two AP's with same SSID for coverage

Having seen several threads about this type of WiFi coverage issue,
I was wondering about having 2 AP's at ends of the house
to provide better coverage....

If we have 2 AP's with the same SSID,
both connected to the same IP segment,
and the DHCP coming from one router,
then what are the issues involved ??

Not really a roaming issue, but it can be...

SO - we have one AP upstairs in the middle of the upstairs bedroom area,
and another down in the family room area.
Will you "associate" with the strongest signal ?
Is that always true ? In the middle, can you ping-pong between them ?

For the most part - you are either upstairs or downstairs.
What happens if you are downstairs, and then walk up to your room
to do homework - will you re-associate with the closer & stronger AP ?

Just wondering how this could/would/should work with multiple AP's
all on the same channel & SSID ?

--
----------------------------------
"If everything seems to be going well,
you have obviously overlooked something." - Steven Wright

"ps56k" <(E-Mail Removed)> wrote in message
news:qF9el.9362$(E-Mail Removed)...
> Having seen several threads about this type of WiFi coverage issue,
> I was wondering about having 2 AP's at ends of the house
> to provide better coverage....
>
> If we have 2 AP's with the same SSID,
> both connected to the same IP segment,
> and the DHCP coming from one router,
> then what are the issues involved ??
>
> Not really a roaming issue, but it can be...
>
> SO - we have one AP upstairs in the middle of the upstairs bedroom area,
> and another down in the family room area.
> Will you "associate" with the strongest signal ?
> Is that always true ? In the middle, can you ping-pong between them ?
>
> For the most part - you are either upstairs or downstairs.
> What happens if you are downstairs, and then walk up to your room
> to do homework - will you re-associate with the closer & stronger AP ?
>
> Just wondering how this could/would/should work with multiple AP's
> all on the same channel & SSID ?
>
> --

The multiple APs should be set to DIFFERENT channels and the same SSID.
This is the standard configuration for multiple APs and wireless roaming.
It generally works well, though is best with identical APs and the same
protocol (e.g. both 11g). Yes, you will re-associate with stronger signal
AP. You will not generally "ping pong" between them is signal strengths
are similar. It's generally no problem to move from one coverage area to
the other though if you have an active download (for example) going on it
could be interrupted by the momentary delay in the changeover.

"Peter Pan" <(E-Mail Removed)> wrote in message
news:srydnQ-(E-Mail Removed)...
> Jeff Liebermann wrote:
>>
>> Huh? Is this one router or two, each with it's own SSID? I can see
>> how it would work with two routers, but not one.
>>
>
> actually 3, but i assume you only care about the first b/g and the third n
> only... Tivo's and one of the PDA's and tablets will only do G, and one
> pda and smartphone b only, so I tend to leave old working stuff working,
> and just add new as I aquire it
>
> oddly enuf, i've been getting them donated/collecting old ones (if nothing
> else, now I have 7 I can play with/brick, and many spare/backup walwarts,
> i'm retired, so now time to putter and fix instead of destroying , and
> just change the starting ip address for both wired and wireless, so they
> don't conflict, leaving the dhcp servers on etc, and plugging the router
> parts/port to each other to daisy chain em

I thought there were perf probs when letting a B signal exist
within the same range as a G - and therefore probably N ?

In message <(E-Mail Removed)> "Peter Pan"
<(E-Mail Removed)> was claimed to have wrote:

>ps56k wrote:
>> Having seen several threads about this type of WiFi coverage issue,
>> I was wondering about having 2 AP's at ends of the house
>> to provide better coverage....
>>
>> If we have 2 AP's with the same SSID,
>> both connected to the same IP segment,
>> and the DHCP coming from one router,
>> then what are the issues involved ??
>>
>> Not really a roaming issue, but it can be...
>>
>> SO - we have one AP upstairs in the middle of the upstairs bedroom
>> area, and another down in the family room area.
>> Will you "associate" with the strongest signal ?
>> Is that always true ? In the middle, can you ping-pong between them ?
>>
>> For the most part - you are either upstairs or downstairs.
>> What happens if you are downstairs, and then walk up to your room
>> to do homework - will you re-associate with the closer & stronger AP ?
>>
>> Just wondering how this could/would/should work with multiple AP's
>> all on the same channel & SSID ?
>
>why the same?

Specifically to allow roaming.

>To eliminate the prob I have different SSID's (don't roam,
>just associate with whichever gets the best signal, PPinUSA and PPinUSB)...
>As to your q, depends on your client software, no standards if it hangs on
>to the last one, or trys to reassociate with a stronger one).... Actually
>have a 3rd now (PPinUSN, mimo/pre-n, so devices that can do it's n can see
>that too)

That works, but it ensures that clients will see a disconnect when
jumping from one to another, as well as requires additional
configuration upfront. Maybe that's a problem, maybe not.

In message <(E-Mail Removed)> Jeff Liebermann
<(E-Mail Removed)> was claimed to have wrote:

>On Sat, 24 Jan 2009 20:38:56 -0500, "Peter Pan"
><(E-Mail Removed)> wrote:
>
>>Jeff Liebermann wrote:
>>> Incidentally, even if you decide to use two 40MHz channels (wideband
>>> mode), all it takes is one local 802.11g radio, and the system
>>> switches to CTS (flow control) mode to prevent collisions and
>>> interference. The only way it works fast is if everything is using
>>> 802.11n. There are some comments on this and other issues at:
>>> <http://en.wikipedia.org/wiki/802.11n>
>
>>or plan B.... (using different ssid's one for normal and one for High Speed)
>
>Won't work. As soon as the 802.11n router hears *ANY* 802.11g signal,
>regardless of SSID, it will switch on flow control to prevent
>collisions. In the 40Mhz mode, it can't decode the preamble or
>beacons from an 802.11b/g AP or client. So, it has to treat
>everything as a potential co-channel user and give them some airtime.

So move to 5.8GHz. There are so few 802.11a networks out there that
802.11n 5.8GHz is, at least for now, the holy grail of wireless
performance.

DevilsPGD <(E-Mail Removed)> wrote in
news:(E-Mail Removed):

>>Won't work. As soon as the 802.11n router hears *ANY* 802.11g signal,
>>regardless of SSID, it will switch on flow control to prevent
>>collisions. In the 40Mhz mode, it can't decode the preamble or
>>beacons from an 802.11b/g AP or client. So, it has to treat
>>everything as a potential co-channel user and give them some airtime.
>
> So move to 5.8GHz. There are so few 802.11a networks out there that
> 802.11n 5.8GHz is, at least for now, the holy grail of wireless
> performance.

You'd think....but....

Last fall while doing some field site surveys for 5.8Ghz gear in Buffalo,
NY, in two different locations/nodes/subnets (whatever you want to call it,
several different PtP links within city limits), both locations tested
showed 3 or 4 different AP's already operating on 5.8Ghz. Only one was a
duplicate from one location to the other. Oddly enough, these same
locations showed the same number of AP's operating in 2.4Ghz also. I would
have expected a dozen.

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

> Most of the time, I'm mildly drugged, which seems to have an effect on
> my judgement and memory. I've reduced the frequency of my postings,
> but apparently that's not sufficient to insure accuracy. I think I'll
> abstain from any more postings that require much thinking until I'm
> off the pain killers.

Reminds me of a scene in "Taxi" where Christopher Lloyd's character
(Ignatzki?) has to renew his driver's permit, and the clerk querying his
qualifications asks "Drug addiction or mental illness?" and he replies
"Hmm, that's a difficult choice."

Get better soon, J.L. Anyway, even stoned you make more sense than most
people on Usenet.
--
W. Oates

Jeff Liebermann:

> On Sat, 24 Jan 2009 11:47:42 +0000, LR <(E-Mail Removed)> wrote:

<snip>

>>Could you clarify the "use 2 40MHz channels" as I was under the
>>impression that 802.11n could use 2 20MHz channels to produce one
>>40MHz channel and that they haven't as yet tried to bond 2 40MHz
>>"channels".
>
> Sorry. You're right. It should be "use two 20MHz channels" which
> are bonded to produce one 40MHz channel.

And you mean operating channels. The "channels" (1, 2, ... 11 for the
US) are 5 MHz wide.


--
"Well, opinions are like assholes. Everybody has one." Dirty Harry