Fronthaul: Small Cells’ New Best Friend (pt 1)

Cellular Trends Impact Network Architectures (Part 1 of a 2-Part Series)

Wireless network operators are evolving their network strategies to cope with the explosive growth in data traffic. All are making significant capital investments to build out the network, provide sufficient capacity and to put in place scalable architectures to handle ever-increasing needs.

In particular, new backhaul methods and technologies are currently being evaluated, and will be deployed in the coming years. Industry analysts estimate that more than $30 billion will be spent on backhaul equipment between 2012 and 2016. Starting with the mobile radio access, development and deployment of LTE and soon LTE-Advanced provide a clear and standardized roadmap. However, the access technology is only the visible tip of the iceberg and now operators are focusing their attention on the next network bottleneck: the transport links connecting the mobile transceivers and the rest of the mobile network.

The latest fourth generation (4G) mobile technologies such as LTE can deliver speeds and capacities that will soon exceed 100 Mbps per sector. But still this will not be sufficient to handle high-speed data’s explosive growth in mobile networks. (See Figure 1.) Carriers also need to deploy an increasing number of sectors or cell sites in order to provide even more capacity and high quality coverage. Indeed, the incremental capacity gains of the latest radio access technologies and additional spectrum bands is far from sufficient to handle the onslaught of data traffic expected in the next few years and the only alternative is to deploy more sites per square mile.

The combination of high bandwidth and denser networks is therefore forcing carriers to rethink their deployment and backhaul strategies. Sites need to become smaller, simpler and the nature of their connection to the network is changing. (See Figure 2.) While connecting every cell site with fiber may seem like a desirable solution, this is just not an economically viable or practical solution for many sites and especially the newer ones. Indeed, wherever fiber is not readily available, deploying new fiber whether by trenching or not takes time and a large amount of capital. In addition, fiber in itself does not solve all the issues facing the network operator: getting authorization for a new cell site is an extremely time consuming and expensive process and operators need as much flexibility as they can get to go around the many obstacles they face in this process.

Figure 2. Smaller cells evolution in cellular networks.

There are various network topologies used to backhaul to include not just FTTH or FTTC but also hybrid microwave — fiber architectures to Ethernet over Copper, and even PONs. Each has its own business case depending on the network architecture, bandwidth requirements and respective investment and infrastructure in place or planned to deliver access. Then again, those backhaul topologies do not solve all the issues facing the network operator.

Radio Access Networks Architectural Trends

In order to properly assess the optimal solution, one also has to factor fundamental architectural trends in mobile networks. Starting in the mid 2000s, a first fundamental evolution was the transition to an all IP core and radio access network. This made the core and radio access networks more flexible and able to operate much more economically with data as well as voice traffic. More challenges remain as mobile networks rely on increasingly dense networks of smaller cells. While the cost of electronics and radio equipment keeps dropping, installation, operational costs (including electric power) and the need for constant technological evolution remain a key obstacle. At the same time, cellular sites are becoming increasingly difficult and expensive to find and get approved.

Meanwhile, the emergence of Remote Radio Head (RRH) networks in the market enabled by fiber fronthaul solutions delivers built-in scalability to achieve cell density and supports rapid deployment at a reduced cost.

The second fundamental evolution is therefore a concentration of the radio access network processing (ultimately into "the cloud") while keeping the actual RF transceivers (the Remote Radio Heads or Units — RRHs/RRUs) distributed at their ideal locations (where customers require the most capacity).

Carriers are looking for solutions that will offer:
• High capacity in order to offer scalability to next-generation radio access technologies.
• Maximum deployment flexibility to be deployed near high traffic areas and near available sites.
• An evolution path from current network architectures and topologies.

Fronthaul Architectures

Carriers are challenged today by being able to transport data between a multitude of RRHs ranging from high power to small form factors and lower power, and the baseband processing units (generally called Base Band Units or BBUs). Indeed, while all integrated "small cells" offer an attractive indoor or "hot zone" solution, they also introduce high levels of cell edge interference as well as new operational challenges (especially when interacting with the main existing macro networks).

This explains why operators are now looking at deployment architectures using fiber from the base station to the RRH and antenna. By contrast to the traditional backhaul architecture, this new method is called fronthaul. (See Figure 3.)

Figure 3. Backhaul and fronthaul network architectures.

Fronthaul is similar to traditional Distributed Antenna System (DAS) architecture: in both cases, the baseband processing and transceivers are separated by a wide area transport network. However, a fundamental difference is that only fronthaul provides the needed additional capacity and coverage where and when it’s needed. Indeed, fronthaul-based systems operate as point-to-point multicast connections between the baseband processing and the RRH/RRUs. Traditional DAS operates on a simulcast basis between the DAS hub and the distributed antennas. In the case of the fronthaul-based system, the full channel capacity is available at each radio head, whereas the channel capacity is distributed among all antennas in a traditional DAS.

Fronthaul provides several key advantages over backhaul: an easier and less risky evolution path from macro networks, higher wireless performance and the prospect for a more open network architecture. It does, however, come with a few technical challenges: the need for transporting high quality RF signals in real time puts a high requirement on data rates and latencies for the fronthaul transport network. For this reason, fiber has traditionally been seen as the only solution for fronthaul transport medium.

Unfortunately, in reality, fiber is not nearly as ubiquitous as would be required for a full fronthaul network. Even in the best of cases, fiber may be present to the curb, but may not be economically available to the rooftop, pole or towertop where the RRH/RRU needs to be deployed. In other cases, installing fiber on an existing tower may prove to be a challenge.

Read more about options providers have in the second part of this series in OSP® magazine’s November 2013 issue.

Frédéric Leroudier is General Manager, North America, EBlink Inc. He has more than 20 years of experience in the wireless industry, and has led network design, technology selection and business development for a variety of international carriers and system integrators. For more information, visit www.e-blink.com.