Congestion and block rate

[dacoder]

Can anyone please explain to me the difference between sdcch congestion rate and sdcch block rate ( also, tch congestion and block rate). Do they both mean the same thing?

Also what is meant by sdcch mean holding time?

[s52d]

Can anyone please explain to me the difference between sdcch congestion rate and sdcch block rate ( also, tch congestion and block rate). Do they both mean the same thing?

Also what is meant by sdcch mean holding time?

In short: yes.
I find it funny: over years I've seen several definitions/statistics formulas, giving
simmilar results. Different enough so I can present any number, and defend it.

As example, I might say for a cell: CellX had 36 seconds congestion, meaning
in peak busy hour 36 out of 3600 seconds there were no free TCH in a cell.
This maps to 1% of time, and roughly 1% of calls can not be made.

Same call has 3 calls out of 1000 blocked due to no TCH available in tha same peak busy hour.
How come it is not 10? Others were passed to another cells (direct assignment) or some other
GSM mechanism was involved. So, it is 0.3% blocking.

Both numbers are correct. Both are different. Both measure same end user experience.


Mean holding time measures average session length on SDCCH.
It is good indicator on traffic mix: LA update, SMS or voice call each take different time.
Also, cell MHT should be lower as BSC MHT, because of handovers.
MHT is calclulated by deviding traffic (erlangs in seconds) with number of calls to get average call duration.

Years ago we found EIR SS7 problem, when we started investigating SDCCH mht increase.
(In fact, we noticed higher SDCCH traffic, checked mht and so on).

BR
s52d

[dacoder]

Thanks s52d....
but i'm confused in this: if there is high sdcch congestion then there obviously will be sdcch block. But i think sdcch block can occur even when there is no sdcch congestion due to environment problem or hardware problem. Is it so?

About MHT: suppose for voice call if MHT is 2sec then does it mean that when voice call is attempted, sdcch will be occupied for 2sec and then released thereafter?

[tsiz]

In short: yes.
I find it funny: over years I've seen several definitions/statistics formulas, giving
simmilar results. Different enough so I can present any number, and defend it.

As example, I might say for a cell: CellX had 36 seconds congestion, meaning
in peak busy hour 36 out of 3600 seconds there were no free TCH in a cell.
This maps to 1% of time, and roughly 1% of calls can not be made.

Same call has 3 calls out of 1000 blocked due to no TCH available in tha same peak busy hour.
How come it is not 10? Others were passed to another cells (direct assignment) or some other
GSM mechanism was involved. So, it is 0.3% blocking.

Both numbers are correct. Both are different. Both measure same end user experience.


Mean holding time measures average session length on SDCCH.
It is good indicator on traffic mix: LA update, SMS or voice call each take different time.
Also, cell MHT should be lower as BSC MHT, because of handovers.
MHT is calclulated by deviding traffic (erlangs in seconds) with number of calls to get average call duration.

Years ago we found EIR SS7 problem, when we started investigating SDCCH mht increase.
(In fact, we noticed higher SDCCH traffic, checked mht and so on).

BR
s52d

No. As far as I know, both are not the same. TCH congestion is soft blocking and TCH blocking is the actual hard blocking where the user is actually blocked.

If you see BH data..cells will have TCH congestion but not have blocking although after a certain time, when there are no time slots available, extremely high TCH congestion will lead to TCH blocking.
Correct me if I am wrong.

[rasar00]

No. As far as I know, both are not the same. TCH congestion is soft blocking and TCH blocking is the actual hard blocking where the user is actually blocked.

If you see BH data..cells will have TCH congestion but not have blocking although after a certain time, when there are no time slots available, extremely high TCH congestion will lead to TCH blocking.
Correct me if I am wrong.

nice informationthank you everyone

[dacoder]

If congestion and block rate are two different things, can anyone provide their respective formula?

[no_1'sfasterthanradiowave]

Dear dacoder,

as explained you can asume congestion as time during which all resources were occupied, therefore kpi would be: time_without_free_resources/total_measurment_period [%], e.g. you had all channels available occupied simulataneously for 6 minutes during one hour that makes 6/60 or 10%.
blockinr rate would be number_of_blocked_calls/number_of_requests [%] during the measurment period.
hope that helps!

[uel888]

Nice to know, i believed its not the same the congestion and block. :-)

[plannerguy]

Hi

: Blocking---% of channel requests blocked or rejected due to no available channel(Attempts)

Congestion-Total time during which no channel was available.(Blocks)

Congestion = time when all resources are occupied (no free TCH available)
Blocking = rejected (blocked) attempts over all attempts in %.

Blocking dives you the non served calls
congestion gives the time when no resource are available (it is possible that nobody needs them so no blocks)
there is a possibility to have high blocking and low congestion - this means that you have a peak of the attempts.

Congestion is time based (i.e at particular time say IN Busy Hour) and blocking is resource based (can be any time due to some issues like H/w, no TCH available) .


In simpler words:
Suppose there is cell named "Beta"....if all Channels are full then system will report it as congestion.Block occurs when a attempt is made into cell beta....
it may be direct ot handover attempts



Regards
Plannegruy:(

[dacoder]

So high congestion rate means channel resources are unavailable as they are fully utilised. Does this mean that higher the congestion rate, the more beneficial to the Service provider as they get to generate more revenue ? But this also means customer needs to repeatedly try to establish a call. Am i right?

[tsiz]

So high congestion rate means channel resources are unavailable as they are fully utilised. Does this mean that higher the congestion rate, the more beneficial to the Service provider as they get to generate more revenue ? But this also means customer needs to repeatedly try to establish a call. Am i right?

Higher the congestion rate does not mean beneficial to the operator. its the opposite. You need to add resources to counter it. Customers do not like to be waiting for the call to connect and be blocked.