UC Riverside

This dissertation is concerned with information security of wireless networks with full-duplex radios, which covers three topics. The first topic is a development of optimization algorithm for a full-duplex relay system where the source node, the destination node and an eavesdropper each have a single antenna while the full-duplex relay node has multiple antennas. We optimize two beamforming vectors that are concurrently used at the relay node: one for signal transmission and the other for artificial noise used to jam eavesdropper. The second topic is an analysis of the secret key capacity achievable from an anti-eavesdropping channel estimation (ANECE) method which was designed for networks of multiple cooperative full-duplex radio nodes each with any number of antennas. We compare the secret key capacities based on ANECE with that based on the conventional channel training method, which are shown to have the same degree of freedoms. The third topic focuses on an analysis of the secrecy capacity of information transmission between multi-antenna nodes that apply ANECE-based channel estimates. Both a one-way scheme and a two-way scheme are considered. Unlike the information transmission schemes based on conventional channel training, our information transmission schemes based on ANECE have a non-vanishing secrecy capacity even if the number of antennas on eavesdropper increases to infinity.