Parabolic dish with usb wireless

hi,

>From what i've read about usb wireless on
http://www.usbwifi.orcon.net.nz/, it looks like you stick a usb
wireless stick in the focal point of a reflective parabolic dish. Is
that with no fiddling with the usb adapters built in antenna? Because
this might be the road that i go down rather than searching for
pigtails for various different cards and being ripped off.
Thanks in advance.

On 12 Dec 2004 14:34:33 -0800, (E-Mail Removed) wrote:

>>From what i've read about usb wireless on
>http://www.usbwifi.orcon.net.nz/, it looks like you stick a usb
>wireless stick in the focal point of a reflective parabolic dish. Is
>that with no fiddling with the usb adapters built in antenna? Because
>this might be the road that i go down rather than searching for
>pigtails for various different cards and being ripped off.

Fiddling is done with a fiddle or violin. What manner of butchery
were you trying to avoid? As for getting ripped off, wanna try to
build your own from parts? You won't like the connector prices.

The basic problem with shoving a USB dongle at the focus of a dish or
corner reflector is illumination efficiency. Maybe an optical analogy
is easier to understand. The USB dongle does a fair job of radiating
equally in all directions, very much like a light bulb. I you put a
reflector behind the light bulb, you'll have some of the light emitted
by the light bulb reflected from the reflector, but the greatest
majority of the bulbs light will end up going in directions where you
don't want it to go. This is called "overspray". The same thing
happens to 2.4Ghz wireless. Much of the signal is lost and not
reflected. Oh yeah, just to make life interesting, the best
efficiency a mesh or solid reflector can do is about 50% (losing -3dB
gain). Obviously, a reflector will improve the gain of the insipid
tiny PIFA antennas used in most USB dongles. Anything is an
improvement. However, don't expect spectacular gain and you will see
LOTS of side lobes.

See:
http://www.w1ghz.cx/antbook/chap4.pdf
for a great tutorial on dish design that includes quite a bit on
illumination issues.

Personally, methinks the way to go with USB is removing the PIFA
antenna and replacing it with an SMA connector. The connector would
then go to a proper panel, patch, biquad, or dish antenna. At least
all the RF goes to where it's suppose to end up.


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

On 12 Dec 2004 14:34:33 -0800, (E-Mail Removed) wrote:

>hi,
>
>From what i've read about usb wireless on
>http://www.usbwifi.orcon.net.nz/, it looks like you stick a usb
>wireless stick in the focal point of a reflective parabolic dish. Is
>that with no fiddling with the usb adapters built in antenna? Because
>this might be the road that i go down rather than searching for
>pigtails for various different cards and being ripped off.
>Thanks in advance.

I made a parabolic reflector for war driving using the below pdf
template, and use it with a Microsoft USB wireless adapter that
has a flip up antenna. Works pretty good and has noticable gain
using the netstumbler signal strength display. Noticable
improvement connecting to distant spots. Made it with a piece of
thin cardboard to hold the parabolic shape, a piece of manilla
folder with aluminum foil attached for the reflector, hot glue,
and a hole in the cardboard to slip over the antenna at the
appropriate point. It all depends on what your needs are
(business vs. hobby) and how much $$$ you want to spend.

http://www.freeantennas.com
http://www.freeantennas.com/prod01.htm
http://www.freeantennas.com/parabolic.pdf

>The basic problem with shoving a USB >dongle at the focus of a dish or
>corner reflector is illumination efficiency. >Maybe an optical analogy
>is easier to understand. The USB dongle >does a fair job of radiating
>equally in all directions, very much like a light >bulb. I you put a
>reflector behind the light bulb, you'll have >some of the light
emitted

So you mean that it is okay for focusing the signals that its receiving
BUT the improvement in transmission is not that much. And what's the
point of receiving a wireless signal if you can't send back a reply
from a distance (unless you just want to use it to capture packets).
Therefore you need a parabolic dish at both ends to pick up each others
weak and misdirected signals. Am i right in interpretation of what you
said?

My initial thoughts was that it would be not much better at sending a
signal but i thought that i was wrong. Based on my knowledge of
constructing solar concentrators/furnaces.

How is it good for wardriving? (Not that this is what i want it for but
another guy mentioned this) You would only be seeing an increase in
reception ability and not transmission strenght therefore what's the
point if you can see all these networks but can't talk to them any
better.

Slowly i'm piecing together the general workings of the field at
practical level.

BTW, i can get one rp-sma + one n type connector for around $6-7.00
which is what you need for a lot of cards (eg. netgear) therefore
building a pigtail isn't that expensive BUT you probably need a crimper
for the rp-sma.. or do you?

Thanks,
Stephen

(E-Mail Removed) wrote:
> So you mean that it is okay for focusing the signals that its receiving
> BUT the improvement in transmission is not that much. And what's the

The reflector works in both transmit and receive.

There will be noise received from the "back" side of the USB dongle as well
as from the reflector side. Some other radiator would make more effective
use of the reflector. That's not to say that the reflector doesn't help.
The New Zealand site has some annectdotal data that says that a reflector
improves the connection. I don't find the USB dongle to be very omni. I
think it is heavily loaded toward the label side of the DLink-DWL-122.
That would make the reflector more helpful than if it were a good omni.

To your other question, Netstumbler only shows receive signal and noise.
Some client programs provided with WiFi clients show "this signal" and
"partner signal". That would be a better evaluator of the dish.



I haven't used a reflector, but I have used a coffee can for a USB dongle.

http://www.rahul.net/dold/clarence/u...42-800x600.jpg
http://www.turnpoint.net/wireless/cantennahowto.html

---
Clarence A Dold - Hidden Valley (Lake County) CA USA 38.8-122.5

On Mon, 13 Dec 2004 17:23:08 +0000 (UTC),
(E-Mail Removed) wrote:

>(E-Mail Removed) wrote:
>> So you mean that it is okay for focusing the signals that its receiving
>> BUT the improvement in transmission is not that much. And what's the
>
>The reflector works in both transmit and receive.

Not exactly. Let's go back to my light analogy. If you're receiving
light from a distant source, all the light that bounces off the
reflector is reflected back to the detector at the focus. This is
good. However, if you're transmitting light, from a non-optimum
illuminating source such as an ordinary light bulb, much of the
"transmitted" light will end up going everywhere except to the
reflector. If half the light goes off in some direction other than
the reflector, you lose 3dB. My guess(tm), assuming a spherical
radiation pattern, and a typical 0.5 f/D ratio barbeque grill dish,
only about 1/8th of the signal (light) makes it to the dish for a loss
of 9dB. Yes, this means that the "gain" of such an antenna is
different in transmit and receive. You don't see different tx/rx
specifications for commerical dish antennas because they have properly
designed feeds, where most of the TX signal hits the dish (and uses
the full surface area of the dish). In such cases, the difference
between tx and rx gain is negligible.

Read the part on reflector illumination and spillover:
http://www.w1ghz.cx/antbook/chap4.pdf

>There will be noise received from the "back" side of the USB dongle as well
>as from the reflector side. Some other radiator would make more effective
>use of the reflector. That's not to say that the reflector doesn't help.
>The New Zealand site has some annectdotal data that says that a reflector
>improves the connection.

Dunno. Any kind of reflector, that redirects signal in the desired
direction, is a good thing (as long as it doesn't cancel or interfere
with the direct signal).

>I don't find the USB dongle to be very omni. I
>think it is heavily loaded toward the label side of the DLink-DWL-122.
>That would make the reflector more helpful than if it were a good omni.

True. Note that the DWL-122 antenna is backed by a ground plane (the
main circuit board) and therefore has less radiation in that
direction:
http://www.LearnByDestroying.com/pic...122/index.html

>To your other question, Netstumbler only shows receive signal and noise.
>Some client programs provided with WiFi clients show "this signal" and
>"partner signal". That would be a better evaluator of the dish.

Yep. Don't be suprised if the received signal, when compared to the
internal antenna or rubber ducky antenna, increases different amounts
at each end of the link, with a reflected antenna using an inefficient
feed.

>I haven't used a reflector, but I have used a coffee can for a USB dongle.
>http://www.rahul.net/dold/clarence/u...42-800x600.jpg
>http://www.turnpoint.net/wireless/cantennahowto.html

I hear expresso and caffinated has more "punch" than decaf, which is
rather inspid.

>Clarence A Dold - Hidden Valley (Lake County) CA USA 38.8-122.5

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