Statistical Analysis of CP-Based Artificial Noise in Narrowband and Broadband PLC Systems

This paper presents a statistical analysis of artificial noise (AN) designed based on the degrees of freedom provided by the cyclic prefix to enhance physical-layer security in power line communication (PLC) systems. A comprehensive description of the AN design is provided, and the AN is modeled as a random vector while its analysis is based on the first- and second-order statistical moments. Numerical results in both narrowband and broadband PLC channels show that, despite being theoretically zero-mean, the expectation may deviate from zero in practice when the number of realizations is low. Moreover, the analysis reveals non-negligible cyclic correlation structures in the AN, which could potentially be exploited by an eavesdropper, thereby compromising the physical layer security of PLC systems.