Proactive Eavesdropping of Suspicious Non-Orthogonal Multiple Access Networks

In this correspondence, legitimate surveillance of a downlink suspicious non-orthogonal multiple access (NOMA) network with multiple groups of suspicious users (SUs), in the presence of a full-duplex monitor equipped with a successive interference cancellation (SIC) decoder, is investigated. Proactive eavesdropping is adopted, i.e., the monitor simultaneously eavesdrops the suspicious communications, and sends jamming signals. The problem of maximizing the number of successfully eavesdropped SUs by jointly optimizing the jamming power allocation and the decoding order at the monitor is investigated considering both imperfect full-duplex self-interference cancellation, and imperfect SIC. By exploring the particular problem structure, a heuristic iterative algorithm is proposed, where the optimal decoding at the monitor is proved to be equal to that at the SUs. Extensive simulations confirm that the proposed algorithm achieves performance very close to the optimal one but with significantly lower complexity, and outperforms various benchmark algorithms.