10 questions on 5G

6 Jan

As 2014 approaches, conversations around “4G” almost seem antiquated. That’s not to say 4G is old news, but the buzz and excitement around the promise of 4G isn’t as prominent as years past. In fact, the new trend seems to be circling around “5G.” However in a world where we’re still in the very early stages of 5G and assuming history repeats itself, we should only expect recognized standardization to begin happening towards the end of this decade. In fact, in terms of being an everyday technology like W-CDMA and the way LTE is soon becoming, it is probably reasonable to say that sometime after 2025 is a fair time horizon for 5G becoming an everyday technology.

So what should we expect in the next decade? What needs to be done in order for 5G to be a reality? And which companies will set themselves apart from others as leaders in the space?

Question No. 1: What technologies are considered as “5G”? Is it too early for these technologies to exist due to the fact that 4G is still in the early stages of adoption?

I don’t think the future just suddenly happens like turning a corner and magically a next generation appears. There are always signs and trends along the way that provide directional indicators as to how the future will likely take shape. So-called 5G will be no different than previous generations whose genesis was seeded in societal challenges and emerging technologies often conceived or identified decades earlier. 5G wireless will be driven by more efficient network architectures to support an internet of everything, smarter and new approaches to spectrum usage, energy-centric designs and more intelligent strategies applied to the handling of content based upon context and user behaviors. From these perspective technologies/trends like the cloud, SDN, NFV, CDN (in the context of a greater move to information-centric networking), cognitive radio and millimeter wave all represent interesting first steps on the roadmap to 5G.

Question No. 2: Has the 5G standardization process begun?

No. However it is very reasonable to say that as an industry we are at the beginning of what might be described as a consensus building phase. If we reflect on wireless history seminal moments, they may be where the next “G” began. The first GSM networks rolled out in the early 1990’s but its origins may be traced back as far as 1981 (and possibly earlier) to the formation of Groupe Spécial Mobile by CEPT. Third-generation and 4G share a similar history where the lead time between conceptualization and realization has been roughly consistent at the 10-year mark. This makes the formation of 5G focused industry and academic efforts such as the 5GPPP (in Horizon 2020) and the 5GIC (at the University of Surrey) in 2013/14 particularly interesting. Assuming history repeats itself, these “events” may be foretelling of when we might realistically expect to see 5G standards and later deployed 5G systems.

Question No. 3: What are the requirements that need to be met for 5G development?

Requirements for 5G are still being discussed. However the best first stab at such requirements is reflected in the good work of the 5GPPP (in Horizon 2020). Some of the requirements that have been suggested thus far have included:

–Providing 1,000-times higher capacity and more varied rich services compared to 2010.

–Saving 90% energy per service provided.

–Service creation from 90 hours to 90 minutes.

–Secure, reliable and dependable: perceived zero downtime for services.

–User-controlled privacy.

–Latency drastically reduced (one millisecond has been suggested).

Question No. 4: What are key components of 5G technologies?

I think 5G will bring profound changes on the both network and air interface components of the current wireless systems architecture. On the air interface we see three key tracks. The first track might be called the spectrum sharing and energy efficiency track wherein a new, more sophisticated mechanism of dynamically sharing spectrum between players emerges. Within this new system paradigm, it is quite reasonable to discuss new and more suitable waveforms. A second track that we see is the move to the leveraging of higher frequencies, so called mmW applications in the 60 GHz bands and above. If 4G was the era of discussing the offloading of cellular to Wi-Fi, 5G may well be the time when we talk of offloading Wi-Fi to mmW in new small cell and dynamic backhaul designs.

A final air interface track that perhaps bridges both air interface and network might be called practical cross layer design. Context and sensor fusion are key emerging topics today and I believe that enormous performance improvements can be realized through tighter integration of this myriad of information with the operation of the protocols on the air interface. While real infinite bandwidth to the end user may still remain out of reach in even the 5G time frame, through these mechanisms it may be possible to deliver a perception of infinite bandwidth in a very real sense to the user. By way of example, in some research and development labs today organizations have developed a technology called user adaptive video. This technology selectively chooses the best video streams that should be delivered to an end user based upon user behavior in front of the viewing screen. With this technology today bandwidth utilization has improved 80% without any detectable change in quality of experience perceived by the end user.

Question No. 5: How will 5G impact the networks?

I believe 5G will be shaped by a mash up (and evolution) of three key emerging technologies: software defined networking, network function virtualization and an ever deeper content caching in the network as exemplified by the slow roll of CDN technology into GGSN equipment today (i.e. the edge of the access network). This trend will continue deeper into the radio access network and, in conjunction with the other elements, create a perfect storm where an overhaul to the IP network becomes possible. Information-centric networking is an approach that has been incubating in academia for many years whose time may now be right within these shifting sands.

Overall, the network will flatten further and a battle for where the intelligence resides either in the cloud or the network edges will play out with the result likely being a compromise between the two. Device-to-device communications in a fully meshed virtual access resource fabric will become common place within this vision. The future may well be as much about the crowd as the cloud. If the cloud is about big data then the crowd will be about small data and the winners may well be the players who first recognize the value that lies here. Services in this new network will change. A compromise will be struck between the over-the-top and carrier worlds and any distinction between the two will disappear. Perhaps, more than anything else 5G must deliver in this key respect.

Question No. 6: How do organizations in the 5G space play into 5G millimeter waves?

For some organizations that have seen early on the need for higher frequency spectrum, mobile research has always been driven by staying ahead of the curve. This has been a historical trend since 1G that the industry didn’t see stopping at 4G. Therefore they have built up a comprehensive set of research tools (propagation models, link level and network level simulations) to allow testing of millimeter-wave architecture vision. In order to succeed, these organizations must work with industry partners to build consensus for how millimeter wave will be used for 5G.

Question No. 7: How much would capacity increase with 5G millimeter waves? How would it change cellular architecture?

There are many times that more spectrum is available in millimeter wave bands than is currently used in all of cellular and TV/radio broadcasting combined. The capacity gains based on spectrum availability could easily exceed 1,000-times today’s networks. Changes to the architecture along with associated costs will likely set the realized capacity gains. Dense “small cells” will be needed, due to the short propagation ranges at millimeter wave. Wireless backhaul will be essential in many deployments where wired backhaul is not available and costly to install. Mesh networking can be used to provide connectivity of millimeter wave hotspots. Since these will initially be deployed as hotspots, dual connectivity between established 4G and higher capacity 5G mmW bands will need to deliver a seamless experience to the user.

Question No. 8: What will set companies apart in the development of 5G?

Companies who have a “knack” for working on industry problems before the industry knows that they are problems will be set apart from others. Specifically, companies who are doing research and development work in areas that might be considered foundational to 5G will prevail. Wi-Fi in white space is an early embodiment of a new more efficient spectrum utilization approach that is highly likely to be adopted in a more mainstream manner in 5G. More than this, companies are also exploring new waveforms (new proverbial “four letter acronyms” that often characterize a technology generation) that outperform LTE “OFDM” in both energy efficiency and operation in this new emerging dynamic spectrum sharing paradigm. The next generation will have many touch points in some of today’s emerging technologies. These are just two examples of 5G R&D currently going on in R&D labs today, but there are many others.

Question No. 9: What are the uncertainties and challenges of 5G development?

The next generation of wireless technology will likely roll out into a world of considerably changed business models from its predecessor generations and this raises perhaps the greatest uncertainty and challenge. What will these business models look like? It is clear that today’s model where carriers finance huge infrastructure investments but reap less of the end customer rewards is unsustainable over the longer term. Some level of consolidation will inevitably happen but 5G will also have to provide a solution for a more equitable sharing of the infrastructure investment costs. Just how these new business models take shape and how this new thinking might drive technological development is perhaps the greatest uncertainty and challenge for 5G development.

Question No. 10: What are the benefits of 5G development?

Even the most conservative traffic forecast projections through 2020 will challenge the basic capabilities and spectrum allocations of LTE-A and current generation Wi-Fi. Couple this with a recognition that energy requirements in wireless networks will spiral at the same rate as the traffic projections and add the chaos of the emergence of the 50 or 100 billion devices – the so called “Internet of everything” all connected to a common infrastructure and the value of exploring a fifth generation should quickly become apparent. The benefits of 5G at the highest level will simply be the sustaining of the wireless vision for our connected societies and economies in a cost effective and energy sustainable manner into the next decade and beyond.




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: