Ronny Haraldsvik, VP Marketing for Spidercloud, explains what their enterprise RAN solution does, why it differs from Enterprise Femtocells, explains their go-to-market approach and gives his view of the mobile enterprise in the years ahead. Their system does use small 3G RF transceivers around the building, but concentrates the traffic in a central services node which can also directly offload to the local network/office phone network.
Where does SpiderCloud originate from?
The core of the company is made up of people from Cisco, Flarion Technologies, Juniper and Qualcomm (which bought Flarion). We’re based in Silicon Valley, but have offices in London (UK), Ohio and Westfield, NJ (USA). Unlike other new startups we have existing operator relationships from our previous businesses.
We’re focused solving two major problems for the operators: indoor capacity and coverage for the emerging enterprise market .
What is your product?
We’ve been in product development for 2 years and are entirely focused on building an Enterprise Radio Access Network – what we call an E-RAN. Our SmartCloud E-RAN, an emerging category we believe, is a system is deployed inside business premises by the mobile network operator. It provides excellent radio coverage and capacity for any 3G UMTS mobile device when onsite with a low-cost and turnkey system.
The radio heads are fully fledged NodeB basestations supporting voice and HSPA data which support any 3G UMTS mobile phone. Each radio node can transmit at 100mW (20dBm) RF power, with an external antenna increasing this to EIRP of 22dBm, giving a typical indoor range of 30-50 meters in a real-life environment, This is perhaps 3-4 times that of a typical 20mW (13dBm) femtocell.
We’ve written our own MAC/link protocol layers and have flexibility when it comes to silicon choices. We can’t say enough good things about picoChip and Percello who provide the chipsets in this field. Along with TI, it’sgreat news that Qualcomm are entering the market too. The number of users supported per radio head isn’t that important to us as a differentatiors– we’ve seen a lot of public statements about this recently - what’s critical is how the subscribers are managed overall. We hear that all the time from the operators
The radio noded is managed centrally by via Ethernet by the SmartCloud ™ Services Node, which concentrates these and connects into the mobile operator’s core network. We use Power over Ethernet, so a separate power supply/cable is not required for each radio node.
Our GA (General Availability) version of the product will also include WiFi within the radio node.
We have designed this system to fit office, floors, buildings and campuses and ensure it’s ideally suited for 150 to 15,000 people or more. The system can be quickly deployed with a payback of the cost within 2 years with as little as 25% subscriber penetration for businesses of most sizes.
What about the SmartCloud Services Node?
We’ve designed this architecture specifically to address the indoor enterprise market – this is not a domestic femtocell adapted for the purpose. The SmartCloud system is a complete network in a box, including RNC, SGSN and GGSN functionality. For our Beta product, voice calls are handled by the mobile network’s existing MSC, but data sessions can be locally routed and offloaded without going through the operator’s core – significantly reducing backhaul bandwidth and improving latency. Our GA version of the product will also support local routing of voice calls – they will still be controlled by the central MSC, but directly switched within the building.
Each Services Node can handle about 50 radio nodes. Our architecture allows for multiple serviced nodes so it can scale up as required.
Our approach is not to change the user experience or interfere with the macrocellular network, just provide excellent local coverage and capacity inside the building. This is all about customer acquisition and retention.
Is this standards compliant?
Our system supports standard 3G UMTS phones and connects into the operator’s standard 3G core network using any required core interface by the operators,such as Iuh over IP, CS, PS and Iuh.
How far along the development path is this product?
We are currently in lab trials with a European operator and expect this process to take another 4-5 months. We plan to limit the number of concurrent lab trials with our very own internal test lab to help operators who cannot set aside lab trials in their own facility. We expect this lab to help expedite our time to market . Our goal is to have a commercial deployment somewhere in Europe before the end of 2010.
Of course every small startup is enthusiastic about its market potential. We believe our approach is unique and will get a lot of attention. Major operators are asking for a centralised RAN architecture for indoor deployments. We have that and more with our wireless platform.
There are existing enterprise solutions such as GSM picocells, why go for 3G?
The majority of phones sold today are now 3G and I think if you fast forward by 12 months you’ll find the proportion of 3G phones in use will be dramatically increased. Operators are much less worried about 2G support in these systems than before.
Other technical solutions such as DAS [Distributed Antenna Systems] are expensive and time consuming to install. It can take up to 9 months to install DAS throughout large buildings, whereas our system can be in place much more quickly (over a weekend for smaller buildings) and at much lower cost. But note, we think there’s ample room for all solutions, that being DAS, Femto, Picocells or our very own approach, an E-RAN. For SpiderCloud Wireless to succeed, no company has to fail.
What is your channel to market?
Like any new startup with a different approach, first we need to convince an operator about our system. Then we’ll sit down with OEMs and System Integrators who can distribute and install our system worldwide. We know that we’ll need larger partners to scale up our business, and so our business plan includes extensive training and knowledge transfer program during 1st half 2010.
What geographies are you planning to sell in?
Our first market will be in Europe, then the US and then Asia-Pacific. After a few trials, we will develop the right relationships to grow the business.
What’s unique about your approach?
Our central RF management solution includes several interference migitation techniques and the ability to self-organize and auto configure the system We have several patents in this area and a great deal of expertise.
Mobility: hand in and hand out as well as Radio Node to – Radio Node seamless handoff. [Ed. Note: Hand in – seamless transfer of active calls when entering the building – is typically not supported by femtocells today. SpiderCloud indicated that they achieve hand-in through considerable co-ordination and configuration of the macrocell network, but would not elaborate further]
Our local routing, which avoids backhauling substantial voice and data traffic through the operator’s core network.
Our system is designed from the ground up as an indoor enterprise wireless platform.
At the end of the day, vendors can put many claims in the powerpoint slides, but we are all measured on what we can deliver.
What is your vision of the enterprise telephone service in the years ahead?
Over the next two years, enterprises will be buying many fewer new wired desk phones. The reliability of the mobile network is already becoming good enough to depend on it. In some situations, you can do this yourself (such as with UMA, the WiFi dual mode system) or you can ask an operator and relieve yourself of the cost and pain associated with managing a wireless network.
Our studies (and those of others) have found that most IT teams don’t want or require deskphones anymore. They are moving about the office too much. Cost remains an important factor, with unlimited voice and data calls onsite being seen as extremely attractive.
I believe that within 5 years, mobile devices will outnumber fixed devices in the workplace and be the primary communication tool.