Alcatel-Lucent’s recent reorganization was all about becoming an IP specialist, regarding the wireline/wireless divide as, nowadays, artificial and taking on Cisco as much as Ericsson. So R&D and many other activities now span all media, and IP is charged with delivering the growth, while mobile infrastructure is regarded largely as a cash generator, not an expansion driver.
This is a smart move that revolves around ALU’s biggest recent success story, its edge and core routers. It has built a strong business around the 2003 acquisition of TiMetra, and this delivered a 24% revenue increase in 2012, while most other divisions were flat. It is arguable that ALU’s new CEO, Michel Coombes, should have been even bolder and got out of wireless RAN altogether, but that was apparently a bridge too far and, while the mobile division has executed poorly in some areas recently, it does have valuable customers in LTE and the opportunity to weave 4G tightly into its IP tapestry.
An important area in which the reinvented ALU must shine is the last mile, whether that is to the home, enterprise or small cell site. As fiber proliferates and lurks every closer to the user, the last mile remains a hotly contested area, and one in which carriers of all hues will have to invest heavily to support users’ rising data consumption, whether at home or on their mobile devices.
Fiber and wireless have, of course, taken center stage in this respect, but as most operators look to integrate fixed and mobile services ‘ as seen in Vodafone’s race to buy up cable and broadband providers in its formerly wireless-only territories ‘ they will be using a whole menu of technologies. And all of these require investment by the vendors, to keep up with the challenges of the data explosion, as well as from carriers.
Along with fiber, LTE and the array of wireless backhaul options for the last mile, there is considerable effort going on in cable and DSL. The former industry is working on the next iteration of its standard, DOCSIS 3.1, which will rely on the OFDM techniques which have also boosted cellular networks to new levels of spectral efficiency in LTE (and Wi-Fi and WiMAX). The DSL sector has pinned its hopes on the G.fast system, and this is one area where ALU is demonstrating its determination to drive performance boosts across all platforms.
The company is turning up the heat on G.Fast trials. Telcos in the US and Europe have had to take a back seat recently as cable companies have claimed the top spot in terms of broadband speeds ‘ but perhaps not for much longer. ALU has a trial on live telco lines at Telekom Austria, which are in a traditional wire bundle, using G.Fast and operating at speeds over Gbps – indicating that G.Fast will be ready and waiting shortly after or even before DOCSIS 3.1 delivers even higher speeds for cable companies.
The G-fast standard should be finalized next year and appear in commercial chips in 2015, and so far ALU and Huawei seem to be the most active OEMs. It is one of the important attempts to keep DSL relevant in the broadband picture though it is only intended for distances up to 100 meters. Over those short ranges, it can almost match fiber speeds (1.1Gbps over 70 meters and 800Mbps over 100 meters in the Austrian test, using a single, good quality cable, while on older, unshielded cables, it managed 500Mbps over 100 meters). G.fast can also use vectoring to work over multiple copper lines simultaneously.
Some momentum is growing around G.fast as a technology for the last mile to small cells, indoor femtocells and Wi-Fi, as well as access. However, the trials also highlight the high level of investment telcos need to make in order to keep up with cable (or, some scenarios, fixed wireless). Cablecos simply have to replace their DOCSIS modems and cable modem termination systems (CMTS) to upgrade to 3.0 and then 3.1, but telcos have to physically move fiber closer to each home, which means digging up the roads and either resiting every cabinet they own or placing a new class of smaller box closer to homes, which handle the transition from fiber to G.Fast. They also have the same replacement costs of the CPE and termination systems that cable operators have.
DOCSIS 3.1 claims it will offer 10Gbps down and 1Gbps up. The specification, like G.Fast, is still being developed. Two technologies which are widely used in other markets, especially radio, are harnessed – OFDM and LDPC, a form of forward error correction. Installations are not expected until 2015 and full scale deployment will be phased in over a 4-5 year period from then. This suggests that ALU can propel telcos, if they will invest the capex, to beat the cablecos by about a year, though both camps are likely to accelerate their schedules, eager to claim the top speed spot in broadband.
ALU makes the point that any telcos that don’t want to wait for G.fast can shorten its copper and use VDSL2 plus vectoring, and still get speed gains while they wait for the full G.Fast standard. But early trials like the Austrian one are vital to push the standards committee in the direction ALU wants and to establish its credentials among operators.