Understanding 5G Beamforming Using Phased Arrays

[kevinisfrom]

TL;DR Summary

My hypothetical question is, is it possible to suppress all side lobes (side lobes -->0) and create a beam with a very small angle (theta -->0) with antennas -->infinity? A more practical question, if the beam always spreads as it travels, does this mean the receiving antenna must be much larger than the transmitting antenna?

The general concept of 5G beamforming using phased arrays makes sense from the videos I've seen. Something I've noticed was that there are always side lobes and also the beam spreads as it travels. My hypothetical question is, is it possible to suppress all side lobes (side lobes -->0) and create a beam with a very small angle (theta -->0) with antennas -->infinity? A more practical question, if the beam always spreads as it travels, does this mean the receiving antenna must be much larger than the transmitting antenna?

 

[Dale]

is it possible to suppress all side lobes (side lobes -->0)

No, the best you can do is what is called diffraction-limited. The larger your source the less diffraction you will have, but it will always be present.

if the beam always spreads as it travels, does this mean the receiving antenna must be much larger than the transmitting antenna?

No, it means that the transmitting antenna will transmit a lot of energy and the receiving antenna will receive very little energy.

 

[Baluncore]

The simple rule of thumb is that the beamwidth of the main lobe will be 360°/ 2Pi = 57°, divided by the width (measured in wavelengths) of the antenna, or array of elements.

 

[sophiecentaur]

does this mean the receiving antenna must be much larger than the transmitting antenna?

Just re-read what you wrote. It would imply that every home TV aerial would need to be the same height as the cylinder at the top of the transmitting mast. How big is your antenna? Do you know of any localised radar systems that use different sizes for Tx and Rx? How could it be better to use a wider Tx beam than Rx beam? A shared antenna makes total sense in pretty much any radar system you would want to instal on a ship.
You could have made the time to read around about this topic and avoid asking some really flawed questions. The strength of received signal depends on every link in the comms chain. But a very small amount of reading around would have made all this sort of thing clear to you. There is no point in trying to reverse engineer in an attempt to learn about a subject.

On the subject of sidelobes. The rules of thumb formulae which give an idea of the width of an antenna beam are only approximate. Sidelobes can easily be suppressed by altering the weighting of the element feed powers. This will be slightly at the expense of the width of the main beam but it can fill in the nulls, too. Depending on what you actually want from the array, that can be an excellent solution for the available aperture. Weighting of array elements is dealt with in many textbooks and this link is an up to date paper which shows how sidelines can be dealt with, usefully.

 

[tech99]

No, the best you can do is what is called diffraction-limited. The larger your source the less diffraction you will have, but it will always be present.

I think there is a subtlety here. The width of the main beam of an antenna is dependent on antenna size, but the side lobes lie outside this beam. By using a binomial distribution of energy across the antenna, sidelobes can be theoretically reduced to zero but at the expense of gain (ie broadening) of the main lobe. There are other distributions across the antenna which allow us to trade forward gain for sidelobe level.

 

[sophiecentaur]

There are other distributions across the antenna which allow us to trade forward gain for sidelobe level.

Indeed. In the case of a service like 5G (which is just the same as broadcast TV and other systems, the nulls can be a worse embarrassment than the sidelobes as the vertical radiation pattern in particular will need to have a main beam slightly tilted to the ground (a max that points just short of the wanted service area limit - sometimes the horizon) and all the nulls filled so 'everyone gets a bit' in the service area.
The 'best' antenna depends entirely on the requirement.