Author: Scott Wakelin
Mobile operators worldwide are investigating new architectures in the Radio Access Network to increase capacity and reduce costs. The C-RAN architecture is the leading solution to this challenge. With C-RAN, baseband processing is moved out of the cell site and into a central location, which creates a new challenge: how to cost-effectively extend the Common Public Radio Interface (CPRI) to the centralized baseband location – a function the industry has named Fronthaul.
Different solutions for fronthaul have been proposed, including: point to point fiber, microwave, passive and active WDM, as well as OTN. To date, equipment vendors both small and large have gravitated towards passive WDM creating a crowded market for a solution that doesn’t meet mobile operators’ operational requirements.
Meanwhile, few, if any OTN based solutions have made it to market due to concerns that it can’t meet CPRI’s complex performance requirements.
In this blog series, I’ll outline new solutions that are emerging that prove an OTN based mobile fronthaul network can meet operators’ key performance and operational requirements.
Before diving into the requirements for a fronthaul network and why OTN is the best choice in comparison to the alternatives, let’s begin by exploring the market drivers influencing mobile operators Radio Access Network. Click the video link below for a quick overview or read on for an in-depth explanation.
Challenges Facing Mobile Operators
Anyone who follows trends in mobile networks is likely acutely aware of the various predictions for the approaching tsunami of mobile data traffic about to flood mobile networks around the world. For example, the Ericsson Mobility Report from December 2014 predicts traffic will increase nearly 8x between 2014 and 2020, after already increasing 11x from 2010 to 2014.
I sometimes take predictions like these with a grain of salt. But that changed for me in November 2014, when I traded in my Blackberry Torch for an Apple iPhone 6 Plus. With access to an LTE network, for the first time I had truly entered the age of mobile broadband. In the six days that remained in the billing cycle, I consumed an amazing 454.5 MB of mobile data – versus only 35 MB in the preceding 24 days. With access to a great collection of apps and mobile video, my mobile data usage increased 10-20x almost instantly.
As more and more mobile subscribers switch to the rich user experience that LTE offers, how will mobile operators cope given the economic constraints of an essentially flat CAPEX and revenue environment? Clearly, to keep pace with the expectations of their subscribers (and ahead of their competitors down the block), mobile operators need to deploy a higher capacity network, at a lower overall total cost of ownership.
Let’s take a closer look at this problem and why mobile operators are looking at C-RAN as the solution.
Today, most mobile networks are built around a distributed architecture, where each site has an equipment room that houses baseband processing (BBU) as well as air conditioning and power supplies. On the tower, antennas connect to one Remote Radio Head (RRH) per sector, and the RRH is then connected to the BBU via fiber-optic-based CPRI.
In this architecture, the BBU is sized to handle the peak load for the site. However, user load on the network varies greatly over the course of a day and exhibits a “tidal effect.” In the daytime, load on cell-sites covering residential areas is minimal but peaks in business districts. In the evening, the opposite is true.
The net result is that on average, only 30 percent of the BBU’s processing capabilities are utilized. Regardless of load, the cell site equipment room is fully powered and air conditioned. China Mobile estimates that up to 46 percent of cell site power consumption is devoted to air conditioning. That’s a lot of wasted power.
To grow capacity to meet these anticipated bandwidth demands, a mobile operator has two primary solutions at their disposal:
- Deploy more spectrum – support carrier aggregation
- Densify – add more cell sites and deploy small cells
But deploying capacity in this way results in a network with:
- A proliferation of under-utilized baseband units;
- An over-allocation of precious CAPEX towards non-traffic related costs (like site engineering and air conditioning);
- An over-consumption of power (46 percent of which is going to the air conditioning of the cell site equipment room); and
- Increased interference, which lowers the capacity of the network.
As a result of these challenges, there is growing momentum among mobile operators worldwide to reconsider the RAN architecture and look at alternatives such as C-RAN.
Introduction to C-RAN
The fundamental idea behind C-RAN is to move the baseband processing resources for all cell sites within an area to a centralized location. In the C-RAN architecture, operators can now take advantage of the tidal effect and size their baseband investment for the average needs over a region. By doing so, a 50 percent reduction in baseband processing capacity can be achieved. In their whitepaper on C-RAN, China Mobile claims that C-RAN can result in CapEx savings of up to 30 percent and OpEx savings of up to 53 percent.
The following table summarizes the CapEx and OpEx advantages offered by a C-RAN architecture.
Beyond economic advantages, C-RAN also has several important technical advantages. By co-locating baseband units, support for interference busting techniques like Coordinated Multipoint (CoMP) can be employed. In their paper, “Recent Progress on C-RAN Centralization and Cloudification,” China Mobile has shown uplink CoMP gains of up to 100 percent at the cell edge. Additionally, concentrating baseband into the CO creates the opportunity to co-locate services such as video CDN with the base station. The result is reduced network bandwidth requirements, reduced latency, and a better overall user experience.
C-RAN Drives the Need for a Mobile Fronthaul Network
The fundamental challenge to unlocking these economic and technical advantages is the rollout of a cost-effective and fiber-efficient “Fronthaul” network that interconnects the RRHs at the cell sites with the Baseband Units in the central office. Whereas a backhaul network is used to interconnect an eNodeB with the mobile core using IP/Ethernet-based networking, fronthaul transports the constant bit rate CPRI signals between the RRHs and the BBU.
Fronthaul represents one of the most significant barriers to wide-scale C-RAN deployment. Today, carrier transport networks worldwide are built around a G.709-based Optical Transport Network (OTN) infrastructure. While there is no doubt that OTN provides best-in-class OAM, demarcation and scalability, the industry has gravitated towards passive- or active-WDM-based solutions for fronthaul due to uncertainty over OTN’s ability to meet the strict 3GPP and CPRI latency and timing requirements. As a result, the industry has suffered from a lack of cost-effective and carrier-grade fronthaul solutions.
That’s no longer an issue with PMC’s new OTN-based fronthaul solution. It enables our field-proven HyPHY Flex OTN processor family to meet the stringent CPRI and 3GPP performance specifications while providing network operators with the many advantages offered by OTN based fronthaul.
My next post in this series will explore the requirements for a fronthaul network and how OTN-based fronthaul can reduce operators’ total cost of ownership.
Leave a Reply
You must be logged in to post a comment.