physical location of AP

Hi,

I got a new project on WIFI technology. I am facing a problem. Th
problem is when user is accessing my desktop application throug
internet. I have to identify which Access point he is using. In othe
words I want to find out the current physical location of the member
Is there any way to find out?

jite

--
jiten das
brought to you by http://www.wifi-forum.com

On Thu, 18 Nov 2004 12:38:12 GMT, jiten dash spoketh

>
>Hi,
>
>I got a new project on WIFI technology. I am facing a problem. The
>problem is when user is accessing my desktop application through
>internet. I have to identify which Access point he is using. In other
>words I want to find out the current physical location of the member.
>Is there any way to find out?
>
>jiten

No. It's just another IP connection to your computer...

However, knowing the IP address, you might be able to track it down on
your own network if you have managed switches and each WAP is connected
to a different port on the switch. You should then be able to track down
which MAC address are connected to which port, thus you'll know which
access point the connection came through.

Sadly, if you are talking about wireless hot-spot usage, ie your sales
manager is slacking off somewhere, and you want to know which Starbucks
he's at, then you'll have a hard time figuring that out... You'll have
to track down his public IP address, and that might still give you
nothing...

Lars M. Hansen
http://www.hansenonline.net
(replace 'badnews' with 'news' in e-mail address)

jiten dash <(E-Mail Removed)> wrote:
>I got a new project on WIFI technology. I am facing a problem. The
>problem is when user is accessing my desktop application through
>internet. I have to identify which Access point he is using. In other
>words I want to find out the current physical location of the member.

This homework assignment seems to come up a lot. There's really no
way to know unless you know the entire topology of the network.

On Thu, 18 Nov 2004 12:38:12 GMT, jiten dash
<(E-Mail Removed)> wrote:

>I got a new project on WIFI technology. I am facing a problem.

Yawn. I posted an answer to your previous variation on the same
question. See:
http://www.google.com/groups?selm=p3...6bqq%404ax.com
Asking the same question 7 times is not very good form.


--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

On Thu, 18 Nov 2004 09:44:02 -0500, William P.N. Smith wrote:

>jiten dash <(E-Mail Removed)> wrote:
>>I got a new project on WIFI technology. I am facing a problem. The
>>problem is when user is accessing my desktop application through
>>internet. I have to identify which Access point he is using. In other
>>words I want to find out the current physical location of the member.

>This homework assignment seems to come up a lot. There's really no
>way to know unless you know the entire topology of the network.

He also doesn't have a very clear description or understanding of the
problem.

My guess(tm) is that he wants to play TDOA (time difference of
arrival) direction finder on the internet. This has been done before
and is actually quite easy if you make some rather marginal
assumptions. If you assume that:
1. A ping return comes back via the same route that it is sent.
2. Latency for a given ICMP echo responder is constant.
3. Latency on backbones is fairly constant over short periods of
time.
4. The router doesn't delay ICMP or UDP echos using QoS.
5. You have a ping with better resolution than the usual 1 msec.
6. Whatever else I forgot.

What's necessary is to ping the IP address from a variety of distant
locations. I watched it work at a local university to locate roaming
hackers and student illegally deployed servers. By mapping the ping
times from various nearby known locations, the unknown responder can
be found by comparing the patterns. I was amazed to see a solution
converge in less than a second. It's not foolproof and requires some
tweaking and guesswork, but it does work.

As for identifying the access point in use, that can't be done with
simple tools from the internet (WAN) side. The local router/firewall
will block most probes aimed directly at the AP (which is on the LAN
side of the router). If there's any information to be found on the
WAN side, it must be sent (initiated) by the client or access point
and not polled from the internet (WAN) side.


--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

Jeff Liebermann <(E-Mail Removed)> wrote:

> What's necessary is to ping the IP address from a variety of distant
> locations. I watched it work at a local university to locate roaming
> hackers and student illegally deployed servers. By mapping the ping
> times from various nearby known locations, the unknown responder can
> be found by comparing the patterns. I was amazed to see a solution
> converge in less than a second. It's not foolproof and requires some
> tweaking and guesswork, but it does work.
>
> As for identifying the access point in use, that can't be done with
> simple tools from the internet (WAN) side. The local router/firewall
> will block most probes aimed directly at the AP (which is on the LAN
> side of the router). If there's any information to be found on the
> WAN side, it must be sent (initiated) by the client or access point
> and not polled from the internet (WAN) side.

And if the user's router does not respond to pings, wouldn't that also
defeat the triangulation technique described in the first paragraph
above?

On Thu, 18 Nov 2004 23:39:06 GMT, (E-Mail Removed) (Neill
Massello) wrote:

>Jeff Liebermann <(E-Mail Removed)> wrote:
>
>> What's necessary is to ping the IP address from a variety of distant
>> locations. I watched it work at a local university to locate roaming
>> hackers and student illegally deployed servers. By mapping the ping
>> times from various nearby known locations, the unknown responder can
>> be found by comparing the patterns. I was amazed to see a solution
>> converge in less than a second. It's not foolproof and requires some
>> tweaking and guesswork, but it does work.
>>
>> As for identifying the access point in use, that can't be done with
>> simple tools from the internet (WAN) side. The local router/firewall
>> will block most probes aimed directly at the AP (which is on the LAN
>> side of the router). If there's any information to be found on the
>> WAN side, it must be sent (initiated) by the client or access point
>> and not polled from the internet (WAN) side.

>And if the user's router does not respond to pings, wouldn't that also
>defeat the triangulation technique described in the first paragraph
>above?

Correct. It certainly would. Some possible workarounds.

1. Use UDP ping (udps) instead of ICMP ping.
2. Use arping instead of ping.
http://www.habets.pp.se/synscan/prog...hp?prog=arping
Arping doesn't work through routers. It works nicely on University
and ISP networks that are not broken up by routers.
3. ping the nearest gateway router. It won't locate the user, but
will certainly locate how they connect to the internet.
4. For managed routers, use SNMP ping. Even if you don't have the
SNMP community name, the NACK from a failed open can be timed.
5. A few more I don't wanna talk about.

I've found that most wireless hot spot routers do respond to pings as
whomever is managing the hot spot wants to know if it's up or down.


--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831.336.2558 voice http://www.LearnByDestroying.com
# (E-Mail Removed)
# 831.421.6491 digital_pager (E-Mail Removed) AE6KS