Opportunistic fair resource sharing with secrecy considerations in uplink wiretap channels

In this paper, we propose two opportunistic scheduling algorithms for uplink wiretap networks with multiple legitimate users (LUs) and eavesdroppers. Different from the existing works on scheduling algorithms with secrecy considerations, we focus on the practical scenario where LUs experience diverse path-loss to the base station (BS). Hence, both secrecy throughput and fairness among LUs are crucial design considerations. In the proposed scheduling algorithms, the feedback information generated by each LU is designed to reduce the probability that the LU is selected by the BS when the eavesdroppers overhear much information. It is proved that fairness among LUs can be achieved in arbitrary fading channels. In order to investigate the efficiency of our proposed scheduling algorithms, the normalized secrecy throughput, i.e., the secrecy throughput for a given LU normalized by the probability of it being selected, is analyzed and proved to achieve double-logarithmic growth when the number of LUs in the network increases to infinity.