Securing Task Offloading and Service Caching in Multi-Tier Computing Networks With Untrusted Relays

Abstract

Due to the rapid development of the Internet of Things (IoT) applications, which generate vast volumes of data at high speeds, security and privacy issues have become challenging. IoT devices use the advanced encryption standard algorithm before transmission. However, since the system communicates through amplify-and-forward relays, the data may be leaked through the untrusted relays. Depending on a mathematical tool may affect the data security vulnerability. Thus, to enhance the system security, physical layer security is used to transmit a jamming signal to confuse the untrusted relay nodes. Hence, the proposed framework ensures security by combining the physical and data layer security which comes with a cost regarding system complexity, system latency, and energy consumption. The proposed framework addresses the joint problem of physical and data layer security, multicell association, task offloading, users’ power allocation, and service caching in multitier communication and edge computing networks. The objective is to minimize the system latency and energy consumption under the secrecy capacity constraint. Due to the NP-hard nature of the joint problem, we use a low-complexity Lyapunov drift-plus-penalty optimization technique based on the Gibbs sampling algorithm. The simulation results demonstrate the proposed framework’s superiority over the state-of-the-art in terms of high secrecy capacity and low computational complexity. When the secrecy capacity threshold increases, the secrecy capacity is enhanced by approximately 5.72%, while the system latency and energy consumption increase by 38.18% and 69.99%, respectively, compared to the literature.