Click to See Complete Forum and Search --> : Longest link possible without breaking European regulations?


LaGa
11-15-2002, 02:28 PM
I'm wondering if someone can help me with the following problem:

From what I can tell, the ETSI regulations say that you can only have 100mW e.i.r.p. (i.e., including antenna gain). Unlike the FCC, there seems to be no special rules for point-to-point with high gain antennas.

I obviously want to stay within the regulations, and I want an 11MBps connection - what is the longest link possible, and what equipment would I need to use?

Will this require a solution with separate send (low dBi) and receive (high dBi) antennas? Seems like radios with separate transmit and receive ports are a wee bit more expensive than traditional gear. :-/

Thanks in advance,
LaGa.

Sumaryo
11-15-2002, 07:51 PM
To answer this question, let's do some simple math.
The relation between power output (eirp), distance and operating margin given by:
SOM=Received signal - Receiver sensitivity
Received signal = EIRP - FSL + Receiver gain
FSL = 32.45 +20 log f (freq in MHz) + 20 log D (distance, in km)
Using standard output WLAN (typical 12 dBm) with built in indoor antenna (2 dBi gain) -that gives up EIRP of 14 dBm , and 11 Mbps troughput sensitivity at - 73 dBm (Dont bother with manufacturer claimed of 300 m open space range, that is at 1 Mbps, not full 11 Mbps over the air), and SOM of 10 dB (good figure to have a solid link) giving us a distance of 90 m. (an excel sheet could help calculate this easily).

As we know that ETSI allowed us to have EIRP up to 100 mWatt or 20 dBm, theoretically we can just replaced the above setup with an antenna with higher gain, lets say 8 dBi instead of 2 dBi rubber duck dipole that comes with the units. Now, our distance become 350 m (still with the other parameters being equal and assuming no other parties using the same channel /interference free).

Now, what happen if we decide to select lower power output radio and higher gain directional antenna, lets say 0 dBm radio and 20 dBmi flat panel antenna. Now, our distance become 1.4 km (still with the other parameters being equal and assuming no other parties using the same channel /interference free).

Now, for the purpose of theoretical excercise , lets say we can have very low power output radio -20 dBm and very high gain antenna, 40 dBm gain (so that eirp still 20 dBm or 100 mWatt). With this setup, the distance could reach 14 km. Ofcourse this is not a practical case, no radio at -20 dBm because average white noise is already at -2 dBm or higher, depending on circumstances and a full parabolic antenna at 2.45 GHz having 40 dBi gain has a size of 5.25 m !!!

Radio with separate transmit and received ??? It's exist, it's not a TDD, hence it's not a 802.11xx comply, however the calculation above still valid.

Yohanes

LaGa
11-15-2002, 09:25 PM
<i>11 Mbps troughput sensitivity at - 73 dBm</i>

<i>Now, what happen if we decide to select lower power output radio and higher gain directional antenna, lets say 0 dBm radio and 20 dBmi flat panel antenna. Now, our distance become 1.4 km</i>

Kind of what I expected. Now, if we go with something like the Alvarion BU-DS.11 which has -85dBm at 11Mbps (is it generally safe to believe the sensitivity numbers that manufacturers claim, btw?), I have 12 dBm extra on the budget.

The units have an integrated 16dBm antenna, so something like 4KM should be about the limit.

A lot less that the numbers you can read about from the US, but regulations are regulations.

Sumaryo
11-15-2002, 11:35 PM
>which has -85dBm at 11Mbps (is it generally safe to believe the sensitivity numbers that manufacturers claim, btw?),

At what packet error rate level ? Probably the figure above defined at a 20% error rate (useless for our uses ofcourse).
Some manufacturer defined receiver sensitivity together with error rate, for example -76 dBm at 11 Mbps at 2x10e-6 error rate ( 2 bits over million bits error), which increase to -85 dBm at 11 Mbps at 2x10e-1 error rate ... and so on.

So, don't be confused with marketing gizmo's

adsdos
11-19-2002, 05:28 AM
>> A lot less that the numbers you can read about from the US, but regulations are regulations.

hmmm.... not only on US:
"se establece una comunicación inalámbrica entre Tenerife y Gran Canaria de 70,5 kilometros."
A wireless connection stablished between Tenerife and Gran Canaria at 70,5 Km.
No amplifiers used only 2 pc card D-Link, pigtails and Yaggi antennas of 24 and 18 Dbi.
More Info at here (http://www.sincables.net/modules.php?name=News&file=article&sid=148&mode=thread&order=1&thold=-1) Only in Spanish

Sumaryo
11-19-2002, 12:48 PM
"A wireless connection established between Tenerife and Gran Canaria at 70,5 Km, without the use of any signal amplifier"
FSL at 70.5 km at 2450 MHz is 137 dB
Dlink PCI card has output power 15 dBm and sensitivity -80 dBm at 11 Mbps at 10e-5 error rate (which may increase to -91 dBm at 1 Mbps at 10e-5 error rate).

If we fed this parameters to our equation with short transmission line (1 dB loss max) and 24 dB antenna on both side, we got SOM=15 dB.

That is very possible that a good link can be established (LOS an free 0.6 R fresnel zone is a must)