As in all things 5G, most of what AT&T said this week about its new DirecTV Now trials in Austin Texas was a lot of nonsense. It said in effect that it was taking its 5G wireless trial out of the lab and into the streets. But it tried also to imply that this meant 5G was nearer, it isn’t; that this was a specific delivery mechanism for DirecTV Now – when it’s just low latency, high packet rate IP delivery.
AT&T is leveraging the decision which it pushed for at the March plenary meeting of the 3GPP in Croatia, where it was agreed that a non-stand-alone version of 5G could be completed ahead of the full 5G standard. This allows for an existing LTE network to co-ordinate 5G New Radio services, especially for fixed wireless broadband.
What it has not said, but we feel the need to add, is that the economics have never been right for fixed wireless in the US. AT&T is suddenly trying to get us all to believe that with the arrival of 5G, and the pending decisions on how 28GHz and 37-39GHz will be licensed, that all changes. But does it?
Since the only country which is experimenting in 5G fixed wireless services during 2017 is the US, with a view to 2018 launches, the equipment will not have then built up the pricing advantages of global volumes. Also since AT&T itself is on a parallel path to explore SDN and Virtualized base stations and services, which will not mature until late 2018, those cost advantages will not be with the company until 2019. This type of service is entirely reliant on getting a 1 Gbps signal to a home any way you can, and then spraying it around using WiFi, not 5G. So this type of service will not require the root and branch transition of devices which people associate with 5G. It will be as if nothing has happened and AT&T just re-joined the fixed line broadband race after an absence of 10 years.
On that subject, AT&T has had no broadband net wins since 2007, and it is going to have to learn how to sell against cable opposition, at a time when they will already have the lion’s share of the fixed broadband market, and when they have the economies of scale in DOCSIS 3.1 to continue to drive down pricing in the market. AT&T may be ready to compete technically, but with expensive equipment, and with an arrogance that has allowed in the past for it to match 300 Mbps cable services, with a 30 Mbps ADSL2+ experience and charge much the same for it, or more.
The only thing going AT&T’s way is the idea that it will also offer zero rating for data used on DirecTV Now, which in effect means that it can outsell cheaper SVoDs and skinny bundles. If at the next election the US voters decide that the Trump experiment was not to their liking, it is likely that Net Neutrality will be back – big time, to prevent even this.
AT&T described this as its second trial using millimeter wave (mmWave) technology, after last year running a fixed wireless business customer trial in Austin.
This is in the form of an ultra-fast internet connection to residential, small business and enterprise customers using Ericsson’s 5G RAN and the Intel 5G Mobile Trial Platform. Neither of these platforms are ready to be delivered for real, with Intel using a two part system made of an RFFE (Radio Frequency Front-End) and an FPGA, which today operates at just 100 MHz to 200MHz only, in the place of a part that Intel hopes will eventually reach speeds of up to 3Gbps using 100MHz to 800MHz of bandwidth. The Intel part also features 4×4 and 8×8 antenna arrays.
The trial involves DirecTV Now, but also any other app that the customer cares to try. AT&T may reach 1 Gbps, but that’s nothing new since DOCSIS 3.1 can reach this and beyond and both Verizon and AT&T have fiber services which extend beyond this, as does Google. It’s just that this should be wireless and easier to set up. It is also supposed to be cheaper.
Earlier in the year AT&T said it delivered DirecTV Now using mmWave technology at its Middletown, New Jersey lab as well.
AT&T also tried to suggest that the low latency would mean faster loading web pages, but the internet is a complicated place and that depends on so many other variables across the entire Internet, that this is unlikely to materialize during this trial or even after.
AT&T says that data traffic on its mobile network has grown 250,000% since 2007. Which just so happens to be the date when AT&T last invested in its fixed line infrastructure.
These fixed wireless trials will last several months, and later AT&T plans to experiment with a new virtualized-RAN core in late 2017.
AT&T told us last year that it had arranged for a two-year experimental license to conduct trials in 3.4GHz, 28GHz, 37GHz, as well as 64-71 GHz and 71-76 GHz and it has set up an outdoor base station, to nearby residences and test vehicles within three kilometers.
In February AT&T revealed similar AirGig fixed trials, which relied on power stanchions holding mmWave antennas and reaching across up to 100 meters of air space to get a signal to nearby homes. It has since entered negotiations with power companies over the use of their power line infrastructure. Confusingly this used different spectrum I the 65 GHz range.
Also Verizon has managed to purchase XO Communications and snatch Straight Path out from under AT&T’s nose, both of which hold key 28 GHz assets. However AT&T acquired Fiber Tower, which has also obtained some such assets of its own.