Networks: the TIP of Facebook's iceberg

Earlier this year, Facebook gave us a TIP (Telecom Infra Project*) about what could come next for them at the infrastructure level. Most members of the project, starting with Intel, Nokia, SK Telecom, Deutsche Telekom, are also on board of Facebook's much wider Open Compute Project, which also claims top coopetitor Google.

Naturally, Menlo Park wants a front seat for next gen network premieres - typically, early bird SKT is expected to be the first on 5G. But recent announcements*** at the F8 days confirm that the company intends to reach deeper into the value chain to secure and strenghten its leadership, particularly where innovation in network performance can prevent bottlenecks in these pervasive big data / heavy content times.

Terragraph (WiGig WiFi on 60 GHz) proposes ultra efficient W-LAN coverage for dense urban areas. You only need a station every 200 to 250 m. Of course it doesn't provide seamless handover, but a 200 m radius allows you more mobility than your typical hotspot.

Project ARIES (Antenna Radio Integration for Efficiency in Spectrum - MIMO) is more into W-WAN for rural areas. It combines 96 antennas to manage 24 simultaneous data streams, and aims at over 100 bps / Hz compared to 71 in early trials.

Here too, for a fraction of the price of a standard infra...
 
Could Facebook become a non-2-to-5G MNO, or a player in the game of standards? Not so fast. But an even more competitive value aggregator, certainly.

mot-bile 2016

* telecominfraproject.com
** opencompute.org 
*** "Introducing Facebook's new terrestrial connectivity systems — Terragraph and Project ARIES" (Facebook 20160414) 

Do Seamless 5G Communications Start With Flawless 5G PR?

Last March, NTT DoCoMo, KT, and China Mobile announced technical collaboration on 5G, each one with an Olympic event in mind: Pyeongchang 2018 for Korea, Tokyo 2020 for Japan, Beijing 2022 for China.

China Mobile, who started 5G development 5 years after DoCoMo, is clearly the laggard here. But by its sole presence, the World's biggest operator already brings considerable leverage for the standardization as well as for the ramp up phases, securing the full involvement of an ever progressing Huawei*.

NTT DoCoMo just presented at the Yokosuka Research Park the results of its recent trials**:
- on October 13 with Nokia Networks (2 Gbps transmission on 70 GHz, in Roppongi Hills, with the help of "two new technologies: beamforming, which focuses radio waves in a specific direction, and beam tracking to control beam direction according to the mobile device's location" - beamform me up, Scottie!),
- on October 26 with Fujitsu (multi-BS, multi-device 11 Gbps on 4.6GHz),
- on November 12 with Samsung (2.5 Gbps reception on 28 GHz, in a moving vehicle in Suwon),
- on November 18 with Huawei (MU-MIMO transmission of 43.9bps/Hz/cell, outdoor in Beijing), and
- on November 19 with Ericsson (MIMO 10 Gbps reception on 15 GHz).

A concise and impressive display, and a nice PR operation for a leader willing to show it covers all the bases. I'm curious to see how KT follows. In a nation already 70% converted to 4G LTE, Korea's #2 mobile operator has been the most vocal about 5G, and even about 4.5G following the launch of GiGA LTE last June. The names of Samsung, Ericsson-LG, Alcatel-Lucent and Huawei have been announced for collaborations at the new 5G Research and Development Center (in Umyeon-dong, Seocho-gu, Seoul), and such technos as MIMO, Millimeter Wave, 3D beamforming have already been tested, like multipoint transmission in a moving car (with Ericsson last May, Distributed MIMO, 2 Gbps).

KT also demonstrated 'core technologies' for 5G (5G drone, 5G hyper-virtual reality, 3D hologram), which got a few Korean media confused between 5G-friendly services and 5G 'networks'***.

GiGA LTE or 5G? "Drone Ski Jumping with 5G Network" feeds the confusion

That's one of the differences between DoCoMo's 'engineer/product/provider' PR and KT's 'marketer/service/user' PR. 

We already saw how the pedagogy of 2.5G, 3G, or 4G worked, and we already know that quantum leaps at the operator's level can seem mere upgrades at the user's level. Let's not fret over hair-splitting debates between 'post-4G', 4.5G and 'pre-5G'.

mot-bile2015

* ZTE focuses on 4.5G ("pre-5G" deal with SoftBank)
** "DOCOMO Successfully Conducts 5G Trials in Actual-use Environments - Trials carried out with Nokia Networks, Samsung, Ericsson, Fujitsu and Huawei" (NTT DoCoMo 20151126) - see also "DOCOMO 5G White Paper" (NTT DoCoMo 201407)
*** e.g. "KT demonstrates 5G network at National Assembly" (Korea Times 20150902)

4G take-offs and SOFT landings

Captain Kirk obviously paid a visit to Japan, where "4G" trials are announced by ever-the-customer-oriented NTT DoCoMo : "the 1Gbps real-time packet transmission was realized through Variable Spreading Factor-Spread Orthogonal Frequency Division Multiplexing (VSF-Spread OFDM) radio access and 4-by-4 Multiple-Input-Multiple-Output (MIMO) multiplexing using "adaptive selection of surviving symbol replica candidate" (ASESS) based on Maximum Likelihood Detection with QR decomposition and the M-algorithm (QRM-MLD)".
This nuclear mushroom is supposed to impress people listening to the sound of miserable 3G+ trials under way across Asia (ie HSDPA in China and Taiwan, HSUPA sites under construction everywhere - websites, that is).
This time, no one is wondering why high speed is needed : back to Paris I badly miss my 100 Mbps home. Jet lag is OK the other way : that's what I call the "SOFT landing" (Seoul Optic Fiber Transition).


PS : I don't know whatever happened to Club Nokia but here are a few figures from Samsung's Anycall Land : 180,000 hits per day for 4M members.
PS2 : I don't know whatever happened to KTF's investment fund devoted to content but here is how they start countering SKT's shopping spree : a minority stake in a movie investment fund set up by a key player in distribution (among Showbox Inc assets : Megabox multiplexes, On*Media cable TV...)
PS3 : sucks (according to an independant Redmond based group of Xbox consumers and developpers).

Finding MIMO ? after MoMu and MoMa

July 2004 : long before the opening of the new MoMa on the West 53rd, the MoMu opens its virtual gates to the world. SK Telecom's online Mobile Museum encompasses 20 years of wireless experience and is in itself a broadband experience : even with huge Hanaro or KT Megapass bandwidth, browsing the site takes some time. The thing is all major architects were drafted for the Leeum, Samsung's Chairman Lee Geon-Hee's Xanadu and a place to see if you love Korean ceramics (ie a unique Buncheong ware collection) and traditional art - as far as modern art is concerned, Leeum's impressive check-list sounds like a turn of the millenium's Frick collection.
Well. R&D people have other worries than art, worries worthy of the age of mobile broadcasting : will "Finding MIMO" (Multiple Input Multiple Output) be the next blockbuster, will this uncertain roadmap "Trow mamma off the train" or will they have to bury "The Mummy" ?