This paper studies robust resource allocation algorithm design for a multiuser multiple-input single-output (MISO) cognitive radio (CR) downlink communication network. We focus on a secondary system which provides unicast secure wireless layered video information to multiple single-antenna secondary receivers. The resource allocation algorithm design is formulated as a non-convex optimization problem for minimization of the total transmit power at the secondary transmitter. The proposed framework takes into account a quality of service (QoS) requirement regarding video communication secrecy in the secondary system, the imperfection of the channel state information (CSI) of potential eavesdroppers (primary receivers) at the secondary transmitter, and a limit for the maximum tolerable received interference power at the primary receivers. Thereby, the proposed problem formulation exploits the self-protecting architecture of layered transmission and artificial noise generation to ensure communication secrecy. Semidefinite programming (SDP) relaxation is employed to derive a resource allocation algorithm which finds the global optimal solution to the formulated problem. Simulation results demonstrate significant transmit power savings and robustness against CSI imperfection for the proposed resource allocation algorithm for layered transmission compared to baseline schemes with traditional single-layer transmission.