Sliced Spectrum Sensing—A Channel Condition Aware Sensing Technique for Cognitive Radio Networks

Cognitive radio technology, which helps to alleviate spectrum scarcity problem, plays an important role in future wireless communication systems. Spectrum sensing is the key technique of cognitive radio. The upcoming fifth generation (5G) communications need to deal with fast time-varying channels (i.e., channel state is non-static within a sensing window), while existing sensing methods are based mainly on conventional quasi-static channels. Consequently, traditional sensing methods may not work well in 5G communications. This paper aims to propose a sliced sensing technique, which can support 5G cognitive radio applications working in fast time-varying channels. We analyze the impact of non-static channels on the performance of traditional sensing methods and disclose a negative factor, i.e., out-of-phase feature distortion effect. Accordingly, we propose a sliced spectrum sensing scheme and probe its feasibility via several relevant trials. Based on the aforementioned works, a complete version of the sliced sensing technique is designed, which can adapt to channel variation and choose the optimal slicing strategy. Simulation results manifest that the sliced sensing technique outperforms traditional sensing methods in both Rayleigh fading and 3GPP spatial channel models.