Joint Optimization for Covert Communications in UAV-Assisted NOMA Networks

• Published in: IEEE transactions on vehicular technology Vol. 73; no. 1; pp. 1 - 15
• Main Authors: Deng, Dan, Dang, Shuping, Li, Xingwang, Ng, Derrick Wing Kwan, Nallanathan, Arumugam
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Airspeed; Automobiles; Autonomous aerial vehicles; Convexity; Covert communications; Error detection; Jamming; joint optimization; Mathematical analysis; NOMA; Nonorthogonal multiple access; Optimization; Trajectory; UAV-assisted communications; Unmanned aerial vehicles; Uplink
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2023.3308718

In this paper, we investigate the design of the trajectory of an unmanned aerial vehicle (UAV) and the transmit power of ground users to improve covert communications against a flying warden in UAV-assisted non-orthogonal multiple access networks, where the legitimate UAV can simultaneously collect the messages from the multiple ground users in a secure manner. Taking the channel uncertainty into account, we derive the analytical expressions of the optimal normalized detection threshold, the minimum detection error probability (DEP), and the security-guaranteed transmit power constraint for the ground users that are exploited to ensure a high DEP at the warden UAV. Subsequently, the design to maximize the average covert achievable rate (CAR) subject to the constraints of flight speed, initial and final locations, transmit power, and detection performance is formulated as a non-convex optimization problem. To obtain a high-quality solution to the design problem at hand, an iterative block coordinate descent-based successive convex approximation method is proposed. From the theoretical analysis, we find that a greater channel estimation uncertainly or a lower effective received power ratio at the warden UAV is beneficial in enhancing the system covertness towards the warden UAV. Moreover, when the successful detection probability of the warden UAV is sufficiently small, the maximum effective received power ratio is linearly and positively correlated to the detection success probability. Besides, extensive simulations are presented to verify the covert performance advantages brought by the proposed method.