Secure Resource Allocation and Trajectory Design for UAV-Assisted IoT With Double Cluster Head

Abstract

Unmanned aerial vehicles (UAVs) enable flexible data collection from Internet of Things (IoT) nodes in remote areas, but the data of IoT nodes face security threats. In the proposed data collection strategy based on a double cluster head (CH) framework, we exploit the inter-user interference (IUI) of uplink non-orthogonal multiple access (NOMA) to improve the security of IoT nodes. Specifically, inter-CH interference in NOMA is used as jamming signals to hide confidential data. Then a CH selection scheme is designed to alleviate the unbalanced energy consumption among member nodes in a cluster. Based on the CH selection scheme, we maximize the secrecy energy efficiency (SEE) via joint optimization of power, time scheduling, and trajectory. Due to the highly coupled variables and non-convex constraints, an alternating optimization method is used to decouple the original problem into subproblems and they are solved iteratively. In each iteration, Dinkelbach’s method is used to tackle the fractional objective function; the successive convex approximation technique is used to transform the non-convex subproblems into convex forms. In numerical simulations, our proposed data collection strategy shows effectiveness in improving SEE and hindering wiretapping. Furthermore, the proposed CH selection scheme efficiently extends the lifetime of energy-constrained IoT nodes.