gmichels
join:2005-06-01
| reply to gmichels
Re: [Help Me] Antennas suggestion
Thanks for your replies 
peter_m, all walls are concrete. On the upper floor, there won't be more than 3 concrete walls separating clients to the antenna. The lower floor is far more open and only one room with special walls (we call them here 'dry-walls', I'm not sure if you know the term). Anyway, wifi coverage on the lower floor is low priority.
Dan Koerner, excuse my ignorance, but I don't know what would be a 'collinear array antenna'. Could you provide more information about them?
Anyway, I was analyzing some omni-directional antennas and it looks like the best ones for my case would have between 5 and 9 dBi. Also, as peter_m suggested, I am looking for antennas with a base and cord, so I can put it over the computer case.
Currently, with the default 2 dBi antenna, signal strength from the AP to my laptop's 2915ABG wifi card (2 walls away, ~10m or 33ft) is around 50-60%, while the rate fluctuates heavily between 11 and 48 mbps.
Do you think a 5-9 dBi omni-directional antenna would help on this case?
Thanks! |
|
Dan Koerner
join:2000-08-05
Clinton, TN
| Any omni-directional antenna with a gain greater than 2 dBi is a collinear array, i.e., it's 1/2 wavelength elements are aligned end-to-end with a 1/2 wavelength phasing device between each element.
A 5 dBi vertical antenna has 2x the ERP (Effective Radiated Power) of a 2 dBi vertical antenna along any radial in their respective ground planes. This works both directions, sending and receiving. I have no idea as to the signal strength indicator calibration of the wireless cards... each manufacturer probably has a different system of measurement.
BTW, I really hate dBi (isotropic gain) references. The 0 dBi reference antenna is a hypothetical antenna that radiates or receives equally in all directions, i.e., it's radiation pattern is equal at all points on a surrounding sphere. dBi numbers are not real... just computer generated number by some unknown algorithm and a theoretical design. See »www.its.bldrdoc.gov/fs-1037/dir-···0317.htm. It is just to easy to hype a number that cannot be measured/proofed.
--
Dan |
|
peter_m
Premium
join:2005-07-13
Canada, QC
1 edit | reply to gmichels
Concrete, ouch! That is a tough one. I would get a 9db on a base/cord to get the most on the main floor and just live with what ever you get on the other floor. Just don't get too much cord. The longer it is, the more signal is wasted since 2.4ghz doesn't like wires. Good luck
Peter |
|
