Groups | Search | Server Info | Login | Register


Groups > comp.arch > #6073

Re: Bandwidth for smart phones

Date 2012-02-25 10:30 +0100
From Terje Mathisen <"terje.mathisen at tmsw.no">
Newsgroups comp.arch
Subject Re: Bandwidth for smart phones
References <XTY1r.14910$kv1.8255@newsfe03.iad>
Message-ID <damm19-7bi2.ln1@ntp6.tmsw.no> (permalink)

Show all headers | View raw


Robert Myers wrote:
> It would be natural that I'd be thinking, "But how are you going to feed
> all those devices?"
>
> As far as I can tell there is no physical way to make wired Internet
> connections completely obsolete.

You need them to feed all those base stations...
>
> There are a couple of ways you could still have smartphones selling in
> unlimited numbers forever, but they all seem too complicated and limited
> to me.

A sufficiently dense grid of small (pico/nano/femto) cells, using MIMO 
antennas and backstopped with mesh networking can give us at least a 
couple of orders of magnitude more than we have today.
>
> Net result: the notebook (and it's relatively power-hungry processor)
> aren't going to go away any time soon.

If by that you mean local processing/storage/visualization capability, I 
agree.
>
> In other words:
>
> Server market (still exists)
>
> Notebook/desktop market (isn't going away any time soon)
>
> Super-clever trendy smart phones (will eventually be only for the very
> well-off because scarce bandwidth will be too expensive)

Bandwidth isn't that hard to share if you can reuse it over and over 
again, i.e. small enough cells/low enough xmit power levels.
>
> Any thoughts?

I've been working on the new cell phone grid here in Norway for the last 
two years, today I am writing this from my mountain cabin (where I have 
had fiber since it was new in 2004), but I am using a USB dongle over 
3G/Turbo3G which is giving me 11-14 Mbit/s download and 2-3 Mbit/s 
upload speeds.

The main limiting resource in a modern cell phone network probably isn't 
raw bandwidth, that is relatively easy to supply as long as your base 
stations have a fiber connection. Using steerable antennas can let you 
reuse the same bandwidth for multiple users. It is somewhat more 
expensive to provide more base stations, and buying more wireless 
bandwidth from the government can be very expensive.

Administrivia (registration, cell-to-cell handover, crypto 
(re-)negotiations, roaming, etc) is a major concern, and the reason 
behind the infamous ATT meltdown, as well a smaller but similar problem 
here in Norway last June.

Another issue is the fact that most (all?) phone companies are built on 
classic Erlang models where most subscribers are idle most of the time:

A modern always-connected smart phone (or laptop/netbook/ipad/etc) on a 
3G/4G network needs a lower-overhead way to stay connected, while the 
crypto protocols would like to renegotiate all the keys much more often.

Terje
-- 
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

Back to comp.arch | Previous | NextPrevious in thread | Next in thread | Find similar


Thread

Bandwidth for smart phones Robert Myers <rbmyersusa@gmail.com> - 2012-02-24 22:50 -0500
  Re: Bandwidth for smart phones Robert Wessel <robertwessel2@yahoo.com> - 2012-02-24 22:35 -0600
    Re: Bandwidth for smart phones Morten Reistad <first@last.name> - 2012-02-29 19:40 +0100
  Re: Bandwidth for smart phones Andrew Reilly <areilly---@bigpond.net.au> - 2012-02-25 05:22 +0000
  Re: Bandwidth for smart phones Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-02-25 10:30 +0100
    Re: Bandwidth for smart phones Robert Myers <rbmyersusa@gmail.com> - 2012-02-25 20:49 -0500
      Re: Bandwidth for smart phones del cecchi <delcecchi@gmail.com> - 2012-02-25 21:50 -0800
        Re: Bandwidth for smart phones Robert Wessel <robertwessel2@yahoo.com> - 2012-02-26 02:23 -0600
          Re: Bandwidth for smart phones "Andy (Super) Glew" <andy@SPAM.comp-arch.net> - 2012-02-28 19:39 -0800
            Re: Bandwidth for smart phones Robert Wessel <robertwessel2@yahoo.com> - 2012-02-28 22:38 -0600
            Re: Bandwidth for smart phones "Paul A. Clayton" <paaronclayton@gmail.com> - 2012-02-29 06:18 -0800
              Re: Bandwidth for smart phones "Andy (Super) Glew" <andy@SPAM.comp-arch.net> - 2012-02-29 07:56 -0800
                Re: Bandwidth for smart phones George Neuner <gneuner2@comcast.net> - 2012-02-29 14:31 -0500
                Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-03-01 00:03 -0700
                Re: Bandwidth for smart phones Robert Wessel <robertwessel2@yahoo.com> - 2012-03-01 02:07 -0600
                Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-03-01 15:44 -0700
        Re: Bandwidth for smart phones Brett Davis <ggtgp@yahoo.com> - 2012-02-26 04:30 -0600
        Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-02-26 11:43 -0700
          Re: Bandwidth for smart phones Emil Naepflein <netnewsegn@kabelmail.de> - 2012-02-27 06:29 +0100
            Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-02-27 18:12 -0700
              Re: Bandwidth for smart phones Emil Naepflein <netnewsegn@kabelmail.de> - 2012-02-28 06:37 +0100
                Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-02-27 22:54 -0700
        Re: Bandwidth for smart phones jgk@panix.com (Joe keane) - 2012-02-28 15:52 +0000
          Re: Bandwidth for smart phones Thomas Womack <twomack@chiark.greenend.org.uk> - 2012-02-28 17:15 +0000
          Re: Bandwidth for smart phones Robert Wessel <robertwessel2@yahoo.com> - 2012-02-28 14:14 -0600
            Re: Bandwidth for smart phones Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-02-29 08:28 +0100
              Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-03-01 00:35 -0700
                Re: Bandwidth for smart phones Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-03-01 18:24 +0100
                Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-03-01 15:53 -0700
                Re: Bandwidth for smart phones del cecchi <delcecchi@gmail.com> - 2012-03-01 11:22 -0800
                Re: Bandwidth for smart phones George Neuner <gneuner2@comcast.net> - 2012-03-01 16:06 -0500
                Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-03-01 16:01 -0700
                Re: Bandwidth for smart phones Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-03-02 13:25 +0100
                Re: Bandwidth for smart phones nmm1@cam.ac.uk - 2012-03-02 12:36 +0000
                Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-03-02 12:08 -0700
                Re: Bandwidth for smart phones Piotr Wyderski <peter.pan@neverland.mil> - 2012-03-02 10:49 +0100
                Re: Bandwidth for smart phones George Neuner <gneuner2@comcast.net> - 2012-03-04 14:16 -0500
      Re: Bandwidth for smart phones Robert Wessel <robertwessel2@yahoo.com> - 2012-02-26 00:31 -0600
      Re: Bandwidth for smart phones Brett Davis <ggtgp@yahoo.com> - 2012-02-26 04:01 -0600
    Re: Bandwidth for smart phones Morten Reistad <first@last.name> - 2012-02-29 20:00 +0100
  Re: Bandwidth for smart phones BGB <cr88192@hotmail.com> - 2012-02-25 15:51 -0700
  Re: Bandwidth for smart phones Piotr Wyderski <peter.pan@neverland.mil> - 2012-02-28 13:37 +0100

csiph-web