- This topic has 31 replies, 1 voice, and was last updated 11 years, 12 months ago by Mania.
24th July 2010 at 10:40 #63755PaulGuest
How to reduce Dropcall/TCH radio drops in alu in which the user distribution is far from the covered site (TA)?
What parameters do we need to adjust?
Paul24th July 2010 at 12:29 #63756pixGuest
adjust the antenna orientation (tilt and azimuth)
improve link budget (better antennas, better feeders, space diversity, less combiners…)
ensure the zone is well-covered by the best server and that no ping pong HO is taking place there.
pix24th July 2010 at 18:25 #63757ALUGuest
This is a generic problem and not only ALU Specific.As pointed out by Pix earlier aswell….the best solution is to prevent overshooting by Hard Optimization(tilt/azimuth steering) and confine the coverage of the cells to their logical footprint 🙂27th July 2010 at 16:21 #63758PaulGuest
Dear Alu & Pix,
Thank you for your responses.
Well, that solution of hard optimization is pretty much will reduce the traffic (e.x downtilt) 🙂
In ALU, there is MS_TXPWR_MAX. In my network, it’s set 33 dBm. Can we set this up to 43 dBm? What will be the effect? will the device of the user will easily hotter? 🙂
Paul27th July 2010 at 17:16 #63759pixGuest
choosing 43dBm will not force the MS to emit stronger (they cannot exceed 33dBm, physically), but it will strongly impact your C1 and C2 criterion for reselection.
So keep it at 33dBm.
My solutions below intend to improve the coverage, not reduce it. I didn’t say to downtilt, but “adjust” the tilt. Perhaps you need to uptilt, or perhaps to downtilt, in order to increase the coverage in that area 🙂
pix27th July 2010 at 18:34 #63760PaulGuest
Thanks, what a fast response from you! 🙂 It’s 1.25 am in my country, how about yours? 🙂
So, it means that for far user distribution which experience call drop radio, chance to reduce it via parameter is less, better with hard optimization. Okay, If I have more questions related to this, i’ll post it.
Paul28th July 2010 at 04:07 #63761pixGuest
6am here 🙂 i live at the intersection of belfast, madrid & luxembourg… take a guess !
yes, to improve coverage the only way is hard optim. Unless you forgot to declare the neighboring relationship with the best server that covers this area..? That would be the only possibility.
now, wait… there is one “soft” way to improve coverage:
frequency diversity : activate hopping with more than 4 hopping channels spaced by more than 1MHz each, and that will reduce the effects of fast fading. That’s the theory, now i’ve never measured the effect myself. I used to activate FH in remote area to enhance coverage, but in such area there is not much fast fading. Anyway, it’s still better than doing nothing. I could say I tried 🙂
pix28th July 2010 at 17:04 #63762AksGuest
Hi Pix,Are you talking of Frequency Hopping for Coverage Improvement? But usually we use non Hopping(Non Combining mode) to have 3db gain…
Can you explain yor way….28th July 2010 at 18:02 #63763PaulGuest
Thanks for your response 🙂
Related to hopping channel, I want to share about our frequency planning in my network.
Exmple, FLP GSM
For BCCH : 107 – 124
Guarband : 106
For TCH: 88 – 105
Sect-1 88 91 94 97 100 103
Sect-2 89 92 95 98 101 104
Sect-3 90 93 96 99 102 105
MAIO Sect 1 : 0 2 4
MAIO Sect 2 : 1 3 5
MAIO Sect 3 : 0 2 4
Those are my network FLP planning.
What do you think? Is that frequency allocation secure enough from interference (MAIO)?
Then, what your sugestion (Activate MAIO Hopping more than 4 …), i haven’t tried it, but what I’ve tried was like this :
Cell A Sector -1 :
Old setting of MAIO : 0 2 4
OLD MAlist : 88, 91, 94, 97, 100, 103
New MAIO : 4 6 8
New MAList :88, 91, 94, 97, 100, 103, 107, 113, 119, 122
And that’s working! 🙂 I didn’t add MAIO up to 4 or more than 4, but added MAlist and changed the MAIO period only.
In your suggestion, we added new MAIO or change the MAIO with hopping channel more than 4. So how should I arrange the MAIO without adding the MALIST? If we set MAIO 0,2,3,4, then the hopping channel is too close (impact on interference). Any Idea brother Pix? Well, what’s your suggestion towards my network FLP allocation?
I would be grateful to discuss it with you and all experts in this forum 🙂
Paul28th July 2010 at 18:38 #63764pixGuest
what more do you want to know ? frequency hopping creates frequency diversity which fight fast fading.
i don’t understand the second part of your message.
you increase the MAlist, which now includes all the BCCH arfcn as well. what happened to your BCCH ?
I assume you are now doing BBH ? But in this case the MAlist should contain only 4 frequencies…
So I need you to clarify…
Regarding the MAIO, you can just keep doing:
xxxxxx TRX 1 2 3 4
sect1 MAIO 0 2 4 6
sect2 MAIO 1 3 5 7
sect3 MAIO 0 2 4 6
And it will work fine, with your new Malist.
But again, I’m not sure I understood you correctly. I never talked about MAIO in mly previous post…
pix28th July 2010 at 18:40 #63765AksGuest
How do we recognize Fast fading is happening in any network?28th July 2010 at 22:59 #63766ALUGuest
The intersection of belfast, madrid & luxembourg should be somewhere in “France”…am i right ? 🙂
Secondly, i need clarification for the b/m statement of urs
“I used to activate FH in remote area to enhance coverage”
That means that normally Frequency Hopping is disabled in ur network?? I cannot imagine a Network running without Frequency Hopping activated :). Plz clarify!29th July 2010 at 05:02 #63767pixGuest
fast fading, rayleigh fading… google it. It is the cancellation of phases of a signal that travelled through multiple paths.
good geographic skills !
FH was not activated everywhere : in remote area, the operator used NH.
FH is well known for helping with frequency planning & interference, but less known for its coverage enhancement properties. So in remote areas, I do’nt blame anyone for not activating FH.
On top of that, FH on a cell with 2 TRX sounds useless:
either BBH over 2 frequencies
or SFH on 1 TRX. But then the BCCH TRX doesn’t get the frequency diversity.
what do you think ?
pix29th July 2010 at 23:16 #63768ALUGuest
Thanx for the compliment Pix.I wish my technical skills would be as good as my geographic ones 😛
Well our remotes have mostly 4-TRX cells so thats why i was surprised at not using Hopping. For 2-TRX, you are right that it wouldnt be that beneficial. But if you have got a big Frequency Band, why not use DFH with more than 2 Frequencies(3 or 4) for 2-TRX cells. It should be more beneficial than NH.What do you think? 🙂30th July 2010 at 10:54 #63769pixGuest
dfh is just a smart name for SFH when there is a little amount of trx in the cell, right ?
but doing sfh in a 2trx cell means the bcch freq is not hopping… Applying the sfh would “only” bring better rxlev during a call on 2nd trx. Which is of course a good thing anyway since most calls will be allocated on 2nd trx 🙂
Now, going even further : it means you need to have the SDCCH on the 2nd TRX too, to fully benefit from FH during the call setup. And since you must have a SDCCH ts on 1st TRX too, it means you need 2 SDCCH ts on such cell. (not really a problem) and by default, it is the sdcch on the last trx that is used preferably.