Atoll Reference Signal Threshold

[kotarominami]

Hi,
I am newbie with Atoll
Now, I am working on LTE coverage planning with Atoll...

I confuse about parameter "Reference Signal C/N Threshold"...

Anyone can help me to explain about this parameter for LTE planning? How to determine the value of Reference Signal C/N Threshold for LTE?

In Atoll, default value is -19.5 dB...

I was capture the picture...I found this parameter at "station template" and when I set the "transmitter"...

Thank you for attention and for help...

http://farm6.static.flickr.com/5268/5626555353_a8c8681815.jpg (http://www.flickr.com/photos/55078200@N08/5626555353/)

[Resistor]

Hello Kotaro Minami,

Here is the answer to your question:

1) WHAT THIS PARAMETER IS
----------------------------
This is a parameter used in simulations only. When users are simulated in Atoll their C/I is calculated at each iteration. If it happens to be above that threshold then the diversity mode for that user is set to SU-MIMO, if the value is below this threshold then the diversity mode is set to STTD/MRC. This parameter is only relevant if:

- the infrastructure has diversity antennas;
- the network is allowed to use MIMO;
- a function called AMS (Automoatic MIMO Selection) is activated.

Without any of the above three conditions the settings for this parameter will not be used.

This parameter is per transmitter because the decision to switch from STTD/MRC diversity mode to MIMO is made per user on transmitter level.

2) HOW TO CHOOSE A VALUE FOR THIS PARAMETER
-------------------------------------------------
Setting this parameter affects the throughput for the given simulated user and thus the KPIs of the network at the end of the simulation. If this value is increased (say from -19.5dB to -18.5dB) then the average throughput per user at the end of the simulation will go down. Respectively, if this value is decreased (say from -19.5dB to -20.5dB) then the average throughput per user will go up. In order to decide what value to use, you should obtain data from the equipment manufacturer about how effective MIMO is and what is the minimum recommended C/I for the MIMO to be effective. Then you should use this minimum value as the said threshold. I personally do not have yet experience with what values of this parameter will make sense -- one good source for this is to find drive-test data (I have previously posted one such example of a drive test by NSN in this forum here (http://www.finetopix.com/live-optimization/17382-lte-performance-initial-deployments-white-paper-nsn.html)). As a gut feeling, I would expect the values to be mainly positive, the practical range being from -5dB to about +20dB. From what I understand about the physical link of LTE, I would expect this parameter to be between 1-3dB and 6-9dB. (Why? Because unlike in CDMA, where Ec/Io is the ratio of the (useful signal) to (the sum of the useful signal and the interference) and therefore it by definition is negative, when expressed in dB, here in LTE C/I is the ratio of the useful signal to the interference, and therefore it is expected to be be mainly positive, just like it used to be in GSM, AMPS, TACS, NMT, etc.). I am a bit perplexed that Atoll defaults to such a low value as -19.5 dB. This effectively means that the setting will allow MIMO to be used at all times in all cases. From technical point of view here is what you would do, in order to select the value for this parameter:

- find from your equipment provider when MIMO becomes efficient (at what C/I value) and use that value for this threshold;
- if you do not have access data about MIMO, then find from the link budget the sensitivity of your typical UE, from its receiver bandwidth find the thermal noise and combine the two to produce C/N (carrier to noise threshold). Then using the formula for hexagonal cells and the average cell radius and Okumura-Hata (or another model) find the interference created by adjacent cells. Use this number to get C/(I+N) (carrier to interference and noise). Then apply to it a margin to compensate for uncertainty in the calculations and use this value as this threshold. This number is likely to be between 1-3dB and 6-9dB, I would guess from experience with other system using broadband signals and a combination of FDMA/TDMA like LTE does. Sorry that I cannot give you any more precise estimate than this

3) MY PERSONAL RECOMMENDATION
----------------------------------
I personally would recommend that you leave this value to its default setting. Why? Because:

- its default setting will make your modelled network look more efficient with higher throughput (good thing for your project -- it will look better);
- it will make all calculation on the basis of using MIMO 100% of the time in 100% of the cases (using MIMO all the time appears to be a good idea, people would think so, otherwise why at all implement it?)
- by the sound of it you do not have much data to make an informed decision (I certainly do not have it myself).

Regards,
Resistor

[kotarominami]

Thank you very much for your reply & explanation,,,:D
I think, better I follow your recommendation for my simulation...

[rval25]

ok this is a very reasonable answer

some days ago I was thinking about this parameter