Novel Outage-Aware NOMA Protocol for Secrecy Fairness Maximization Among Untrusted Users

TL;DR

This paper introduces a new secure NOMA protocol that enhances secrecy fairness among untrusted users by optimizing power allocation and decoding order, with analytical and numerical validation of its effectiveness.

Contribution

A novel QoS-aware secure NOMA protocol with a new decoding order and optimized power allocation to maximize secrecy fairness among untrusted users.

Findings

Achieved approximately 55% to 69% performance gain over fixed and individual optimal power allocations.

Derived closed-form expressions for outage probabilities and optimal power allocation.

Provided analytical insights into the tradeoff between secrecy fairness and QoS demands.

Abstract

Observing the significance of spectrally-efficient secure non-orthogonal multiple access (NOMA), this paper proposes a novel quality of service (QoS) aware secure NOMA protocol that maximizes secrecy fairness among untrusted users. Considering a base station (BS) and two users, a novel decoding order is designed that provides security to both users. With the objective of ensuring secrecy fairness between users, while satisfying their QoS requirements under BS transmit power budget constraint, we explore the problem of minimizing the maximum secrecy outage probability (SOP). Closed-form expression of pair outage probability (POP) and optimal power allocation (PA) minimizing POP are obtained. To analyze secrecy performance, analytical expressions of SOP for both users are derived, and individual SOP minimization problems are solved using concept of generalized-convexity. High signal-to-noise ratio approximation of SOP and asymptotically optimized solution minimizing this approximation is also found. Furthermore, a global-optimal solution from secrecy fairness standpoint is obtained at low computational complexity, and tight approximation is derived to get analytical insights. Numerical results present useful insights on globally optimized PA which ensure secrecy fairness and provide performance gain of about 55.12%, 69.30%, and 19.11%, respectively, compared to fixed PA and individual users' optimal PAs. Finally, a tradeoff between secrecy fairness performance and QoS demands is presented.