SINR & effective SINR

[K.TruongDinh]

Hi there,
what is the different between these 2 parameter ? SINR and effec_SINR?
is there any place in the LTE-Sim, that SINR is computed ? or it had
been computed before ?
as i see here ? double SINRForCQIIndex[15] = {
-4.63, -2.6, -0.12, 2.26, 4.73,
7.53, 8.67, 11.32,
14.24, 15.21, 18.63, 21.32,
23.47, 28.49, 34.6
};
is there any method implemented here to compute SINR in depending on
PR(power trasnmision) ?
thank you
Khoa

[Giuseppe Piro]

Hi there,

2011/7/25 K.TruongDinh <ktruo...@googlemail.com>

Hi there,
what is the different between these 2 parameter ? SINR and effec_SINR?

The effective SINR is a SINR value obtained as a weighted average among SINRs of a set of sub channels .

It is used in two cases:

- into the packet scheduler for choosing the MCS to use for data transmission, knowing the SINRs (or, if you prefer CQIs) sent by the scheduled UE for the assigned sub channels

- into the error model for selecting the BLER knowing both the effective SINR and the MCS used during the transmission

 

is there any place in the LTE-Sim, that SINR is computed ? or it had
been computed before ?
 as i see here ?              double SINRForCQIIndex[15] = {
                                      -4.63, -2.6, -0.12, 2.26, 4.73,
                                     7.53, 8.67, 11.32,
                                     14.24, 15.21, 18.63, 21.32,
23.47, 28.49, 34.6
                                        };

The previous code represent a set of SINR thresholds assigned to each CQI value. I.e., if you receive a CQI equal to 1, it means that the SINR is at most equal to -4.63 dB. 

 

is there any method implemented here to compute SINR in depending on
PR(power trasnmision) ?

Yes. have a look to the UeLtePhy::StartRx function. 

Cheers,

Giuseppe

[Florian Schlembach]

I would also like to put out the SINR value depending on the received
power level? As I am running the MultiCell scenario there is no sinr
put out if I am defining the AMC-MAPPING feature in global_config/load-
parameter (btw: which one is loaded exactly?).
As I am writing some couts to the StartRx function for debugging there
is nothing given to console. When is the StartRx() function invoked?
Does the UEs have to receive something to get the StartRx() fct get
invoked?

To sum up: How could i print out those sinr easily?

On Jul 25, 3:03 pm, Giuseppe Piro <pe...@giuseppepiro.com> wrote:
> Hi there,
>
> 2011/7/25 K.TruongDinh <ktruong1...@googlemail.com>

[Giuseppe Piro]

Hi there,

2011/7/31 Florian Schlembach <fschl...@web.de>

I would also like to put out the SINR value depending on the received
power level? As I am running the MultiCell scenario there is no sinr
put out if I am defining the AMC-MAPPING feature in global_config/load-
parameter (btw: which one is loaded exactly?).

Did you compile again LTE-Sim ? 

By enabling AMC-MAPPING you should have an output like this one:

RX INF_BUF ID 2338 B 0 SIZE 1424 SRC -1 DST 2 D 0.001

sinr: 14.087 14.1027 14.12 14.1385 14.1578 14.1775 14.1971 14.2161 14.2339 14.2502 14.2644 14.2761 14.2851 14.291 14.2937 14.2931 14.2892 14.2822 14.2722 14.2595 14.2446 14.2278 14.2097 14.1909 14.1718 14.1531 14.1353 14.1189 14.1043 14.0919 14.0818 14.0741 14.0687 14.0656 14.0646 14.0652 14.0672 14.0702 14.0739 14.0782 14.0827 14.0874 14.0923 14.0974 14.1028 14.1089 14.1158 14.1239 14.1334 14.1446 

CQI: 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 

MCS: 15 15 15 15 15 15 15 15 15 17 17 17 17 17 17 17 17 17 17 17 17 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 

TB: 230 230 230 230 230 230 230 230 230 289 289 289 289 289 289 289 289 289 289 289 289 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 230 

time: 2.84

position: 642.222 57.179

 

As I am writing some couts to the StartRx function for debugging there
is nothing given to console. When is the StartRx() function invoked?

The function is called by the channel after the propagation. Probably your modification didn't cause any effect because of you didn't compiled the code again. 

 

Does the UEs have to receive something to get the StartRx() fct get
invoked?

If the UE is attached to the channel, the method StartRx is always called, also when there are no packets sent to that device. 

 

To sum up: How could i print out those sinr easily?

Probably provided debugging functionalities are complicated for your purposes. You ave to modify the code in order to print only values that are interesting for your wirk. 

How ?  

- compute the value or take it from a variable already present into the code

- add std::cout << ..... << std::endl; codes

- compile again LTE-Sim

Cheers,

Giuseppe

[Florian Schlembach]

Thanks a lot for your reply! I already did what you have mentioned:
-recompiled after making any changes (i.e. make clean and then make)
-enabled AMC-MAPPING

What i want to do?
I have now implemented a new mobility model (Manhattan) to LTE-Sim and
want now evaluate it by shifting the UEs through a multicell scenario
and plot the SINR.
Therefore, i tried modifying the multicell scenario by commenting out
the applications stuff. But now there is nothing put out. I just found
out that the AMC-DEBUGGING stuff is print out if i am switching on any
application (e.g. VOIP) like:

./LTE-Sim MultiCell 1 1 1 1 0 0 0 1 1 30 500 256 1

Hence, I asked if there has to be send any traffic to get the StartRx
function invoked because after switching of the applications by

./LTE-Sim MultiCell 1 1 1 0 0 0 0 1 1 30 500 256 1

i am just getting this output:

time: 7.783
position: -393.006 322.273

To see whether the StartRx() function was invoked or not, i added a
simply cout:

m_measuredSinr.clear();
std::cout << "BLA" << endl;
//COMPUTE THE SINR
std::vector<double> rxSignalValues;
std::vector<double>::iterator it;

There is no "BLA" print out. :-( What do you think is missing to get
the StartRx() invoked?

On Aug 1, 9:50 am, Giuseppe Piro <pe...@giuseppepiro.com> wrote:
> Hi there,
>

> 2011/7/31 Florian Schlembach <fschlemb...@web.de>

[Giuseppe Piro]

Hi there,

2011/8/1 Florian Schlembach <fschl...@web.de>

Thanks a lot for your reply! I already did what you have mentioned:
-recompiled after making any changes (i.e. make clean and then make)
-enabled AMC-MAPPING

What i want to do?
I have now implemented a new mobility model (Manhattan) to LTE-Sim and
want now evaluate it by shifting the UEs through a multicell scenario
and plot the SINR.
Therefore, i tried modifying the multicell scenario by commenting out
the applications stuff. But now there is nothing put out. I just found
out that the AMC-DEBUGGING stuff is print out if i am switching on any
application (e.g. VOIP) like:

ok. The UE is attached to the channel (hat it means the UE is able to receive signals) if there are one or more active applications. 

In your case, you must force the attachment by using the following code ,after the creation of the propagation loss model:

eNBs->at (j)->GetPhy ()->GetDlChannel ()->AddDevice(ue);

Cheers,

Giuseppe

[Florian Schlembach]

Thanks a lot, that works fine now!

On Aug 1, 6:20 pm, Giuseppe Piro <pe...@giuseppepiro.com> wrote:
> Hi there,
>

> 2011/8/1 Florian Schlembach <fschlemb...@web.de>

[Tamer Darweesh]

Hi Giuseppe,

It seems the calculation of effective SINR is useless in the current implementation. The UE sends only 1 CQI value (not 1 CQI value per sub carrier). So in the function SelectUsersToSchedule, I see that we only set this single CQI value in m_channelContition. Hence this vector will only have a single value, there will be no averaging taking place in the function GetEesmEffectiveSinr and the effective SINR will be always the same as SINR.

Regards,

Tamer

[Giuseppe Piro]

Il giorno domenica 9 dicembre 2012, Tamer Darweesh ha scritto:

Hi Giuseppe,

It seems the calculation of effective SINR is useless in the current implementation. The UE sends only 1 CQI value (not 1 CQI value per sub carrier). So in the function SelectUsersToSchedule, I see that we only set this single CQI value in m_channelContition. Hence this vector will only have a single value, there will be no averaging taking place in the function GetEesmEffectiveSinr and the effective SINR will be always the same as SINR.

This is true only when 

- the wideband CQI reporting scheme is used

- the selected channel model and the low speed do not guarantee a variation of CQIs estimated for each sub channel.

 Cheers,

Giuseppe

Regards,

Tamer

On Monday, July 25, 2011 3:03:22 PM UTC+2, Giuseppe Piro wrote:

Hi there,

2011/7/25 K.TruongDinh <ktruo...@googlemail.com>

Hi there,
what is the different between these 2 parameter ? SINR and effec_SINR?

The effective SINR is a SINR value obtained as a weighted average among SINRs of a set of sub channels .

It is used in two cases:

- into the packet scheduler for choosing the MCS to use for data transmission, knowing the SINRs (or, if you prefer CQIs) sent by the scheduled UE for the assigned sub channels

- into the error model for selecting the BLER knowing both the effective SINR and the MCS used during the transmission

 

is there any place in the LTE-Sim, that SINR is computed ? or it had
been computed before ?
 as i see here ?              double SINRForCQIIndex[15] = {
                                      -4.63, -2.6, -0.12, 2.26, 4.73,
                                     7.53, 8.67, 11.32,
                                     14.24, 15.21, 18.63, 21.32,
23.47, 28.49, 34.6
                                        };

The previous code represent a set of SINR thresholds assigned to each CQI value. I.e., if you receive a CQI equal to 1, it means that the SINR is at most equal to -4.63 dB. 

 

is there any method implemented here to compute SINR in depending on
PR(power trasnmision) ?

Yes. have a look to the UeLtePhy::StartRx function. 

Cheers,

Giuseppe

--
You received this message because you are subscribed to the Google Groups "lte-sim" group.
To post to this group, send email to lte...@googlegroups.com.
To unsubscribe from this group, send email to lte-sim+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lte-sim?hl=en

--
Giuseppe Piro, PhD
Post Doc Researcher
DEE, Politecnico di Bari
via Orabona 4 - 70125 (Bari), Italy.
email: g.p...@poliba.it
phone: +39 080 5963301
web: telematics.poliba.it/piro