Talk:Generic Access Network

From Wikipedia, the free encyclopedia

WP:TEL This article is within the scope of WikiProject Telecommunications, an attempt to build a comprehensive and detailed guide to telecommunications on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project as a "full time member" and/or contribute to the discussion.

The word "connexion" is used... A typo?212.143.17.66

You changed it already, but no, it's not. Look it up in your dictionary, or even Wikitionary. ----Broccoli 21:57, 30 June 2006 (UTC)

Contents

[edit] Roaming onto PSTN?

A GAN mobile cannot roam from 802.11 to PSTN. That would require a physical connection to the local loop. Rather, it can move from 802.11/bluetooth/any IP provider to the GERAN (GSM) or UTRAN (UMTS) network.

[edit] Expand please

Customers can enjoy one bill for their internet and cell phone plans.

Why? I don't see why the ISP and cellular service provider would even have to be the same company.
All it says is "Can", though it's a dubious benefit. Customers of BellSouth can "enjoy" one bill for their internet and cellphone plans already. Billing isn't really that relevent. This part really needs to be deleted. Squiggleslash 15:34, 29 September 2006 (UTC)

Because handsets must listen to two different radio technologies, they must have two radios on board. Both radios must be scanning for networks at all times, in case the user roams into an area where a Wi-Fi network exists. Battery life may be limited.

Battery life is limited as a direct result of this, right? How much power is used by each radio? Can they be turned off if not in use?
At this stage, the question is being asked too early. GSM radios under normal circumstances usually last anything from three to ten days depending on the battery and manufacturer. My Wifi-enabled Nintendo DS can sit listening for 802.11 traffic for around eight hours in my experience, but that's some other type of battery, and the DS doesn't shut off most of the hardware when it's listening for "One cartridge games".
At the same time, realistically, both radios do not need to be on all the time. While the 802.11 connection is strong, the GSM radio can be turned off completely. The 802.11 connection can be polled periodically, if the GSM signal is low, or if network activity is about to start. While I'm not intimate with the precise details of the UMA specification, it's hard for me to believe at least some of these options aren't available. Squiggleslash 15:34, 29 September 2006 (UTC)

All data from the handset goes through the carrier's servers and is chargeable. Subscribers might wonder why they are being charged for data that is going over their own internet connections, when they can use other devices, such as laptops, for no extra charge.

Why would data which passes through my own personal wireless network go through my cellular provider's servers? So phones like the Nokia 6136, Samsung SGH-T709, and the T-Mobile SDA cannot connect to regular wireless networks? They can only connect through the cell provider's servers? Is there any technical reason for this or is it just a way to make money for nothing? — Omegatron 00:38, 18 September 2006 (UTC)
Data that you route through your phone will be routed through the provider's servers.
The thing to understand is that GAN/UMA (you know, I'm just going to call it UMA. Only the 3GPP and this article call it "GAN") aims to make your 802.11 AP just another tower on the GSM network. It has limitations, but that's the gist of what it's trying to do. The upside is that the whole transitioning between the "real" GSM towers and your AP is seamless. You can be on the phone. You can be downloading a web page. Nothing will be interrupted. You'll not have to restart anything.
The downside is that occasionally it's not optimal. If two people are using UMA, it makes sense to route calls between them directly rather than via the cellphone provider's servers, but that's difficult to implement without far more layers of complexity. If someone is surfing the net, argubly it's better they receive the data directly from their WLAN than routed over the cellphone operator's network, but if you tried to do that, every time they walk out of range of the WLAN, the data is going to get lost and nobody is going to even realise that's what's happened.
It's a great looking technology, but many of the optimal "hacks" people want for it aren't going to work or be easy to implement in anything approaching an elegant fashion. Perhaps 4G and IPv6 will fix many of these issues by creating a clean architecture we can start afresh with, but for now, this is way better than nothing at all Squiggleslash 15:34, 29 September 2006 (UTC)

[edit] GAN on 3G networks (EGAN)

Is there a website (official if possible) about extension of GAN to 3G UMTS / HSDPA / HSUPA networks? With advent of HSDPA, 2G GAN is now really obsolete. Please expand this part of the artcle or report a good website with these infos. Armando82 14:52, 5 March 2007 (UTC)

I investigated the topic of GAN on 3G networks. These are the results:
  • UMA Today / Kineto Wireless whitepaper ([[1]], [[2]])
"3G/UMTS Handset Support: It is widely known that UMA enables the development of dual-mode 2.5G (GSM/GPRS/EDGE)/Wi-Fi handsets. What is less known is that the existing 3GPP UMA standard also supports the development of tri-mode 2.5G/3G/Wi-Fi handsets and allows for seamless handovers between 3G and Wi-Fi networks. In fact, there are several UMA-enabled trimode handset projects in development at this time."
  • Wikipedia Generic Access Network article
"GAN is designed around connecting directly to an existing 2/2.5G GSM network. As such it is only suitable for network operators who have an existing GSM network. For operators such as '3' in the UK this is not the case. To address the issue a development which may eventually be referred to as EGAN kicks off in January 2007.[1]"
  • 3GPP Enhanced Generic Access Networks Study (EGAN); (Release 8) ([[3]], [[4]])
"GAN Enhancements shall provide optimized support for operators using GSM-only, UMTS-only, or combined GSM/UMTS networks"
What is sure from official sources is that EGAN, which will be in 3GPP Release 8, will supports UMTS networks. What is unclear is if support was already added and, if yes, starting with which 3GPP release (5, 6 or 7?). If you are well-informed of when 3G support was added (or not added) please notify your source. I think this information could be find in 3GPP specifications TS 43.318 and TS 44.318 but the help of an expert is needed.
Finally, Wikipedia assertion that 3G is not yet supported in the standard is unsourced so I think we should modify it with the following:
"Dual-mode 2.5G(GSM/GPRS/EDGE)/Wi-Fi handsets are available. The current 3GPP specification allow 3G/Wi-Fi handover too and several tri-mode 2G/3G/Wi-Fi handset are in development. Until these became available operators who haven't a GSM network, such as '3' in the UK, cannot deploy GAN.
A development referred as EGAN (Enhanced Generic Access Networks) kicks off in January 2007.[2]"
Do you agree with this change? Armando82 18:03, 2 April 2007 (UTC)

[edit] Oy, carrier scam

So basically GAN is a scam by carriers to keep their stranglehold on the voice business. Can't let those overinflated prices fall! How about some open source products that might evolve into alternatives?!

http://www.a-la-mobile.com/company/vision.html

http://www.trolltech.com/products/qtopia/greenphone

http://wiki.openmoko.org/wiki/Main_Page —The preceding unsigned comment was added by 24.222.217.206 (talk) 15:55, 16 March 2007 (UTC).

If by "keep their stranglehold on the voice business" you mean "make their networks suck less", then, erm, yeah.

I've really never understood the mentality of those who argue against everything that's done to improve a technology to make it more usable on the grounds that the only reason they're doing it is, er, to make it usable and thus cement their "monopolies". The same crap has utterly destroyed one article (IP Multimedia Subsystem), let's not let it kill this one. --Squiggleslash 15:59, 16 March 2007 (UTC)

[edit] Question to introductory paragraphs of Generic Access Network

In the introduction on GAN, 3rd paragraph, it says

"The wide network is alternatively GSM/GPRS or UMTS mobile services."

Shouldn't it better read

"The wide area network is alternatively GSM/GPRS or UMTS mobile services."

Matthi2 08:57, 4 May 2007 (UTC)

[edit] "The first UMA deployment in the United States..."

I doubt it was Cincinnatti Bell. T-Mobile's hotspot@home was deployed well before CinBell even announced their service in June 2007. —Preceding unsigned comment added by 216.9.243.111 (talk) 12:28, 31 October 2007 (UTC)

H@H was trialed in select areas before CB's roll-out, but not officially made available to everyone until a few weeks afterwards. I assume CB also trialed their service before June, it's not clear who was trialing it first, but either way, CB's first public, non-beta, roll-out came before TM's. --Squiggleslash 13:33, 31 October 2007 (UTC)