develop a deep learning model for the proposed schemes to evaluate the system SOP. Numerical results show that the BPA scheme greatly outperforms the BRE one, showing the efficiency of the best-path selection approach. Moreover, the deep learning model is capable of predicting the SOP of all schemes with high accuracy while it drastically reduces the execution time, arising a real-time configuration for multi-hop energy harvesting IoT networks.