Training and Data Structures for AN-Aided Secure Communication

For physical-layer security with multiple antennas, artificial noise (AN) aided secure beamforming scheme is considered. An ergodic secrecy rate considering a channel-feedback mechanism and channel estimation is derived. In this formula, two power-splitting factors in the training and data parts (one is between a training signal and AN, and the other between an information signal and AN) are included. In a high signal-to-noise ratio region, the power-splitting factors, which maximize the approximated ergodic secrecy rate, are derived. An alternative optimization process is proposed to obtain the convergent solutions. The asymptotic characteristics of the secrecy rate and power-splitting factors are analyzed.