Artificial Noise-Based Physical-Layer Security in Interference Alignment Multipair Two-Way Relaying Networks

This paper introduces two novel physical-layer security algorithms for interference alignment (IA)-based multipair communication systems with a single half-duplex relay and a single eavesdropper. According to these proposed physical-layer security algorithms, users mix their information signals with jamming signals, and broadcast them at the multiple access phase, while the relay forwards the mixed signals at the broadcast phase. Moreover, the relay and users' precoding and decoding matrices are designed in a way which enables the legitimate receivers to eliminate the jamming signals while the hidden eavesdropper is unable to eliminate these jamming streams. In this context, the proposed algorithms are designed to transmit the information streams with minimum power, preserving the user received signal to noise ratio above a pre-determined threshold and utilizing the remaining power for the jamming signals. Therefore, the user and relay power budgets allocation is formulated as a joint optimization problem that can be solved using an iterative optimization algorithm and semi-definite programming. In such fashion, four transmission models are proposed to manage the artificial noise transmission among the different users to achieve a tradeoff between the users sum-rate and secrecy rate. Extensive simulation results are provided to show the efficiency of the proposed algorithms and the transmission models in achieving the transmission security for IA-based multiuser relaying networks.