Enhancing the physical layer security of uplink non-orthogonal multiple access in cellular internet of things

This paper investigates the physical-layer security of uplink non-orthogonal multiple access (NOMA) in the cellular Internet of Things (IoT) with invoking stochastic geometry. Poisson cluster process-based model is applied to characterize the NOMA uplink transmission scenario, where IoT terminals are located around the serving base station. Considering the severe interference brought by a large number of IoT terminals, inter-cell interference is also taken into account. To enhance the physical-layer security of uplink NOMA transmission with limited overhead increment at IoT terminals, the base stations not only receive the signals from IoT terminals but also keep emitting jamming signals all the time to degrade the performance of any potential eavesdroppers. In order to characterize the physical-layer security performances, we derive expressions of coverage probability and secrecy outage probability. To obtain further insights, network-wide secrecy throughput (NST) and network-wide secrecy energy efficiency (NSEE) are analyzed. It is demonstrated that the security performance can be improved by the proposed full-duplex base station jamming scheme at the cost of reliable performance. The analytical and simulation results show the effects of BS intensity and jamming power on network performances. We also verify that NST and NSEE can be significantly enhanced by our proposed scheme. Using these results, the security of confidential information transmitted by low-complexity IoT terminals can be protected from overhearing.