- This topic has 40 replies, 1 voice, and was last updated 11 years, 9 months ago by Styrax.
-
AuthorPosts
-
26th August 2011 at 08:38 #67028VanGGuest
Hi Pix,
is it possible to help me with Repeated FACH/SACH feature activation in B10/B11?
Which are the parameters related for activation of this features and thresholds that to be modified?>>rrb
1st September 2011 at 07:25 #67029VanGGuestIs there somebody to know ??
3rd September 2011 at 18:11 #67030pixGuesthi vanG,
I was on vacations, sorry for the late reply. There are only few parameters, I will list them by monday (approx.. 🙂 )
regards
pix21st September 2011 at 08:25 #67031VanGGuestHi Pix,
we tried to active this feature for some cells but I dont see any performance changes after activation…
What are better values for thresholds REP_DL_FACCH_THRES_AMR_FR/HR? In our configuration we use the highest thresholds, this is good or not?
Is it possible to give me your e-mail to send you our settings to say whether are ok or not?26th September 2011 at 08:38 #67032VanGGuestIs there someone who has activated this feature in ALU?
26th September 2011 at 11:53 #67033pixGuesthi vanG,
I totally forgot…
For the repeated FACCH – impact is on the Outgoing HO Efficiency Rate, and possible on the RTCH Drop Rate.
Here are my recommendations, for RFACCH, to see if there is an impact or not:
EN_REP_DL_FACCH = 2: enabled for LAPDm command frames + LAPDm response frames
REP_DL_FACCH_LEGACY_SUPPORT = 1: repeated DL FACCH enabled for all AMR mobile stations
REP_DL_FACCH_THRES_AMR_FR = 8: 12,2 kbit/s
REP_DL_FACCH_THRES_AMR_HR = 5: 7,40 kbit/s
————————–
and for the RSACCH:EN_REP_SACCH = 1: enabled
RADIOLINK_REP_DL_SACCH = 14
L_RXQUAL_DL_P_AMR_RXACCH = 3.5
L_RXQUAL_UL_P_AMR_RXACCH = 3.5REP_DL_SACCH_THRES = 3
REP_UL_SACCH_THRES = 3
REP_DL_SACCH_WS = 10
REP_UL_SACCH_WS = 10—————————
now, keep in mind that RxACCH will be visibly useful only if codec is changed to lower codecs (in FR, around 7kb/s, in HR around 4kb/s)
If you stay at 12.2kb/s, then the rep. xACCH will show no improvement at all !!regards
pix27th September 2011 at 08:34 #67034VanGGuestThank you Pix, I will checked and send you results…
I have one more question about Call drop. In my region there are cells with higher call drop due to other causes..Did you tried to change the Timers related to this type of drops? What is the “good” value for T3101,T3103,T3107,T3109 and T8?>>rrb_bg
28th September 2011 at 09:17 #67035VanGGuestHi Pix,
I changed our SACH/FACH parameters as your recommendation, but I think that the RADIOLINK_REP_DL_SACCH = 14 is not so good because after changes the call drop due to cause “too low level in DL” increased almost triple…
And I can’t find and chack in RNPO the REP_DL/UL_SACCH_THRES and REP_DL/UL_SACCH_WS parameters.p.s.What do you think for the timers that I mentioned in previous post?
>>rrb_bg
28th September 2011 at 14:32 #67036pixGuestvanG,
regarding the timers, I have no experience. increasing them might reduce the drops but will increase the congestion. Here are defaults values:T3101 = 3s Supervision of the Immediate Assignment procedure.
T3103 = 16.5s Supervision of the Internal Handover procedure.
T3107 = 14s Supervision of the Assignment procedure.
T3109 = 12s Supervision of the Channel Release procedure.
T9113 = 20s Supervision of the External Handover procedure in the target BSC.
If you decrease, then you might reduce congestion but might increase drops and failures. And vice-versa. That would be a good trial to attempt.
——-
About the RADIOLINK_REP_DL_SACCH = 14, the default value is 10.
When you set it back to 10, is the Call Drop Level % going down ?That’s a weird behaviour… it should not!
What are your values for :
RADIOLINK_RECOVERY_THRES (a.k.a N_BSTXPWR_M)
RADIOLINK_TIMEOUT_BSforget about REP_DL_SACCH_THRES (etc.) those parameters are statically set in the system. You cannot modify them. 3 and 10 are good values.
Regards
pix28th September 2011 at 19:56 #67037VanGGuestHi Pix,
about this timers, what do you think how I can decrease the call drop due to cause “Other causes”. As I checked in documentation this “other causes” are due to Handovers and these timers are related to them…Is it possible to improve these type of drops and how?
About RADIOLINK_REP_DL_SACCH I’ll say tommorow, because I changed back to previous value= 30 and waiting for more statistics..
Thank you for your co-operations !!
Best regards,
p.s. What is your e-mail?29th September 2011 at 07:15 #67038pixGuestRADIOLINK_REP_DL_SACCH = 30 ??
how strange it is…why didn’t you put the default value ?
What are the values of the parameters I asked you, one post ago?About the email@, if you need to share some of your qos reports:
pix_erlang
*-at-*
yahoo *dot* combut please continue to post in this forum also 😉
29th September 2011 at 08:18 #67039VanGGuestHi Pix, thank you for your answer..
If RADIOLINK_REP_DL_SACCH is not related to my drop increase, what can be the problem changes?
I just changed RADIOLINK_REP_DL_SACCH=30,
REP_DL_FACCH_LEGACY_SUPPORT = 1;This my current parameter for cell that we tried repeated SACCH/FACCH :
RADIOLINK_RECOVERY_THRES=33;
RADIOLINK_FAILURE_THRES=36;
RADIOLINK_FAILURE_THRES_AMR=36;
RADIOLINK_TIMEOUT=32;
RADIOLINK_REP_DL_SACCH=30(at the moment);
EN_REP_DL_FACCH = 2;
EN_REP_SACCH=1(ENABLE);
REP_DL_FACCH_LEGACY_SUPPORT = 1;
L_RXQUAL_DL_P_AMR_RXACCH=4.0;
L_RXQUAL_UL_P_AMR_RXACCH=4.0;
REP_DL_FACCH_THRES_AMR_FR=12.2;
REP_DL_FACCH_THRES_AMR_HR=7.4;>>rrb_bg
29th September 2011 at 21:16 #67040pixGuestvanG,
OK, your radiolink timers are way too high ! I understand now why my previous settings had such impacts :))
Anyway, it proves that the repeated sacch is indeed working – by almost disabling it, your call drop % increased !
regards
pix30th September 2011 at 09:22 #67041VanGGuestI changed to the previous values L_RXQUAL_DL (UL) _P_AMR_RXACCH = 2.0 and REP_DL_FACCH_LEGACY_SUPPORT = 1 and now “Call Drop_DL level” decreased to values before the changes …
What do think about my other question? How can I improve the “Call Drop_Other cause”?
Thanks for your help …
Regards,1st October 2011 at 08:26 #67042pixGuestyou can’t improve this drop, because it contains all the drops where a HO cause where not verified during the drop.
it’s the default classification…regards
pix -
AuthorPosts
- The forum ‘Telecom Design’ is closed to new topics and replies.