Unexplained Download Speed versus Upload Speed - Any thoughts Appreciated

Merry Christmas and thanks in advance for any responses. My parents
have constantly struggled with poor wireless access in their living
room, driven in part by the fact that the wireless modem is in their
study (opposite side of the house). The stats are as follows:

Direct connection to wireless access point: 17mpbs down (expected),
3up (expected)
In living room: 0.8mpbs down, 5up (unexpected)

The 0.8mpbs down is expected and the main problem. I have no idea how
we can be getting the 5up (tested using several sites).

Two questions:

1) Any idea why there might be the anomaly in the living room?

2) Does anyone know a good reference piece for setting up a second
wireless access point (we'd hard wire it in the living
room) . . . .the key being that we want seamless transitioning between
the two access points, which I'm told is harder to achieve. Just to
explain further through an example, if I were downloading a large file
on my computer in the living room and walked back to the study with my
computer, I'd want the computer to be able to seamlessly switch from
one WAP to another without any hickup (the file would keep
downloading). We have this setup in our corporate headquarters but I
don't know how difficult it is to accomplish in more of a home
setting.

Again, any thoughts would be much appreciated.

crbutler01@yahoo.com (for it is he) wrote:

> Direct connection to wireless access point: 17mpbs down (expected),
> 3up (expected)
> In living room: 0.8mpbs down, 5up (unexpected)
>
> The 0.8mpbs down is expected and the main problem. I have no idea how
> we can be getting the 5up (tested using several sites).

Test locally with iperf to a PC on the wired LAN instead.

--
<http://ale.cx/> (AIM:troffasky) (UnSoEsNpEaTm@ale.cx)
22:47:08 up 35 days, 5:48, 6 users, load average: 0.07, 0.05, 0.05
"People believe any quote they read on the internet
if it fits their preconceived notions." - Martin Luther King

crbutler01@yahoo.com wrote:
> Direct connection to wireless access point: 17mpbs down (expected),
> 3up (expected)
> In living room: 0.8mpbs down, 5up (unexpected)

You need to check your units. I'm going to assume you mean 17 Mb/s and 3
Mb/s for the direct connection. But what units are you using to provide
the 0.8 and 5? (M is not m, and B is not b, so this is really important).


> 1) Any idea why there might be the anomaly in the living room?

It's possible that there isn't an anomaly but that you are misreading
(or misunderstanding?) the results. Or that they are being reporting in
the wrong units. BICBW.


> I'd want the computer to be able to seamlessly switch from
> one WAP to another without any hickup (the file would keep
> downloading). We have this setup in our corporate headquarters but I
> don't know how difficult it is to accomplish in more of a home
> setting.

Either run a cable between the two APs and have them broadcast on
different channels, or else run the cable only to the first and configure
them as master/slave on the *same* channel (this will halve the wireless
throughput so is not an ideal solution). In both cases they must have
the same SSID and the same WPA/WEP key, and must not be "hidden".

Chris

On Sat, 24 Dec 2011 07:40:30 -0800 (PST), crbutler01@yahoo.com wrote:

>Merry Christmas and thanks in advance for any responses. My parents
>have constantly struggled with poor wireless access in their living
>room, driven in part by the fact that the wireless modem is in their
>study (opposite side of the house).

How many walls are they going through between the unspecified model
wireless router and whatever style of computer they're using in the
living room. In general, you can go through one wall fairly easily.
Two walls are a crap shoot. Three or more are very difficult. If
there's aluminum foil backed insulation in the inside walls, forget
it.

>Direct connection to wireless access point: 17mpbs down (expected),
>3up (expected)

Ok, you have cable modem service. Thanks.

>In living room: 0.8mpbs down, 5up (unexpected)

That sucks.

>The 0.8mpbs down is expected and the main problem. I have no idea how
>we can be getting the 5up (tested using several sites).

The basics... Find the connection manager on your unspecified model
wireless client adapter and get the wireless connection (or
association) speed. With 802.11g, it will vary between 1Mbit/sec and
54Mbits/sec. The actual thruput is half or less the connection speed.
Note that the connection speed can be different in each direction.

When I see the upload greater than the download, I look for RF
interference at the client end. Wireless camera, TIVO, security
camera, cordless phone, etc?

>1) Any idea why there might be the anomaly in the living room?

It might be haunted. Please consult an exorcist or building
contractor for remedial actions.

>2) Does anyone know a good reference piece for setting up a second
>wireless access point (we'd hard wire it in the living
>room)

It's easy. Just find either a wireless access point or ANY wireless
router. Run a CAT5 cable between your existing unspecified model
wireless router and the 2nd wireless box, from LAN port to LAN port.
On some ancient models, you might need to wire one end of the cable as
a cross-over cable.

The setup on the 2nd wireless box is detailed here:
<http://wireless.navas.us/index.php?title=Wi-Fi_How_To#Use_a_wireless_router_as_a_wireless_acce ss_point>
You'll also find the same thing on the manufacturers support web pile.

There are some decisions to make. You can use the same SSID for both
wireless access points, and hope that you'll get seamless roaming. The
reality is that you probably won't unless the access points are
identical, and support 802.11r:
<http://en.wikipedia.org/wiki/IEEE_802.11r-2008>
With a random mix of access points, it probably won't work. Note that
this is fast BSS switching, not seamless roaming. You'll loose a few
bits during the disconnect and reconnect cycle, but it won't loose the
connection.

The client software must also be compatible for roaming to work. Many
clients will tenaciously stick to the initial access point connection
and not switch to a stronger signal, unless you first disconnect. Even
worse, some will remember where they connected last, and will
reconnect even if it's the weaker signal.

> . . . .the key being that we want seamless transitioning between
>the two access points, which I'm told is harder to achieve.

It can be done. Have your checkbook ready.

Instead, I suggest you give up now while you're still sane. Pick two
different SSID's and two different non-overlapping channels (1, 6, 11)
for your two devices. If you want to switch location, disconnect, and
reconnect to the desired SSID. If your parents have a difficult time
with this, you might be able to automate it with a script.

>Just to
>explain further through an example, if I were downloading a large file
>on my computer in the living room and walked back to the study with my
>computer, I'd want the computer to be able to seamlessly switch from
>one WAP to another without any hickup (the file would keep
>downloading). We have this setup in our corporate headquarters but I
>don't know how difficult it is to accomplish in more of a home
>setting.

Yeah, that's a worthy goal. One should always be downloading
something while walking around the house. However, I don't know of
any consumer grade hardware that will do 802.11r. I can probably find
something if you give me a clue as what you currently have to work
with and how roughly how much money you want to spend in order to
waltz around the house while downloading.


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

There shouldn't be foil backed insulation or wallboard on interior
walls. Generally the radiant heat barrier faces an exterior wall. I
suppose if you had multi-zone heating and wanted to overkill everything,
you could put radiant heat barriers on the inside, but I bet that isn't
common. Most construction doesn't insulate interior walls unless they
are done to "party wall" standards (two sets of studs,etc).

Note when you build a SCIF, they do have that foil backing on one layer
of the drywall. ;-) A friend of a friend bought a house that turned out
to be a former CIA "safe house". Lots of odd tweaks in it.

On Sat, 24 Dec 2011 16:51:53 -0800, miso <miso@sushi.com> wrote:

>There shouldn't be foil backed insulation or wallboard on interior
>walls.

Several of my customers have that. It's called an interior firewall
designed to prevent a fire from spreading across the barrier. It's
required when a house shares a wall with a garage. For interior
walls, it's optional. I've also seen it in apartment buildings
between adjacent apartments. It's also considered an energy saver as
you can heat parts of a house, while letting the rest freeze. Works
nicely, but is hell for wi-fi penetration.

>Generally the radiant heat barrier faces an exterior wall. I
>suppose if you had multi-zone heating and wanted to overkill everything,

These are all high end custom houses. I wouldn't expect it for cookie
cutter houses.

>you could put radiant heat barriers on the inside, but I bet that isn't
>common. Most construction doesn't insulate interior walls unless they
>are done to "party wall" standards (two sets of studs,etc).

Standard 2x4 interior studs. R13 insulation as I vaguely recall.
Close the connecting door and you can't hear anything from the other
size.

>Note when you build a SCIF, they do have that foil backing on one layer
>of the drywall. ;-) A friend of a friend bought a house that turned out
>to be a former CIA "safe house". Lots of odd tweaks in it.

Tempest qualified house? Nice. Goes well with the foil backed
wallpaper.
<http://www.theregister.co.uk/2004/08/12/wifi_wallpaper/>
Ask your friend to search for the hidden microphones.

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

crbutler01@yahoo.com wrote:
> Merry Christmas and thanks in advance for any responses. My parents
> have constantly struggled with poor wireless access in their living
> room, driven in part by the fact that the wireless modem is in their
> study (opposite side of the house). The stats are as follows:
>
> Direct connection to wireless access point: 17mpbs down (expected),
> 3up (expected)
> In living room: 0.8mpbs down, 5up (unexpected)
>
> The 0.8mpbs down is expected and the main problem. I have no idea how
> we can be getting the 5up (tested using several sites).
>
> Two questions:
>
> 1) Any idea why there might be the anomaly in the living room?
>
> 2) Does anyone know a good reference piece for setting up a second
> wireless access point (we'd hard wire it in the living
> room) . . . .the key being that we want seamless transitioning between
> the two access points, which I'm told is harder to achieve. Just to
> explain further through an example, if I were downloading a large file
> on my computer in the living room and walked back to the study with my
> computer, I'd want the computer to be able to seamlessly switch from
> one WAP to another without any hickup (the file would keep
> downloading). We have this setup in our corporate headquarters but I
> don't know how difficult it is to accomplish in more of a home
> setting.
>
> Again, any thoughts would be much appreciated.
Hmmm,
No mention of router model or which band,channel, mode being used.
Signal can be affected by moving router little bit in any direction.

On 12/24/2011 6:04 PM, Jeff Liebermann wrote:
> On Sat, 24 Dec 2011 16:51:53 -0800, miso<miso@sushi.com> wrote:
>
>> There shouldn't be foil backed insulation or wallboard on interior
>> walls.
>
> Several of my customers have that. It's called an interior firewall
> designed to prevent a fire from spreading across the barrier. It's
> required when a house shares a wall with a garage. For interior
> walls, it's optional. I've also seen it in apartment buildings
> between adjacent apartments. It's also considered an energy saver as
> you can heat parts of a house, while letting the rest freeze. Works
> nicely, but is hell for wi-fi penetration.
>
>> Generally the radiant heat barrier faces an exterior wall. I
>> suppose if you had multi-zone heating and wanted to overkill everything,
>
> These are all high end custom houses. I wouldn't expect it for cookie
> cutter houses.
>
>> you could put radiant heat barriers on the inside, but I bet that isn't
>> common. Most construction doesn't insulate interior walls unless they
>> are done to "party wall" standards (two sets of studs,etc).
>
> Standard 2x4 interior studs. R13 insulation as I vaguely recall.
> Close the connecting door and you can't hear anything from the other
> size.
>
>> Note when you build a SCIF, they do have that foil backing on one layer
>> of the drywall. ;-) A friend of a friend bought a house that turned out
>> to be a former CIA "safe house". Lots of odd tweaks in it.
>
> Tempest qualified house? Nice. Goes well with the foil backed
> wallpaper.
> <http://www.theregister.co.uk/2004/08/12/wifi_wallpaper/>
> Ask your friend to search for the hidden microphones.
>

Yes, I could see this for fire protection. Also at some point in
construction, people will use what is handy. Certainly you can use foil
backed drywall where it isn't required. Or they just buy one type of
drywall to cover multiple applications. Less line items to screw up,
plus it may not make much difference when you consider labor.

You can troll fbo.gov to get SCIF construction guidelines.

I haven't been there in years, but UPS in Sunnyvale would rent space in
a nearby building during Christmas to handle the overflow from
customers. The building still had all the spook notices. Things like "no
classified conversations by this door" and such nonsense. Weird Stuff
back in the day would often be peddling old TEMPEST qualified computer
items like printers.

On Tue, 27 Dec 2011 18:25:42 -0800, miso <miso@sushi.com> wrote:

>> Tempest qualified house? Nice. Goes well with the foil backed
>> wallpaper.
>> <http://www.theregister.co.uk/2004/08/12/wifi_wallpaper/>
>> Ask your friend to search for the hidden microphones.

>Yes, I could see this for fire protection.

For those interested in turning their house into a RF shield room:
<http://www.usg.com/sheetrock-gypsum-panels-foil-backed.html>
<http://www.askthebuilder.com/601_Reflective_Foil_Insulation.shtml>
You can also get foil backed wallpaper, foil backed fiber board, and
doors laminated with foil between layers.

There's also titanium nitride low-E window coatings, that are quite
good at blocking cell phone and wi-fi signals.

Of course, we can't leave the roof unshielded:
<http://www.atticfoil.com/radiant-barrier-instruction.htm>

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

On 12/27/2011 7:40 PM, Jeff Liebermann wrote:
> On Tue, 27 Dec 2011 18:25:42 -0800, miso<miso@sushi.com> wrote:
>
>>> Tempest qualified house? Nice. Goes well with the foil backed
>>> wallpaper.
>>> <http://www.theregister.co.uk/2004/08/12/wifi_wallpaper/>
>>> Ask your friend to search for the hidden microphones.
>
>> Yes, I could see this for fire protection.
>
> For those interested in turning their house into a RF shield room:
> <http://www.usg.com/sheetrock-gypsum-panels-foil-backed.html>
> <http://www.askthebuilder.com/601_Reflective_Foil_Insulation.shtml>
> You can also get foil backed wallpaper, foil backed fiber board, and
> doors laminated with foil between layers.
>
> There's also titanium nitride low-E window coatings, that are quite
> good at blocking cell phone and wi-fi signals.
>
> Of course, we can't leave the roof unshielded:
> <http://www.atticfoil.com/radiant-barrier-instruction.htm>
>

I suggest vibrating the glass with white noise.

There is a bit of controversy with the attic foil, mostly because it
makes the roof hotter. But some states mandated it IIRC. Foil against a
conditioned surface makes sense.

For someone in need of a refresher, there are three types of heat
transfer: conduction, convection, and radiation. Those "emergency"
covers you see in camping stores are to reduce radiation, but they don't
insulate. I only know one person that attempted to use one instead of a
sleeping bag and he said he froze all night. Anyway, the radiant heat
barrier can also cut down on convection it sealed.

I'm trying to recall the court decision, but there was a case where the
cops tried to use thermal imaging to see if someone is in the house. Now
this is like seeing through walls, so it went to court to determine the
legality. I think foil would nix thermal imaging. Or the cop would see
his own reflection.

On Thu, 29 Dec 2011 14:50:58 -0800, miso <miso@sushi.com> wrote:

>I'm trying to recall the court decision, but there was a case where the
>cops tried to use thermal imaging to see if someone is in the house. Now
>this is like seeing through walls, so it went to court to determine the
>legality. I think foil would nix thermal imaging. Or the cop would see
>his own reflection.

Think isothermal contour. They'd see bugger-all.