Long Short-Term Memory Based Trust Security Protection in Edge Computing Environment
WU Qing, Wang Junyang, Kang Qu Zhou, Wei Bowen, Huiping Wu, Haifeng Wang · 2024
The security and privacy are the significant anxieties connected to Edge Computing, as it is an open-access as well as self-organized network. Through the combination of the Internet of Things (IoT), and 5G-class large networks, traditional enterprise Cloud Computing (CC) systems are not appropriate process with enormous data produced through integrating with network edge electronic devices. So as to efficiently address this difficult issue, Edge Computing (EC) comes into existence. So, this research proposed the hybrid learning approach of Long Short-Term Memory (LSTM) for security approach along with a trust model was developed to manage the EC network from familiar as well as unfamiliar attacks, while reduced a False Detection Rate (FDR). The effectiveness of the proposed method attains good results and it is estimated by various evaluation indices such as accuracy, False Positive Rate (FPR), F1-score and Packet Loss Rate (PLR) of values about 99%, 75%, 0.9727 and 0.08 respectively when compared to the existing methods like Generative Adversarial Network (GAN) and UAV-Trust based Task Offloading (UAV-TTO).