Introduction
The fourth-generation (4G) mobile communication systems are commercially operated worldwide mainly based on Long Term Evolution (LTE) Release 8. As it evolved, LTE Release 9 was published in 2009 with the introduction of multimedia broadcast multicast services, home eNodeB (eNB), location-based services (LBS), and so on. The Third Generation Partnership Project (3GPP) Release 10, published in the beginning of 2011, made great progress through the introduction of carrier aggregation, multiple-input multiple-output (MIMO), and relaying. LTE Release 11 introduced the coordinated transmission and reception of base stations. LTE Release 12 (R12) mainly concentrated on small cell enhancement. However, there are still lots of open issues to be addressed. One of the most important topics in 5G networks is the ultra-dense network (UDN), where small cells will be ultra-densely deployed in hotspots. Thus, the downlink intercell interference will become much more serious, leading to huge energy consumption and intolerable delay. Although the concept of control and user plane (C/U) decoupling has been proposed for a long time, detailed approaches to splitting signals and channels are not discussed for the UDN scenario. In addition, existing new systems based on C/U splitting do not focus on the UDN scenario, and also face severe challenges in interference cancellation and energy saving. Thus, a new carrier designed for base stations in the ultra-dense scenario is vital for the next generation system to minimize interference and contribute to a “softer” and “greener” network [1].