Archived from groups: alt.internet.wireless (
More info?)
On Tue, 07 Dec 2004 15:51:28 GMT, purple
<purple.1gwi4y@WiFi-Forum_dot_com> wrote:
>sorry im very newbie ...can you explain a little bit more please?? how
>can i use that calculator that you sent the link??
Ummmm... I'll try. You get the short version. What the link
calculator does is estimate the "fade margin" or "system operating
margin" for a wireless link. Everything is in dBm (decibels above 1
milliwatt reference level).
Looks like the YDI site go hijacked so we'll switch to:
http://www.tcstx.com/software/Flash/ThermalFadeMargin.swf
the Flash version.
You start at one end of the link with the transmitter. It puts out
+15dBm (about 32 milliwatts) of RF power at 2.4GHz. That's the
starting point.
The +15dBm gets reduced by about half (3dB) by loss in the coaxial
cable pigtail and connectors. That's the TX cable loss figure. The
signal is now down to +12dBm.
The signal then goes to the cantenna. The best of the breed cantennas
have a gain of about 8dBi. That's 8dBi above a reference theoretical
isotropic radiator. 8dBi is 6.3 times gain over this theoretical
point source radiator or about 6dBi gain over a cheezy dipole antenna
as found on many cheezy access points. The nice thing about using the
isotropic numbers is that I can just add the gains. So, we take the
+12dBm signal, add 8dB gain to it, and end up with a signal level of
+20dBm.
The +20dBm signal level is at the transmit antenna output but hasn't
gone anywhere useful. Mother nature and her accomplis, physics, has
conspired to make RF communications expensive. Inverse square law
says your +20dBm signal gets weaker as it blunders toward the
receiver. 5km will yield a free space loss of about 114dBm. So our
+20dBm signal is now decreased to 20 - 114 = -94dBm. The minus sign
means that it's 94dB *BELOW* the 1 milliwatt reference level. That's
a very weak signal.
The cantenna on the receive end adds 8dB gain for a signal level of
-86dBm.
The receive cables and connects eat 3dB so that the signal at the
receiver input is now -89dBm.
Digging into the specifications on the WAP11 and similar radios, I
find that the receiver sensitvity is -82dBm at 11Mbits/sec. Note that
the sensitivity varies with modulation method and connection speed.
I'll do this at 11Mbits/sec although a slower speed may be more
useful. The difference between the received signal (-89dBm) and the
receiver sensitivity (-82dBm) is called the fade margin or system
operating margin. In this case, it's -7dBm, or 7dBm short barely
functioning. That won't work at all.
Even if the fade margin were 0dB, it still would not work.
Sensitivity is measured at a BER (bit error rate) of 1*10^5. That's a
1 bit error every 10,000 bits. That's a rather high error rate and
you would not want to run your system at this level. There are also
the effects of signal bounce, atmospherics, birds sitting on the
antenna, and such to ruin the signal. You need some kind of margin to
insure that you have a strong enough signal at all times. 10dB fade
margin is considered the absolute minimum. 20dB is a tolerable
target. There's also a direct relationship between fade margin and %
reliability.
You're 17dB short of such a marginal system using WAP11's and
cantennas. You need more antenna gain or more tx power or you don't
have a chance at 5Km.
--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558
Facebook
Twitter
Instagram