- This topic has 12 replies, 1 voice, and was last updated 11 years, 7 months ago by NA.
29th September 2010 at 23:06 #64470rfengineer999Guest
I have a sector(Alpha) on top of a hill with traffic at the foot of the hill and traffic another 5 miles from it.
Alpha sector ant configuration:
ANT1 Tilt: 9
ANT2 Tilt: 5
ANT3 Tilt: 5
What’s the problem with doing multiple tilt?
I know this might not be best practice; however it seems logical to cover to area with this setup.
I would like to hear your feedback. Thanks!30th September 2010 at 06:00 #64471BijoyGuest
There is no problem as long as things(KPI,traffic,C/I.etc etc) are right.
The practice which gives best output is the best practice.If with this set up as long as things are met,its OK.
Bijoy30th September 2010 at 06:50 #64472ManiaGuest
Well in terms of coverage it might be ok. but the problem is that BCCH is sensed to setup a call and then it is assigned to any of the TRXs and your coverage would be discribed by the tilt of the BCCH for all practical reasons for setting up calls / handing over. now if other TRXs are connected to different tilt antennas then when BCCH levels are good at a point and allocated a call to another TRX its Levels might not be good so. call failures and drops would result ..better approach would be to have coolocated cells with different tilts and different frequencies like DCS having overage of closer calls down tilted and GSM taking far away calls .. not downtilted..30th September 2010 at 08:39 #64473BrotherGuest
Decisions are taken according to BCCH level, and BCCH rxlev is like a reference to the other TRX’s rxlevel, if you falsify (like bcch good level other trx’s bad level, vise versa is not a problem) this will create problems during call setup, handover intercell and intracell.30th September 2010 at 09:48 #64474PixGuest
how can you have 3 antennas for one cell ?30th September 2010 at 12:59 #64475korgiGuest
It’s unusual application but it can be by using splitter. 🙂1st October 2010 at 05:07 #64476pixGuest
yes, why not… but let’s assume it’s what rfengineer999 is doing :
the signal (of 4 TRXs, let’s say) is combined over two antenna feeders at the output of the BTS, which are -each- split in 3.
Instead of going to one xpol antenna, the signal is split over 3xpol antennas.
In this case, each (cross polar) antenna broadcasts the signal of all 4 TRXs equally. There is no coverage difference between the trx bcch and the other trx, as stated below.
But such configuration does not make much sense… why would ANT2 and ANT3 have the same tilt ?? Why would they all be oriented towards the same azimuth ? (note: korgi, those are only rhetorical questions, i don’t expect you to answer 😀 )
I think rfengineer999 is either talking about 1 sector with 3 cells from different band (2G 900, 2G 1800, 3G 2100), or is talking about sector 1, sector 2 and sector 3. So any clarification on the actual azimuth, frequency band, number of trxs, power of trxs, etc is very much welcome 🙂
pix2nd October 2010 at 19:51 #64477EagleGuest
This aplication is strange and i dont think is helpfull for what u want to achieve. instead of giving different tilt for each TRX u can use concentric cell and so u can cover the far area with the overlaid TRX and the near place with the
Hope i was clear4th October 2010 at 12:57 #64478BrotherGuest
can it be due to high configuration of the cell? there is a configuration(correct me i am not sure) like using 2 antennas and two combiners (4 port) instead of using 1 combiner 1 antenna (8 port) for 8 trx in 1 cell, this may be a logic one, two antennas for one band and 1 antenna is for other band.4th October 2010 at 13:38 #64479pixGuest
yes, i could understand the “lox loss” configuration : 2 antennas for one cell. That’s possible, and you gain 4dB in DL TX. It’s pricey too 🙂
But 3 antennas for one cell, that is very special. Rfengineer999 is not replying to this post, so I think we can close the topic.
pix22nd November 2010 at 08:06 #64480NAGuest
here’s the question is we r using 2G(900 & 1800) band per sector, NSN is the equipment vendor. two independent antennas are used, so if there is any problem regarding the tilt of the two is different.
and if there is no problem then what practice would be better : 1800 antenna more downtilted w.r.t. 900 band antenna
thanx22nd November 2010 at 08:54 #64481pixGuest
no problem at all, this is standard practice to use different tilts in 1800 and in 900.
The strategy is very often to tilt the 900 more, and tilt the 1800 less.
For example : 900 = 10° tilt and 1800 = 6° tilt.
1/ 900 are smaller
2/ 1800 are same size as these “small” 900.
The point is to reduce the 900 coverage in the city, in order to minimize 900 interference, since 900 coverage is usually very good, but the quality is not so great.
And it also allows to get a similar coverage between 1800 & 900, to be able to efficiently share traffic between both bands.
On the other hand, you can also use the same tilt for both antennas : this will allow a smaller 1800 overlap with its 1800 neighbours. Therefore, you can pack the 1800 with more TRX’s, because the frequency planning is cleaner. The 1800-cells are considered as a smaller inner zone of the 900-cells.
The strategy really depends on your network topology.
– intersite distance is small (< 500m) => tilt 900 = 8°, tilt 1800 = 6° (assumption : the vertical HPBW of the antennas is 8°)
– intersite distance is large (> 1.5km) = use same tilt for 900 (5°)and for 1800 (5° or 4.5°).
The idea is that when tilt becomes small, a difference of 0.5° makes a huge difference.
pix24th November 2010 at 14:35 #64482NAGuest
Pix- thanx for your kind response.
Whatif we inverse the tilt as i stated before.