- This topic has 8 replies, 1 voice, and was last updated 4 years, 8 months ago by ummair khaliq.
22nd November 2013 at 11:46 #69342Irfan KhanGuest
We just conducted a refarming exercise in one of the customer networks, some weird problems we are facing now which were least expected. Appreciate if somebody can please illuminate.
1. Network is NSN, FlexiMR BTSs, Flexi BSCs
2. Initially BBHOP to 1×1 SFH implemented, DL RxQUAL degraded and Drop Call (dcr_8h) degraded
3. Tried Antenna Hopping with SFH, Drops degraded more and quality degradedcmore
4. AHOP disabled and Ad-HOC SFC implemented… DL RxQUAL improved significantly but TCH drop didn’t imorove at all (not at the benchmark level)
5. Implemented BBHOP with less frequencies RxQUAL and TCH Drops both improved.
Now the question is –
1. Why SFH is not working while frequencies are the same
2. Understand that 1×1 didn’t work as there is overshooting in the network but why Ad-Hoc didn’t work ?
3. Ad-HOC SFH is improving the RxQUAL but TCH Drops didn’t responded at all. Why ?
Thanks for your support !23rd November 2013 at 10:20 #69343pixGuest
could you please describe the difference between the four phases :
1/ BBHOP 1×1 SFH
4/ BBHOP with less frequencies.
Please provide the number of freq used in each case, or some typical site config examples.
I have a good knowledge of those methods, but I need to ensure we’re on the same wavelength 🙂
pix24th November 2013 at 20:30 #69344pixGuest
except AdHocSFC, which i have no clue what it is ! i’m curious to know though.
pix25th November 2013 at 14:38 #69345Irfan KhanGuest
Thanks for your prompt response…
Below are the required details, This is all about G900 layer. The network is both DCS1800 & G900, While in both layers we did the retune/refarming but the damage is only in the G900 layer. There is no underlay-overlay implemented and all sites are stand alone. Another thing is most of G900 sites are using power sharing with U900 (40 Watts – G900 / 20 Watts – U900).
1. BBHOP – Base-band hopping was in the network since beginning, it was using 29 Frequencies. Most of the frequencies were planned manually by the operator.
2. 1×1 SFH – We turned the network into SFH post refarming, with 23 Frequencies in single MA list implemented across the network using MAIO 0-6-12 & MAIO Step 2, All sites using different HSNs (same HSN used in all three cells of a site). DL Qual and TCH Drop both degraded in this case.
3. AdHoc SFH – Planned the MA lists through MR based tool (NetAct Optimizer), based on site configurations the tool planned the MA lists, Fractional load target was given to the tools was 0.75. The planned looked very good from the Delta values (cost) and also while analyzing from MapInfo. No Adjacent frequencies in same cell, even in the same site. Still using the same 23 frequencies. Now the DL QUAL improved by 25% but the TCH drops still remain degraded. Tried Antenna Hopping also, after enabling the DL QUAL also degraded significantly.
4. BBHOP with more frequencies – Now we planned baseband frequency plan again and included 10 EGSM frequencies (total 23+10 = 33 ARFCNs), both DL QUAL and TCH drop improved, even better than before refarming scenario.
The network is using Packet Abis 100%, average TRX configuration is 4+4+3.
Thanks again for your time and efforts.27th November 2013 at 14:52 #69346AyeGuest
I am a bit confused on how 1×1 SFH is implemented.
Can you clarify the following info:
1. Are the 23 Frequencies dedicated for the TCH TRX. Meaning NO BCCH channel is included on the 23 Frequencies?
2. Assuming it is all for TCH TRX. 23 Frequencies will give you MAIO Values equal to 0,2,4,6,8,10,12,14,16,18,20,22 (No adjacent MAIO). Why you are only using 0,6 and 12?
3. How many TRX per cell?
4. Assuming you have 4 TRX per cell…1 BCCH TRX and 3 TCH TRX. Your 12 MAIO values are more than enough to cater this TRX configuration.
5. In 1×1 SFH network HSN of all the sectors/cells in one site should be the same, this is to ensure that there will be no MAIO collisions since there is only one MAL being shared by each sectors/cells in one site.
Aye27th November 2013 at 19:35 #69347pixGuest
Thanks for all the details.
The differences between Adhoc SFH and BBH are… minimal, aren’t they ?
Adhoc SFH : if the number of TRX > 3 then the total number of frequency = the number of TRX. The TRX0(chgr0) does not hop. The RXn (chgr1) hops. Am I correct ?
BBH : the nb of freq = the nb of TRX (same as adhoc SFH), but here, the BCCH is included in the hopping sequence.
Am I still correct?
If that’s true, then you ought to test something : take one small cluster of cells, and change their configuration from the current BBH to a SFH, using the same frequencies. This way, we can verify that whether the SFH itself is causing DL QUALITY issues.
Also, another thing : did you activate the BS Power Control ? Did you reduce the DL Qual target in case of SFH ? The Qual target should be 5.0 or 5.5 in SFH, and 4.0 for BBH.
However, for adhoc SFH however (if my understanding of n freq = n Trx), the qual target should be same as BBH.
Ok, let me know if there are any useful infos in my reply. How can we proceed further ?
pix28th November 2013 at 11:12 #69348Irfan KhanGuest
Hi pix, thanks for the reply. yes we have initiated for one small cluster SFH trial approval, let’s see if we get any clue there.
We tried both DL QUAL thresholds, no significant impact was there.
Will keep you posted if anything moves.25th July 2015 at 05:35 #69349PAKAJ AGRAWALGuest
ACTUALLY I HAVE LITTLE KNOWLEDGE ON HSN, MAIO AND MA LIST CAN ANYBODY TELL ME WHAT THEY ATUALLY ARE. PLS EXPLAIN WIT VARIOUS CONDITIONS .
PLS EXPLAIN I NEED THIS KNOWLEDGE
THANKS IN ADVANCE3rd November 2015 at 09:50 #69350ummair khaliqGuest
please give me two r three examples of baseband frequency hopping and synthesized frequency hopping. and please differentiate it with the help of examples.. thx