SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Nokia Corp. (NOK) -- Ignore unavailable to you. Want to Upgrade?


To: waitwatchwander who wrote (2848)3/17/2003 3:39:24 PM
From: Eric L  Respond to of 9255
 
Trevor,

Born Digital GSM platforms v. Born analog ANSI-41 Core Networks

<< How does an analog RAN (or digital for that matter) impact a core network design or operation? >>

It sort of the other way around.

You can do what the GSM fraternity did with their born digital and you can design, standardize, and implement a complete end to end digital PLATFORM consisting of a core network consisting of numerous interconnected core components and peripherals but also and most importantly consisting of underlying protocols and services designed to support digital voice and digital data (however rudimentary) as well as the digital RAN (with no initial concern about interoperability with a legacy analog network of any type, and then as a community you can implement that platform more or less uniformly network to network to network.

You can also do what the ANSI-41 community did with D-AMPS and IS-54 TDMA guys did with IS-54 which is essentially what the CDMA (IS-95) guys did. You can take a (justifiable) shortcut designed to protect legacy networks and attempt to shoe-horn a digital RAN onto a born analog ANSI-41 network, and then go back and try to implement protocols and data services at some time in the future.

The classic work on GSM Phase One which laid the foundation for GSM Phase 2 which is what all American GSM networks launched with. and the design philosophy and documentation of interplay between the digital core and digital RAN is Mouley and Pautet's "The GSM System for Mobile Communications).

A decent short treatise about was written by John Scourias of the University of Waterloo. this can be contrasted with Tom Farley's (or others) works on the evolution of ANSI-41 based technologies.

ccnga.uwaterloo.ca

<< Your post makes it appear that moving from an analog to digital RAN has a major impact on the core network. I find that hard to believe. >>

Well don't believe it. I'm sure you have some explanation of why technology adoption of cdmaOne/CDMA2000 by carriers worldwide has been so abysmal, and why so many carriers who make platform decisions have chosen GSM/3GSM over QCDMA, 5 or 6 to 1.

My post was a response to a good friend and consummate GSM basher who has been saying "GSM is Toast" for so long that his noodle is frazzled.

<< Networks handle circuit and packet switching. They've being implemented with digital technologies since the 60's. >>

Say what and pardon me? Mobile wireless networks have been implemented with digital technologies since the 60's? What am I missing here?

Would you please give me an example of digital mobile wireless network technology (cs or ds or mobile IP) implemented in the 60's, 70's, or 80's.

- Eric -



To: waitwatchwander who wrote (2848)3/17/2003 5:26:10 PM
From: 49thMIMOMander  Respond to of 9255
 
It is still difficult to get a digital ISDN call connected through without going analog inbetween, or
messed up 64/56kbps as well as "stolen bits" for controll purposes. That is, nothing much
is left of the original data if the 64/46kbps stream is converted to some noisy analog voice,
and then converted back to digital.

That is, the "30 years of promise of ISDN" gives an idea on how well the whole network is, or was,
digitlized.

Note also that one can convert a 8-13kbps voice-coded datastream to analog voice, route it
through consecutive analog plus 64/56kbps digital channels, and back to the same voice-coding
system for the other handset, but then one is back to the "old days", when local calls were OK
but long distance were sometimes, sometimes not, gradually worse or just dropped.

Additionally the digital networks must be synchronized to actually have error-free 64kbps
transfer. GSM was designed to utilize ISDN as well as tolerate transfer between
non-synchronized ISDN networks.
This means one buffer-full of data will disappear or maybe be "repeated twice" every now and
then, not just bit-errors, but messing up frame synchronization, control bits,etc

That is, to move digital data from the east to the west coast through some intermediate nodes and
networks, all have to be synhronized for this not to happen.
Adding buffering "helps" but then the data is delayed, which turns out as voice-echo, same as for
satellite calls (especiilly if 2-3-4 satellite links are used, by accident or on purpose)

This is easy if the call is made from and to the same operator which probably has their
own network,etc "all the way" , but worse between, for example, two small local operators which
use two or even more other larger operators/networks for the long distance parts.
(depending on what time of the day, etc)

Another example are SMS-messages which get delayed minutes or hours instead of
seconds, and sometimes totally disappear due to inter-operator problems (or activites).
(remember my old stories on 10 second back-forth delay to Africa for SMS-messages,
at the time Wap was CRAP using SMS in UK)

Ilmarinen

Another sign of "digitalization", other than the availability of long distance ISDN, is
if the local operator provides ADSL services. When they provide that they have:

- figured out what kind of copper-network they have, moved around T1 connections
to cables where they do not kill ADSL signals, or the other way around
- Bought and installed, once again, new Subscriber equipment, 256-2Mbps packet for former ordinary
just POTS, Plain Old Telepnone customers
- installed an internet backbone, fiber, servers and access points in the same cellar
that all of above is (CISCO + much more)

A third example might be one of my Quack-favorites, this Quack-thread guy whose 56kbps
modem had done a maximum of 28kbps, that is, he had actually never been able to use
the "56kbps" function, just the plain old analog V32bis modem, and not even its 33kbps
speed.

Doubly funny as the max speed for a "56kbps" is 48kbps.
That is, a "56kbs" works by using the 56/64kbps (logartminc) DA-converter in the subscriber
equipment for downstream, modem detects which of those 64,000 levels it transmitted.
However, to actually work, that 64kbps, normally voice-only channel, must not be converted
to analog anywhere else.(and additionally be "good enough")

If that does not work, the modem falls back to plain old analog modem, at 19-28kbps and
if a really good network, 32kbps, but then 48kbps at "56kbps" should work too..

Another funny story was when one USA long distance operator advertized cheaper
long distance calls in the early 90s, by using ADPCM to put two voice channels on
one 64kbps channel. One minor problem, fax modems did not work anymore.
(And AT&T went busy advertising that "we support faxes")

When widely available, the operator has additionally "decentralized" the old switch,
moved out small "parts of it" in small boxes with 0.5-1.5 miles range from "clusters
of homes"



To: waitwatchwander who wrote (2848)3/20/2003 5:46:32 PM
From: Eric L  Read Replies (1) | Respond to of 9255
 
Tremblay's Fido Restructuring Approved

>> Creditors Approve Plan to Restructure Microcell

Toronto
Global Wireless
March 18, 2003

Microcell Telecommunications, owner of the Fido cell-phone network, has been taken over by its creditors after they agreed yesterday to a restructuring plan proposed by the carrier.

The deal, accepted by representatives of banks and other debt holders, lets them take control of the company in return for forgiving C$1.76 billion (US$1.2 billion) in Microcell debt.

“We are pleased the plan received such a strong endorsement from our creditors,” said Microcell Chief Executive Andre Tremblay.

Microcell is the smallest of the four Canadian mobile phone carriers and the only one to seek bankruptcy protection. It has close to 1.2 million subscribers.

Company officials said operations will continue as before. But the company will not attempt to expand its customer base, which reaches more than 60 percent of the population. <<

- Eric -