Hi Daniel
I know your H and V diversity numbers below are very encouraging, but H
and V isn't anywhere near as good as decent frequency diversity......
You are much better with a circular poalrized transmit antenna if you
are going to do that. Try the H and V diversity in a variety of urban,
suburban, rural etc channels and you'll find H and V diversity sucks....
I know this apart from all the text books that show-so, 25 years in FM
radio broadcast (pretty much all circular). Whereas circular might limit
fades to 10dB deep, freq diversity fixes them all together.
The issue is, if you transmitting to a vertical mobile from a
combination of V and H antennas, you are pretty much wasting your time
with the H pol (because there is ZERO coupling to the vertical antenna)
until the path gets very nasty and flips polarization.
Polarization is in general preserved until the wavelengths get much
smaller than the features (say >1 GHz)
A few check questions......
The other thing- now pure is your H pol antenna for your test ? Was the
HPOL antenna mounted near ANY vertical structures at all ? Was your HPOL
antenna exciting these vertical structures ? How far up in the air are
you antennas ? how far apart ?
That and it is much more difficult at say 2m to build an omni colinear
antenna for H pol.
On 18/02/2015 10:53 AM, Daniel Mundall wrote:
Hi Glen,
I've got some exciting news, I just tried cross polarity (H and V) and
discovered that it seems to do almost as good as frequency diversity.
Inline image 1
The horizontal polarity (bottom) is attenuated by about 10db, this is
partly because the h-pol antenna is not entirely omni-directional and
also my receiving antenna is vertical.
In any case this method allows both channels to be aligned very
closely frequency wise, even to the point that they are orthogonal to
each other..
*Downlink* *Uplink*
*Freq A* Frame 1 Frame 1
*Freq B* Frame 1
Freq A and B could be like 2khz apart or even less, uplink it doesn't
matter because you only need one uplink channel and the two RX
antennas for each pol.
I think this method has big gains for mobile devices as they are not
locked as vertical or horizontal.
A few thoughts on TDD, there's no reason why the repeaters can't be
GPS synced so all the repeaters share a common TX slot, you could have
may repeaters at one site without any affect. This could be done at
very minimal cost. A $10 dollar module can give a pulse signal that is
tied to the UTC.
One thing that I like about TDD is the ability for the CE to go full
duplex voice. If you start having to play with duplexers 600khz
becomes useless for mobiles, and even 1.2mhz would be a pain to work
with. FDD is great in the sense that you have a 3db advantage over
TDD, but it comes at a big cost.
One last thought on this, if we can get low cost hardware out there
that WAY out dose existing FM. Even the "Old boy's" will want to switch.
We stand to win a few ways:
* Low Cost Repeaters - even a handheld could be used
* More services - messaging etc
* Longer range - way longer
* Way better quality - as soon as the SNR supports it we can have HD
voice.
* Smart routing.
* Full-duplex voice
When people start realizing how great it really is, we should be able
to get a lot more people on board.
We just need some more bright people to get involved to help out with
getting this GMSK code all finished up. It would be really nice to
actually see a BER for each method to see what it's doing.
Daniel Mundall VA7DRM
On Tue, Feb 17, 2015 at 2:29 PM, glen english <g...@cortexrf.com.au
<mailto:g...@cortexrf.com.au>> wrote:
Hi Daniel
I'll be looking forward to your single antenna split result- just a T
piece on a BNC will do the job (costs 6dB rather than 3dB) I
think you
will find a higher diversity gain compared with two antennas.
Now, you need to set up a test course etc, that you always drive in
order to compare apples with apples day after day.
Adding another carrier is a nice demonstration but likely to
provide the
great gain for the effort.
It is pretty obvious by visual inspection how good it is.
I think that TDMA duplexorless repeaters should be avoided because
they
do not enforce good filtering . But maybe that is like bad government
policy from an old guy like me.
Now, the next thing is, how to do two TX frequency , two RX
frequency 2m
repeater duplexor !
hmmm that is harder. I wouldnt recommend it at 600kHz offsets. 5 megs
70cm is a bit easier.
Let's put aside the 2nd diversity channel for the moment.
The issues with single frequency TDD (time division duplex) are a
sharing the frequency with other users. All current comms hardware is
two frequency duplex, that is there is a transmit band and a receive
band, and they are separated enough to make filtering easy.
Therefore operating single frequency TDD on a comms site is very
unfriendly , and therefore because you will NOT be able to be in any
existing repeater frequency plan, a new bandplan will be required that
gets you out on your own away from TX and RX repeater pairs, and far
enough to have practical filtering.
At 2m, for same antenna, for 600kHz offset, about 100dB
rejection is
needed of the TX noise. You will need to have the TDD operation at
least 600-1000 kHz away from any other repeater input/output
(otherwise
you will not get the rejection from an affordable set of filters)
So, sounds easy in theory, not in practice. Filtering is not that
hard,
so I would suggest sticking with two frequency duplex.
How would you integrate the diversity system into a standard 2m
bandplan. Not easily and it would require extra cavity filters. ef the
existing was RX 146.3, TX 146.9, might become RXA146.1, RXB 146.3, TXA
146.9, TXB 147.1
but... whatever you come up with is going to have to fit in the
bandplane, and with other users, and their egacy expectations.
certainly 70cm there is more room and scope, and also 70cm
benefits even
more form diversity because the nulls are much much deeper than 2m.
Compatibility with existing bandplans and users etc is harder
than the
technology.
with TDD, then there is the problem of multiple users in the one area.
Because multiple users may be transmitting while others are receiveing
just a few tens of kHz apart, the system performance will be heavily
degraded by desense between the users.
At least with two frequency duplex, there is a decent frequency sep
between those trying to listen and those talking....
glen.
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and
Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration
& more
Get technology previously reserved for billion-dollar
corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
_______________________________________________
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
<mailto:Freetel-codec2@lists.sourceforge.net>
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
--
-
Glen English
RF Communications and Electronics Engineer
CORTEX RF
&
Pacific Media Technologies Pty Ltd
ABN 40 075 532 008
PO Box 5231 Lyneham ACT 2602, Australia.
au mobile : +61 (0)418 975077
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
_______________________________________________
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
