Change timer T305 and T308 in E///

[auto_art]

hi
one of my collegue changed timer T305 and T308 in E/// from 30s to 1s.

and he has observed in improvement in tch rop rate from 0.9 to 0.8 at nbh.

pl share your views.

[zhanglw268]

I assume that your coleague changed the MSC side timer T305 and T308, your coleague did a "clever" trick to let the MSC clear the call early before the BSS side peg the drop call statistics. In long run, your coleague just plays around these timer parameters and doesn't improve the network quality. I saw similar actions before.

Here are the explanation what these 2 timers are used for.

T305 timer is started at the MSC/MS when DTAP message DISCONNECT is sent, this will guard the DISCONNECT procedure run correctly in T305 period.

T308 timer is started at the MSC/MS when DTAP message RELEASE is sent after the DISCONNECT procedure, this timer guards how long the RELEASE procedure runs.

(GSM 04.08 section 5.4 call clearing)

T305 and T308 at MSC should be set to no less than 11 seconds to guarantee that the layer 2 T200 is timing out N200 times on the air interface, reducing these timers to less than 11 seconds will pull them below the related BSS timers[T200*(N200+1) in this case] at the air
interface that are used to ensure a successful transmission of a message (DISCONNECT and RELEASE in this case) to/from the mobile over the air interface. Normally the higher level MSC timers would be set longer than the related lower level times in the BSS, using a strategy that lower level timers should expire first.

Setting T305 and T308 to less than 11 seconds here is forced the higher level MSC timers to expire before the BSS timers[T200*(N200+1) and link_fail(only for MSC issuing DISCONNECT/RELEASE first)] when there are air interface transmission problems (RF interference) for DISCONNECT and RELEASE signalling, and as a result the BSS statistics SD/TCH drop doesn't get a chance to peg for T200*(N200+1).

The end user experience is probably not that relevant here as they were trying to clear the call anyway. The end user would see no difference.

However the network operator could be fooled here though thinking that he has a clean RF interface with a lower drop call rate, when in fact there could be RF interference issues to be resolved. The low MSC timers are just masking it.

[auto_art]

I assume that your coleague changed the MSC side timer T305 and T308, your coleague did a "clever" trick to let the MSC clear the call early before the BSS side peg the drop call statistics. In long run, your coleague just plays around these timer parameters and doesn't improve the network quality. I saw similar actions before.

Here are the explanation what these 2 timers are used for.

T305 timer is started at the MSC/MS when DTAP message DISCONNECT is sent, this will guard the DISCONNECT procedure run correctly in T305 period.

T308 timer is started at the MSC/MS when DTAP message RELEASE is sent after the DISCONNECT procedure, this timer guards how long the RELEASE procedure runs.

(GSM 04.08 section 5.4 call clearing)

T305 and T308 at MSC should be set to no less than 11 seconds to guarantee that the layer 2 T200 is timing out N200 times on the air interface, reducing these timers to less than 11 seconds will pull them below the related BSS timers[T200*(N200+1) in this case] at the air
interface that are used to ensure a successful transmission of a message (DISCONNECT and RELEASE in this case) to/from the mobile over the air interface. Normally the higher level MSC timers would be set longer than the related lower level times in the BSS, using a strategy that lower level timers should expire first.

Setting T305 and T308 to less than 11 seconds here is forced the higher level MSC timers to expire before the BSS timers[T200*(N200+1) and link_fail(only for MSC issuing DISCONNECT/RELEASE first)] when there are air interface transmission problems (RF interference) for DISCONNECT and RELEASE signalling, and as a result the BSS statistics SD/TCH drop doesn't get a chance to peg for T200*(N200+1).

The end user experience is probably not that relevant here as they were trying to clear the call anyway. The end user would see no difference.

However the network operator could be fooled here though thinking that he has a clean RF interface with a lower drop call rate, when in fact there could be RF interference issues to be resolved. The low MSC timers are just masking it.

do you kno the command to change the same form MSC side.?

[auto_art]

I assume that your coleague changed the MSC side timer T305 and T308, your coleague did a "clever" trick to let the MSC clear the call early before the BSS side peg the drop call statistics. In long run, your coleague just plays around these timer parameters and doesn't improve the network quality. I saw similar actions before.

Here are the explanation what these 2 timers are used for.

T305 timer is started at the MSC/MS when DTAP message DISCONNECT is sent, this will guard the DISCONNECT procedure run correctly in T305 period.

T308 timer is started at the MSC/MS when DTAP message RELEASE is sent after the DISCONNECT procedure, this timer guards how long the RELEASE procedure runs.

(GSM 04.08 section 5.4 call clearing)

T305 and T308 at MSC should be set to no less than 11 seconds to guarantee that the layer 2 T200 is timing out N200 times on the air interface, reducing these timers to less than 11 seconds will pull them below the related BSS timers[T200*(N200+1) in this case] at the air
interface that are used to ensure a successful transmission of a message (DISCONNECT and RELEASE in this case) to/from the mobile over the air interface. Normally the higher level MSC timers would be set longer than the related lower level times in the BSS, using a strategy that lower level timers should expire first.

Setting T305 and T308 to less than 11 seconds here is forced the higher level MSC timers to expire before the BSS timers[T200*(N200+1) and link_fail(only for MSC issuing DISCONNECT/RELEASE first)] when there are air interface transmission problems (RF interference) for DISCONNECT and RELEASE signalling, and as a result the BSS statistics SD/TCH drop doesn't get a chance to peg for T200*(N200+1).

The end user experience is probably not that relevant here as they were trying to clear the call anyway. The end user would see no difference.

However the network operator could be fooled here though thinking that he has a clean RF interface with a lower drop call rate, when in fact there could be RF interference issues to be resolved. The low MSC timers are just masking it.

Hi pl go through attachement..it is detailed explaination.

i recieevd it from E*******.!

it includes commands to change too..!

[auto_art]

hey did u people aplied the same in your network.?

it is realy giving good results.

[parsarka]

Hi,Could you please tell me what are the names of these timers in NSN systems and the command to execute the same...please let me know as i need it urgently..thnx

[justdream]

(GSM 04.08 section 5.4 call clearing)

T305 and T308 at MSC should be set to no less than 11 seconds to guarantee that the layer 2 T200 is timing out N200 times on the air interface, reducing these timers to less than 11 seconds will pull them below the related BSS timers[T200*(N200+1) in this case] at the air
interface that are used to ensure a successful transmission of a message (DISCONNECT and RELEASE in this case) to/from the mobile over the air interface. Normally the higher level MSC timers would be set longer than the related lower level times in the BSS, using a strategy that lower level timers should expire first.

Dear, could you please share GSM Spec. DOc that contains this information?

[fmomani]

Hi Guys,
we tried to change the timers as below in our network (E/// Core):
T305: 30 --> 2 Sec
T308: 30 --> 1 Sec

we have witnessed around 13% decrease in E/// BSC TCH Drop call rates and 40% decease in Huawei BSC TCH Drop Call rates.
however, the reduction in call drops were noticed in all kind of drops (for Huawei) and not only T200 expiry.

[takeiteasy612]

I assume that your coleague changed the MSC side timer T305 and T308, your coleague did a "clever" trick to let the MSC clear the call early before the BSS side peg the drop call statistics. In long run, your coleague just plays around these timer parameters and doesn't improve the network quality. I saw similar actions before.

Here are the explanation what these 2 timers are used for.

T305 timer is started at the MSC/MS when DTAP message DISCONNECT is sent, this will guard the DISCONNECT procedure run correctly in T305 period.

T308 timer is started at the MSC/MS when DTAP message RELEASE is sent after the DISCONNECT procedure, this timer guards how long the RELEASE procedure runs.

(GSM 04.08 section 5.4 call clearing)

T305 and T308 at MSC should be set to no less than 11 seconds to guarantee that the layer 2 T200 is timing out N200 times on the air interface, reducing these timers to less than 11 seconds will pull them below the related BSS timers[T200*(N200+1) in this case] at the air
interface that are used to ensure a successful transmission of a message (DISCONNECT and RELEASE in this case) to/from the mobile over the air interface. Normally the higher level MSC timers would be set longer than the related lower level times in the BSS, using a strategy that lower level timers should expire first.

Setting T305 and T308 to less than 11 seconds here is forced the higher level MSC timers to expire before the BSS timers[T200*(N200+1) and link_fail(only for MSC issuing DISCONNECT/RELEASE first)] when there are air interface transmission problems (RF interference) for DISCONNECT and RELEASE signalling, and as a result the BSS statistics SD/TCH drop doesn't get a chance to peg for T200*(N200+1).

The end user experience is probably not that relevant here as they were trying to clear the call anyway. The end user would see no difference.

However the network operator could be fooled here though thinking that he has a clean RF interface with a lower drop call rate, when in fact there could be RF interference issues to be resolved. The low MSC timers are just masking it.

Hi zhanglw268
Your information is very usefull, do you have the document that contain the information you've post ? If you know the document's name, it's very usefull for me.

[mirtunjay_1975]

Hi firends coud u send your analysis after change T305 from 30 to 2 sec an T308 from 30to 1 sec.

your support will be highly appreciable.