Secure cognitive radio system with cooperative secondary networks

In cognitive radio (CR) systems, we often use jamming schemes to improve the physical-layer security in the presence of passive eavesdroppers. However, it also needs additional signal sources to transmit artificial noise. In this paper, we investigate the multiple-input single-output (MISO) cognitive radio system where two secondary links secure each other´s communication. Each secondary transmitter (ST) sends intended messages to its own receiver while acts as a friendly jammer to the other CRN. Our target is to maximize the worst-user´s secrecy rate while limiting the interference in the direction of the primary receiver. We solve the problem through a sequence of convex semidefinite programs (SDPs) under individual power constraints. We also introduce two sub-optimal zero-forcing (ZF) beamforming methods, one of which solves the problem based on SDP and the other adopts the approach of Pareto-optimal parameterization in a more computationally efficient way.