FTTx :: FTTH

FTTx stands for Fiber To The x, where x could be H(=Home), C,B,N,P or perhaps other letters. It is basically a form of "broadband" access technology and refers to "a generic term for any network architecture that uses optical fiber to replace all or part of the usual copper local loop used for telecommunications" (Wikipedia).

Types of FTTx:

• FTTCurb | FTTNode | FTTCab | FTTKerb? – Fiber to Street Cabinet --> distribution and drop cables: telco copper (DSL) or new copper; categorised as DSL technologies
• FTTBuilding – Fiber to Apartment / Office Building --> in-building cables: building copper or fiber; regarded as a transitional stage to FTTH
  
• FTTHome – Complete Fiber Path to Home --> in-building cables: house copper or fiber or wireless; may require more initial investment (capex), but promises considerable savings in operating costs (opex)
  

Note: The difference between FTTN and FTTC is quite subtle and is mostly that the latter is nearer the customer than the former. The broadly-defined term fiber to the premises (FTTP) is sometimes used to describe FTTH and/or FTTB.

According to FTTH council, as of mid 2008 the distribution of FTTH/FTTB subscribers is as follows:

• North America : 3.4 mn subs
• Europe : 1.4 mn subs
• Asia Pacific (mostly East-Asia) : 27.8 mn subs
  

APAC's leaders in FTTH/B deployments are South Korea, Japan, Hong Kong, Taiwan and Singapore. How about Indonesia ? :-D It's quite far away but last year Biznet has claimed to have "a first Fiber To The Home (FTTH) network in South East Asia" (SG isn't in South East Asia???)

Source/Courtesy : Wikipedia/Riick

FTTH Statistics (as of June 2008) - Source FTTH Council

Highlight (as of June 2008) :

• FTTH is THE mainstream fixed access technology ( over 32M FTTH/B subscribers globally )
• Governments and Regulators in all countries are working towards the large scale rollout of FTTH ( RFP’s underway in Malaysia, Singapore, Australia, ..)
• Asia-Pac leads the world in FTTH depoyment !

Singapore

• NetCo RFP awarded to OpenNet (Axia, Singtel, SPH, SPT)
• Government Grant of $750M
• OpCo RFP – submission 14 Nov 08, award Q1’09

Malaysia

• HSBB to build in selected regions, 20Mbps, fiber-based
• BBGP for all Malaysia, 2Mbps, multi-technology
• Telecom Malaysia will build HSBB

Asia Pacific Submarine Cables

I've just read Peter Evans's article on the new WFN issue dealing with cables in APAC region and mentioning some optimistic/pessimistic views of this development. Short speaking, almost everybody agrees that oversupply happens for intra Asia segment (though, it doesn't mean that prices will decline! - history has shown this!) but the arguments for trans-pacific are varied; while some players/observers see some positive things, the other predict that it could be a "dangereous" bubble (something likes that in 2000s).

Each party has many arguments and I don't want to mix with another argument or more :-D But I just want to comment on the tables he has presented on the article. It seems that it isn't very accurate e.g. as of now Nava's cable isn't real yet (Ochre might reinvent that idea, but though it is still a plan!) CMIIW.

Source/Courtesy : WFN/P.Evans@Budde

TGN Pacific

Tata's TGN Pacific is actually a quite "old" system :-D; but it counts to the well-designed systems.

• The system forms a self-healing ring connecting US to Japan
  
• RFS : December 2002
• CAPEX: USD 1.7 bn (For comparison: Tata acquired Tyco for USD 130 mn in 11/2004 !!)
  
• Capacity : 460 Gbps initial ??; 960 Gbps (upgrade in 2006); 1.28 Tbps 2008 ??
• Length : 22,300 km

Source/Courtesy : Primetrica

Exploring Telkom / Telkomnet

I've captured robtex-view of Telkomnet two or three months ago (see below). And this is relevant with the conversation between Sang Pemimpin and Meneer Tutut.

First of all, "the edges of an AS graph represent logical relationship between ASes and do not represent the form of the physical connection" (Lixin Gao). Short speaking, this describes more-or-less the BGP-view from a specific AS.

Telkom currently owns at least 2 ASes : AS7713 and AS17974. All external connections seem to be happened at AS7713, while AS17974 looks to be a "domestic island". As we can see below, BGP-views from these ASes are different. Two or three years ago, I found a good source related with this issue made by Timothy Griffin. I have to check this again, or maybe if you have interest, you could try "google" using his name.

As mentioned by Meneer, the following ASes are adjacent to AS7713 as "upstream" (in the CIDR terminology - not necessarily real Telkom's upstream providers):

• AS24077 --> TMHK-TRANSIT-AS-HK-AP TMHK Global Transit
• AS24399 --> ABN-PEERING-AS-AP Asia Broadcast Networks Peering AS
• AS38154 --> PSN-NAP-AS-ID PT. Pasifik Satelit Nusantara
• AS30362 --> HPT-ASN - Hawaii Pacific Teleport LP
• AS4657 --> STARHUBINTERNET-AS Starhub Internet, Singapore
• AS4788 --> TMNET-AS-AP TM Net, Internet Service Provider
• AS6762 --> SEABONE-NET Telecom Italia Sparkle
• AS7473 --> SINGTEL-AS-AP Singapore Telecommunications Ltd

Looking at the number of BGP advertisements, it seems that AS7473 (Singtel) and AS24077 (Telekom Malaysia Hong Kong) plays important roles for Telkomnet's network. Disagree ? Any advise?






*Update 04/09*

Sedangkan untuk kapasitas bandwidth, Nyoman mengatakan, Telkom yang sebelumnya memiliki sekira 13 Gbps, saat ini bertambah menjadi 17 Gbps. Dan pada bulan juli mendatang Telkom akan mendapatkan sekira 40 Gbps sebagai bagian dari keanggotaan Telkom dalam konsorsium Asia-America Gateway (AAG).

Pacific Cables + Completion of TGN - IA

Quotes from TM's talk:

• The increasing trend has seen upgrades and new submarine cable investment in Asia. An increase of 120% in the Trans-Pacific route capacity is already happening in 2008.
• An additional approx 4Tbps would be available; further 5-7 Tbps being planned for 2010
• Whilst these are planned capacity; it still uncertain as to the actual capacity that will be made available for the market
• If the trend is to be followed, then such capacity will be required. Again the issue of cost will creep in whereby can the end user actually be willing to pay for the ever increasing hunger for bandwidth
• The breakdown of the model of end users paying for the bandwidth may see the capacity being less utilised leading to over capacity

Several hours ago, Commsday has reported that "Tata Communications has confirmed that its TGN Intra-Asia cable is undergoing final testing and will be ready to carry traffic from February".

Some other quotes:

• The new cable fills a gap where while it has 100% ownership of cables between India and Singapore as well as Japan to the US it has inferior options in intra-Asia
• The cable connects Singapore and Japan via Hong Kong with branches to Vietnam and the Philippines; the direct link on the cable improves latency compared to other cables that are characterized by multiple hops on that route
• It avoided the route taken by other cables through the earthquake-prone section of Taiwan waters; the cable has been designed to be physically as far away as possible (from the critical regions)
  

• “We’ve always taken a contrarian view that ownership of undersea cables is critical in order to have the right cost structure and the end-to-end manageability that our customers are looking for.” ; “To have that ownership in Asia was important for us.”

• “We also believe that submarine cables are a good business investment if you get the right mix of wholesale and enterprise traffic.”
  

Source/Courtesy TM

Femtocells & Femto BTS

I've just read some articles related with this "femto" issue. Also, maybe some of you have heard about this thing ...

Source/Courtesy FemtoForum.org

At least Sprint and Verizon have already offered early "femto" services (cited from dailywireless.org | lightreading.com). But for me, at a first glance, it wasn't quite clear: what benefits are given to the customers? Whether these femto-APs/BTSs could really with-reasonable-benefits replace existing infrastructures? ....

Let's start with some definitions/quotes:

• "Femtocells, or tiny cellular base stations, ..." (lightreading) --> okay, intuitive, it shall be smaller than pico
• "A femtocell is a low-power wireless access point that operates in licensed spectrum to connect standard mobile devices to a mobile operator’s network using residential DSL or cable broadband connections" (FemtoForum) --> several questions such mentioned above begin to confront !
• "The pint-size base stations go into homes or offices to improve indoor cellular coverage and the traffic is backhauled to the mobile core network over users' own broadband connections" (lightreading) --> it isn't new, heh ?
• “Perhaps best of all, the device sends all calls over the subscriber's home broadband connection, usually DSL or cable, so the carrier doesn't have to pay to carry the traffic from the femtocell to its network. "Backhaul" traffic, which runs calls from cellular towers to the wired network, is a major cost for carriers.” (hp community) --> sure, but could they really replace existing outdoor BTSs? No! (disagree? leave a comment please :-D)
• "Femtocells in our homes connected to our DSL or cable modems where we get our Internet access from the local access provider. We all pay the local access provider for the “backhaul” of our WiFi access points, home routers, PCs, etc. Hence we are subsidizing the backhaul of the femtocell for the wireless operator to obtain coverage footprint in our homes. And yet, Sprint is charging $50 for their femtocell (from Samsung) and $15 per month service for unlimited calls from home…" (hp community) --> agree :-D
  

A better argument

• "At an LTE conference in Berlin I spoke to two Motorola network managers. They explained me a femtocell approach that could be suitable for integrated carriers such as Vodafone, which owns one of the biggest DSL providers in Germany: Arcor.

  People could get a very cheap DSL connection from Arcor under the condition that they install a femtocell at home, which is not only for themselves but can be used by every passing Vodafone user on the street. Vodafone would subsidize my fixed line if I help them with their wireless LTE coverage.

  The Motorola people also told me that the most expensive cost unit of a cellphone base station is the roof rent. I guess they would be happy to cut some antennas in exchange for femtocells. It seems plausible to me." (Markus Goebel Comment @gigaom)
  

First forget about LTE, DSL or Cable connections can not reach 150 Mbps right ? cmiiw! For voice services as Verizon does with its femto-offerings, at least the connections shall be able to reserve 100+ kbps for three parallel conversations. How about QoS ? well, it shall be another thread :-D

Could we soon hear Telkomsel-Femto or Indosat-Femto offerings ??? Questionsmark :-D

*Update - From Wikipedia.org*

In telecommunications, a femtocell—originally known as an Access Point Base Station—is a small cellular base station, typically designed for use in residential or small business environments. It connects to the service provider’s network via broadband (such as DSL or cable); current designs typically support 5 to 200 mobile phones in a residential setting. A femtocell allows service providers to extend service coverage indoors, especially where access would otherwise be limited or unavailable. The femtocell incorporates the functionality of a typical base station but extends it to allow a simpler, self contained deployment; an example is a UMTS femtocell containing a Node B, RNC and GSN with Ethernet for backhaul. Although much attention is focussed on UMTS, the concept is applicable to all standards, including GSM, CDMA2000, TD-SCDMA and WiMAX solutions.

For a mobile operator, the attractions of a femtocell are improvements to both coverage and capacity, especially indoors. There may also be opportunity for new services and reduced cost. The cellular operator also benefits from the improved capacity and coverage but also can reduce both capital expenditure and operating expense.

Femtocells are an alternative way to deliver the benefits of Fixed Mobile Convergence. The distinction is that most FMC architectures require a new (dual-mode) handset which works with existing home/enterprise Wi-Fi access points, while a femtocell-based deployment will work with existing handsets but requires installation of a new access point.

Click here for complete femto story at Wikipedia

*Update - OC*

Femtocells for indoor coverage
+ pushed heavily as a solution, but issues remain
+ unit pricing, standards, spectrum availability, interference management
+ fits best with an integrated fixed/mobile operator

Click here for the story of Verizon-Femto at dailywireless.org

"Alternative" Cable

I still don't get any support for this idea. One of my friend, which is financially strong connected, said that it won't financially be feasible!! So, what do you think? Isn't Malaysia-Singapore-Indonesia market large enough to have a good business with this? Learning from Hengchun earthquake, do we really need route diversity ? Of course, there is "oversupply" on the northern route (EAC-C2C, APCNx, etc.) but is there no enough reason for telcos/companies to buy capacity on this route ? :-D

Below is only a rough drawing; sure you don't have to transit in Sangatta/Borneo. Instead, you could put some BUs for possible connections to the Philippines and perhaps Brunei.

By the way, besides "Alternative", I would call this as SIG (Singapore, Indonesia, Guam) Cable, or may be MSIG (+Malaysia) Cable, or also perhaps SGE (Souteast-asia Guam Express) Cable.

SJC (2)

According to Evans (EGS):

"The supply contract for South East Asia Japan (SJC) seems at the time of writing about to be signed. This pan Asian system has landing sites in Singapore, Malaysia, Thailand, Philippines, Hong Kong and Japan; the system has additional optional segments to Vietnam and Guam. Installation is expected in 2010."

Source/Courtesy : Graham Evans/WFN

Capacitymedia (10/2008):

Google is looking to extend its undersea cable reach across Asia. The company is part of a consortium of operators planning to build a cable from Chikura, Japan to Tuas, Singapore with branches reaching Guam, Thailand, the Philippines, and Hong Kong. The consortium includes Bharti Airtel, Singtel, KDDI, Global Transit as well as Globe Telecom in the Philippines and TOT in Thailand.

“If you look at Google’s current cable, the Unity cable between the US and Japan you see that they have more interest than just connecting to Japan,” says Alan Mauldin, research director at Telegeography. “They will be able to connect with a lot of different locations in Asia. Some of these places are clearly to serve local markets but some of these are strategic locations such as Singapore, which helps the company get to India. If you go to Guam there are options for getting to Australia with the new pipe coming along there.”

The Southeast Asia Japan Cable (SJC) is being planned by a similar consortium to that of the Unity Cable. Only Pacnet, which will have two fibre pairs on the Unity cable, will not be involved. “By doing this Google seems to meet its needs in terms of trying to get capacity at cost to serve its internal demand,” says Mauldin. “The cable actually looks nearly identical to the cable Tata is putting into service in the fourth quarter of this year. It follows a really similar path. It is not a ring-based system. It is a direct path between Japan and Singapore.”

If Alan's statement is true, then it is indeed quite reasonable that SJC will land in Tuas. Though, of course there are many other possibilities.

JaKa2LaDeMa

I've just followed a news entry fed to my reader from Subtel. It's about Jaka2ladema, I've mentioned here several months ago. Well, it's for me quite a bit surprise that Telkom announced officially its project. Or could it be just the suppliers' initiative ? :-D . In any case, my congrats to Fujitsu/NSW! I do hope that NEC does the similar thing, since BSCS is not a secret project, right ? :-D

• Suppliers/Contractors : Fujitsu-Norddeutsche Seekabelwerke (NSW)
• CAPEX: more than USD 100 mn
• Connects Kalimantan-Sulawesi; Java, Bali, Lombok
• RFS schedule : Q1/2010
• Hybrid repeatered/repeaterless system + BUs
• Cables: LW, LW protected, SA, DA -- NSW's MINISUB
  

Source/Courtesy : Telkom/Fujitsu-NSW/Subtel

*Update - Kompas|Bisnis|Detik|Telegeography*

• CAPEX : USD 115.41 mn | IDR 1.26+ tn
• Capacity : 320 Gbps (which segment?)
• Length : 1,836 km (total); (Ring 4 i.e. Pontianak-Ketapang-Pangkalan Bun-Banjarmasin --> 1,198 km; Ring 8 i.e. Sangata-Donggala--> 326.4 km; DMCS i.e. Benduluk-Denpasar-Mataram --> 312 km ??? -- need more explanation; press typo ??)
• Contract signed : 30 December 2008
  

• Primetrica : "Under the deal, Fujitsu is providing terminal equipment, repeaters, branching units’ system integration and related services for the JaKa2LaDeMa project. NSW is providing surveying, marine installation and civil works, and will also integrate Fujitsu’s repeaters into the repeatered cables it is deploying."
  

Planned & As-Laid Pacific Cables

Just want to highlight some interesting issues mentioned by Graham Evans in the new WFN release. In the pictures below, you see some new transpacific cables (planned/currently-installed/in-service).

According to Graham : "TPE is in service, AAG is nearing completion, Unity has been surveyed and will be installed in 2009, and the FLAG Eagle system has been postponed". Well, we might soon hear some formal service/project announcements of these new systems; AAG?, SJC?

Source / Courtesy : Graham Evans (EGS) / WFN

Palapa Ring (8)

Actually, there is no significantly observable progress of this Palapa Ring project since my previous post. Only some minor issues can be found as public releases (as of middle January 2009) :

• "Belum ada perkembangan yang berarti dari Palapa Ring. Sekarang masih pembahasan teknis dan prosesnya sedang dilaksanakan," ujar juru bicara konsorsium Palapa Ring Rakhmat Djunaedi, saat ditemui di Wisma Bakrie 2, Kuningan, Jakarta Selatan, Jumat (Bisnis - 9/1/2008) - quick translated --> no significat progress !
  
• Untuk mengerjakan proyek tersebut, lanjut Rakhmat, sudah terdapat empat vendor jaringan yang berminta untuk mengikuti tender, yaitu Alcatel Lucent, NSW, Tyco, dan NEC. Keempat vendor tersebut akan mengikuti tender untuk semua cluster - Tyco returns ?? How about Fujitsu ??
  
• Consortium members (no changes) : Telkom, ISAT, XL, BTEL
• Agreed consortium budget (no changes) : USD 180 mn (may be enough after network design adjustment)
• Minister of ICT M. Nuh has explicitly said that tax incentives for optical cables will be issued this year - questionmark is that whether these incentives are only for Palapa Ring project?! If not, well, it's time to build your domestic fiber infrastructure ! :-D