UAV-IRS-Assisted D2D Covert Communications With Multiple Eavesdroppers

Unmanned aerial vehicle and intelligent reflecting surface (UAV-IRS)-assisted device-to-device (D2D) communication enhances coverage and improves channel reconfigurability, but it still faces significant challenges in maintaining covert transmission in the presence of multiple randomly distributed eavesdroppers. This paper studies covert communication in spectrum-sharing D2D networks assisted by UAV-IRS. Unlike existing works that focus on single or location-known eavesdroppers, this paper addresses practical scenarios with multiple spatially random eavesdroppers. We propose a novel system model, where the UAV-IRS dynamically reconfigures the wireless environment and a full-duplex base station generates artificial noise to enhance covertness. The minimum detection error probability at the eavesdroppers is adopted as the covertness metric, and a comprehensive joint optimization problem is formulated to maximize the covert transmission rate while ensuring the reliability of the cellular link. To tackle the non-convexity, a hierarchical alternating optimization (AO) algorithm is developed to obtain closed-form solutions for IRS phase shifts and UAV positioning, and to efficiently allocate power under covertness and outage constraints. Numerous analyses demonstrate that the proposed scheme outperforms baseline methods, validating the effectiveness of joint UAV and IRS optimization in enhancing both covertness and covert rate under multiple eavesdroppers environments.