With LTE, the cycle of hope and disappointment has been squeezed dramatically compared to preceding generations. The technology was, for a few brief years, hailed as the answer to the desperate plight of the cellcos ‘ facing exploding data traffic and capex requirements, but with no corresponding promise of revenue hikes. This plight, as well as the kick in the pants administered by WiMAX, accelerated development of LTE and wide-scale roll-out has proceeded at a far quicker pace than it did for 2G and 3G. But only four years after TeliaSonera switched on that first commercial LTE network, operators know that the 4G upgrade is not nearly enough.
Mobile data growth continues to rise sharply, only alleviated to some extent by Wi-Fi offload, compression and optimization systems. And while LTE does enable some new charging structures and revenue streams, these do not match the rise in capex required to meet customers’ demands for more data, with improved quality of experience, and at much the same price as before. A dramatic rethinking of the whole network is needed to drive down cost of delivery, allocate capacity flexibly where it is most needed, and support massive bandwidth affordably. The air interface upgrade itself delivers only just over 15% of the additional capacity and efficiency that will be required by 2020, according to a survey of 27 tier one cellcos by Maravedis-Rethink. Many other techniques will have to be used in combination with LTE, and carriers are currently poring over the long menu of options ‘ offload, small cells, optimization, virtualization, smart antennas and so on.
In this discussion, LTE is already looking pass’, and most operators with significant urban and data-hungry populations are already planning deployments of key elements of the LTE-Advanced family of upgrades (3GPP Releases 10 and 11), in 2014-15 or even this year. And while talk of ‘5G’ is still mainly marketing fluff ‘ hands up SK Telecom, NTT Docomo, Samsung and Huawei ‘ the push towards the next generation of network enhancements, to deliver even better spectral efficiency and even smaller cells, is starting to happen in the labs. As Docomo pointed out at this week’s Broadband World Forum event in Amsterdam, when some attendees scoffed at its promise to deploy 5G for the 2020 Tokyo Olympic Games, ‘2020 is not really that far away’.
A combination of new standards and new ways of planning the network will be needed to optimize the efficient use of spectrum and capacity. The deployment of a layer of small cells will be a common approach, and an important step towards the fully-blown HetNet (heterogeneous network, another overused term but the ultimate goal of most cellcos. This will create a common pool of capacity, harnessing different technologies, spectrum bands, cell sizes and backhaul links in a fully integrated way.
But the macro layer remains the primary source of coverage and capacity, and it too must be modernized to provide better total cost of ownership, cost of data delivery and overall efficiency. There are two main approaches to this. One is Cloud-RAN, in which carriers deconstruct their networks completely, virtualizing the baseband functions in base station hotels or on standard cloud servers, and leaving just low-cost, low power radio/antenna units at the cell sites (which may be increasingly small). The other is to add more intelligence to that cell site and create what the UK’s leading cellco, EE, dubs the ‘super macro’. These approaches need not be either/or ‘ many operators will use a mixture of both, or will converge the two into a hybrid system where super macros act as ‘master cells’ within a Cloud-RAN, localizing some functions like caching and supporting a large number of more basic sub-cells, which may use a distributed antenna design.
Similarly, most operators will use a combination of Cloud-RAN and small cells on the road to HetNet, even though these two are often presented as mutually exclusive (or even confused with one another). Small cells are initially being deployed, in the public access environment, to fill gaps in the 3G network, but their real disruptive effect will be felt when they form a separate layer of capacity, interworking completely with a reconstructured macro layer.
LTE-Advanced functions such as eICIC and CoMP will make such an architecture easier to roll out, addressing some of the issues of interference between layers, and cell edge performance (indeed, many believe the main motivation to invest in Cloud-RAN will be to support CoMP better). According to the latest research by Maravedis-Rethink’s RAN Service, which tracks the network strategies of the top 100 LTE operators, cellcos will spend $3.45bn on Cloud-RAN equipment between 2013 and 2018, in order to reduce cell site costs, allocate capacity more flexibly between sites, and better enable LTE-Advanced in the macro layer.
The report, ‘Tearing the Network Apart: The Economics of the New RAN to 2018’ found that carriers are preoccupied with reducing cell site equipment costs as they ‘densify’their networks to meet capacity demands. That will involve investing in integrated, autonomous small cells, but also shrinking macro layer cells to an extent where traditional base station architectures become unaffordable. The low footprint radio head/antenna units that are coming to market ‘ such as Alcatel-Lucent’s lightRadio or Ericsson’s AIR ‘ indicate how the macro RAN is increasingly distributed, with several cell sites sharing a common baseband resource. C-RAN will take that further, virtualizing those baseband functions in software on standard server platforms, and supporting hundreds of sites per central resource. Operators will slash costs by leaving only ultra-low cost equipment at the cell site, eventually driving the equipment cost down below $100 by 2020.
However, while C-RAN economics are very tempting, but they will only be realized by investing heavily at the back end, and by harnessing standards to ensure interoperability and reduce cost. That, in turn, will create a whole new ecosystem, as carriers are set to deploy almost 2.5m cloud-connected macro layer sites from 2013 to 2018. Many of these macro sites will be equipped with extremely low cost antenna/radio units, creating a significant opportunity for vendors in the antenna and remote radio spaces, and for a reworking of the DAS idea for C-RAN. There is also considerable R&D effort going into the IT-centric back end, notably in the massive China Mobile/Intel project with various RAN partners.
And SK Telecom of Korea, one of the major investors in turning C-RAN into a reality, recently announced a joint project, also with Intel, which is naturally a huge supporter of the architecture, given that it would end up running the functions of the network on x86 servers rather than specialized base stations and routers. The two companies are collaborating on a virtualized base station (vRAN), which can load and reconfigure key network functions in software on a general purpose server, in the same way that apps are loaded and deleted. Under the new agreement, the two firms have agreed to build a vRAN testbed and carry out joint R&D on intelligent base stations and cloud base stations.
However, C-RAN faces many barriers, notably the immature state of telecoms virtualization systems, and the current need for fiber to link the basebands to every radio head. Development of virtualization standards and of microwave fronthaul technologies will be critical to improve the business model.
While a few pioneers, such as China Mobile and Korea Telecom, lay the groundwork for a significant uptake of C-RAN from 2016, others are pushing the super macro as a shorter term option, and one that can prolong the life of investments in existing technologies, even delaying the need to splash out on small cells and/or C-RAN.
Andy Sutton, principal network architect at EE, told the recent Base Station Conference in London that it aims to supersize some of its macrocells as part of a HetNet roadmap which also includes small cells, Wi-Fi and other emerging technologies such as carrier aggregation. EE currently covers about 55% of the UK population with LTE and recently doubled its peak download speed to 150Mbps by refarming more of the 1.8GHz GSM band in which it is deploying its first network. Later, it will adopt carrier aggregation to make best use of additional spectrum in 800MHz and 2.6GHz, acquired at auction early this year.
Sutton told the conference: “Super macro is the first step toward building a HetNet. Evolving the macro is the most cost-optimized way to add capacity into our networks.” A super macro would typically support 2G, 3G and 4G in multiple bands with 3-6 base station sectors and carrier aggregation. Such a site would often be a master cell, acting as a hub for smaller microcells or distributed antennas.
Super macros use some of the techniques which were well tested in 3G for adding capacity, coverage and efficiency ‘ adding more base station sectors, vertical sectorization, attaching microcells, and so on. They will also be increasingly reliant on emerging antenna techniques such as ‘super MIMO’ (4×4 and beyond, or ‘massive MIMO’ as China Mobile is trialling), and AAS (Active Antenna System). Indeed, most ‘5G’ R&D centers almost entirely on antennas, along with software-defined networking and the effective use of many spectrum bands, including ones currently considered unsuitable for mobile broadband access, such as 60GHz. Increasingly, the radio will be integrated entirely into the antenna, allowing for sophisticated beamforming, vertical sectors and multiband, multimode support, with MIMO arrays adding increased range and bandwidth.
Takehiro Nakamura, director of Docomo’s radio system design group, defines 5G as 1,000-fold increase in access network capacity, needed to support the proliferation of connected devices. The cellco’s research centers on using large amounts of high frequency spectrum (as high as 20GHz), coupled with massive MIMO arrays. Takehiro said his firm’s simulations showed a 30-fold capacity increase by using 100MHz of 3.5GHz spectrum and dividing a cell between 12 smaller ones. The same number of cells in 400MHz of 10GHz spectrum would deliver 125 times more capacity, and in 1GHz of 20GHz spectrum with massive MIMO, they could reach the magic figure of 1,000 times more capacity.
Ericsson CTO Ulf Ewaldsson is also considering what 5G might look like, and says it will need a new approach in order to support key trends like cloud networks and the internet of things. “It would not be meaningful to just go out and develop a new radio,’ he told the recent GigaOM Structure event in London. He is looking, like most 5G thinkers, at harnessing very high frequency bands, above 30GHz and advanced antenna arrays ‘ but also preserving backwards compatibility with the other 3GPP generations, of course. However, he thinks the 3GPP and other such bodies must take a back seat in 5G. “The requirements on 5G will come from the applications, the users, and the devices, whereas the requirements on the previous systems have been from think-tanks of standardization people,” he suggested.
In Ericsson’s view of 5G, services will reside in clouds and the wireless networks will need to provide high quality, ultra-precise, ultra-high speed delivery from those clouds to any kind of device. The speed and precision could be adapted automatically by the network according to the type of device or application and its urgency ‘ or even according to a user’s mood.
There are many 5G R&D projects such as the EU’s METIS program, the UK’s 5G center at the University of Surrey, and a host of operator and vendor initiatives. And Docomo is right ‘ time is not on anyone’s side. With basic LTE running out of steam almost before it has launched, it will nevertheless have considerable potential for enhanced performance and efficiency with the LTE-A and subsequent upgrades. But around the nominal deadline for 5G, the end of the decade, there is general agreement that current assumptions about radio and network design will be nearing their ceiling and a world of massive antenna arrays and software-defined networks will have to be accommodated in standards.
All that will kick off the cycle of capex investment again, so the emerging business cases around M2M and other areas will need to deliver convincingly if carriers ‘ traditional ones or the Googles of the world ‘ are to be tempted into commercial 5G. Alcatel-Lucent’s corporate CTO Marcus Weldon put it best in Amsterdam this week, saying: “The problem in front of us is not technology, but confidence and ambition. The technologies are within our grasp. The architectures are within our grasp. It’s really a question of choice and the economic reality.”
For more information about Maravedis-Rethink’s new RAN Service report, ‘Tearing the Network Apart: The Economics of the New RAN to 2018’ please contact Adlane Fellah on [email protected].
Pushing the LTE standards -recent breakthroughs:
‘ Australia’s Telstra, always in the forefront of the latest network advances, is to start trials of LTE-Broadcast by the end of this year. LTE-B can broadcast multimedia content and software to many users simultaneously, allowing operators to save on bandwidth and improve efficiency. Networks director Mike Wright said the trial would aim to determine how the technology could be best used, whether cost savings would ensure, and what the impact on the LTE systems would be.
‘ NSN has enhanced its Smart Scheduler – which it calls the ‘crown jewel’ of its LTE implementation and its Flexi Multiradio family ‘ to support carrier aggregation (CA) for FDD or TDD spectrum, tapping into the experience of working with the early-stage deployments of CA by the three Korean cellcos. The upgraded scheduler promises to deliver up to 50% higher uplink data rates at the cell edge and up to 20% more uplink cell capacity, in FD-LTE or TD-LTE networks. The upgrade can be done in software on Flexi MultiRadio Base Stations.
‘ SK Telecom and Ericsson claim to have solved the main cause of signal degradation in a small cell environment, developing a technology that could eliminate the need for handover in LTE-A networks. The technology enables multiple base stations to broadcast the same signal to terminals at a cell edge, removing the need for inter-cell handover. This could improve transmission rates by 1.5 to two times, and total cell system capacity by 5%-10%, said SKT, which plans to implement the technique next year using the core LTE-A technology Transmission Mode 9 (TM-9), which was first demonstrated by the operator, with Ericsson, in January. TM-9 allows base stations to send different demodulation reference signals to each mobile device located within a coverage area, and enables each base station to use its unique channel information reference signal to communicate with terminals. The partners have also developed a successor to the eICIC element of LTE-A, called RPS (Reduced-Power Subframe) which claims to improve cell system capacity by as much as 62% compared to its predecessor.
NSN addresses the small cell layer:
To fulfil their maximum disruptive potential in the new RAN, integrated small cells need to have comparable performance to a conventional macrocell, but at far lower cost, power and footprint. This is a challenge since, unlike the stripped-down C-RAN cell site, the metrocell has the full complement of functionality integrated into its diminutive box (though in compensation, it does not have the super-high capacity fronthaul requirements of the distributed base station).
The latest vendor to claim a breakthrough in this regard is NSN, with its Flexi Zone LTE microcell and picocell. These are ‘true’ small cells ‘ not just traditional designs shrunk-down, but based on a single SoC (system-on-chip), one of the key aspects of small cell economics. ‘We’ve worked very hard with our chipset manufacturer to combine different chipsets on the same silicon in the same small cell unit,’ St’phane Daeuble, head of small cell and Flexi Zone product marketing, told Mobile Europe. ‘This means we can make the right chipset at the right place to give us the processing power we need.’
Both units come in a five kilogram, five liter box with 5W output power, but claim the same processing power as a macrocell, and can run the same software as their larger cousins in the NSN Flexi base station range.”This makes it easy to integrate them into complex HetNets and drastically simplifies software upgrades across the network so that the same features are available everywhere at the same time,” said the company.
Initially, NSN expects the products, which ship early next year, to be installed singly to address capacity hotspots in 3G/4G networks, but later they will become part of dense hotzone roll-outs and eventually a full small cell layer within a HetNet. The company already provides Flexi Zone, a platform for integrating cellular and Wi-Fi small cells, and will take this a step closer to the HetNet next year when it releases a new Flexi Zone controller. This will turn clusters of small cell hotspots into hotzones, making all the cells appear on the network as a single eNodeB to simplify network management.
Daeuble added: ‘You can have small cells without macro parity, but you’re going to spend lots of money going back into the macro tweaking the network and moving antennas. It’s going to get quite complicated. Operators cannot afford to spend the same time optimizing small cells as they spend with the macro network.’