Have Small Cell pioneers ip.access gone completely crazy! They want you to install their small cells without connecting them to the network or carrying any voice/data traffic. We asked CTO Nick Johnson to explain why anyone would want to do this, uncovering a radically different approach that promises to liberate a huge amount of value from the mobile operators' radio networks.
So what is a "presenceCell"?
It's a fully functional 3G small cell which can be installed standalone. The hardware is the same as our latest 3G SOHO product, the S2-08, however using a completely different software build. There's no need for a small cell gateway or connection to the mobile operator's network. Each presence cell is autonomous and uses our existing Self-Organising Network features to verify its location and synchronise with nearby macro and small cells.
Typically we'd position a small number of them at critical locations throughout a retail area or central business district. These briefly interact with 3G phones passing by, smartphones and feature phones alike, capturing anonymised identity information and RF measurement data. This data is uploaded, aggregated and enhanced to provide a range of analytics for a variety of business uses such as retail analytics, m-commerce and mobile promotions.
Gartner estimate this data could be worth an additional $300 million a year to the average mobile network operator.
How does it work?
3G devices constantly seek out new/alternative cells for a better signal.
When they find our presenceCellTM, there is a brief exchange of signalling messages which allows us to capture data while rejecting the phone camping onto the cell. We can use a few clever tricks to increase the level of interaction, ensuring that each cell has the opportunity to capture data. We can also make use of RF signal measurements to triangulate the location more accurately than just to within the coverage footprint.
Cell ranges can be large or small, and you can set the size of footprint - anything from 1m to 100m – to match the requirement you are serving; for example it could be small enough to determine exactly which of several checkout tills you were standing at in a large supermarket. .
Why not use other methods to capture precise location data?
It's true that Wi-Fi has been adopted for this purpose, but many smartphones have Wi-Fi turned off. Very few people turn off their 3G service when out and about.
Bluetooth beacons are another potential competitor, but again require the smartphone to have Bluetooth enabled which is often not the case.
The beauty of our solution is that it works with any 3G device, and it doesn't need any special configuration, or App to be downloaded, or the consumer to do anything at all. A recent study by Maravedis Rethink also confirmed that small cells using licensed radio spectrum were the favoured approach for presence technology among vertical market players.
Where would you position the presenceCells? Do you need advanced radio planning for that?
Installed locations will be determined by a business's requirement to track presence, for example at retail entrances, points of sale or special attractions.
This choice will have quite different drivers from the traffic hotspots or coverage notspots targeted for normal cellular planning, and installation doesn't require any specialist RF knowledge.
What about power and backhaul?
The minimal power requirements can be met using a power adaptor or Power over Ethernet where available. Backhaul requirements are also minimal. The data logged is securely encrypted and compressed, with small files being uploaded periodically – say once an hour. This could easily be done using a cellular modem without the need for any wired backhaul.
Isn't this restricted to a single mobile network operator?
Normally this would only capture data from subscribers of a single network but that is a commercial restriction not a technology one. With the co-operation of multiple operators we could make use of the standard MOCN (Multiple Operator Core Network) feature to track all subscribers from two or more networks.
This would involve some configuration parameters in the macro network neighbour lists as well as in the presence cells but it could be done. The presenceCell would work the same way and would reject camp on requests from subscribers of all networks.
What about customer privacy issues?
The data captured is encrypted and securely uploaded to a central server. At this point, we only have the IMSI (SIM card number) and not the MS-ISDN (mobile phone number), so don't know the subscriber's identity.
Aggregated, anonymous information in this form doesn't directly impact individual subscriber privacy but is useful for businesses, retailers and organisations to use. We can report not just on volumes of footfall through each zone, but also their progress and pathways taken. Dwell time - how long on average people are stopping to look at a sign or shop window – is a popular metric.
The next step is to decode the IMSIs into customer identities using the HLR or CRM system. This would filter out those who had not opted-in, allowing the data to be used to tailor and improve the interaction with marketing programs. Retailers already know when any of their loyalty card customers make a purchase, but at that stage they are already leaving the store. However, they are may not be aware of those times when they visit stores without buying anything – and they would certainly like to know when someone arrives in their store. This combination of presence tracking and loyalty card subscriptions is very powerful enhancement.
Another case in point is one very large electrical retailer that reports up to 20 per cent of potential revenue is lost simply because their customers can't find the products they are looking for. Being able to guide them around the store using a standard mobile phone could quickly solve that problem.
Is this a cop-out because your new 3G Small Cell doesn't work properly?
Definitely not – in fact if it did not work properly as a small cell it couldn't work properly as a presenceCell either!
Launching a small cell that doesn't carry traffic may seem counter-intuitive and a surprising strategy for a small cell vendor. However this is a product which is targeted to specific market needs that we have identified working with trial operator partners.
At Small Cells World Summit, Stuart Carlaw from ABI commented that the lack of added services being offered by 'traditional' small cells is actually limiting their growth and wider adoption.
We really do believe that there is an additional revenue stream to be had from location and presence analytics. We've seen considerable interest in the concept from retailers and banks, and are trialling the solution today with a number of Tier one Mobile Operators.
Of course there is no problem combining the Presence feature with fully operational 3G small cells, and indeed there is an upgrade path from an initial Presence only deployment to a fully functional Small Cell network.
So is this ip.access withdrawing from the Coverage and Capacity space?
We are by no means pulling out of the small cells for coverage and capacity market – quite the contrary. We are fully committed to our Consumer [C-Class], SOHO [S-Class] and Enterprise [E-Class] product lines, with a complete end-to-end portfolio of 2G, 3G, 4G and Wi-Fi integrated small cell solutions.
We are simply the first vendors in the small cell space to have designed and brought to market a standalone presence-technology solution, which is no surprise really considering that we were the first vendors to supply the market with GSM Enterprise small cells back in 2002 to T-Mobile USA.