Despite all the talk about small cells, the real architectural shift of 2013 has been to break down the traditional base station and distribute radio and antenna resources in every part of the network. The most dramatic example has been in the next generation macro layer, where cutting edge operators are moving to Cloud-RANs that virtualize baseband processing on central servers, which serve stripped-down base stations combining just the radio and antennas. Those base stations may, in turn, support ‘client’ small cells or standalone antennas to distribute coverage and capacity further at low cost. This basic approach is not confined to the big macro roll-outs though. It is also seen in new approaches to indoor wireless, in enterprises and venues. Small cells, DAS (distributed antenna system) and Wi-Fi offload have been seen as rival approaches to bringing the cellular network indoors effectively. But as companies from the start-ups, like Spidercloud, to the giants, like Ericsson are demonstrating, techniques from all three may be combined to support optimal cost and flexibility.
Spidercloud’s E-RAN, for instance, creates a fully self-contained carrier class network in an enterprise, with a central services node (SCSN) managing up to 100 3G/LTE/Wi-FI radio nodes. All these small cells appear as a single node to the LTE evolved packet core or 3G Iuh-standard gateway. The SCSN acts as the local control point for the small cell enterprise network, handling functions like cell coordination and inter-small cell signalling.
Ericsson has also taken a highly distributed approach with its new Radio Dot offering for the enterprise market. Like C-RAN architectures outdoors, this relies on centralizing most of the baseband processing on indoor base stations or access points, which support almost 100 Dots apiece. The Dot itself is the size of a smoke detector, and is a highly simplified 3G/LTE unit with an antenna and power amp. The resulting network is like a miniaturized, indoor version of the increasingly distributed, virtualized architecture taking over the macro layer, and shows how distributed antenna techniques are being adapted for new networks whether the cells are large, small or private/enterprise. “We think this will redefine the small cell space,” said CTO Ulf Ewaldsson.
And Johan Wibergh, head of Ericsson Business Unit Networks, added in his statement: “With the Radio Dot System we lower the threshold to building indoor coverage. The dot is the most cost-effective, no-compromise solution to the indoor coverage challenges expressed by our customers. It is ultra-small but can scale to virtually unlimited capacity;
The 300-grams Dot can be stuck on any wall or ceiling to address difficult corners where there is limited coverage or where a capacity boost is required. Multiple Dots can share workload and connect to a shared base station via Cat-5 cabling ‘ up to 96 can be linked in this way, with no need for expensive fiber, a feature that Ericsson says can slash 60% off cabling costs and 75% off installation. Indeed, the cable aspect is one of the most important differentiators, according to the vendor – Radio Dot has been in development for two years and is supported up by 14 patents, many of them related to the cabled interface. This is a contrast to most enterprise small cell solutions which work over the switched LAN, and does offer deployment flexibility, but in some situations it may also require new cable to be installed.
A Dot emits 100mW of ERP (effective radiated power) and can cover 500-800 meters, with higher power versions planned in future. As new LTE releases and functions like carrier aggregation emerge, the Dot system can be upgraded simultaneously with the macro network using the same software. Wibergh also promises close integration with Ericsson’s Wi-Fi portfolio, supporting real time traffic steering between the two networks.
The Swedish giant unveiled its new product with great fanfare, claiming it would transform the enterprise small cell space by moving the price points for coverage and capacity. The gadget itself may be simple and low cost, but it is highly strategic to Ericsson, which knows the enterprise cellular network is where it is vulnerable to Cisco’s attempts to steal some of its carrier market. And while much of the coverage referred to small cells ‘ an architecture which Ericsson has often treated with disdain ‘ in fact the OEM is just extending the network design it prefers, the distributed RAN, rather than a design based on large numbers of small base stations with local baseband processing. Its flagship outdoor offering, the AIR (Antenna Integrated Radio) offers low cost, low power, stripped-down kit for the macro cell site (whether the cell itself is large or small in radius), which then attaches to separate, and often centralized, basebands.
The same is true, on a very small scale, of its new enterprise indoor platform. Neither offering places heavy reliance on homogeneous mini-based stations, and Wibergh could not resist criticizing the technology which threatens to disrupt its base station economics and margins. ‘Femtocells and picocells still encounter performance challenges and require a great deal of improvement’, especially in the areas of interference and distortion when large numbers are involved, he claimed, while Ewaldsson argued that small cells which include the baseband functions have higher TCO as they have to be upgraded when baseband standards change.
However, Ericsson will be unable to avoid small cells altogether as operators start to build out their small cell layers, often in separate spectrum from their macro networks and C-RANs. AT&T has been a flagwaver for small cells while publicly cautious about C-RAN, at least in the short term. However, it is also a great investor in DAS and will deploy a combination of integrated small cells and distributed antennas to achieve its coverage and capacity goals, especially indoors. (As part of its Project VIP, it plans to deploy 40,000 small cells and 1,000 distributed antenna systems by the end of 2014.)
Ericsson will hope the timing of Dot ‘ which will not be fully commercial for almost a year ‘ will be right for the 2014 wave of AT&T’s huge roll-out, which will be heavily weighted to indoor coverage in the first years. “We came out with this now because it will make a lot of changes to the small cell market, and we need to let our customers plan for it,” Ewaldsson said of the timing. “Most operators make one-year plans so they are working on their 2014 investments.’
Ericsson is one of the operator’s two main RAN suppliers and will want to secure a position in the small cell as well as the macro layer. Indeed, AT&T’s SVP of architecture and planning, Kris Rinne, expressed support for the Dot in the press release announcing its launch, though suggesting it would be an additional option, rather than replacing existing plans. ‘A solution like the Ericsson Radio Dot System gives AT&T another tool to choose from in its next generation toolkit,’ she said.
It will be interesting to see how far Radio Dot provokes a response from Ericsson’s large rivals, particularly Cisco, which has an aggressive small cell and indoor wireless strategy of its own. This is designed to defend its enterprise networking territory by extending its platform to indoor cellular, but the company may also start to go after Ericsson’s carrier customers in the public access network too, by supplying the small cell layer.
There is still heated debate about whether a multivendor HetNet ‘ in which the small cells can come from a different vendor from that of the macro layer, but still interoperate seamlessly ‘ will really prove workable. If it does, it will be down in part to Cisco, which could claim a strategic place in the cellco network via its small cell activities, centered on carrier Wi-Fi, and its acquisitions of Ubiquisys and Intucell. Before thinking about the public network though, it will be eager to remain dominant in the corporate world, where it will come under siege from the RAN suppliers which once stayed outdoors. Huawei, for instance, announced its LampSite solution for deep indoor coverage at Mobile World Congress in February.
Ericsson is always, predictably enough, more cautious about the idea of multivendor layers. Wibergh said the Dot platform could be used alongside non-Ericsson macro networks, but he suggested there might be some limitations on the functionality compared with an end-to-end deployment. As in all its approaches to new mobile platforms, Ericsson has a card to play which Cisco and the start-ups and Wi-Fi companies do not ‘ integration with existing macro networks. In summarizing the benefits of Radio Dot, Wibergh stressed: ‘It is easy to install, future proof and it is 100% integrated with existing mobile networks.” The emphasis on the last point highlights the impending battle between the end-to-end HetNet ‘ heterogeneous in air interfaces, cell sizes and spectrum bands, but not suppliers ‘ and the multivendor approach. That battle will decide whether the established OEMs control the new distributed networks too, or whether there will be a way in for Cisco and other others ‘ and it will not just be fought in the public access HetNets, but in the enterprise too.