Most of the debate about LAA (Licenced Assisted Access) vs Wi-Fi focuses on the technical RF aspects. We ask whether anyone has thought about how to price such a service, discuss the issues and suggest a way forward.
Contrasting wireless vs wireline charging approaches
The majority of cellular contracts are based on usage, with capped levels of data consumption per month of typically a few Gigabytes. Attempts to charge more for faster speeds or 4G technology haven't been widely successful. Operators recognise that higher speeds and greater spectral efficiency lead to higher consumption, which itself should increase revenues.
Wired broadband on the other hand is generally charged on the basis of peak throughput, with unlimited or very high usage caps.
Wi-Fi is priced differently depending on how and where it's used
Residential Wi-Fi is considered "free" because there is no incremental cost for use. Many opt-in to schemes such as FON that allow others to share their Wi-Fi, allowing them in turn free access when visiting other users on that same network. Fixed broadband and Cable networks also offer free access by their subscribers to large numbers of public Carrier Wi-Fi hotspots as part of their package. Some mobile network providers have also been investing in Carrier Wi-Fi too, usually allowing their customers to use it for free and not consume their allowances.
Enterprise Wi-Fi is provided by many businesses for their own staff and visitors.
Amenity Wi-Fi may be provided free of charge by venue owners as a benefit to their visitors. Some venues such as hotels and airports charge for access, or for faster data speeds. Several Wi-Fi aggregation services offer seamless access to huge numbers of public Wi-Fi hotspots for a monthly subscription fee.
Where charges apply, they may be based on connection time or data usage or both.
A combination of charges
Smartphone users access the majority of their data using Wi-Fi. The vast bulk of that will be at home and office incurring no additional charges. Data consumed via cellular is tracked and the amount of current allowance remaining is easily accessible by web, sms or voice enquiry. Many cost conscious users have trained themselves to turn on/off cellular data unless they actively need it. Some battery conscious users may choose to selectively turn off Wi-Fi.
Worldwide, the charging concepts for wireless data are generally well understood.
Managed Wi-Fi Offload
Standards are making it much more seamless to connect to these various public and private hotspots. The rollout of HotSpot 2.0 and Passpoint are perhaps slower than I had envisaged but already running their course. There is still some work to do to decide whether to connect to a specific hotspot or not – it might not always be the best connection.
Several systems are available from end-to-end Wi-Fi solution providers which manage the selection, connection and quality of service from Wi-Fi. Vendors such as Aptilo, Devicescape and Birdstep spring to mind. These provide clients on the device which will take advantage of Wi-Fi where possible and sensible.
David Fraser, CEO of Devicescape, told me their system is configurable to be biased towards Wi-Fi or LTE at the users discretion. Some users will prefer to remain on the LTE system until the signal quality (not necessarily strength) degrades to where Wi-Fi would offer a better service. Others might prefer to switch to Wi-Fi at any opportunity.
In both cases, the end user has an understanding of the preference and how that would impact their mobile phone bill.
How will LTE/Wi-Fi Link Aggregation (LWA) affect charging?
The simpler form of combining Wi-Fi and Cellular is to operate both in parallel, splitting the traffic and sending it through multiple paths. Smartphones would operate both Wi-Fi and LTE cellular radios simultaneously. The primary link would always use LTE with heavier data traffic overflowing onto Wi-Fi as needed. The system determines when and how traffic should be directed in real-time, based on available resources and the service required. For example, I'd expect a video broadcast stream to be offloaded to Wi-Fi with buffering whereas a real-time voice call might be retained on cellular.
I can't see that the charging aspect of the traffic routing would be taken into account. As with the Devicescape example above, perhaps a user preference could be set to bias Wi-Fi usage should they choose to.
The result might be that in some places, where Wi-Fi is easily accessible, watching a video might use up very little of your cellular data bucket. Elsewhere, when Wi-Fi isn't usable, it might all be charged that way. Access to Wi-Fi can vary for a wide variety of reasons – the property owner might have switched it off, disconnected the backhaul, the area might be congested at that time etc.
The point is that the end user cannot then determine in advance what the charge will be to use the service. The optimist (glass half full) might take the view that they'd expect to pay in full, and any offload to Wi-Fi is a bonus. The pessimist (glass half empty) might say they'd only want to use the service if it was (almost) free. One view might be to suggest that cellular data is turned off in preference to Wi-Fi for those with that attitude, which can be done selectively for individual apps/services on modern smartphones.
Using LTE directly in unlicenced spectrum is likely to result in a much more deterministic way of charging – all data sent this way would simply come out of your cellular data bucket. It's not Wi-Fi and so customers wouldn't expect it to be free.
Perhaps this is what makes it more attractive that Carrier Wi-Fi to mobile network operators.
Careful market messaging and positioning with end users will be essential when combining licenced and unlicenced spectrum use. LWA offers the potential to deliver higher speeds and capacity quite quickly by combining existing cellular and Wi-Fi network assets. Some see it as a natural evolution from today's mix of cellular and Wi-Fi usage. However, there may also be potential for confusion and misunderstanding about usage costs.
LTE-U is very much a clearer prospect from a charging perspective. This may also initially appear more lucrative to network operators but may result in more Wi-Fi offloading by price sensitive customers.