Can anyone explain wireless antennas to me?

[Talcite]

Well I'm heading to res this september, and there's no wireless. I read the TOS and there's no restriction against plugging an access point into the wall. I don't know if it's even going to work yet, so I have to test it out with a cheap router in access point mode.

First I guess i should also ask is this the right way to go about solving the problem?

And second of all, if this works, what kind of antenna should I be considering for this? I know signal strength is a combination of transmission power and of the antenna's dBi rating. Should I be looking at an omni directional or a directional antenna? I want maximum transmission range. I don't know where my dorm room will be exactly, but I can assure you it will be off the ground, perhaps 2-4 floors up.

I heard there are problems with directional antenna? And what is this doughnut syndrome? How do the broadcasting arcs work?

Have I missed any other factors in considering an antenna?

What brands are usually good for this kind of thing?

 

[Born2bwire]

I would not worry too much about antennas in regards with wireless. Most of the antennas for sale on the market are probably all going to have the same transmission power because that is regulated by the FCC. The classic "cantennas" that you find on the internet are used in place of commercial antennas because they are directional but also because they are generally more powerful since they violate FCC regulations. As for directional over omnidirectional, directional would probably have the most radiated power in your direction. I cannot recall offhand what are the regulations on wireless antennas and that would matter for this. For example, with radio station transmitters, if I recall correctly, the amount of power radiated is what is regulated. But some stations directionalize their signal to reach out more into a specific direction to get around the transmission distance limitations incurred by limiting radiated power or overall directionality. So if the regulation is on radiated power, then a directional antenna would probably have a slightly longer range since more power can be concentrated into the beam.

Doughnut syndrome, never heard of it. But it does not take much imagination to guess what it is referring to. The radiation pattern for a dipole antenna, which is probably what the standard "omnidirectional" antenna is, is a toroid (or donut). The null in the radiation pattern is along the axis of the dipole. So you would not receive any signal if your receiver was above or below the dipole. But you can always tilt the dipole to move the null. Specifically, the strongest beam will be in the plane normal to the axis of the dipole.

Omni directional would make it much easier to set up the wireless points and to daisy chain them as necessary. You may have to daisy chain the wireless points if you are going through floors.

 

[stevf]

easy version of his explanation of donut syndrome - dont point the end of the antenna at the transmitter or receiver - this is for omni-directional antennas like dipoles and 1/4 wave stubs - ie. - those that look basically like a piece of wire(or thick wire wrapped in plastic) e

 

[Born2bwire]

On a quick retrospective of this, if commercial means fail you, which I think may be the case for the amount of floors and walls it seems you will need to penetrate, you may need to make your own antenna. Other than your classic waveguide cantenna, I have knocked around the idea of building a simple Yagi-Uda printed antenna. I had a friend in another lab talking about wanting to do geolocation using wireless access points and I thought that a moveable antenna with a high directionality would be good for him. A Yagi-Uda is similar to the classic TV antenna and it would be an easy thing to make a print in a PCB program, print up a few boards, and then use them as an antennas. This would be pretty easy since there are a few standard antenna books that have the specific dimensions for antennas of a given frequency. Wish I could remember the name of the text offhand though. I think you can expect maybe 12 dB gain, which is going to be much more than what you will achieve with a single commercial omnidirectional antenna. A half-wavelength dipole has a directivity of 2.15 dB, so your ideal gain is 2.15 dB.

 

[Talcite]

so if i were to make my own antenna... I should do a yagi? Arn't those directional though? does that mean I have to hook up several of them to get good coverage? What's a simple pcb printing program? I'm pretty interested in doing this now haha. How would you connect the pcb to the TNC connector or whatnot?

 

[ScottMac]

How much coverage d'ya figure you're gonna need for a dorm room?

Chances are you could run with no antenna and get adequate sgnal strength and quality.

Go with the stock antennas, work up from there.

FWIW

Scott

 

[Talcite]

Lol, oh I know i don't need alot of coverage for a dorm room... the problem comes when I get bored and decide the ENTIRE dorm should have wireless internet =p. I was actually hoping for some of the area outside the dorm, that way I can be outside and stuff have the luxury of wireless.

Anyways, I looked up yagi in wiki, and there's ALOT more than I thought there would be. I'm seriously confused now, because I have no idea what all those wire configurations in wiki were. Btw, what frequency am I looking for? I know it's the 2.4GHz range, but the exact frequency?

Btw, I don't want to spend TOO much, because the entire campus will be wireless by 2009... plus residence is only for 1 year. anyways, there's still 3 years to go! =o. Lol I'll be on fiber by then... (seriously, the campus is upgrading right now 0.o)

 

[Born2bwire]

Originally posted by: Talcite
Lol, oh I know i don't need alot of coverage for a dorm room... the problem comes when I get bored and decide the ENTIRE dorm should have wireless internet =p. I was actually hoping for some of the area outside the dorm, that way I can be outside and stuff have the luxury of wireless.

Anyways, I looked up yagi in wiki, and there's ALOT more than I thought there would be. I'm seriously confused now, because I have no idea what all those wire configurations in wiki were. Btw, what frequency am I looking for? I know it's the 2.4GHz range, but the exact frequency?

Btw, I don't want to spend TOO much, because the entire campus will be wireless by 2009... plus residence is only for 1 year. anyways, there's still 3 years to go! =o. Lol I'll be on fiber by then... (seriously, the campus is upgrading right now 0.o)

Well, the way I envisioned you using the Yagi was that it was attached to an access point in your room and directed to another access point elsewhere in the dorm that also has a Yagi antenna for receiving (and transmitting) that you wanted to hook up to. If you are transmitting wireless for a general device, like your laptop, then such a highly directionalized antenna would not be too useful since you would only be able to get a signal at a specific place. So if this is not your intended usage then you may want to abandon using a directional antenna. Except if you want to go outside and have line of sight from your window. Then it would be easy to direct the antenna to a chosen spot and then head out there.

I think the easiest solution would be one of the cantennas since they are easy to build and there is a lot of information already out on the web. The Yadi Uda project would only be something to consider if you are really interested in doing an EE project.

Paper
Thesis

The linked paper is an example of a printed Yagi Uda dipole. You would basically lay out a design of copper in a PCB program like Eagle and then you can send that design to be fabricated on PCB like FR4 (this would be like $30). Then solder a SMT coax connector and you are done. One thing though, a Yagi Uda is basically a dipole antenna with dummy dipoles in front of the radiating element (to act as directors) and a dummy dipole behind the radiating element (to act as a reflector). So for a dipole antenna, you need to have a balanced signal. Normal signals are unbalanced, the signal is referenced to the ground. In a balanced signal, the signal on the signal line is 180 degrees out of phase with the "ground" signal. You can easily just put what is called a balun down on the PCB as well to take an unbalanced signal, to make it balanced, and then feed it to your Yagi Uda. Page 40 of the thesis (which is based upon the linked paper) has a picture of a Yagi Uda printed on a dielectric with a balun.

It seems to me that the output to the antenna for your router would be balanced, so you do not need to worry about that. You could just do a wire Yagi Uda antenna instead of printing one on PCB. You could easily use NEC to simulate and fine tune your wire antenna. The advantage of the PCB printed antenna was that you would be able to precisely simulate it in something like HFSS or ADS Momentum and the resulting printed antenna would be very close to your simulations since they can control the dimensions of the result to a high degree.

Example

For frequency range, I think you need from 2.4 GHz to 2.9 GHz. Basically this would just require scaling a design since it really all based upon wavelengths. I think the book I was thinking of was the ARRL Handbook. In there I would expect them to have an example printed Yagi Uda antenna that you could adapt (or at least a wire antenna design).

 

[herm0016]

different solution. sounds like you should use omnis on the router and build a dirctional for you computer, a cantena or something. that way you have better recive/transmit on the lappy and you just point that in the direction of your router. for everyone: set the antennas on your router 90 degrees to each other to rid yourself of the "doughnut " problem.

 

[Frangelina]

An antenna is like a telescope. To be more powerful, you need to reduce the angle.

 

[spidey07]

Originally posted by: Frangelina
An antenna is like a telescope. To be more powerful, you need to reduce the angle.

Exactly. You can't make more power. But you can focus that power in a particular direction/rediation pattern.

OP - you're looking for a high gain omni-directional antenna, or preferrably a very good commercial access point (this yields better results because of higher quality radios). Building your own cantenna is straight forward if you want to do that for the client.

As for the freqencies...
<1-2462> One of: 1 2 3 4 5 6 7 8 9 10 11
2412 2417 2422 2427 2432 2437 2442 2447 2452 2457 2462
least-congested Scan for best frequency

just remember that once you get over 20 clients associated to an AP your performance can really suffer. Even if they aren't using the network. Espeically if you are running 80211g in compatability mode (allowing 802.11b clients)