collaborative wi-fi note-taking -- possible ????

Archived from groups: alt.internet.wireless (More info?)

On Sat, 28 Aug 2004 02:29:36 +0900, Greg Conquest <gregconquest.com>
wrote:

>I am an English teacher in Japan, and in some of my classes the
>students and I are sitting around a table discussing current events,
>news, etc. and I make notes of new or useful phrases pertinent to the
>topic at hand. I currently use the dry-erase board in the classroom,
>but I think a collaborative note-taking application would allow me to
>type in the phrases, avoid interupting the flow of the conversation,
>and have more comprehansive notes efficiently going out to all the
>students (and maybe even get some notes back from the students as
>well.)
>
>Is there anything currently available that I can use for this? It
>seems that all chat software needs to be connected to a chat server
>(yahoo, AIM, MSN, etc.) in able to work.
>
>I am using a win2000 notebook with built-in 802.11b. Something
>inexpensive enough for students to use would be ideal. Apple's
>Rondezvous appears close to what I'm looking for, but it is not a
>complete package, from what I can tell, and it is not even available
>yet on the PC platform.

I watched a student sit through a lecture with some Apple "white
board" program and audio, while in a coffee shop (via wireless). Why
bother to attend a lecture when you can both hear and see the lecture
remotely? Every once in a while, he would ask a stupid questions via
the instant messaging window or headset, probably to remind the
lecturer that there was someone listening. The chat window showed
about 10 other participants. It was kind cute with an electronic hand
scribbling the drawings on the white board screen. Unfortunately, I
didn't recognize the software or bother to ask.

This looks close, but is probably not the same thing:
http://www.e-beam.com/products/
Ugh... Expensive.

Anything that works over ethernet will also work over wireless. So,
if your chosen internet white board software will play on an ethernet
LAN, it will play via wireless.


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

Archived from groups: alt.internet.wireless (More info?)

In article <q6rui0terdh08v5vu4ku38bf0hls2h672s@4ax.com>,
Greg Conquest <gregconquest.com> wrote:
:I am an English teacher in Japan, and in some of my classes the
:students and I are sitting around a table discussing current events,
:news, etc. and I make notes of new or useful phrases pertinent to the
:topic at hand.

When I -first- read the title, a completely different topic came to mind.

Suppose one had a group of wi-fi transmitors in an area. Suppose
further for the sake of simplicity that they are all the same -kind-
of equipment (so that we don't get sidetracked into discussions of
how different makes/models have different characteristics). Suppose
that the group was generally within radio sight of each other (e.g.,
a class trip or discussion table, in which the grouping can mostly
communicate directly with each other).

Now, given those conditions, can the transmittors effectively form
a "mesh" transmittor / antenna, in order to increase the distance
to which they can communicate to a fixed-point remote node?
Similar in effect (but in reverse) to grid radio telescopes, in which the
correlation of received signals over an area gives the effect of having
a larger antenna.

I might, for example, have a laptop transmitting at the maximum legal
power through the largest omni that can be *conveniently* attached to
the laptop, and with that setup I would be able to effectively communicate
a certain distance. But if (say) 5 or 6 other people nearby with
the same equipment are running a program in co-operation with me,
all transmitting the same thing (though with perhaps slight time skews
due to different positioning and different radio bounce paths) then
we could, it seems to me, effectively get further.

The FCC limitations are on single transmittors, and might perhaps be
interpretted to apply in integration to multiple transmittors owned by a
single person -- but if one is running a W2W (Wireless to Wireless)
cooperative program amongst whomever happens to be present, then
I suspect the FCC regs would not cover the situation. [And if one
is talking about convenient and mobile portable equipment then
one is probably not transmitting at the FCC limits anyhow -- a
parabolic antenna gyroscopically stablized to point in exactly the
right direction is not something that would fit in most people's
pockets.)


Now, if the nodes are wandering around (Bluetooth-like, whoever is
nearby), the calculations involved to co-ordinate the timing of
the packet transmissions to account for different paths, might be
somewhat difficult [though for a first-pass algorithm, one could
just all transmit at the same time to within the limit that one can
synchronize the clocks involved]. And the scheduling of packets
could get interesting when there are multiple senders involved
(all the nodes have to agree on the timing of a packet before
it can be wide-casted).

The reception part might be much more difficult -- correlating
the raw signals of N different stations would allow better
S/N ratios, but possibly one would have to work -at- the raw
signal level, rather than having the various systems involved
distribute around packets that say "Well, I dunno, this is what
the bitstream sounded like to me; I don't recognize it as a valid
packet myself". I'm not sure that "voting" on what each bit position
is would work, even with ECC codes -- one might have to look at
the raw signals and look for timing markers, sync them up, do
phase corrections, and so on, to get useful data.


I'm not saying cooperative WiFi would be easy... I'm just thinking
it could be somewhat interesting if it could be done.
--
Come to think of it, there are already a million monkeys on a million
typewriters, and Usenet is NOTHING like Shakespeare. -- Blair Houghton.

Archived from groups: alt.internet.wireless (More info?)

Walter Roberson <roberson@ibd.nrc-cnrc.gc.ca> wrote:
> In article <q6rui0terdh08v5vu4ku38bf0hls2h672s@4ax.com>,
> Greg Conquest <gregconquest.com> wrote:
> :I am an English teacher in Japan, and in some of my classes the
> :students and I are sitting around a table discussing current events,
> :news, etc. and I make notes of new or useful phrases pertinent to the
> :topic at hand.
>
> When I -first- read the title, a completely different topic came to mind.
>
> Suppose one had a group of wi-fi transmitors in an area. Suppose
> further for the sake of simplicity that they are all the same -kind-
> of equipment (so that we don't get sidetracked into discussions of
<snip>
> Now, given those conditions, can the transmittors effectively form
> a "mesh" transmittor / antenna, in order to increase the distance
> to which they can communicate to a fixed-point remote node?
> Similar in effect (but in reverse) to grid radio telescopes, in which the
> correlation of received signals over an area gives the effect of having
> a larger antenna.

In principle, yes.
In practice, it's going to be tough.
Forget about it being possible with current transmitters.

Problems.
Each transmitter has to know its distance (or relative distance between
it and the other nodes) to the main antenna, to better than a quarter
wavelength or so.
Say 3cm.
Now, they take this distance, talk amongst themselves to discuss what
packet to send, and at an instant timed to a ten billionth of a second,
everyone sends an identical packet. (at slightly different times depending
on distance).

The reverse can be done with recieved packets.

Well before this, we'll see sectional antennas that are spread over the
surface of a laptop/... to act as beam-forming antennae that actively
point at the target, rather than radiating omnidirectionally.

Archived from groups: alt.internet.wireless (More info?)

On 27 Aug 2004 22:25:37 GMT, roberson@ibd.nrc-cnrc.gc.ca (Walter
Roberson) wrote:

>Now, given those conditions, can the transmittors effectively form
>a "mesh" transmittor / antenna, in order to increase the distance
>to which they can communicate to a fixed-point remote node?

You're fairly close to 802.11n. See:
http://www.pcmag.com/article2/0,1759,1635778,00.asp
http://www.wi-fiplanet.com/news/article.php/3394291
http://www.wirelessweek.com/article/CA445108?spacedesc=...
Here's the WWISE TI/Broadcom proposal:
http://www.wwise.org/11-04-0935-00-000n-wwise-complete-...
There's also the MIMO proposal for 802.11n:
http://www.nwfusion.com/news/tech/2004/072604techupdate...
My appologies for the terrible collection of references, but I
couldn't find anything better.

>Similar in effect (but in reverse) to grid radio telescopes, in which the
>correlation of received signals over an area gives the effect of having
>a larger antenna.

Something like that exept the purpose is different. 802.11n in the
form of MIMO, WWISE or whatever, is designed to deliver more bandwidth
using multiple independent paths (and different data rates) through
multiple beam forming antennas or reflections.
http://www.acmqueue.com/modules.php?name=Content&pa=pri...
See illustration at the bottom. The most probable application will be
wireless streaming video.

>The FCC limitations are on single transmittors, and might perhaps be
>interpretted to apply in integration to multiple transmittors owned by a
>single person -- but if one is running a W2W (Wireless to Wireless)
>cooperative program amongst whomever happens to be present, then
>I suspect the FCC regs would not cover the situation.

The FCC has been there. The logic is one transmitter with multiple
antennas is perfectly legal. One transmitter with multiple phase
shifters, power amplifiers, and antennas, is basically a phased array,
which is also legal. It got messy when the pageing companies tried to
licence multiple phase locked (to GPS) transmitters on multiple
mountain tops as one transmitter. The FCC drew the line and specified
some geographic limitations. If interested, I can dig out the
references to that mess, but not now, as this is a diversion. In Part
15 (somewhere), the FCC clearly does NOT allow multiple sychronized
transmitters (which would be so nice in multiple radio access points
to prevent self-interference).

>Now, if the nodes are wandering around (Bluetooth-like, whoever is
>nearby), the calculations involved to co-ordinate the timing of
>the packet transmissions to account for different paths, might be
>somewhat difficult [though for a first-pass algorithm, one could
>just all transmit at the same time to within the limit that one can
>synchronize the clocks involved]. And the scheduling of packets
>could get interesting when there are multiple senders involved
>(all the nodes have to agree on the timing of a packet before
>it can be wide-casted).

That's exactly the problem. The process is called "training" and it's
the major headache.

>The reception part might be much more difficult -- correlating
>the raw signals of N different stations would allow better
>S/N ratios, but possibly one would have to work -at- the raw
>signal level, rather than having the various systems involved
>distribute around packets that say "Well, I dunno, this is what
>the bitstream sounded like to me; I don't recognize it as a valid
>packet myself". I'm not sure that "voting" on what each bit position
>is would work, even with ECC codes -- one might have to look at
>the raw signals and look for timing markers, sync them up, do
>phase corrections, and so on, to get useful data.

Yep. That's a different proposal using OFDM-MIMO. OFDM already has
multiple carriers and a messy DSP number cruncher to put the various
delayed parts of the signal back together. In its present incantation
as 802.11a/g, it it primarily used to deal with the effects of
multipath, but from a single antenna at each end. It is also rate
adaptive offering the most thruput that a "path" can deliver along
with power management to only use as much power per carrier as
necessary (and to remain legal). By extending OFDM to multiple
antennas at each end, one could dramatically increase the bandwidth.

>I'm not saying cooperative WiFi would be easy... I'm just thinking
>it could be somewhat interesting if it could be done.

Cooperative anything is difficult. Ever sit on a committee?


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

Archived from groups: alt.internet.wireless (More info?)

In article <4130d0cc$0$59023$ed2619ec@ptn-nntp-reader01.plus.net>,
Ian Stirling <root@mauve.demon.co.uk> wrote:
:In principle, yes.
:In practice, it's going to be tough.
:Forget about it being possible with current transmitters.

 roblems.
:Each transmitter has to know its distance (or relative distance between
:it and the other nodes) to the main antenna, to better than a quarter
:wavelength or so.

I don't see why that would be the case. All the discussion about
Freznel zones and how reflections can reinforce signals and so on --
those all imply that the signal is travelling different distances
over different paths, and that *way* more than 1/4 wavelength in
path difference is acceptable for satisfactory decoding.

In particular, does not the waveshapes used by 802.11a and 802.11g
hold particular symbols for much longer periods, precisely to account for
the smear in arrival times for multipath?

:Now, they take this distance, talk amongst themselves to discuss what
 acket to send, and at an instant timed to a ten billionth of a second,
:everyone sends an identical packet. (at slightly different times depending
 n distance).

Somehow it doesn't seem likely to me that 802.11g is built to handle
merely "all the multipath that you can get in one ten billionth of
a second".

[Subnote: which "billion" are we using here? You are posting from
the UK and I'm posting from Canada, where "billion" is a different
number (10^12) than "billion" is in the USA. (10^9)"

Even if we suppose 10^(-9), "a ten billionth" would be 10^(-10),
which would be 1/10,000 million. Even if we're talking about one
of the 802.11g "dual" transmission formats of about 100 megabits per
second. Even if the bitstream were done as 1 bit per baud (i.e.,
a single binary bit at a time is in transmission), a "ten billionth
of a second" is still 100 times smaller than the bit timing. But
with quadrature and subchannels and the increased bandwidth that
is needed for all of the "dual" transmission formats, the actual
bits per baud is noticably more, the symbols are held for noticably
longer, and more of them are being sent in parallel -- to the point
where "a ten billionth of a second" would be something like 5000 times
more precise on the timing than would actually be needed.
--
Live it up, rip it up, why so lazy?
Give it out, dish it out, let's go crazy, yeah!
-- Supertramp (The USENET Song)

Archived from groups: alt.internet.wireless (More info?)

On Fri, 27 Aug 2004 11:02:33 -0700, Jeff Liebermann
<jeffl@comix.santa-cruz.ca.us> wrote:

>This looks close, but is probably not the same thing:
> http://www.e-beam.com/products/
>Ugh... Expensive.

Yeah, this seems to be overkill for me.

>Anything that works over ethernet will also work over wireless. So,
>if your chosen internet white board software will play on an ethernet
>LAN, it will play via wireless.

Yeah, I didn't realize this at first. I just need to be able to
refrain from having to have an internet connection for the setup to
work. The chat apps need this.

Jabber might do the trick -- if I could run a jabber server on my host
PC.
http://www.jabber.org/
I'm checking into this more now -- and trying to expand my search to
the more general LAN collaborative apps.

Thanks,
Greg



------ Greg Conquest ------
http://gregconquest.com