- This topic is empty.
26th November 2012 at 14:08 #68740HendrikGuest
I have got a question concerning DTX in UL/DL.
I did some measurements with a mobile phone in a GSM network and took a look at the measurement reports. I often got weird results for RXQUAL before a handover took place. The RXQUAL-values were 0 for quite a long time and than suddenly fell to 7. It looked like this 0-0-0-0-0-0-0-7-7-handover.
These values are the RXQUAL_sub values. However, I dodn’t know wheter DL DTX is turned on.
My question no is whether UL DTX could affect the measurement of DL RXQUAL. Actually, this should not be the case but I am asking to be sure.
Thank you very much for your support.
/Hendrik26th November 2012 at 19:26 #68741pixGuest
UL DTX cannot affect DL Quality (sub or full)
DL DTX affect DL Quality full but not DL Quality Sub.
Are the two last measurements show rxqual dl = 7 ?
Is it more than 2 measurements or less ?
I would say that the Bad Quality in DL caused the HO to be triggered. Is the HO COMMAND (layer 3) is received before the QUAL=7 or after ?
Please check your Qos stats : HO DL Quality Rate. Is it high ? (> 30%)
Are there a lot of drops ?
a drive test mobile would actually give you more than the rxqual : it will also show the BER. If you can find it, then what’s its value ?
pix27th November 2012 at 09:28 #68742HendrikGuest
thank you very much for your support. I got a lot situations where the quality was like this 0-0-0-0-0-0-0-0-7-7-handover-0-0 (values of RXQUAL sub).
However, when I use the RXQUAL_full values (which I have to if DTX DL is disabled), the values are, e.g. 2-3-2-1-handover-2-2-2-0-0 which I can explain.
The handover command came before the bad RXQUAL values. I don’t know why I haven’t had a look at this earlier…
It seems that my evaluation software or the phone does soem strange things since the serving cell changes tree measurement reports after the HANDOVER COMPLETE message.
Unfortunately, I have to rely on a “normal” phone. Thus, I cannot provide any BERs.
Thank you very much again! I just wanted to make sure that I don’t have any wrong ideas.
/Hendrik27th November 2012 at 19:20 #68743pixGuest
“The handover command came before the bad RXQUAL values. ”
ok, that’s the key. I notice this kind of behaviour too, and there’s nothing wrong with it.
Right after a HO CMD, the MS is not listening to the serving cell anymore, so measurements are not valid.
So problem closed 🙂