Better antenna diversity 1

[biff44]

If I had limited package size, and wanted two antennas for some spatial antenna diversity, which option would be better: Option A or B?

I assumed Option A, but then I realized I was not sure.

http://www.MaguffinMicrowave.com

Maguffin Microwave wireless design consulting

 

[VEBill]

Everything that I've read typically states that most of the radiation from a linear monopole antenna arises from the higher current region near the feedpoint. Reciprocity indicates that this would be true for receive mode as well. Thus, option A has a wider effective physical diversity spread than option B.

Another observation. Since the two monopoles are aligned the same (collinear) there's no polarization diversity. Depending on your application, it might be better to have them at an angle.

 

[Comcokid]

I would go with option B. The very end of the antenna is effectively a much higher impedance than the feedpoint. In option A, the two ends might be located close enough that efficiency of either antenna may be lowered through some mutual coupling.

But VE1BLL point on the polarization is a much more important issue for good diversity. You need to rotate one of the antennas by 90 degrees, or maybe both by 45 degrees so they are more orthogonal to each other. Due to path interference from objects and reflections off surfaces, or the ground, the polarization of the signal may be change.

 

[VEBill]

Comcokid makes a very good point about coupling. But there's a lot of undrawn 'stuff' on the circuit card (not shown on sketch) that would probably have greater effect. So it's an issue worth considering.

 

[biff44]

I am planning some polarization diversity on the far end, so it is less imprortant in the hand held device. I am more using it for when the operator blocks one antenna with his body.

It is good that someone else agrees with Option A being "farther apart", although since this is all withing one wavelength...not sure what that actually means. As far as coupling of one antenna to the other, I will build two and see what the proximity effect is.

http://www.MaguffinMicrowave.com

Maguffin Microwave wireless design consulting

 

[VEBill]

If I recall correctly, one-quarter lambda spacing is commonly used. If one antenna is in a multipath null, the other should be on a peak.

(The next big thing in multi-antenna systems is MIMO.)

 

[biff44]

as far as antenna cross coupling, placing two inverted F antennas so that the open circuited ends are closest, at 0.5" gap between the ends I am seeing -18 dB coupling. That level should not upset my system at all.

As far as spacing...I will go with the most my package will allow, which looks like around 3/4 wavelength between feed points.

http://www.MaguffinMicrowave.com

Maguffin Microwave wireless design consulting

 

[biff44]

BTW, test at 915 MHz.

http://www.MaguffinMicrowave.com

Maguffin Microwave wireless design consulting

 

[VEBill]

Does your model include anything within the 'empty' white rectangle shown on your sketch? If there's anything conductive and parallel to the two antennas, then you might see increased coupling via the "third party" element.

But even in the worst case, it's not going to be *that* much higher than -18 dB. Most of the RF has to go in other directions...

 

[biff44]

Yep, you are right. There is plenty of metal stuff down there, switches, power planes, etc. Also has a plastic housing covering it all. Could be a problem. But, thats why we still have engineering jobs--if it was easy nobody would need us!

http://www.MaguffinMicrowave.com

Maguffin Microwave wireless design consulting

 

[Higgler]

for electrically small antennas, alot of the antenna radiation is from current resonant on the groundplane. The chip antenna can be thought of as a tuning element, to resonate the ground plane, although current on the antenna does radiate energy.

Hence, say the ground plane is vertical, even rotating a chip antenna 90 degrees to horizontal (with the ground plane still vertical) will still have most radiation being Vertical polarization.

Wider spaced is preferred, current on ground plane is more separated.

kch

 

[biff44]

Yeah, I figured that out already. The F antenna is basically untunable (for vswr) until I make the ground plane bigger. I am doing some experiments to see if the motherboard ground plane can be used to augment the size of the rf daugtherboard's ground plane area.

http://www.MaguffinMicrowave.com

Maguffin Microwave wireless design consulting