This website uses cookies. If you continue to browse the site, we assume you are fine with how we use cookies on the web. Read More

{ Markettagged:True , MatchedLanguageCode:True }

Podcast: Mobile Network Transport Requirements for 5G

With 5G technology moving out of the lab and into the marketplace, the transport network must evolve in alignment with the 5G radio access and core networks. Now is the time for those transport networks to come out of the shadows and into the limelight. Listen to Ericsson expert Shane McClelland outline the requirements for transport in a fast-moving and synchronized network evolution to 5G.

Main Block

Laurie Spiegel talks to Shane McClelland (VP of Strategy for Transport Solutions in North America) about the three 5G RAN drivers of transport transformation.


Google Play


Laurie SPIEGEL: Welcome back to 15 Minutes with 5G, a bi-weekly discussion with the industry’s big brains about the latest developments in 5G, and what they mean for consumers, businesses, and society in general. I’m Laurie Spiegel, 5G Campaign Manager at Ericsson North America, and your host for this podcast. Today we’ll explore a part of the mobile architecture that I call the “middle child”. That would be the transport network – connecting the mobile cell sites to the core network. You know, the middle child seems to get the least attention, and perhaps so with transport.  With 5G technology moving out of the lab and into the marketplace, the transport network must evolve with the shift to a 5G radio access network, or RAN.

I’m delighted to be joined by Shane McClelland, VP of Strategy for Transport Solutions in North America. Shane, thanks so much for taking the time to talk with me today.

Shane MCCLELLAND: My pleasure Laurie.

SPIEGEL: How about we start with a basic understanding of the transport domain in mobile networks.

MCCLELLAND: RAN Transport in the 2G/3G/4G world, traditionally called mobile backhaul, is relatively straight-forward.  We’ve all seen the network diagrams.  Some of you may have designed and built a RAN backhaul network in the past.  It’s been basically the same for quite a while.  If we break it down, traditionally we have:

  • A fully contained macro cell site with an Antenna, Radio and Baseband – all the radio components in one physical location, with a cell side router.
  • That cell site router then connects to the Mobile Packet Core by way of an IP cloud, over a fiber optic cable or a point-to-point microwave radio link.

It looks more or less the same in each mobile network operator, with some slight variations on the overall theme, of course. 

SPIEGEL: That’s great. So, then what IS happening in the 5G RAN that’s going to drive changes to today’s mobile backhaul networks?

MCCLELLAND: From my perspective, there are 3 major drivers in 5G that will impact RAN transport in the coming years:

  • First, there’s the race to 1Gbps over-the-air to a user device,
  • Second, 5G will coexist with today’s LTE networks for quite some time, and
  • Third, is radio densification.

SPIEGEL: These drivers seem to be very radio-centric, so what does this mean for transport?

MCCLELLAND:  My view is that in general, the underlying transport architecture we’ve become accustomed to over the years will have to evolve, just as the RAN evolves in 5G.

Let’s take the first one … what I’m calling the race to a gigabit per second (Gbps) over-the-air to a device.

  • Devices hitting the market today are UE Cat 15/16 ready (which means they’re capable of Gbps speeds). For example, the iPhone 8 is capable of 1Gbps down, HTC has models as well, and the other device manufacturers will follow.
  • Also, we have Massive MIMO radios that are in trial phase now, some with multiple ports requiring greater than 10 Gbps connectivity.
  • Additionally, basebands that support 10G capacity for backhaul are already being deployed now.

One can argue how many use cases will require a Gbps over the air in the coming years.  However, it is clear this ecosystem is coming together now.  We already know that our customers’ unlimited data plans are stressing the networks today.  And, you can read in the recent headlines from operators globally, touting their gigabit capacity advancements. 

When doing RAN transport planning, this will drive changes to today’s RAN transport architectures, especially in the last mile and the aggregation parts of the network.

We’re collecting empirical data now as well.  For example, in the recent Massive MIMO trial Ericsson successfully completed with Sprint, the radios were pushing over a gigabit per sector.  That is a significant increase in transport BW as compared to LTE per sector bandwidth today.

SPIEGEL: OK, I get the need for speed.  Why don’t we turn to your second point – the fact that 4G and 5G will coexist with each other. In my mind, this is very different from what we had when we were upgrading from 2G to 3G and then again from 3G to 4G. These were substitutions of one generation to the next.  So, tell me what’s changed with rolling out 5G? 

MCCLELLAND: 4G and 5G will play different roles. For the mobility use case, deployment models are looking like 4G will continue to be a great coverage layer, and 5G radios will be deployed for capacity augmentation. And then you have other specific use cases like Fixed Wireless Access, and Massive Machine-to-machine type communications.

SPIEGEL: So, if 4G and 5G will coexist, that must mean that the transport will have to coexist as well.  What has to happen to make that work seamlessly?

MCCLELLAND:  Basically, the 4G radio and 5G radio networks elements must become aware of each other.  The RAN Transport network essentially becomes the underlying glue enabling radio coordination, which is key to implementing advanced radio features like Carrier Aggregation, Uplink and Downlink Coordinated Multipoint, and LTE License Assisted Access.  For these features to work best, ultra-low latency between RAN network elements is vitally important.

So, not only will 5G RAN transport networks need to support capacity demands, but must also support low latency network requirements enabling functions like network-based timing and sync distribution, and distributed IPSec gateway.   

SPIEGEL: OK, good. So, we’ve talked about download speeds, and the fact that 4G and 5G will have to be working together as partners.  So why don’t we turn to your third item, cell densification. Can you explain that one?

MCCLELLAND: Basically, we’ll have smaller cells.  This is a significant change and a major challenge for Transport brought on by the 5G wave. As a macro cell site disaggregates, radios will get smaller, and more densely deployed.

We know the FCC-approved frequencies for 5G are in the millimeter bands (3.5 Gigahertz, 28Ghz, 39Ghz, etc.). Going up-spectrum means smaller Radio coverage areas. Hence, radios will get smaller and more densely deployed in a variety of locations – lampposts, telephone poles, sides of buildings, etc.

SPIEGEL: With smaller cells, I’ve read that in some cases, it makes sense to move the functionality in the baseband away from the radio location to a more centralized configuration.

MCCLELLAND: Yes, implementing 5G RAN will fundamentally change the RAN architecture we’re used to today.

The traditional macro site as we know today will disaggregate. In situations where it makes sense, antenna/radio combinations will be geographically separated from the basebands.  This idea of co-locating basebands is referred to as Centralized RAN. Additionally, the baseband itself will disaggregate further, using virtualization technology, where Radio Control Functions will move to the Cloud, and Radio Processing Functions can stay close to the radio enabling concepts like “the network slice”.

So, you see, with 5G there will be several RAN architectures, depending on business needs and serving geography.  Distributed RAN, Centralized RAN, Virtual RAN will all be viable 5G RAN architectures that have to work together.  Each of these architectures has varying transport interfaces specified, and each interface has specific requirements for the transport network in terms of design, capacity and latency, and of course, we factor in security as well.

SPIEGEL: That’s a lot of things to think about. Let’s turn to the business implications of all this. So, with C-RAN and V-RAN architectures, can you talk a little bit about those business impacts to the operator?

MCCLELLAND: Advantages for mobile operators are:

  • Lower TCO in terms of operational expenses like lower real estate rents and lower AAV fiber leasing fees, and in some cases, even lower CAPEX.
  • Also, we’re exploring how these new RAN architectures, working in a coordinated fashion, will improve radio performance and subscriber quality of experience.

SPIEGEL: It’s certainly a different architecture and it means a lot more transport links to reach all of those radios.

MCCLELLAND: Yes, availability of transport facilities can be a problem, as well as the ability of the transport network to deliver the capacities and latencies necessary for 5G RAN.  But this also represents an opportunity in terms of increased customer satisfaction and reduce churn.

SPIEGEL: Let me sum up. First, 5G transport must meet strict requirements for performance, capacity, latency, and security, and it also has to support the coordination between many, many more cell sites including those that will coexist with 4G technology.  But we haven’t explored the implications on the physical transport infrastructure. Can you comment on that?

MCCLELLAND: Yes, that’s true. In terms of physical connectivity, there are additional considerations for transport networks in terms of fiber availability, fiber distances, and how microwave radios can be used for transport.

SPIEGEL: Are you saying that Dark Fiber is the only way to meet stringent demands on 5G RAN Transport?

MCCLELLAND: Dark fiber is one solution, but there are several other options that work just as well.

Oftentimes, what we’ve learned in our deployments, is that the actual fiber distance between a Remote Radio Unit and a corresponding baseband can be many kilometers long, or not available when needed.  Long fiber runs can result in round trip latencies exceeding 5G RAN interface specifications … not to mention the sheer number of dedicated fibers that will have to be available.  In cases where dark fiber can’t meet the functional requirements or is unavailable or is too costly, Dense Wavelength Division Multiplexer systems are a low cost, high performance option.

Additionally, where fiber is not available or too costly to lease or deploy, Gigabit capable, point-to-point microwave systems provide a reliable, ultra-low latency connectivity for 5G RAN as will.

SPIEGEL: It seems that with 5G, there are many options and alternatives to consider.  But to deliver on the true promise of 5G, the requirements for transport must be factored in on-par with the radio and core aspects, and not be relegated to the status of what I call “the middle child”. Shane, thank you so much for enlightening us all. I really do appreciate your sharing your expertise and insights with me today.

MCCLELLAND: You are very welcome Laurie. Thanks for having me.

SPIEGEL: And thank you all for listening. If you would like more information, please visit us at And if you like what you’ve heard, please subscribe to our podcast series, “15 Minutes with 5G”.