Joint effect of artificial noise and primary interference on security performance of cognitive radio networks

Underlay mechanism allows concurrent communications of primary users and secondary users in cognitive radio networks (CRNs), causing mutual interference between them. However, current literature neglects primary interference or considers it as Gaussian noise. In addition, artificial noise, which is intentionally generated to interfere eavesdroppers, can improve security performance of CRNs. This paper analyzes security performance of CRNs, accounting for artificial noise and considering primary interference as non-Gaussian noise, under maximum transmit power constraint, interference power constraint, and Rayleigh fading channels. The security performance is evaluated through proposed exact expressions of secrecy outage probability, non-zero achievable secrecy rate probability, and intercept probability, which are verified by Monte-Carlo simulations. Various results demonstrate that CRNs suffer security performance saturation in the range of large maximum transmit power or large maximum interference power, and primary interference significantly deteriorates security performance while artificial noise is useful in enhancing this performance.