Idée fixe

Posted on Tuesday, August 28, 2018

This past weekend I read the abridged version of Ericsson’s Fixed Wireless Handbook, a 40-page brochure for the Fixed Wireless Access (FWA) opportunity. No Ericsson products are mentioned so the purpose is purely client education. I want FWA to work this time, really I do.

The 40-page handbook is the linear extract from a much longer “interactive” webpage that allows a reader to get a quick overview and then explore in more depth. A recent Ericsson ‘quarterly review’ webinar talked about FWA and mentioned that the Handbook was an extract from a much larger document, but I could not find the latter on the website.

The Handbook does a good job of making the high-level case for FWA, based on wireless’s ability to provide decent broadband service without the need to lay fiber (or have existing copper) on the last 100 meters and without the need for complicated indoor wiring. The recommended outdoor solution would need a truck roll (needs line-of-sight to the cell) but that would presumably be far cheaper than laying FTTH.

Three broad segments are described:

  • Wireless Fiber – 100 to 1000 Mbps which would be sold into the same neighborhoods that FTTH would be sold, presumably as a competitive offer or as an interim solution until fiber could be laid;
  • Build with Precision – 50 to 200 Mbps which would be sold into the same neighborhoods that xDSL solutions would be sold, perhaps as a way to extend existing xDSL infrastructure;
  • Connect the Unconnected – 10 to 100 Mbps which would be sold or perhaps provided (depending on available government subsidy programs) into neighborhoods where fixed broadband solutions are completely uneconomic.

The ‘Connect the Unconnected’ is the segment I am most interested in although ‘Build with Precision’ is also relevant to the Emerging Markets challenge: extending broadband beyond smartphone connectivity and into neighborhoods that do not have the purchasing power to justify fiber and may not even have decent copper.

There is not enough detail in the 40-page document to build a business case and the numbers will depend on individual circumstances, however, I can recommend the handbook for ‘executive reading’ on the opportunity.

And to repeat, no Ericsson products are mentioned so it is vendor agnostic. It is a good ‘story’.

Unfortunately, I have history with FWA and none of it pretty.

I lost a ton of money in the late 1990’s on Bell Canada International options when the Brazilian regulator refused to follow the Indian regulator’s lead and allow FWA licenses to be upgraded to full mobility. Turns out that there was never a business case for FWA at that time. Companies just said they wanted to do FWA to get the spectrum and then hoped to offer full mobility on cheaply acquired spectrum. The plan worked perfectly in India. Not so much in other countries.

Fast forward a fewyears and I am starting out as a consultant. My client was an enterprise-oriented carrier who wanted to get into SME and household Internet. I did a lot of work to try and convince them to use WiMax, a 4G technology when 3G had not yet deployed in the region. However, they were suspicious of WiMax’s sustainability, especially when a certain Swedish company was doing its best to cast FUD (fear, uncertainty and doubt) over its viability. Mi client shelved the project.

Then I spent a number of brain cycles and some Excel and PowerPoint time looking at how to bring the Internet to a number of upscale gated communities that had started up on the outskirts of Bogotá. Despite high purchasing power, these households were well beyond the reach of DSL or Cable Modem. We were going to use microwave backhaul and either WiMax or Super WiFi for distribution. The project failed to attract investor interest, mostly because of fears that the existing operators would simply decide to build fiber after we had demonstrated the demand.

Most recently, a few years ago I did a series of speaking engagements for another vendor and one of my ‘Brave New World!’ messages was on the benefits of LTE-A for FW Broadband. There have been no implementations by major operators in the region in the intervening years.

Someday I may get this right.

Ericsson certainly thinks there is no time like the present.

So based on the handbook what has changed from 20 years ago?

Apart from two decades of Moore’s Law improving the economics on the hardware rather dramatically, the biggest change is much higher data rates.

To be fair to myself, in my last series of FWA ‘sales’ presentation, I was saying that carrier aggregation would be the key to getting higher data rates, to making customer experience over wireless comparable to that achieved by cable modem or DSL. One reason we have seen no extensive FWA broadband implementations is probably because, until now, there have been limited deployments of LTE-A.

Now we have LTE-A in the continent except for Paraguay, Bolivia and (sadly) Venezuela. Operators at least have the option of leveraging the technology, assuming they see the business case in particular sites. (Thanks to 5G Americas for the infographic.)

Small cell technology is another key difference. Higher data rates are helped by being much closer to the source. Ericsson’s preferred implementation needs line-of-sight and since there is limited ‘wiggle room’ for the home antenna (especially for apartments where much of the target market lives), then the cells have to get closer to the home. Small cells are the clear solution and the cost of these has dropped dramatically in recent years.

It is too detailed an issue to go into here, but fixed broadband economics are improved if infrastructure can be deployed simultaneously with sales. With fiber or copper that is nearly impossible – operators must build and hope ‘they will come’ – however the granularity of small cell architectures makes just-in-time provisioning much easier.

As with my fringe urban upscale neighborhood, one of the main challenges to better broadband is improved backhaul. There is no point in deploying Gigabit LTE if the channels back to the Internet cloud are too narrow. Today, all operators are looking at driving fiber deeper into their networks, expanding the addressable market for FWA broadband solutions.

I believe that 5G will come later rather than sooner to the region at least for the mobile broadband use case. However, I also believe that 5G Industrial IoT applications could come sooner and maybe there are locations where the US-carrier FWA-over-5G business case could be justified (for example those upscale gated communities).

In any event, governments want 5G sooner-rather-than-later and that is encouraging them to look at high capacity spectrum bands (above 2Ghz), at the very high capacity bands required for more advanced versions of 5G (the so-called mm-wave bands) and, simultaneously, at low bands (600Mhz, 700Mhz) to improve both LTE and, eventually 5G coverage. Spectrum to do things like FWA is going to become available over the coming years. Higher-capacity means better broadband experience – but it also increases the need for small cell architectures because of shorter wave lengths.

So now that I am all excited about FWA again, a caveat.

Higher frequency, lower wave length spectrum needs more antennas. There is no other way to go (except boosting power which leads to other issues). This could mean more towers but more likely, small cell architectures. Governments have to find a way to make siting much easier and the ‘business model’ for those leasing sites has to change. To ‘Connect the Unconnected’ as the Ericsson handbook describes it, the today painful and expensive antenna siting process has to get very much faster, very much easier and very much cheaper.

Public paranoia about radio waves – which leads to NIMBY-ism over cell sites – has to be resolved.

Or the business case simply will not happen.

Finally, I want to raise an old ‘hobby-horse’ of mine about fixed broadband. It is another topic that deserves more space than I want to give it here – in an article that is already far too long.

But what if the fixed broadband paradigm goes away in the coming years? What if better smartphone technology and the demographic pressure of all those millennials means that ‘large-screen broadband’ as we once called it, becomes irrelevant. Apart from linear TV (for which there are abundant solutions), we do gaming on AR or VR headsets and use our smartphones for everything else (except maybe business applications).

Then (making it personal) the investment issue for operators and their vendors is how to prove me wrong about 5G mobility deployment (sooner rather than later) instead of overcoming my long and here-to-fore unfruitful history with FWA.


No Comments »

Leave a Reply