WLAN connection

Hi! I live in a flat and I'd like to create a WLAN to share files and Internet connection with my brother who lives 6 flats next to me (about 50 meters): is there a possibility to create a wireless connection? We are il the same "side" of the building and there are no obstacle between us. What's your experience about this? What kind of device I could need? Professional devices? Or just consumer devices? Maybe I need two access-point with bridge capabilities. My budget is not much than 200 $... Thank you!

Reply to
Loading thread data ...

WiFi needs LOS (Line Of Sight) between the two antennae. While a single drywall-on-wood wall will only attenuate the signal a bit, trying to ram a WiFi signal through 6 or more walls is unlikely to work. Maybe, if you and your bro' can arrange to stick your antennae out your windows, you have a chance (and a little experiment can tell for sure); else, I'd look for a way to run a Cat5 cable, or try to piggyback your network onto the building's power lines.

Reply to
Bob Willard

All those walls will I expect become a major impediment

is the ceiling suspended ?

then transmitting the link there might be a point or perhaps if line of sight externally is available ?

Reply to
atec 77

Thanks for your messages! I'll try to explain you better: I didn't think to send the WLAN signal through maybe about 25 walls!!! I was thinking to put the devices near windows, o better, outside in the terrace. My first problem (and question!) if is possible to create a WLAN link of about 50 meters...and wich kind of devices. I was thinking to put antenna fixed over external railing. For example, here

formatting link

(sorry, it's a link to Italian web site...in effects I didn't find the same range in English site!) "declare" this device can cover up to 790 meters! They say ("...ideal to cover distances up to 790 meters!"). How? With which device? I have experience with US Robotics access- point but the signal is not so "strong"! 10-15 meters and you have no more signal! What about your experience? Thanks again!

Reply to

English version is

formatting link
just a basic 5dBi patch antenna. You would need one at both ends of the link and arange them so that they are both vertical as they are polarised. Free space path loss is 36.56 + 20Log10(Frequency) + 20Log10(Distance in miles) Which is about 75dB for 50m and about 54dB for 5m.

Reply to

It should work fine for 50 meters outside if nothing is in the way. That's if. Who knows what railings and BBQs and a complex full of interfering devices might do.

Consider the Buffalo WHR-HP-G54 for the router and a Buffalo WLI-TX4- G54HP as the client/ ethernet converter.

formatting link
They are inexpensive:

formatting link
?Item=N82E16833162168 Note: If you are a bit of a geek and want more possiblities, consider installing the optional DD-WRT firmware on the Buffalos and setting one up as a client. In that case, you would want two WHR-HP-G54s instead.

Installation: They may work well enough with their stock antennas. Buy the router and adapter first and set them up at your place then test them on the balconies.

If they work, install them in outside in weatherproof boxes (tupperware works) and run power and ethernet cables inside to the PCs and/or internet modem.

You can connect your own pcs via cable or wireless to the router. Your brother will connect by cable to the adapter.

If your test doesn't give a good enough signal, then you can try different channels to reduce interference and maybe buy one or two directional antennas. But you may not need them. 50 meters is not far IF the line of sight is good and interference is low.

Be sure to secure your wireless LAN with WPA.

Reply to

Thank you very much for all your "precious" answers!!!

Reply to

On Thu, 01 Mar 2007 07:51:02 -0500, Bob Willard wrote in :

The actual attenuation of a 4" hollow wall is about 5 dB, which is a pretty big loss. See

Reply to
John Navas

John Navas hath wroth:

I'll call your 5dB and raise you a decibel for the drywall. See:

The 3com list you provided looks fairly reasonable, but has problems. Normally, a range of values are specified. I've used various signals strength indications to measure attenuation. The problem with the typical WiFi client is that the signal strength is derived from the RSSI from a value of 0 to 256. That's not very good dynamic range and is guaranteed to result in some rather strange numbers. The same test done with a spectrum analyzer is much better.

Life also gets interesting when assuming that wall attenuation is merely a measure of the number of walls times the attenuation per wall along the straight line of sight. That works fine for 1 or 2 walls, but fails miserably at 3 or more. What happens is that the RF tends to reflect, bounce, refract, and generally follow the path of least resistance. An open door or window offers less attenuation than a solid wall or two. So, as the attenuation through the wall goes up, the relative signal loss through the doorway or window becomes more significant. For example, I was exactly such a test measuring wall penetration in an office building. After about 5 walls, the signal strength was almost constant, regardless of the number of additional walls introduced. I eventually determined that the signal was going out a window, bouncing off the adjacent office building, and returning through another window.

Another clue is to look at the original 3com document, which included the 802.11a (5.7GHz) attenuation. In some cases, the attenuation at

5.7GHz is more than 2.4GHz, in other cases, it's the other way around, which seems rather odd. In some cases, the difference is spectacular, in others, nominal, which is also odd. I've tabulated the differences below (and fixed the dBi label to dB).

BUILDING MATERIAL 5GHZ 2.4GHZ ATTENUATION (dB) (dB) (diff dB) Solid Wood Door 1.75" 10 6 +4 Hollow Wood Door 1.75" 7 4 +3 Interior Office Door w/Window 1.75"/0.5" 6 4 +2 Steel Fire/Exit Door 1.75" 25 13 +12 Steel Fire/Exit Door 2.5" 32 19 +13 Steel Rollup Door 1.5" 19 11 +7 Brick 3.5" 10 6 +4 Concrete Wall 18" 30 18 +13 Cubical Wall (Fabric) 2.25" 30 18 +13 Exterior Concrete Wall 27" 45 53 -8 Glass Divider 0.5" 8 12 -4 Interior Hollow Wall 4" 3 5 -2 Interior Hollow Wall 6" 4 9 -5 Interior Solid Wall 5" 16 14 +2 Marble 2" 10 6 +4 Bullet-Proof Glass 1" 20 10 +10 Exterior Double Pane Coated Glass 1" 20 13 +7 Exterior Single Pane Window 0.5" 6 7 -1 Interior Office Window 1" 6 3 +3 Safety Glass-Wire 0.25" 2 3 -1 Safety Glass-Wire 1.0" 18 13 +5

Note the huge variation in the difference between 5.7GHz and 2.4Ghz attenuation. If we eliminate structures with reflective content (anything with metal), the theoretical difference in attenuation for purely absorptive materials should be about +3dB. There are a few such "solid" materials listed but only a few seem to follow theory. Glass and concrete seems to be a problem.

Anyway, I just wanted to point out that there are HUGE variations in such numbers, that they are difficult to reproduce, that testing methods cause substantial errors, and that your mileage may vary.

Reply to
Jeff Liebermann

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.