TCH Time Congestion - E*******

[rezapa]

Hello guys,
Can anybody explain to me in which conditions in E******* BSC, the counter "TCH congestion times" (TFTCONGS) gets triggered? Specially I'm interested to know if GPRS traffic and 3G can cause triggering this counter?
How does paramters DTHAMR/NAMR affect this? I have a cell that returns too many TCH time congestion samples when I set DTHAMR/NAMR to low values (encouraging less HR), while it gives no such sample when I set them high.
My perception is that when the MS requests a FR-TCH and we don't have it, the TCH time congestion must increment. So how come when I set DTHAMR/NAMR to 99/99, which makes all the TCHs HR, there is no TCH time congestion?

[wsoong]

Hi,

I found the library explain the reason to increment the counter TFTCONGS as below:

When the last idle channel of the whole cell, used for signalling, data or speech is allocated and afterwards when the first time a new allocation attempt is done irrespective if preemption is successful or not, the TCH congestion time counters are started incrementing each second until a channel becomes idle.

When DTHAMR/NAMR is set to low value, meaning HR triggered is lower and only FR channel is available for new calls. Thus you can easily hit the last idle channel of the whole cell and counter TFTCONGS is incremented.

However when you set DTHAMR/NAMR is set to high value, meaning HR triggered is fast, thus last idle channel of the whole cell will not be easily hit since HR is indirectly double up your existing TRX capacity. You will still have a least last idle channel of the whole cell unless the site is really blocking and congested with 100% HR, i think you will still see TFTCONGS incremented.

Hope this help. :)

[Leonardo Camilo]

TFTCONGS Soft congestion time counter for underlaid subcell. The counter starts to increment when a channel is requested but no idle channels are available. The corresponding half-rate counter for overlaid subcells is named THTCONSUB. In the case of GPRS no consideration is made as to whether on-demand PDCHs exist in the cell or not i.e. both on-demand and fixed PDCHs are regarded as busy.TFTCONGS Soft congestion time counter for underlaid subcell. The counter starts to increment when a channel is requested but no idle channels are available. The corresponding half-rate counter for overlaid subcells is named THTCONSUB. In the case of GPRS no consideration is made as to whether on-demand PDCHs exist in the cell or not i.e. both on-demand and fixed PDCHs are regarded as busy.TFTCONGS Soft congestion time counter for underlaid subcell. The counter starts to increment when a channel is requested but no idle channels are available. The corresponding half-rate counter for overlaid subcells is named THTCONSUB. In the case of GPRS no consideration is made as to whether on-demand PDCHs exist in the cell or not i.e. both on-demand and fixed PDCHs are regarded as busy.
Hope to help you!

[electron]

Hello guys,
Can anybody explain to me in which conditions in E******* BSC, the counter "TCH congestion times" (TFTCONGS) gets triggered? Specially I'm interested to know if GPRS traffic and 3G can cause triggering this counter?
How does paramters DTHAMR/NAMR affect this? I have a cell that returns too many TCH time congestion samples when I set DTHAMR/NAMR to low values (encouraging less HR), while it gives no such sample when I set them high.
My perception is that when the MS requests a FR-TCH and we don't have it, the TCH time congestion must increment. So how come when I set DTHAMR/NAMR to 99/99, which makes all the TCHs HR, there is no TCH time congestion?

Hi Reza PAR...,

Suppose in a cell the last TCH channel occupied and till the time 1 TCH channel becomes idle ; actually as you know this is definition of Time congestion.

Regarding the counter TFTCONG and THTCONG(for FR and HR TCH channel request) they are step up whenever a request comes to a cell but there is no idle TCH channel to assign to user in this case based on channel type that requested each counter can step up.

About the condition for GPRS there is relation between GPRSPRIO parameter and this counters it depend which setting you selected. As practically observed when you change GPRSPRIO from 0 (default) to 1(On demand PDCH as busy) HR TCH Time congestion will decrease also HR utilization in CS domain will increase.

Draw back of changing DTHAMR/NAMR to 99 is even the time that cell does not have PS/CS traffic still is giving HR to users which will decrease SQI also connection will be more sensitive to interference and provability of TCH drop will increase. If you check some time when we change FR/HR setting TCH drop also will increase .

Hope it helps

[Leonardo Camilo]

Try to use this formula:
Formula: (TCH_CONG_FR*( TCH_T_FR_ER/TCH_T_ER)) + (TCH_CONG_HR * (TCH_T_HR_ER/ TCH_T_ER))

Where:

TCH_CONG_FR à (TFTCONGS|TFTCONSUB)/3600; where 3600 = PERLEN * 60
TCH_CONG_HR à (THTCONGS|THTCONSUB)/3600; where 3600 = PERLEN * 60
TCH_T_FR_ER à TFTRALACC / TFNSCAN
TCH_T_HR_ER à THTRALACC / THNSCAN
TCH_T_ER à (TFTRALACC / TFNSCAN) + (THTRALACC / THNSCAN)

[rezapa]

Hi,

I found the library explain the reason to increment the counter TFTCONGS as below:

When the last idle channel of the whole cell, used for signalling, data or speech is allocated and afterwards when the first time a new allocation attempt is done irrespective if preemption is successful or not, the TCH congestion time counters are started incrementing each second until a channel becomes idle.

When DTHAMR/NAMR is set to low value, meaning HR triggered is lower and only FR channel is available for new calls. Thus you can easily hit the last idle channel of the whole cell and counter TFTCONGS is incremented.

However when you set DTHAMR/NAMR is set to high value, meaning HR triggered is fast, thus last idle channel of the whole cell will not be easily hit since HR is indirectly double up your existing TRX capacity. You will still have a least last idle channel of the whole cell unless the site is really blocking and congested with 100% HR, i think you will still see TFTCONGS incremented.

Hope this help. :)
WSOONG, your reply was 100% clear and useful. Thanks very much.
BTW, do you have any doc that explains the counters in detail, including their trigger condition?
Thanks again.

[rezapa]

Thanks every one, answers were all so helpful. If you have a doc that explains counters in detail, I'd appreciate if you share it.