In [30], a flexible framework is proposed where the centralized BBU pool is connected to the SDN controller to make another level of intelligent abstraction for joint coordination and increase scalability. The novelty of the work is a definition of an architecture where given the practical constraints of real-time RAN, a Multiaccses coordination algorithm for joint processing is proposed, and this processing is centralized. It is also capable of end-to-end connectivity with enhanced QoS for cell-edge users; those are affected by ICI without CoMP.

In [31], new functional split architecture between RRHs and BBUs is investigated to reduce transmission throughput. Two novel architectures are proposed in this work. It was shown that in a fully loaded system, 30%-40% bandwidth could be saved, and if half of the radio channel is used, then up to 70% bandwidth can be saved using this new architecture. It is seen that when transmitting the LTE guard sub-carrier, CP makes a significant portion of redundant data. At the same time, some other redundant info is being sent, called Physical Resource Block (PRB), which is also sub-carrier. Generally, not all PRBs are occupied in LTE transmission. If this information can be inferred at the RRH, then only used PRB can be sent, thus less data to be sent. At the same time, if CP removal and FFT is done in RRH, then 50% of less data will be transmitted. Another way to reduce uplink data is first to change the signal from the frequency domain to the time domain. Then PRB usage state detection and finally the quantized data is sent to the BBU.

C-RAN has many advantages over traditional RAN, such as low CAPEX and OPEX, less cell interference, easy for maintenance, and fast deployment due to less civil work. Over the past years, C-RAN has been evolved, ranging from “centralized”, “cooperative” to “Cloud”. Being the final evolution, cloud RAN includes all the benefits from centralized and collaborative architecture. Due to the centralization of BBU at the BBU pool, on-demand resource allocation is possible, and resources are effectively utilized through resource sharing. BBU pool also helps in the mobility management of UEs. In the BBU pool, If any BBU is heavily loaded or fails, a backup BBU is always ready to connect through Software Defined Radio (SDR) switch. In traditional RAN, even though the CoMP scheme improves the cell edge throughput by reducing inter-cell interference, but the implementation of CoMP becomes a cumbersome task due to the extensive system overhead and the delay between the BSs [3]. Since in Cloud RAN, all the BBUs are located at a single location, maintaining the tight synchronization between the BBUs has become easy. In this way, as C-RAN structure is best suitable for the implementation of CoMP, and hence spectral efficiency can be improved [32].

A study on greenhouse emission in [33] says, 2% of the $CO_{\rm 2}$ global emission is contributed by information and communication technology (ICT) sector. By 2020, the carbon footprint is expected to be triple due to the massive growth of the cellular sector. Thus an energy-efficient system design is highly required to reduce $CO_{\rm 2}$ emission. C-RAN is an energy-efficient system as compare to traditional RAN. For a centralized BBU, the power consumption required for cooling the BBUs has come down due to a single room. Due to many to one mapping between RRH-BBU, a few BBUs are required in the BBU pool. It saves power consumption.

In C-RAN, RRH becomes very light, and the footprint is reduced due to the shifting of all layers and its functions to BBU. It becomes easy to install RRH, especially in dense urban and hilly areas. The requirement of less construction space for RRH installation helps in reducing CAPEX. In a short period, the installation of RRH is possible; hence network expansion can be easily done to increase the system capacity. To have an insight into the requirement of C-RAN and its benefit, we have chosen two typical areas, which are depicted in Figure. 2. The chosen areas are football stadium and airport, where better network coverage with high throughput is essentially required. By applying C-RAN architecture, an operator can get more benefits as compared to tradition RAN.

By implementing Network Function Virtualisation (NFV) over BBUs, it becomes possible to share the same resource among multiple operators. Thus, compared to traditional RAN, fewer resources are required in C-RAN. Hence, it helps in reducing the CAPEX. Cost efficiency is a main driver of NFV. Due to NFV, load-balancing, network scale up and down, move functions across distributed hardware resources become easy in C-RAN. It improves the flexibility of the network service provisioning and reduces the time to deploy new services.

Pages ( 5 of 12 ): « Previous1 ... 34 5 67 ... 12Next »

This site uses Akismet to reduce spam. Learn how your comment data is processed.