Last week I had the opportunity to attend a Global VSAT Forum (GVF) conference, chaired by ip.access partner Huckworthy, on the topic of cellular backhaul. The GVF typically looks to promote the satellite communications industry, and given the overlap in vertical markets between satellite and small cells, their invite to attend a cellular focused event was too good to refuse.
The day consisted of around 15 speakers from industry (including our own Nick Johnson), each talking on a topic roughly aligned to the conference subject of satellite backhauling for cellular.
As each speaker presented and then engaged in Q&A with the audience, a theme emerged.
How can we find a compelling value proposition for satellite backhaul for cellular
Satellite isn’t deployed just anywhere of course – the participants were referring specifically to areas beyond the range of economical fibre or copper links, with central Africa often presented as the archetype of the hard to reach place, but including rural areas of established and emerging economies.
The cellular driver should be obvious – people want to be connected. The mobile industry consumes ever more data annually and yet revenues for mobile operators are stagnating or even dwindling. Add to the mix of countries where average revenue per user can be around $3 and you start to see the problem: data use is increasing exponentially yet revenues are dropping. Mobile operators are having to focus capex spend, and concentrating on covering the populace in urban and suburban areas in those areas where ARPU is high enough to justify the cost of traditional cellular infrastructure.
It’s simply not a good business case for them to build their own backhaul infrastructure to cover a relatively small percentage of the population, especially one which is not a big revenue generator.
Small cells and satellite. The value proposition.
There are various factors here in the small cell/satellite value proposition. The capex improvements alone contribute to up to a 5x reduction in price compared to a macro deployment.
- Cost of cellular hardware
- Targeted coverage
- Standards based cellular backhaul optimisation
- Proprietary backhaul optimisation
- Reduced cost of satellite backhaul
- Reduced cost of maintenance
- Reduced cost of power requirements.
The first two items go hand in hand. Small cells are up to 10x cheaper than macro cells, and whilst small cells may handle half as many active users (typically 32 users) their short range allows coverage to be targeted in the areas that need them, rather than blanketing a much larger area which in rural deployments is mostly unpopulated. With a typical range of up to 5km radius (when amplified), the small cell provides an approximate 5x price improvement per user than a Macro cell. This includes external encapsulation and antennas. When you add in installation & commissioning, the total cost of ownership of small cells can be 7x better than that of macro cells.
Backhauling costs then come into play with the ‘flat’ LTE architecture and small cell backhaul optimisation serving to reduce the amount of satellite bandwidth consumed.
LTE eNodeBs – small cells included – minimising backhaul signalling by moving certain RAN functions inside the small cell. This was initially intended to speed up call set up times and handover but has the added advantage of lowering the need to communicate over the backhaul to the EPC. It should be noted that 3G small cells also offer benefit here, as the Iuh protocol is significantly more lightweight than Iub.
Taken to a logical extreme, lightweight EPCs can be hosted alongside the small cell. This allows continuation of local voice services in the event the backhaul becomes congested or inoperable.
ip.access small cells benefit from further backhaul optimisation. When the backhaul size is specified within the small cell, it will prioritise the type of traffic – signalling, data etc. – to ensure the best quality of service.
Satellite technology likewise serves to optimise traffic, using compression and acceleration techniques. TDMA links also operate on-demand, rather than being dedicated, and are therefore better suited to rural communities with sporadic traffic.
The final input to this section is that of backhaul cost, with the maturity of Ka and Ku band high throughput satellite (HTS) technology serving to drive down the cost of satellite backhauling, making it an affordable option in areas where fibre is impractical or costly.
Time to Revenue and Site Opex
A small cell can be deployed from start to finish in as little as 3 days thanks to the simplicity of the technology and the relatively short range to be optimised, thus delivering revenue at the earliest opportunity.
Rural locations suffer from un-reliable power sources. A site built around a small cell may consume less than 100 watts, so battery backed solar becomes a viable option for cheap, uninterrupted electricity.
Maintenance is trivial. The plug and play nature of a small cell allows the configuration to be stored and managed centrally, and pushed out to the cell in event of a replacement being installed.
Rent is the final factor here, with Small cells having more installation flexibility than macro cells, meaning a greater selection of installation sites, and lower site rental options.
I’ve already alluded to the targeted coverage of a small cell, and whilst macro vendors would scoff at ‘only’ 5km range, that proximity to subscribes massively improves the link budget and therefore the quality of service.
Typical satellite equipment vendors already have to accommodate 500ms latency due to the 36,000km path the data has to take to the satellite. I’m pleased to note that our eNodeB has been proven to work reliably with over 4s of latency! That’s the benefit of a product whose 2G/3G predecessors have been successfully deployed on commercial aircraft, via satellite backhaul, since 2010.
The ip.access value add
I’d like to add a couple of additions to the paradigm before we conclude. The business model of tower ownership has evolved somewhat in recent years with so-called ‘tower cos’ buying and operating a proportion of telecommunications towers and sites in rural areas. Mobile operators therefore are able to monetise the sale of towers in the short term in exchange for a regular Opex commitment.
ip.access complements this approach via the development of our Viper platform – a virtualised platform supporting the deployment of 3G and 4G small cells. This allows operators and tower cos to minimise their infrastructure costs by using a hosted virtual solution rather than a dedicated hardware solution.
A final addition to Viper is the use of our SUMO multi-operator capability. This enables one small cell access point to provide coverage for all local mobile network operators through a combination of shared spectrum, with a shared Small Cell using a single satellite backhaul and shared gateway.
Through tower-co ownership and our SUMO feature we reach the multi-carrier neutral host model beloved by DAS operators.
I was quite surprised just by how enthusiastic several of the speakers were with the merits of small cells. However given the points above it became easy to see why small cells make a huge contribution in driving down the cost of satellite powered rural cellular deployments. In a world where it’s not currently cost effective to provide service to the last 3 billion inhabitants of the planet, small cells and modern satellite backhauls are a significant step forward.