In some ways, 2015 was the year when the macro base station reasserted itself at Mobile World Congress. All the major OEMs announced significant upgrades to their macro layers (see Wireless Watch February 26 2015 for details), which meant their key suppliers – especially the antenna makers – were also demonstrating big leaps in this market.
Massive MIMO, 3D beamforming, wideband and active antenna systems – all these were hot topics for companies like Kathrein and CommScope, as well as the big OEMs. However, when operators and vendors looked ahead to what might lie in 5G, there was a growing consensus that, at that stage, innovation would be all about small cells. Companies like ZTE are very focused on ever-larger MIMO arrays, but while they may label these ‘pre5G’, they are really about extracting more performance from future releases of LTE.
A new air interface is not required to do that, and a repeated theme of MWC was that, with the 5G dawn around 2020, the macro layer would continue to provide broad coverage based on LTE-Advanced for many years to come, perhaps indefinitely. The LTE specifications, and architectures such as Massive MIMO, would therefore continue to be enhanced well into the 2020s (as EDGE prolonged the life of GSM well after 3G launched, and HSPA+ did the same for W-CDMA). The real 5G activity would focus on multiple layers of additional capacity, perhaps with different air interfaces for different activities (such as internet of things and high performance data), and these would be centered around small cells of various kinds, and often around high frequency bands (see separate item).
The small cells in ‘5G’ will certainly come in many forms, some of them not yet envisaged, though they may be hinted at in innovations like Parallel Wireless’s mini-Cloud RAN (being deployed by EE for rural coverage in the UK); or Artemis’s pCell (which creates personal cells of capacity around individual devices dynamically, and is currently being trialled in Dish’s spectrum in California).
These include features which will be part of the small cell layer well before 2020, and are also not dependent on a new air interface, working quite happily with LTE. Far more important than the 4G-to-5G transition is the shift to software – in virtualization, programmable networks and software-defined radios. The ability to allocate spectrum and network resources dynamically; create and close down multiple instances of a network function as usage demands; move nimbly between different available frequencies and technologies – these are the hallmarks of the future small cell, and they are starting to show through in the architectures. Indeed, the Small Cell Forum confirmed that its next Release Program will focus on virtualization with regard to mini-base stations.
So the key themes for small cells are densification and virtualization. To address these shifts, the cells themselves are becoming more diverse, optimized for different use cases as well as different network philosophies. The larger vendors are bringing out many different form factors within one portfolio – Ericsson and Huawei have their mini-vRANs, Radio Dot and Lampsite, living alongside conventional picocells and compact radio/antenna units.
ZTE claimed the world’s smallest outdoor baseband unit (BBU), as part of its Pad series of base stations, indicating how miniaturization is coming to all the elements of the RAN, not just the self-contained small cell. Like those cells, the compact BBU is designed to be easily concealed in street furniture or other sites, and it can then support multiple antenna/radio endpoints. The Pad range also boasted a new remote radio unit (RRU), tablet-sized and with in-built antennas. These RRUs can be mounted on lamp posts or buildings and several can be included in a single logical cell to reduce handovers. ZTE said its all-in-one micro base station is also one of the smallest in the industry, and with transmit power of 2x5W, it is targeted at urban densification roll-outs.
One of the debates is how far cellular and WiFi will converge in the small cell layer (see separate item). One of the core notions of the HetNet, and of 5G, is that WiFi and cellular should be interchangeable within a network, but that can mean many things – a common small cell, at box or system-on-chip level, like Alcatel-Lucent’s new Wireless Unified Network; separate access points sharing the same spectrum, as with LTE-LAA and WiFi in 5GHz; separate networks managed by a common packet core and BSS/OSS; devices capable of jumping transparently between the best connections available at any one time. For some, too much convergence will lead to a lowest common denominator approach, and they want to keep the strengths of LTE and WiFi, but bring them under common control.
Whatever the chosen approach to convergence, the systems which manage and coordinate the cells will become more software-driven and cloud-based, and so enable new business models themselves. The cells may form clusters with their own controller (local or virtualized) to support an enterprise or a rural deployment.
For the former, the big news was that Cisco will resell the Spidercloud Enterprise-RAN solution, despite its own 2013 acquisition of small cell pioneer Ubiquisys. Cisco will resell the whole Spidercloud portfolio under the USC 8000 Series brand and the smaller firm will also develop custom technology for its partner, including 3G and 4G modules to integrate into the Aironet 3600 and 3600 enterprise WiFi access points. This is significant news for Spidercloud, which has played a patient game since it was founded in 2006 around its Enterprise-RAN concept.
The E-RAN platform promises operators a route to more strategic relationships with enterprise customers, via a scalable in-building network of small cells that is run from a dedicated central services node (SCSN). Each SCSN can handle about 100 cells and appears to the core network as a single entity. The system can handle over 100,000 data sessions and hand-offs a day with just one connection to the carrier’s core, allowing for scalability in large enterprises or venues.
The SCSN handles functions like cell coordination and inter-small cell signaling (soft handover signaling for 3G and X2 signaling for LTE). This saves every small cell having to connect back to the cellco’s gateways individually. The architecture also enables Local IP Access (LIPA), for local switching and local break-out, which is needed to ensure inter-small cell mobility, and localized services or content caching.
Incubated and supported by Vodafone, which has deployed E-RANs in the Netherlands and elsewhere, Spidercloud also has a reseller agreement with NEC. In the past year, its progress has gathered pace, beyond these two key partners, and a US breakthrough came recently when Verizon added the system to its offering. But Cisco is clearly the partner everyone wants in the enterprise space, and a couple of years ago it appeared to be a closed door when it acquired its own small cell technology, via Ubiquisys. That acquisition was interpreted as Cisco’s push into the RAN, using enterprise – and in future, perhaps, outdoor – small cells as its back door into Ericsson’s territory and account control. However, it is now unclear how the giant will position its inhouse technology alongside Spidercloud (which is perhaps the acquisition it should have made), while we can assume its own ‘snap-on’ cellular modules for Aironet access points will be replaced with the rebadged offerings.
Spidercloud has updated its range too, adding dual-band radio nodes which support
simultaneous 3G/LTE or dual-carrier LTE services, and are now software upgradeable to support carrier aggregation with peak rates up to 300 Mbps. A software update also allows the dual-band devices to be shared by two operators, often a key requirement to make the business case for a large indoor deployment in a public space like a mall. The firm has also added integrated Bluetooth Low Energy beacons in the radio nodes to improve installation and maintenance by making it easier to locate nodes.
For the rural market, the Small Cell Forum kicked off its latest Release Program, devoted to easing deployment issues in rural and remote scenarios, from villages to oil rigs to temporary situations such as disaster relief. Quortus, with its virtualized packet core, was one of the first to update its portfolio to target this important area, while Parallel Wireless also relies heavily on virtualizing the baseband functions.
According to research for the Small Cell Forum by RealWireless, about 650m users worldwide could benefit from small cell solutions that would cover rural and remote locations cost-effectively. That is a market estimated to be worth $163bn in equipment and services. This explains the new level of innovation in a sector which has traditionally been the poor relation of RAN R&D, as companies race to achieve the price:performance ratios required for sparsely populated areas. A macrocell with only a few inhabitants was always hard to make profitable, but with small cells, a low cost base station can be used to target small groups of people very specifically (though backhaul can often remain an issue).
One of the companies whose offering fits well with rural and remote requirements is Quortus, which has an implementation of its ECX Enterprise platform for this sector. The system reduces cost and backhaul load by virtualizing many of the functions of the packet core at the network edge – running them as software on the small cell itself or on other local hardware. This enables a self-contained and secure local network to be established, under the control of the local MNO or MVNO, which can offload voice, data and signalling from the macro. It can also be tailored to specific needs, with benefits like IP-PBX and LAN integration for enterprise installations such as oil rigs, and local ‘community of interest’ features for rural communities. It also features a range of features to optimize the efficiency of the satellite backhaul link – the only option in many remote areas – and to reduce its load by offloading data to the local LAN.
Meanwhile, Parallel Wireless’s approach to rural and remote deployment involves a ‘micro network’ or cluster of small cells, connected back to a host base station over microwave in a similar way to Cloud-RAN. The system relies on a mesh of small cells to cover the village itself. These are fairly basic antenna units, which are connected to, and managed by, an LTE access controller which can support outdoor and in-vehicle cells. That, in turn, is backhauled, via an in-band last mile wireless link, to a macro base station almost four miles away (a distance which could be extended up to 10 miles in future, according to the vendor). The orchestration of the access points is run in software on the controller, which supports the NFV standard.
The technology avoids two costs which usually make rural mobile coverage economically inviable. There is no need to install a full-scale cell tower just to serve a small number of people within its range. And there is no need to run underground cables to provide backhaul. Instead, said initial deployer EE, three or four microcells can cover half a square mile and up to 150 buildings. The operator expects users to get average data speeds of 20Mbps outdoors and slightly less indoors. In cost terms, if the villagers do not charge fees for the access point sites, EE says it can connect rural communities for about 30% of the cost of adding a macrocell, as well as reducing backhaul costs by using microwave in its own spectrum.
In more crowded markets, the classic small cell is expanding its reach, seeking to provide greater value than basic coverage and capacity. Ip.access, another of the founders of this industry, has gone as far as to position its Presence Cell purely as the enabler of big data and e-commerce services – and not necessarily connected to the main network at all.
Its approach has convinced Vodafone, which announced that it would deploy the retail-oriented platform in Turkey, where Ilker Serbest, location solutions manager, said: “We have been looking for a reliable and cost-effective solution for network monetization and working with ip.access we have found a high resolution, reliable and rapid deployment solution.” The company has already rolled out presenceCells in its own stores across Turkey and will now move on to third party locations.
The presenceCell captures users’ location and identify information from their mobile devices, within the operator’s permissions framework, and the data can be used to target the user with personalized offers or services, and to feed into analytics packages. The technology is an example of operators looking for more reasons to deploy small cells than just to fill coverage and capacity gaps, and to harness the location awareness inherent in a dense network. ip.access argues that the authentication of the user, via the SIM, and the security of the cellular network, can support higher value services for the retailer and operator than solutions like beacons, including fraud prevention.
Then there were small cells which did not follow the traditional architecture. Stripped-down antenna/radio units for Centralized-RAN; separate antennas optimized to work with urban small base stations, from companies like Kathrein and CommScope; a converged WiFi/cellular unit from Alcatel-Lucent; hosts of carrier WiFi access points and management platforms. This is a segment where all options are open, and in which operators will pick and choose the solutions which suit their individual spectrum, business model and capacity requirements.
The virtualization of the RAN is a more distant prospect, for most operators, than the lower risk decision to run a packet core or even a CPE as software on off-the-shelf hardware. However, some pioneers were demonstrating their vRANs, notably Telefonica and China Mobile.
ASOCS has been a prominent start-up in the Cloud-RAN world, arguing that virtualization will eventually enable a ‘$100 cell site’ since so little will be left to deploy in hardware. At MWC, it was working with Radisys to show off an NFV-compliant virtual base station for localized C-RANs, using Radisys’ CellEngine LTE software with ASOCS’ vBS architecture and running on standard servers. Tom McQuade, general manager of software and solutions at Radisys, said in a statement: “By deploying radios in the network and moving the functionality to the cloud through the deployment of virtual base stations, mobile operators can take advantage of processing aggregation and dynamic allocation of resources from a central processing unit, thereby improving network efficiency and reducing costs.”
Also showing off a vRAN was Alcatel-Lucent, which has been prominent in early trials of the approach with China Mobile and Telefonica, and says its solution will be commercialized and ready for customer trials later this year, with full deployments expected in 2016. Its virtualized baseband unit runs on Intel Xeon processors and supports NFV.
Not everything can be converted to software of course, though even the physical elements like antennas and radios will be increasingly software-defined and programmable. Pushing that trend to its extreme was Cambridge Consultants, which has developed the IP for the first all-digital radio transmitter, Pizzicato.
Unlike conventional software defined radio, it has no analog components, which allows many radios to work together without interference. In the first trial, Cambridge Consultants created 14 simultaneous cellular base station signals at low power, and with the radios “squashed together in a way that analog doesn’t tolerate”. Such solutions can be programmed to generate manhy combinations of signals at any frequency in an adaptive way. The Pizzicato transmitter consists of an integrated circuit outputting a single stream of bits, and an antenna, with patented algorithms to perform the real time computations.
The technology is not a commercial product, and its creator is open to different licensing models – or just to moving the industry closer to a situation where Moore’s Law, which has defined digital processors, can also be applied to radio. This is about manufacturing process – which under Moore’s Law delivers low cost and power consumption in every generation – rather than base technology, said Monty Barlow, director of wireless technology at the consultancy.
The need for such a new architecture is being driven by the rising data rates in mobile broadband and the proliferation of devices in the IoT. Cognitive radio techniques, allowing a device to switch parameters on the fly, could make large amounts of spectrum more usable by mobile services, especially above 10GHz. But at these high frequencies, where beamforming and meshing are needed to offset poor range, the analog elements of the radio will create severe bottlenecks. “Crowding 50 analog radios together on one chip, switching their operational parameters every few microseconds and expecting them to work at 60GHz is an analog designer’s nightmare,” said Barlow. “With Pizzicato, we have created a glimpse of future disruptive technology – a radio built purely from computing power.”
Radio Dot makes progress:
Last year’s MWC was heavily focused on in-building wireless, with most of the DAS (distributed antenna systems) vendors announcing lower cost and digital implementations to target smaller enterprises than their traditional solutions, while other suppliers – such as Ericsson, Huawei and Airvana – were adopting distributed RAN techniques to address the same issues.
This space was less noisy in 2015, apart from the news of the Cisco/Spidercloud deal. But many of the platforms unveiled a year earlier are now coming to market – Kathrein’s K-BOW and CommScope’s ION-E, for instance, are in operator trials, while Huawei updated LampSite to support three frequency bands, and Ericsson announced some initial customers for Radio Dot.
The Swedish giant demonstrated a TD-LTE version of its indoor solution in partnership with China Mobile and also extended the use cases for its platform towards the IoT and smart buildings. It has worked with construction group Skanska on a project to test and build Radio Dot systems to enable smart services indoors, and to extend mobile coverage in an environmentally friendly way. Håkan Stenström, VP of Skanska Sweden, said: “We aim to be the industry leader in sustainable building practices so we are very interested in working with Ericsson to explore ways to improve mobile broadband coverage and capacity in our buildings in a way that complements our focus on sustainability and green construction. This solution makes it possible to optimize the use of energy in the building.”
Meanwhile, Saudi Telecom (STC) became the first operator in the Middle East to deploy Radio Dot, targeting medium and large buildings and venues; and Telefonica demonstrated the technology at the MWC location, supporting carrier aggregation across the 1.8GHz Dot and the Spanish carrier’s 2.6GHz network.