Cognitive radio technology is an effective tool to alleviate spectrum scarcity problems for future wireless communication systems. Spectrum sensing is the core technique of cognitive radio, because it directly influences the utilization efficiency of spectrum resources. However, the fifth generation communications will encounter many scenarios with fast time-varying channels, in which traditional sensing systems cannot maintain a satisfactory performance as that in traditional quasi-static channels. Motivated by this circumstance, in this paper, we propose a novel sensing system, i.e., sliced sensing system, to dedicatedly solve the challenge above. The proposed system includes an operational concept, an architecture, six specific core modules, and detailed workflows for all the modules. Numerical results are provided to testify the effectiveness of sliced sensing system. It demonstrates that the proposed system can obviously strengthen the sensing performance under fast time-varying channels. Furthermore, it also reveals that the sensing performance of the proposed system is robust, even with inaccurate channel information.