Rate Performance Analysis of User-Centric C-RAN with Orthogonal Multiple Access

To support the massive data traffic in the next generation cellular networks, the cloud radio access network (C-RAN) architecture is promising to significantly improve the spectral efficiency. When a massive number of remote radio heads (RRHs) are employed in the C-RAN, substantial rate gain can be achieved over that with co-located antennas. To reduce computational complexity at the central processor, a user-centric cell structure, known as virtual cell, is proposed, where each user is served by a few surrounding RRHs. This paper focuses on the scaling behavior of the average ergodic rate of a C-RAN with user-centric virtual cells. By assuming that orthogonal multiple access method is adopted among the users with overlapping virtual cells, a lower-bound of the average ergodic capacity is derived as an explicit function of the virtual cell radius r and the RRH intensity λA. Simulation results verify that the average ergodic capacity increases monotonically with r and λA in the orders of $O ((\alpha-2)\ log_{2} r)$ and $O((\frac{\alpha}{2}-1) \log_{2} r)$, respectively.