15 NOVEMBER 2018

Interview with Gavin Hayhurst, Product Marketing Director, TEOCO

(Part 1 of a 2-part series)
Click here to read part 2

5G is shaking up the traditional wireless network, which historically involved one antenna, one radio and one baseband unit. All typically provided by one company. That’s not the case any longer. In this 2-part series, TEOCO Product Marketing Director Gavin Hayhurst talks about network slicing, NFV, edge computing, massive MIMO, and the changes – and challenges – of creating tomorrow’s 5G networks.

Q: Gavin, what new innovations and advanced technologies do you feel will be impacting tomorrow’s 5G wireless networks?
Gavin Hayhurst: As we approach the launch of commercial 5G networks, it seems the number of game-changing technologies being touted far outweighs what we experienced in the run up to LTE. The list is long, but some of the interesting ones in my opinion are: network slicing, Network Function Virtualization (NFV), massive MIMO and edge computing.

At the most basic level, all innovations are driven by a desire amongst communications service providers (CSPs) to either increase revenues or decrease costs. And then, being more candid, I think there is also an element of bragging rights or the fear of missing out, which are driving some decisions too.

Q: Let’s look at some of the key drivers and challenges of each of these examples in more detail, starting with the 5G concept of Network Slicing.
GH: The key driver behind network slicing is all about being able to deliver new service offerings and growing revenues. The aim is to dynamically carve up the network capacity into different slices. Each slice can then be configured and optimized for the service it will deliver, and support SLA’s attached to that service delivery. Slicing will enable services to be ring-fenced on a network – meaning premium or critical communication services can be protected from degradation due to other users consuming too much bandwidth from services, such as video streaming.

As with all new technologies, there is always a learning curve and challenges to solve along the way. Network slicing is a new approach that will require much more collaboration between the operations and sales and marketing teams within an operator. Sales and marketing will need to understand what they can promise to end users in terms of SLA’s based on network capabilities, and operations will need to get a handle on delivering potentially thousands of SLA’s. It’s all going to get very complex.

Q: What about Network Function Virtualization (NFV)?  How is this changing communication networks?
GH: NFV is about virtualizing functions on commodity hardware, with the aim of driving down network CAPEX. Another key driver for NFV is its supposed ability to provide a cheaper and faster mechanism for network evolution. The intent of the technology is that it will be much easier for carriers to simply deploy a software upgrade on their network than to swap out critical hardware components.

But NFV is an area that has been facing one challenge after another. For over six years, operators have been touting the benefits of NFV and their plans to adopt the technology, but the Promised Land still seems far off in the distance. For example, Orange Spain took nine months and US$7 million to onboard a single virtual network function. Clearly, this process has not resulted in the hyper-efficient, cost-effective outcome they were hoping for.

But timelines and costs aside, perhaps the biggest challenge to be overcome is in the area of critical communications. As operators implement network slicing and edge computing, and they enable services such as V2X communications and services like remote surgery, we could find ourselves in a situation where a problem on the network could actually result in the loss of human life. This is not something CSPs have had to deal with in the past.

Q: It sounds as though with 5G the stakes are getting higher.  You mentioned edge computing. What is this – and how does it play into these new critical services and other offerings?
GH: Edge computing promises to enable many more services. The benefits are two-fold. The first one is that by providing computing power close to the end-user (the edge), as opposed to sending data all the way back to the core network, processing it and sending it back, the latency of communications is significantly reduced. Time-sensitive applications such as cellular V2X finally become possible. Putting computing power close to the user will also enable more processor-intensive services, such as haptic virtual reality and next generation gaming, where some, or all, of the graphics processing can be handled at the edge – meaning handset processing power will no longer be a limiting factor.

Edge computing will again require operators to collaborate more than they are today, but in this instance, it will likely be with 3rd parties. Will the computing power be owned by the operator, or by a game manufacturer or some other 3rd party? In either situation there will need to be close collaboration to ensure the dream is realized.

Q: The last area of innovation you mentioned was Massive MIMO. This is something that impacts a very valuable commodity – wireless spectrum.  Why is this such a critical component of a 5G Network?
GH: Massive MIMO is all about increasing the capacity available for a given block of spectrum and driving down the cost per MB of delivering data. Massive MIMO will deliver a significant increase in the capacity that can be achieved with a certain frequency block of spectrum. This means operators will be able to deliver more capacity to customers – without acquiring additional expensive spectrum.

But Massive MIMO also presents a number of challenges. On the planning front, accurate modeling of multi-path propagation, (which is about creating optimal pathways for wireless signals to travel), relies upon highly accurate, high-resolution 3D clutter maps. But when cities are constantly expanding and changing, accurate 3D maps are a hard thing to achieve. Even trees growing and losing their leaves over the seasons can impact service.

Implementation will also provide challenges. At lower frequencies, massive antenna arrays will be physically large and heavy, making mounting them on existing sites impractical in many situations.

Interested in learning more?  Stay tuned for Part 2 of Gavin’s interview, where he talks about some of the industry initiatives around OpenRAN.

Gavin Hayhurst OpenRAN

About Gavin Hayhurst:  Gavin maintains overall responsibility for the product marketing of TEOCO’s OSS and BSS software solution portfolio. He has over 18 years of mobile telecoms experience working for both network operators and global vendors of hardware, software, and services, including AIRCOM International, Ericsson, and MTN. Gavin holds a MEng in Industrial Engineering and a BSc in Electrical Engineering and is an avid travel photographer.

Read Part 2