The case for multi-operator small cells and neutral hosting

Multi-operator small cells are still a rarity. Yet building owners are crying in the wilderness for cellular in their buildings. Complaints from owners of buildings regarding the quality of their coverage are met with mystification and inertia by the mobile operators.

We don’t see customer complaints, say the mobile operators. No, say the building owners, because they complain to us. Ah, so it’s your problem, says the mobile operator to the building owner. And the conversation spirals downhill from there.

Why is Wi-Fi so successful indoors, yet licensed radio small cells indoors still seen as exotic? As with most technology arguments, the answer is nothing to do with technology. Let’s follow the money.

The one thing that still distinguishes Wi-Fi access from Licensed Radio access is the commercial model. Licensed radio is predominately a mobile operator play, where the base sites and the coverage are controlled by the operator. Wi-Fi access hardware is more often provided by the site owner or local tenant company, even if the service itself is provided by a WISP (wireless internet service provider).

Wi-Fi neutral host model

With Hotspot 2.0 becoming the norm, every Wi-Fi AP gives the end-user access to their home WISP. That’s to say, Wi-Fi starts with an assumption of neutral host. The Wi-Fi host network is neutral towards the service providers who offer service over it.

But with Licensed Radio spectrum controlled by the operator, it’s still unusual to see multi-operator small cells. In the macro network it is relatively common to see RAN sharing agreements, which may range from simple tower sharing deals to full infrastructure sharing. Even there, though, it’s unusual to be sharing spectrum.

All macro-network sharing that I’m aware of uses a system called MORAN, where the shared base station contains a non-shared radio card that creates the RF signal in the particular operator’s spectrum. The host network is of necessity non-neutral. If it’s not transmitting in your operator’s spectrum, you don’t get service.

One of the major blocking factors in enterprise and indoor deployment of licensed radio cellular today is this host non-neutrality. If a mall owner or a hotel chain wants good indoor coverage it can phone up an operator and they may or may not put in some coverage. But it’s only for their subscribers, so it doesn’t serve all the users in the space.

The operator may see an uptick in customer satisfaction, measured by Net Promoter Score (NPS), but the mall owner probably won’t. Only a fraction of mall users will see their experience improved. To get the improvement across all the mall users, the owner has to call up all the operators and persuade them improve their coverage. It can be an expensive business, piping RF around a building using traditional methods.

For many, the business case isn’t positive – they may see no new revenue from such a move – even if the NPS scores improve a bit. It may be a long time before the operators gets around to it, after dealing with all the other revenue and NPS improvement opportunities that are on the list.

The DAS alternative

Another approach is for the mall owner to commission a Distributed Antenna System (DAS) for the building. In such a way, a third party analyses the building and designs an installation to provide good radio coverage in the building, and provides a feed to connect up a base station in the basement, and off they go.

No design effort for the operator, and no design replication for each operator. One design for the mall works for all the operators. The building owners see a great improvement in coverage and their NPS can improve across the board, but they pay a hefty price. DAS systems are not cheap to buy, install or modify.

The small cell approach is a huge leap forward from DAS. It requires no major capex. It’s as easy to install as Wi-Fi, but it suffers from the non-neutral host problem, like other cellular systems. If you want to offer equal coverage in a facility, you need a small cell per operator.

It looks unwieldy, and the capex advantage is eroded since you have to buy 4x the kit, and it’s just ugly. Despite its advantage over DAS, no-one’s really embarked on this kind of small cell solution.

Until now.

The MOCN solution

There’s a 3GPP feature that’s been around for a while called MOCN (multi-operator core network). It uses one operator’s spectrum to provide coverage for many. One base station broadcasts the network codes for all the network sharers, and the handsets recognise them and register, even though the cell isn’t using their home operator’s spectrum.

At one stroke, it offers neutral host, multi-operator in a single package, offering the deployment simplicity and capex of Wi-Fi and the licensed radio quality of cellular.

Why isn’t everyone using it?

There’s a catch. The feature as specified essentially opens the door to a free-for-all in terms of resources. There’s no commercial benefit to the first mover. All the first mover does is open the door to their competitors. So there are no first movers.

It’s time to inject some commercial reality into the technology.

It is time for a new resource management and visibility paradigm. Heading towards a 5G world of spectrum agility, RAN and Edge Service Virtualisation, we need a management model to match. We need to extend the resource management model of the eNodeB to make it MOCN aware.

We need to extend the management database to allow the host operator to define, impose and report on a resource sharing policy. And we need it to be more dynamic and adaptive than present day ANDSF (access network discovery and selection function). With these crucial components in place, the host operator can monetise the service they’re providing for their competitors in a way hitherto impossible.

Welcome to the ip.access future, opening the door to simple, optimum capex, commercially compelling, resource managed multi-operator small cell deployments.

Image Credit: Flex