Adaptive NOMA-Based Spectrum Sensing for Uplink IoT Networks

With the rapid growth in the requirements of the Internet of Things (IoT), the scarcity of spectrum resources is becoming serious. Non-orthogonal multiple access (NOMA) and spectrum sensing offer the opportunity of addressing spectrum shortage to some extent. In particular, NOMA stably enables multiple users to share the same frequency band, while spectrum sensing dynamically detects the target spectrum utilization. However, due to the characteristics of NOMA, the application of spectrum sensing to the uplink IoT systems makes the process more complex, also brings the challenge of obtaining the target users' states accurately from interfered multi-user signals. Meanwhile, the combination of NOMA and spectrum sensing does not reach the upper bound of spectrum utilization. Under the context, we are motivated to propose an adaptive NOMA-based spectrum sensing method for uplink IoT networks, which aims to flexibly and precisely identify the target frequency usage. The relationship among the threshold, the false-alarm probability, and the transmission willingness are derived in closed form. We also customize a sensing algorithm of the complete adaptive working mechanism. Numerical results clarify that the proposed method achieves stability under different SNRs and transmission willingness, while improves the system throughput which outperforms the existing technologies by 38.20%.