MIMO wireless - stability

[Rob Nicholson]

We use wireless at market research meetings where we use wireless between
the viewing room and the interview room through a one way mirror. The
stability of the wireless connection is cause for concern - it occasionally
drops out. I appreciate that this often happens with 802.11g wireless - it
happens at the office and it happens here at home.

802.11n (MIMO) is touted as having increased range, increased speed and
increased stability. The first two are obvious to demonstrate but what about
increased stability?

I suspect that the problem we get is a combination of sending the signal
through a silver mesh (the one way mirror) plus other surrounding wireless
networks causing the signal to bounce all over the place. 802.11n is
supposed to work with this effect instead of fighting against it so it could
be better.

Comments?

Thanks, Rob.

[Jeff Liebermann]

On Tue, 26 Feb 2008 21:24:12 GMT, "Rob Nicholson"
<nicholson.rob@spamme.com> wrote:

>We use wireless at market research meetings where we use wireless between
>the viewing room and the interview room through a one way mirror. The
>stability of the wireless connection is cause for concern - it occasionally
>drops out. I appreciate that this often happens with 802.11g wireless - it
>happens at the office and it happens here at home.

It may happen anywhere on the planet. It's usually caused by
interference from everything from microwave ovens to radio controlled
airplanes. See a list of possible culprits at:
<http://wireless.wikia.com/wiki/Wi-Fi#Interference>

>802.11n (MIMO) is touted as having increased range, increased speed and
>increased stability. The first two are obvious to demonstrate but what about
>increased stability?

Where did you get the idea that you'll get increased range with MIMO?
Basically, you can trade range for speed. Double the range, and you
have to slow down to 1/4th the speed. Similarly, cut the range in
half, and you can go 4 times as fast. For long range, you'll never
get faster than 802.11g speeds. MIMO is about improved resistance to
reflections and multipath. 802.11g sorta tolerates multipath by being
tolerant to frequency selective fading. MIMO uses the multipath to
create additional data streams (spacial multiplexing), which will
increase the speed, but not the range.

>I suspect that the problem we get is a combination of sending the signal
>through a silver mesh (the one way mirror)

I've never tested a one-way mirror for RF attenuation. My guess is
that it's about the same as aluminized reflective window coatings
commonly used in office buildings for energy efficiency. These block
signals like a brick wall. You're lucky that any signal goes through.
<http://wireless.wikia.com/wiki/Wi-Fi#Attenuation>

>plus other surrounding wireless
>networks causing the signal to bounce all over the place.

The surrounding wireless networks cause interference. If this is the
case, move the access point to a less polluted channel (1, 6, or 11),
move the access point away from windows and doors so that it doesn't
pickup as much junk, or elect public officials that don't support
municipal wireless networks running high power (1 watt) radios.

>802.11n is
>supposed to work with this effect instead of fighting against it so it could
>be better.

Pre 802.11n Draft 2 is suppose to be faster. It's also somewhat more
resistant to reflections. It slows down to 802.11g modes in the
presence of interference and is therefore no better than 802.11g. I
guess you can call that more reliable.

>Comments?

Keep it simple. Play with antenna patterns. Change channels or bands
(802.11a is less polluted for now).

Also, don't elect officials that promote "free" wireless networks that
neither work or pay for themselves.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558 jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com jeffl@cruzio.com
# http://www.LearnByDestroying.com AE6KS

[Rob Nicholson]

> Where did you get the idea that you'll get increased range with MIMO?

Err, articles like this:

http://www.theregister.co.uk/2005/06/07/review_netgear_mimo_router/page2.html
http://en.wikipedia.org/wiki/Multiple-input_multiple-output

> get faster than 802.11g speeds. MIMO is about improved resistance to
> reflections and multipath. 802.11g sorta tolerates multipath by being
> tolerant to frequency selective fading. MIMO uses the multipath to
> create additional data streams (spacial multiplexing), which will
> increase the speed, but not the range.

Range isn't actually important for this requirement - the receiving
equipment is 5m away from the transmitter.

> I've never tested a one-way mirror for RF attenuation. My guess is
> that it's about the same as aluminized reflective window coatings
> commonly used in office buildings for energy efficiency. These block
> signals like a brick wall. You're lucky that any signal goes through.
> <http://wireless.wikia.com/wiki/Wi-Fi#Attenuation>

A one-way mirror isn't solid - it's dots of silver with gaps inbetween and
only works because the viewing room is kept dark whereas the other room is
bright light so the majority of the light reflects back. Turn the lights off
in the meeting room and on in the viewing room and the one-way mirror
reverses. I always wondered how they worked :-)

So that's probably how the signal is getting through but it's probably
causing a lot of degredation. I'll get them to check the signal strength
next time.

> The surrounding wireless networks cause interference. If this is the
> case, move the access point to a less polluted channel (1, 6, or 11),
> move the access point away from windows and doors so that it doesn't
> pickup as much junk, or elect public officials that don't support
> municipal wireless networks running high power (1 watt) radios.

These meetings are often staffed by non-technical people so we have to keep
it as simple as possible. Yes, it's a challenge :-)

> Keep it simple. Play with antenna patterns. Change channels or bands
> (802.11a is less polluted for now).

Another alternative is a video sender - the PC in the viewing room is
currently remote controlling the one in the meeting room. Instead, we're
considering a video sender using the S-Video output of the laptop and the
input of the data projector. This would have similar problems (same
frequencies I think) but I've got one at home and it's rock solid. We've got
a concern over the video quality but most of the time it's big PowerPoint
slides on the screen.

Cheers, Rob.

[Jeff Liebermann]

On Tue, 26 Feb 2008 23:46:40 GMT, "Rob Nicholson"
<nicholson.rob@spamme.com> wrote:

>> Where did you get the idea that you'll get increased range with MIMO?
>
>Err, articles like this:
>
>http://www.theregister.co.uk/2005/06/07/review_netgear_mimo_router/page2.html

OK, I'll bite. It says:
"We were impressed with MIMO's range capabilities and
found it delivered exceptional coverage in a normal
home environment."

The big question is what they mean by "coverage". Also whether they
were testing in the various MIMO modes, or were falling back to the
802.11b/g modes when testing the coverage. I'll leave it an open
question as I fail to see any substantiating numbers. They do
indicate that the MIMO router is more "reliable" which they interpret
as experiencing fewer dropouts. My guess is that they were testing in
a highly reflective indoor environment, where MIMO does do better.

>http://en.wikipedia.org/wiki/Multiple-input_multiple-output

Well, there you have me. The article mentions:
"MIMO technology has attracted attention in wireless
communications, since it offers significant increases
in data throughput and link range without additional
bandwidth or transmit power."

However, searching the article, I find no substantiation of this claim
and no furthur mention of the buzzwords "range", "coverage", or
"distance". In short, it's another recycled claim devoid of any
substantiating numbers.

>Range isn't actually important for this requirement - the receiving
>equipment is 5m away from the transmitter.

Then, you should be getting "MIMO performance", which is a good thing.
The major benifit will be increased speed and a much more reliable
connection with fewer disconnects due to reflections and multipath.

>> I've never tested a one-way mirror for RF attenuation. My guess is
>> that it's about the same as aluminized reflective window coatings
>> commonly used in office buildings for energy efficiency. These block
>> signals like a brick wall. You're lucky that any signal goes through.
>> <http://wireless.wikia.com/wiki/Wi-Fi#Attenuation>
>
>A one-way mirror isn't solid - it's dots of silver with gaps inbetween and
>only works because the viewing room is kept dark whereas the other room is
>bright light so the majority of the light reflects back. Turn the lights off
>in the meeting room and on in the viewing room and the one-way mirror
>reverses. I always wondered how they worked :-)

<http://en.wikipedia.org/wiki/One_way_mirror>
Well, it doesn't say how the mirror is constructed but other
references mention dots and "smears". If they're electrically
connected to each other (they would have to be if the coating was
electrically deposited), then it would probably block RF. My
uninformed guess is that it wouldn't be as bad as a solid wall, but
would still block some RF. I could run a test, but I don't have a
one-way mirror handy.

>So that's probably how the signal is getting through but it's probably
>causing a lot of degredation. I'll get them to check the signal strength
>next time.

Check the signal strength for an equal distance but without the
mirror. Then test with the mirror over roughly the same distance. The
signal strength difference is the mirror attenuation. You may need to
make a series of measurements as antenna positions and room
differences will have a substantial effect.

>> The surrounding wireless networks cause interference. If this is the
>> case, move the access point to a less polluted channel (1, 6, or 11),
>> move the access point away from windows and doors so that it doesn't
>> pickup as much junk, or elect public officials that don't support
>> municipal wireless networks running high power (1 watt) radios.

>These meetings are often staffed by non-technical people so we have to keep
>it as simple as possible. Yes, it's a challenge :-)

Explaining microwave propogation and antenna characteristics isn't too
horribly difficult. If they are totally clueless, an optical analogy
works well enough. Anything light can go through, so can RF. Water
filled balloon for antenna pattern demos.

>> Keep it simple. Play with antenna patterns. Change channels or bands
>> (802.11a is less polluted for now).

>Another alternative is a video sender - the PC in the viewing room is
>currently remote controlling the one in the meeting room. Instead, we're
>considering a video sender using the S-Video output of the laptop and the
>input of the data projector. This would have similar problems (same
>frequencies I think) but I've got one at home and it's rock solid. We've got
>a concern over the video quality but most of the time it's big PowerPoint
>slides on the screen.

I just bought a cheapo 2.4Ghz wireless TV contraption. It rips on my
802.11g thruput and my access point mangles the video. It's anything
but rock solid fading in and out as people walk around between the
camera and the receiver. I guess they vary in quality.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558 jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com jeffl@cruzio.com
# http://www.LearnByDestroying.com AE6KS

[stephen]

"Rob Nicholson" <nicholson.rob@spamme.com> wrote in message
news:Ax1xj.5770$ab5.4820@newsfe1-win.ntli.net...
> > Where did you get the idea that you'll get increased range with MIMO?
>
> Err, articles like this:
>
>
http://www.theregister.co.uk/2005/06/07/review_netgear_mimo_router/page2.html
> http://en.wikipedia.org/wiki/Multiple-input_multiple-output
>
> > get faster than 802.11g speeds. MIMO is about improved resistance to
> > reflections and multipath. 802.11g sorta tolerates multipath by being
> > tolerant to frequency selective fading. MIMO uses the multipath to
> > create additional data streams (spacial multiplexing), which will
> > increase the speed, but not the range.
>
> Range isn't actually important for this requirement - the receiving
> equipment is 5m away from the transmitter.
>
> > I've never tested a one-way mirror for RF attenuation. My guess is
> > that it's about the same as aluminized reflective window coatings
> > commonly used in office buildings for energy efficiency. These block
> > signals like a brick wall. You're lucky that any signal goes through.
> > <http://wireless.wikia.com/wiki/Wi-Fi#Attenuation>
>
> A one-way mirror isn't solid - it's dots of silver with gaps inbetween and
> only works because the viewing room is kept dark whereas the other room is
> bright light so the majority of the light reflects back. Turn the lights
off
> in the meeting room and on in the viewing room and the one-way mirror
> reverses. I always wondered how they worked :-)
>
> So that's probably how the signal is getting through but it's probably
> causing a lot of degredation. I'll get them to check the signal strength
> next time.

put the antenna on the other side of the mirror.

or use one of the powerline networking boxes with a built in AP to give you
a link between the 2 rooms - as long as they are on the same mains
distribution you should get a reasonable link.

or just run some Cat5 patching into the local flood wiring system and avoid
wireless completely?
>
> > The surrounding wireless networks cause interference. If this is the
> > case, move the access point to a less polluted channel (1, 6, or 11),
> > move the access point away from windows and doors so that it doesn't
> > pickup as much junk, or elect public officials that don't support
> > municipal wireless networks running high power (1 watt) radios.
>
> These meetings are often staffed by non-technical people so we have to
keep
> it as simple as possible. Yes, it's a challenge :-)

if you can afford glass with silver, then a bit of cable shouldnt hurt.

if security is an issue then why use wireless anyway?
>
> > Keep it simple. Play with antenna patterns. Change channels or bands
> > (802.11a is less polluted for now).
>
> Another alternative is a video sender - the PC in the viewing room is
> currently remote controlling the one in the meeting room. Instead, we're
> considering a video sender using the S-Video output of the laptop and the
> input of the data projector. This would have similar problems (same
> frequencies I think) but I've got one at home and it's rock solid. We've
got
> a concern over the video quality but most of the time it's big PowerPoint
> slides on the screen.
>
> Cheers, Rob.
>
>
--
Regards

stephen_hope@xyzworld.com - replace xyz with ntl