Nice explanation that takes me way back to my fiber optics days. 

Patrick Leary
Aperto Networks
813.426.4230 mobile

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf 
Of Patrick Shoemaker
Sent: Friday, May 22, 2009 5:16 AM
To: WISPA General List
Subject: Re: [WISPA] Thanking Marlon and multipath experience

Not trying to be pedantic here, but I think some clarification is needed 
regarding reflection, refraction, and diffraction.

Reflection is the phenomenon that occurs when an EM wave encounters a surface 
(larger than approximately one wavelength), hits it, and leaves the surface 
with reduced intensity. Same thing as light hitting a mirror. It's the big one 
for multipath.

Refraction is the effect that "bends" an EM wave when it transitions between 
two different media with different permittivities. We see this effect when you 
put a stick in a pool of water, and the stick looks bent at the surface of the 
water.

Diffraction (knife-edge diffraction) is the phenomenon that allows EM waves to 
travel around objects. It's the reason you can get an AM radio signal in a 
valley obstructed by mountains on all sides, the mountaintops diffract the 
signal and "fill" the valley with RF. It's my understanding that this 
phenomenon is not really worth considering at the frequencies we're using.

Hope that helps!

Patrick Shoemaker
Vector Data Systems LLC
[email protected]
office: (301) 358-1690 x36
http://www.vectordatasystems.com


Marlon K. Schafer wrote:
> Glad to be of help Forbes!  Once in a while I make a lucky guess.  
> grin
> 
> OK, design note time.
> 
> I work to keep all of my links in the -65 to -75 range.  If you look 
> up the specs for your radios you will likely find that this is well 
> above the lowest receive signal level, but not too far above it.
> 
> Why is this important?  As I've been taught over the years (Thanks 
> Jaime, Bob and others) most knife edged refraction (a bounced signal 
> off of an
> object) will be about 30 dB down below the original signal level.
> 
> So if your radio has a max sensitivity of -94 and you have a -65 
> signal you'll be almost out of the possibility of the receive radio 
> even picking up your refracted signal.  Think of the refraction as a 
> reflection or echo.  An echo makes it really hard to understand 
> someone.  Multipath is the same thing.  But if you can make the echo 
> so quiet that you can't hear is it won't hurt anything, even if it's there.
> 
> MOST of my towers put out LESS than 1 WATT!!!!!  A few are still close 
> to 4 watts, but changes in AP's allow me to run ever lower tx powers.  
> Believe it or not I have customers getting over 2 megs of service from 
> systems that are less than 2 watts and at ranges of 18 miles!  Yes I 
> can prove this if anyone wants to come visit....
> 
> If you are picking up your towers at levels above -60 to -65 you'll 
> have a LOT more trouble on your network.  Especially tower to tower.  
> Turn the danged things down.  If you need more power at a customer's 
> place install a bigger antenna!
> 
> Over the years I've run into many companies that try to use a bigger stick. 
> They systems never work well for long.  The more customers go on it 
> the worse that problem gets too.  I fact I have started pulling 
> customers from a competitor in the area.  He's running a system near 
> me that looks to be running about 42 watts (remember we're only 
> allowed 4!!!!).  Why am I getting his customers?  Because his system 
> doesn't work right.  Much of that is due to his design flaws.
> 
> OK, next question is, how do I know what he's running?  All you need 
> to know is what gain the antenna is at your end, what the receive 
> signal is and how far the transmitter is away from you.  From there it's easy 
> with a formula.
> At 42dB I have him at about a 15dB antenna and 1 or so watt TX power.  
> A VERY common config sold by some distributors.  It's too bad, these 
> never work well long term and rotten wisp networks give us all a bad name.
> 
> Anyway, Forbes, try turning that power down.  WAY down.  I'll bet you 
> can go back to g mode and have even more stability than you have now.
> 
> laters,
> marlon
> 
> ----- Original Message -----
> From: "Forbes Mercy" <[email protected]>
> To: "WISPA General List" <[email protected]>
> Sent: Wednesday, May 20, 2009 1:47 PM
> Subject: [WISPA] Thanking Marlon and multipath experience
> 
> 
> I admit that most of you can dance around me in the field of RF.  I'm 
> more of an installer and management type.  Today I had a Nano CPE with 
> a -48 dbi signal which, as you know, is amazing; it should be it was a 
> block away from the tower.  The problem is the customer had long 
> outages and erratic service.  Aiming away from the tower kept the 
> great connection but still crappy pings.
> 
> Marlon suggested Multipath as a possibility and advised to turn the 
> radio to "B" only and reduce the power on the radio. I had heard of 
> Multipath but never experienced it. We changed both settings and while 
> the signal stayed at -48dbi the time-outs and erratic pings went away.  
> 90% of our system is two or more miles from towers, so we never had 
> this problem before. This particular town is the only place where the 
> towers are on buildings all within ½ mile of nearly everyone
> 
> 
> It answered a lot of problems for others in town we have been battling 
> supposed interference with and I just wanted to publically thank 
> Marlon for exposing me to a side of RF I hadn't even thought of.  This 
> is what makes WISPA worth it to me.
> 
> Forbes Mercy
> President - Washington Broadband, Inc.
> 
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.0.237 / Virus Database: 270.10.25/1956 - Release Date: 
> 02/16/09 18:31:00
> 
> 
> 
> 
> ----------------------------------------------------------------------
> ----------
> WISPA Wants You! Join today!
> http://signup.wispa.org/
> ----------------------------------------------------------------------
> ----------
> 
> WISPA Wireless List: [email protected]
> 
> Subscribe/Unsubscribe:
> http://lists.wispa.org/mailman/listinfo/wireless
> 
> Archives: http://lists.wispa.org/pipermail/wireless/
> 
> 
> 
> ----------------------------------------------------------------------
> ----------
> WISPA Wants You! Join today!
> http://signup.wispa.org/
> ----------------------------------------------------------------------
> ----------
>  
> WISPA Wireless List: [email protected]
> 
> Subscribe/Unsubscribe:
> http://lists.wispa.org/mailman/listinfo/wireless
> 
> Archives: http://lists.wispa.org/pipermail/wireless/


--------------------------------------------------------------------------------
WISPA Wants You! Join today!
http://signup.wispa.org/
--------------------------------------------------------------------------------
 
WISPA Wireless List: [email protected]

Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless

Archives: http://lists.wispa.org/pipermail/wireless/


--------------------------------------------------------------------------------
WISPA Wants You! Join today!
http://signup.wispa.org/
--------------------------------------------------------------------------------
 
WISPA Wireless List: [email protected]

Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless

Archives: http://lists.wispa.org/pipermail/wireless/

Reply via email to