On Fri, 17 Feb 2017 20:16:06 +0100, "Carlos E. R."
wrote:

On 2017-02-17 19:53, JF Mezei wrote:
Cell antennas are very directional.

I know. We installed them at a small company I worked with.
But the antenna on the mobile phone is not. The mobile can not know the
direction of the signal from the signal alone, that's what I said.

Not exactly. The handset has to pass an SAR (specific absorption
rate) test in order to convince the FCC that the handset is not going
to fry the users brain with too much RF. To make that work, handset
antennas are usually located on the side away from the users head, at
the bottom of the phone, or backed by a metal shield. On simulations
and in an RF anechoic chamber, the antenna pattern is somewhat
directional favoring the directions away from the users head.

"Mobile Phone in Vicinity of Human Head - SAR calculation"
http://www.wipl-d.com/applications.php?cont=emc/sar-calculation
See Fig 8:

As for the tower antenna patterns being direction, it's a matter of
what you consider directional. In the typical 3 sided tower
configuration, the sector antennas have a horizontal beamwidth of
about 60 degrees. The tower can and does indicate which sector is
being used, but that has a granularity of 120 degrees, which is hardly
accurate enough to determine anyones position. On systems that use
various forms of AGPS (augmented GPS) using TDOA (time difference of
arrival), two different towers can obtain a location fix of a handset.
That requires double the number of available receivers, two towers
that can hear the handset, and the necessary technology. That's why
I've only seen it on demonstration projects. It's also useless for
locating the tower, which I believe was the topic of discussion prior
to this topic drift.

The vertical beamwidth of sector antennas is very sharp. The vertical
beamwidth and downtilt angle are the major contributors to what
determines the coverage area of a cell site. Too narrow, and signal
will go over the heads of users close to the tower. Too wide, and the
tower will be talking to gophers and airplanes, not users on the
ground. For example, a common Andrew HBXX-6517DS-VTM antenna:
http://www.commscope.com/catalog/wir....aspx?id=15654
has a horizontal beamwidth of 66 degrees, and a vertical beamwidth of
4.7 degrees. Draw a 5 degree angle on a piece of paper and you'll see
the problem. It's bad enough that there are products to vertically
align sector antennas to about +/-0.1 degrees.
https://sunsight.com
https://sunsight.com/index.php/products/95-sunsight-instruments/products/199-antenna-alignment-tool
However, that's also useless for locating handsets, unless you want
the altitude.

Dinner... gotta run.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558