Around this time last year, the mobile sector was finally starting to rally around the idea of looking ahead to 5G, even though there wasn’t really a consensus on just what 5G actually was - at least in terms of technology. Previous evolutions to 3G and 4G, while technologically innovative and challenging, were essentially next-gen RAN upgrades with related architectural improvements, which was fairly straightforward compared to the nebulous concept of 5G being kicked around at industry conferences.
However, there was a general understanding that “5G” was far more than a next-generation radio interface - it was that and a collection of other technologies, from existing LTE-A networks and Wi-Fi hotspots to small cells, SDN/NFV and cloud technologies, working together to meet the escalating requirements of mobile data usage. Put another way, the 5G concept was defined less by its technological components and data-throughput capabilities, and more by what kinds of services, applications and devices they would need to be able to support.
That concept was formalized in mid-2015 when the International Telecommunication Union’s ITU-R Working Party defined the overall goals, processes and timeline for 5G development.
Code-named “IMT-2020”, the ITU’s definition of 5G - and its performance parameters - were shaped by the ITU-R Working Party’s understanding of the current trends driving mobile data usage today and where all this is headed over the next five years.
Obviously there are a lot of details to be worked out between now and then - not just within the ITU but among other standards bodies like the 3GPP and IEEE, and industry organizations like the GSM Association and the NGMN Alliance, etc. But from this point on, the mobile sector has a much clearer idea of what 5G should be, and how to get there. A few operators in Asia-Pacific and elsewhere have already started down the road to 5G, although - true to form for the mobile tech sector - it seems not everyone will agree on when they’ve arrived at their destination.
IMT-2020 vision
According to Recommendation ITU-R M.2083-0 (which outlines the framework and objectives of IMT-2020), a key requirement for 5G is extremely low latency and high reliability for apps like cloud services, augmented reality and virtual reality, as well as M2M apps like driverless cars, real-time traffic control optimization, emergency and disaster response, smart grids and healthcare.
5G also must support high user density - not only in terms of crowded areas like shopping malls, stadiums, festivals, traffic jams and emergency scenarios, but also in terms of the Internet of Things (IoT). The IoT has its own requirements that go beyond density, such as power consumption, transmission power, and latency requirements - all of which will vary greatly depending on the specific device and application. A report from the Global mobile Suppliers Association (GSA) classifies key IoT apps into two broad categories: (1) massive machine-type communications (M-MTC), which involves huge volumes of low-cost devices and modules for sensor networks, connected home, smart metering, and (2) mission-critical apps like connected cars, industrial automation and health-related apps like remote surgery, where high reliability and low latency are critical.
5G also has to be able to provide “high quality at high mobility” - which is to say, it has to provide the same (or at least similar) quality of experience whether you’re sitting at home or on a high-speed train.
Not unexpectedly, it will also have to support mobile high-definition multimedia, from Ultra HD displays, multi-view HD displays and mobile 3D projections to immersive video conferencing, augmented reality and “mixed reality” displays and interfaces. And these won’t be limited to TV and games - they will also be used for medical treatment, safety, and security. That in turn means that more and more mobile apps and services will government-related, including e-government, smart cities, public protection and disaster relief communication, education, healthcare, etc.
And all of this will require support for extremely accurate positioning as new location-based services emerge for things ranging from emergency rescue services to drones.
Technology roadmap
IMT-2020 also takes into account technological developments already in the pipeline that will play a role in the evolution of 5G.
On the RAN side, the most well-publicized developments are centered around carrier aggregation, advanced antenna technologies like 3D-beamforming, active antenna systems, massive MIMO and network MIMO that promise better spectrum efficiency. Particularly in Asia, there’s also been a lot of work on dual-mode TDD-FDD and dynamic TDD to enhance spectrum flexibility.
There’s also a lot of activity to develop advanced waveforms, modulation and coding, and multiple access technologies aimed at improving spectral efficiency, such as filtered OFDM (FOFDM), filter bank multi-carrier modulation (FBMC), pattern division multiple access (PDMA), sparse code multiple access (SCMA), interleave division multiple access (IDMA) and low density spreading (LDS).
Other RAN technology developments includes simultaneous Tx/Rx on the same frequency with self-interference cancellation, flexible backhaul and dynamic radio access configurations, higher-order modulation for small cells, joint management of multiple radio access technologies (multi-RATs) and flexible uplink/downlink resource allocation.
On the network side, SDN and NFV are already in play, and are expected to play a key role in enabling operational network efficiency and give mobile networks the flexibility and visibility they need to support the deluge of apps, services and connections coming down the road.
Cloud RAN (C-RAN) will play a major role in ensuring that baseband and higher-layer processing resources can be managed and allocated dynamically on demand. Inter-cell coordination schemes like self-organizing network (SON) technology will enable operators to improve the opex efficiency of multi-RAT and multi-layer networks.
Then there are technologies designed to enhance mobile broadband service QoS, such as small cells, License Assisted Access (LAA), Evolved Multimedia Broadcast and Multicast Service (eMBMS), dynamic adaptive streaming over HTTP (DASH) to accommodate more video streaming content in existing infrastructure, context aware applications for more personalized services, and proximity-based techniques to enable direct device-to-device (D2D) communication.
This is 5G
With all that in mind - and with the key caveat that 5G has to be conceptualized and designed not only for the applications and services the industry knows are coming, but for the apps and services we haven’t even thought of - the ITU has developed a list of eight key parameters that should define 5G’s capabilities:
- Peak data rate: maximum achievable data rate (Gbps) under ideal conditions per user/device.
- User experienced data rate: achievable data rate (Mbps or Gbps) that is available ubiquitously across the coverage area to a mobile user/device. (Note that “ubiquitous” refers to the target coverage area, not literally everywhere in the country.)
- Latency: the contribution by the radio network to the time from when the source sends a packet to when the destination receives it (in ms).
- Mobility: maximum speed (km/h) at which a defined QoS and seamless transfer between radio nodes which may belong to different layers and/or radio access technologies can be achieved.
- Connection density: total number of connected and/or accessible devices per unit area (per km2).
- Energy efficiency: on the network side, energy efficiency refers to the quantity of information bits transmitted to/ received from users, per unit of energy consumption of the RAN (in bit/Joule). On the device side, energy efficiency refers to quantity of information bits per unit of energy consumption of the communication module (in bit/Joule).
- Spectrum efficiency: average data throughput per unit of spectrum resource and per cell (bits per second/Hz).
- Area traffic capacity: total traffic throughput served per geographic area (in Mbps/m2).
When IMT-2020 was first announced, many media reports focused primarily on the first parameter - peak data rate - if only because that’s the one consumers relate to the most. After all, faster data rates from the previous generation were the primary marketing focus of 3G and 4G services. Initial reports claimed that IMT-2020 would provide peak rates of 20 Gbps.
That’s true - in a sense. ITU-R M.2083-0 does assign actual values to each of the eight parameters, and it does say that 5G will support peak rates of up to 20 Gbps - under certain conditions and scenarios. Under other circumstances, peak data rates will be more along the lines of 10 Gbps, and that’s mainly for enhanced mobile broadband services (i.e. accessing the internet on your smart device like you do now, only with a faster and more efficient connection, and better latency, etc).
But data rates for 5G will vary depending on the scenario. The ITU document says: “IMT-2020 would support different user experienced data rates covering a variety of environments for enhanced Mobile Broadband. For wide area coverage cases, e.g. in urban and sub-urban areas, a user experienced data rate of 100 Mbit/s is expected to be enabled. In hotspot cases, the user experienced data rate is expected to reach higher values (e.g. 1 Gbit/s indoor).”
Which is another way of saying that with 5G, the exact peak rates won’t matter, so long as they deliver the best possible user experience for any given app or service.
The same goes for the other parameters. As Figure 1 (below) shows, the ITU has assigned benchmarks for each of them, but cautions: “The values are targets for research and investigation for IMT-2020 and may be further developed in other ITU-R Recommendations, and may be revised in the light of future studies.”
The ITU adds further that the benchmarks for peak data rates, mobility, spectrum efficiency and latency are basically extrapolations of its current benchmarks for IMT-Advanced (the ITU’s codename for LTE-Advanced and related technologies).
And as a further caveat, the ITU notes that the importance of each parameter is app-specific. For example, mobility isn’t that important in a hot-spot scenario, while a massive machine-type communication (MTC) scenario would place higher emphasis on connection density than fast data rates (depending on the app).
Spectrum issues
One of the hottest issues surrounding 5G is, of course, spectrum: not just the amount required but where to get it. According to another ITU report (ITU-R M.2376), “utilizing the bands between 6 and 100 GHz is feasible for studied IMT deployment scenarios, and could be considered for the development of IMT for 2020 and beyond.” And indeed, a lot of research at the moment is already looking at ways to use the higher-frequency millimeter-wave bands as spectrum resources for 5G.
For now, however, most available frequencies for mobile operators are in the sub-6GHz range. After last November’s WRC-15 conference, the GSM Association was able to secure harmonized bands for 700 MHz (specifically, 694-790 MHz) the lower 200 MHz of the C-band (3.4-3.6 GHz) and L-band (1427-1518 MHz). But new frequency allocations above 6 GHz weren’t on the table, and are not likely to be forthcoming until at least WRC-19, where millimeter-wave bands above 24 GHz will be on the docket.
When can we call it 5G?
Even so, no one’s going to wait around until WRC-19 to get cracking on 5G deployments. Many operators are already heading down that pathway by deploying many of the core elements of 5G, such as IoT platforms, NFV, carrier-grade Wi-Fi and carrier aggregation for LTE-A, to name a few. It’s early days, but the foundations for 5G are already being laid, even if most operators aren’t thinking of it as 5G. Meanwhile, there are already a handful of cellcos actively committed to 5G roadmaps and trials. In Asia, that includes all of the Korean operators (KT, LG U+, and SK Telecom), China Mobile, NTT DoCoMo, Singtel, Softbank, Telstra, and Vodafone Australia.
Perhaps inevitably, there is already some debate about just how many parameters an operator should achieve before they can officially declare their network or service “5G”.
This is partly because the above ITU parameters are guidelines, not hard rules. Official 5G standards won’t be ready for a few years, and the standards bodies working on them - including the ITU, the 3GPP and the IEEE, all have overlapping timelines (see Figure 2, above).
Also, 5G technologies won’t be rolled out uniformly or even everywhere. Legacy 3G and 4G will still exist, and 4G will continue to evolve, possibly to the point where it can actually meet some 5G performance criteria on its own. By 2020, many initial 5G rollouts will be highly targeted and limited to specific areas and apps.
As for when it starts counting as 5G, it depends who you ask. A November 2015 survey from the GSA provided operators, vendors, regulators, analysts and consultants with a list of 5G parameter performance benchmarks and asked how many of them operators would have to meet to credibly brand the network 5G. Over 40% of operators said all criteria would have to be met, but some said “a few,” while a small percentage said just one benchmark would do. Vendors were considerably more lax in their assessments of what would count as 5G, with close to 25% saying just a few benchmarks would have to be met.
(For the record, the GSA’s stance is that in order to qualify as “5G”, the two defining network characteristics would be a new air interface alongside the legacy ones, and sophisticated network slicing capabilities via SDN and NFV that intelligently match resources to applications).
The GSA’s forecast for 5G rollouts takes this into account. Accordingly, the GSA says we’ll only see between ten and 15 “early 5G” commercial launches - in other words, 5G offerings that meet some but not all benchmark criteria.
The industry group predicts that commercial deployment will begin in earnest after the standardization process is completed in 2020, with over 270 operators having introduced commercial 5G services by 2025. However, the report says, “this does not mean those networks will be extensive.”
The GSA expects coverage to likely be initially limited to dense urban areas, with cellcos “using whatever frequencies, modes and technologies make sense to an individual operator in any given location in order to deliver indoor and outdoor coverage - but in particular in hot spot locations where capacity starts to run out. We do not anticipate many widespread 5G network deployments within the forecast period.”
This article was first published in Telecom Asia 5G Insights February 2016 edition