A recent experience I had with my ISP initially made me feel the need to do a consumer “rant,” but after calming down, I examined the business case from an operator’s perspective of fixing this situation and considering whether the promise of new 5G services might be the solution I’m looking for.
Not long ago, I took a few days off in Costa Blanca, Spain and I realized very quickly just how much I take the Internet for granted. My fixed WiMAX connection stretches over a clear line of sight of about 1000 meters and is supposed to provide me with a 16Mbps download. While it is nothing extravagant, I noted that after spending several hours on the phone with the provider, it would be easier to locate the Holy Grail than to get a stable connection with the advertised speed. There is no fixed line available and when I enquired with the local Internet provider about a fiber connection, they just laughed and said rather than “mañana” (normally everything is tomorrow in regional Spain) think more three to four years.
So, what are my options today?
Fiber to the home or FTTH would be ideal, delivering speeds in the order of GBps. But, the sad reality is that it remains a dream for most residential sites across Europe, particularly those in rural locations. The cost of civil works versus the ROI remains a major stumbling block.
For those places with fixed phone lines, a half-way house for many operators has been to sweat their existing copper assets even further with the introduction of the latest DSL variants, such as G.Fast, which promises to supply multi 100Mbps over a short ‘last mile’ copper loop from the local wiring cabinet to the home.
BT here in the UK are making a lot of noise about this technology, although the rollout areas are very limited. The conundrum is that whilst this approach negates much of the civil works needed for FTTH, G.Fast technologies are really limited to a useful copper loop length of <300m. Given the average overall loop length in Europe tends to be around 1.5-2km and the distance from cabinets to homes is around 300-500m, this again has limited appeal, particularly for non-city center dwellings.
My other option was to use my mobile as a router and make use of the excellent 4G signal which provides approximately 40Mbps. With the recent EU ruling that European roaming must be treated as domestic data, this seems at first to be the ideal solution; however, the sting in the tail is that my UK based mobile data plan is limited. Binge watching Netflix’s Altered Carbon soon sees my monthly allowance disappear in a puff of IP packets (the fact that I didn’t receive an immediate notification is a subject I’ll leave to another blog about the problem of telco billing solutions). Of course, the other problem with 4G is that in the summer the population in the Costa Blanca rises tenfold, mainly with Europeans enjoying their new data roaming freedom. The infrastructure, however, stays constant which only leads to one outcome, massive contention, reduced bandwidth and unhappy subscribers.
So, if fiber remains elusive, DSL impractical, Wi-Max is on the wane and LTE has its limits, all roads appear to point to the new kid on the block – 5G.
Last year, my Internet service provider introduced a new 5G service and being the early, ever optimistic, adopter of technology transformation, I signed up straight away looking forward to speeds up to 50Mbps. The dream soon turned into a nightmare when I discovered that the 5G connection would never last more than about 10-20 seconds before the session died. This made any type of streaming, video calling, WebEx etc. impossible and after lengthy investigations with no resolution, I grudgingly rolled back onto WiMAX like a kid who has been given an ice cream and then cruelly had it taken away after a couple of licks!
So why isn’t 5G delivering? Or rather, when can it deliver? Speaking to our in-house network expert, I decided a crash course on 5G technology was necessary.
Much of the underpinning technology in 5G NR such as OFDMA, MIMO and Carrier aggregation has been proven in LTE Advanced and Advanced-Pro specifications defined in 3GPP releases 13 and 14. The 5G specifications defined in 3GPP releases 15 and 16 will build on that by extending not only the techniques being deployed but also in the use of spectrum. Techniques such as Massive MIMO, Beamforming and enhanced carrier aggregation will drive a much greater Bits/s/Hz spectral efficiency – effectively sweating the air asset even further and providing the uplift in capacity over LTE so keenly sought.
As for spectrum, there is always an interesting trade-off to be balanced. Higher frequency ranges dramatically increase the bandwidth available; however, the compromise is in reach and propagation – higher frequency RF waves don’t penetrate objects such as buildings as well as those lower in the RF range. Lower frequencies travel much further and have intrinsically better propagation characteristics BUT the bandwidth is reduced. 5G services are being positioned in three different ranges within the overall RF spectrum. The main band being targeted is typically between 3.4 – 3.8GHz which offers a trade-off balance between bandwidth and propagation. Some countries are looking at offerings in the 700-800MHz range which improves signal propagation but at the expense of bandwidth, which could be argued goes against one of the main tenets of 5G’s promise. The exciting area is in the millimeter wave spectrum range – typically frequencies above 25GHz all the way up to 300GHz. This is seriously bandwidth rich territory and could harvest speeds in the GBps range and above which sounds like nirvana until the reality of propagation comes back into play.
A typical 3G/LTE macro cell site operating in the 700-800MHz up to 1.8GHz range can omni-propagate a signal to a distance of ~20km but the bandwidth offered per device may only be in the 10’s Mbps/low 100Mbps range. A millimeter wave cell site with clear line of sight may only be able to achieve propagation distances of ~1-2Km. Clearly in built-up areas or locations such as shopping malls, a dense mesh of small cells operating at that frequency could offer an effective service, again things become more challenging in out of town or rural settings.
The next issue is one around noise or interference. From my crash course it appears that the high encoding designed to increase speeds can be very susceptible to interference from just about anything. In theory, the encoding can be stepped down without causing session loss and only impacting speed.
The conclusion to all this is that certainly the best I can hope for comes all the way back to the 5G fixed wireless access service I mentioned and experimented with earlier. Using 5g in the 3.4 – 3.8GHz band to overcome the last mile challenges of fiber and DSL and provide a ‘beamformed’ 5G service to my home router seems like the way forward, and so my ISP is definitely moving in the right direction.
I also recently read that Verizon is introducing 1Gbps 5G fixed broadband wireless to their customers in the U.S., so it definitely sounds like 5G will become a real alternative to fixed line especially given the lead times for rolling out. However, while the technology appears to available, it is still a rapidly maturing market reminiscent of the wild west with a way to go in terms of the actual usability of the service and the ability of the ISP to support its customers.
This is part 1 of a 2-part blog. In the follow-up to my rant, I will discuss in part 2 some innovative options that CSPs may want to consider for how to better match pricing by allowing consumers to personalize various levers that give everyone the price/performance they need.