- This topic has 15 replies, 1 voice, and was last updated 14 years, 5 months ago by
.
-
Posts
-
Hi all,
we like to know the optimum value for T200 on LAPDm. All vendors have different default values (Siemens 145 ms, Nokia 220 ms, Satellite Abis 400 ms etc.) for this timer – but which value is the best to reduce SDCCH drops and to keep the retransmissions at an acceptable level ?
Example: SDCCH/8
During a 51er multiframe the SDCCH/8 occupies four consecutive TDMA frames (four bursts are sent). Than the MS / BTS has to wait for the next 51er multiframe (i.e. 235 ms) before the next Layer 2 frame could be sent. 145 ms / 220 ms are shorter than the 51er multiframe (235 ms) so in case of an missing acknowledgement this is always a T200 expiry. The SDCCH drop will occur if T200 expired N200+1 times. If the T200 is increased (for example to 500 ms) we have two 51er multiframe to get the acknowledgement and the SDCCH drops are reduced.
Kind Regards
Zbigniew
hi zbigniew,
i may be wrong, or i may have misunderstood your point, so don’t hesitate to correct me 🙂
T200 supervises the reception of the Ack of a Lapdm frame.
when BTS sends a lapdm I frame (= that requires an ack), the MS has an opportunity to send the ack in the next UL TDMA frame.
For the case of SDCCH. We start to count at TDMA Frame = 0
Let’s assume the BTS needs to send “big” L3 message, that requires a segmentation over 4 SDCCH bursts. It means that the message is going to be segmented in 4 Lapdm frames, each Lapdm “L2” frame is then sent in one TDMA “L1” burst.
the MS receives the 1st burst, which should be acknowledged. That was Frame 0.
Then it receives the 2nd one, and the 3rd one, and the 4th one (they are all sent in consecutive TDMA frames because of the way the SDCCH multiframe is multiplexed)
We are now at the end of frame 4.So the MS needs to ack the last burst received, the 4th one. It will be able to do so when it is his turn to send his UL SDCCH frame. Which is in frame 15 !
So T200 is running since Frame 0, awaiting for an ack. The frame 15 is just 69ms later. If MS doesn’t send his ACK now, it can still do it in frame 16, 17 or 18.
After T200 (=220ms), the BTS assumes the ACK is “lost”. It is correct for the BTS to assume so : the ack should have arrived a *long* time ago 🙂
After 235ms, the BTS has the opportunity to repeat the previous SDCCH msg because the ack was not yet received.
If you allow T200 to be longer, then the BTS misses one opportunity to repeat the SDCCH ! Of course, it is possible the MS will acknowledge the 1st message in the second multiframe, but it means the MS is very slow. It is safer to assume that the MS didn’t decode the previous SDCCH msg rather than assume the MS will eventually send an ACK later.I hope I was clear… and I hope I’m not too far off the truth 🙂
cheers and thanks for the interesting question !
pixHi Pix,
you understood my question exactly ! 😉
We did not know that the BTS is able to sent the ack in an earlier frame (frame 15 in your example) – we thought that the earliest acknowledge is possible in the next 51er frame. Do the UL and DL SDCCH frames have different “starting points” which leads to this “gap” ?
We made several investigations with different T200 settings. The SDCCH dropping seems to be reduced (SDCCH radio failures decrease in the counters) in Nokia and Siemens if the timer has larger values. From our point of view this must be related to bad HF conditions: for example the MS can’t hear the BTS, several retransmissions are done and the connection is ended because of Layer 2 failure (after N200+1*T200). If T200 is increased the chance to get better HF conditions (while MS is moving around) is higher because the “tear down time” for Layer 2 is increased as well.
Cheers and Thank you for replying
Zbigniew
Yes indeed, the SDCCH UL and DL multiframes are offset by 15 TDMA frames. The starting point is different 🙂
I’m not sure why you measure less drops when you increment the T200. Are you sure the improvements occur only in bad RF conditions ?
Default value was 145ms, which other values did you try ?Are we talking about the T200 for supervision of SDCCH SAPI0 ack, or for SDCC SAPI3 ack ?
pix
Hi Pix,
here is some more information.
In Siemens there are two parameters sdcchSAPI0 and sdcchSAPI3 which represent the T200. We changed them both equally in steps of 100 ms up to 1000 ms. In Nokia there is only one parameter T200S (T200F is for FACCH call setup which we don’t use) and it was incremented the same way. The range is from 145 (220) ms up to 1260 ms.
There are several counters for the SDCCH drops (i.e. Radio / BTS / Abis Failure, User Action etc). But the only counter that is decreased (if T200 is increased) is “SDCCH Radio Failure”. We guess that after (N200+1)*T200 the L2 is torn down, which will result in an “Error Indication” to L1, which will afterwards be torn down as well. Because the “SDCCH Radio Failures” are the main part in the SDCCH drops we think that this must related to bad RF conditions – but we haven’t found an exact method to make this “bullet proof”.
Why do you want to distinguish between SAPI0 and SAPI3 ?
SAPI0 = RR,MM,CC SAPI3 = SS,SMSThe T200 procedure is the same, isn’t it ?
Kind Regards
Zbigniew
Z,
yes the procedure is the same but the timers could have been different (they are different in alcatel)… i just wanted to make sure.
at which value do you start to see a decrease of radio failures ?
in my opinion, to give the 2nd chance of sending the ack to the MS, the improvement could be visible only starting at T200 = 310ms, approx. Which is 235ms + 80ms.
Then further improvements for the 3rd chance at 235 + 235 + 80 = 550ms…Thanks a lot for sharing all your findings, it is interesting. Even though it is still mysterious *why* it works 🙂
regards
pixHi Pix,
here are some results for a single Nokia cell (daily values):
DAY SDCCH Assign SDCCH Radio Fail SDCCH Radio Fail [%] T200
1 16088 506 3,15 220
2 16231 537 3,31 220
3 13879 372 2,68 300
4 12639 298 2,36 400
5 17751 218 1,23 500
6 15540 208 1,34 600
7 13732 203 1,48 700
8 14836 274 1,85 800
9 14867 296 1,99 900
10 14717 143 0,97 1000
11 12800 101 0,79 1100
12 15288 133 0,87 1200
13 22190 129 0,58 1260I agree with your calculation that improvements are only possibly in “steps” of 51er frames.
In your example you wrote that the MS is able to ack in frames 15,16,17,18. So the minimum timer value must be 18 frames = 82 ms.
But I think that in the opposite direction (in case the BTS has to ack L2 messages) the timer must be set to greater values:MS sends in frames 15,16,17,18
BTS acks in frames 1,2,3,4The minimum timer value will then be 51-15+4=40 frames = 184 ms
The strange thing is that the Siemens default value (145 ms) is below that minimum. The Nokia default value (220 ms) seems to fit if the ack has to be sent in the same 51er frame.
Which are the default values for T200 in Alacatel ? Which average SDCCH Radio Failure Rate do you see in your network ?
Tnank you for this interesting discussion !
Zbigniew
Z,
The Qos Stats show a clear & strong corelation between the KPI and the parameter T200. That’s amazing !
Regarding the Ack from the BTS, I’m not sure (and I’m tired to look in the 3GPP specs 🙂 ). Take the subchannel “0” from the SDCCH ts.
in DL : the BTS sends SDCCH/0 on burst 0, 1, 2, 3 and the SACCH/0 on burst 32, 33, 34, 35
in UL : th MS sends SDCCH/0 on burst 15, 16, 17, 18
and the SACCH/0 on 47, 48, 49, 50And I ***believe*** that the Lapdm acknowledgments can be sent on either the SACCH or the SDCCH, since both of them are sent over the same LapDm link. I’m not sure at all about this though, but it sounds logical.
The value of the SDCCH Drop due to Radio failures (in ALU) is usually around 1% in a fairly good network. The SDCCH Drop due to Radio Failures is a counter that encompasses both the Radio Link Timeout and the “T200*N200+1 times” failures.
I am not able to test your changes because I am not working on a live network (i am a gsm trainer, living in a world of theory…)
T200 = 220ms for sDCCH SAPI0
= 450ms for SDCCH SAPI3
= 900ms for SACCH associated to SDCCHLooking at those timers, I fear that the LapDm link for SDCCH is different than the LapDm link for the SACCH of the SDCCH… I’ll let you look it up though 🙂
pix
Hi Pix,
I think that the higher value for SACCH (900 ms) is related to the different N200 values which depend on the channel type (GSM 04.08)
Channel Type N200
SACCH 5
SDCCH 23
FACCH (FR) 34
FACCH (HR) 29So the products T200*(N200+1 are nearly the same:
(5+1) * 900 ms = 5.4 sec
(23+1) * 220 ms = 5.3 secZbigniew
Hi Zbigniew,
I m supervising Nokia Network. I saw your KPI SD fail decreased by tune T200.But how do you tune T200? And when you increased T200, are there other KPIs change ? ( worse, better..?)
Thanks!so wt effect does it produce if I keep my RET(max no of retransmission) to a higher value(say 4) and reduce the T200 value..
Thanks pix & zbigniew.
It has been a nice discussion.I work with siemens system & i just started working with this t200 last couple of days ago.
So far i’ve seen, this parameter description is not that clear for siemens and i was not getting enough confidence to change this value.
I was changing this value with a step of 5.
After following the calculation provided by u two, i got an abrupt positive result. Thanks again to both of u.My SD problem is solved.
But, according to SPOTS, i’m getting TCH (EIT200EXPF, within NRFLTCH) drop for t200 in case of full rate.Is it facchTCHF to change to decrease TCH drop ??
Plz kindly explain in detail like before.
Titumeer !!
What is T200S ?? I NEED FULL EXPLANATION…..i am very new to this term
hi titumeer,
plz tune the value of facchF/H to get better results
Hi All,
your discussion is very intersting and helful.
I have read that the minimum value for T200 regarding SDCCH should be at the reception of the 18 multli frame meaning 8*0,577*18=83 ms
Huawei default value is 60 msec meaning at the 13th frame.
Have you any explanation???It seems that the T200 expires before the MS has the opportunity to send the ack!
- The forum ‘Telecom Design’ is closed to new topics and replies.
