3GPP TR 36.873 calibration results

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #7442

    Hello? I`m graduate school sutdent of Hanyang University in Korea.
    First of all, I`d like to give you thanks for implementing this system level simulator.
    This simulator is very useful for understanding many things regarding system simulation and very helpful to my research.
    By the way, when I read the book named ‘The Vienna LTE-Advanced Simulators’, I recognized that there are calibration results compared with other company in Chapter 10.
    In TR 36.873, large scale calibration results are coupling loss, wideband SINR. But I cannnot find these results in your book and in ‘reproducibility’ folder of simulator.

    1. Don`t you have these large scale calibration results?

    And In Fig. 10.13, There are CDF of largest, smallest singular value.
    I think largest singular value is reasonable because Viennas result is located between other companys calibration results.
    However I think smallest singular value is a little bit far from other company`s results.

    2. Can you explain What is the reason of this result?

    • This topic was modified 4 years, 1 month ago by heungseop.
    • This topic was modified 4 years, 1 month ago by heungseop.

    Dear user,

    1.Regarding calibration results in general for the 3D model:

    – the phase 1 calibration contains only the geometry (coupling loss) and the correlated large scale parameters from step 4 of the procedure of generating the channel. As we don’t have any wrapping method implemented on our simulator we didn’t look at this part. Note that to compensate for the wrapping, we have a feature to generate the results only for the UEs on the center of our simulated scenario, by excluding the UEs served by eNodeBs on the outer tier of Base Stations (LTE_config.compute_only_center_UEs = true).

    – we perform only phase 2 calibration (which includes the fast fading channel): so correlated large scale parameters and largest-and smallest singular values of the channel. We don’t show the wideband SINR as in our simulator we didn’t include the RSRP, and in TR36.873 the wideband SINR is determined from RSRP (see eq. 8.1 of TR 36.873).

    2. For the smallest singular values, as you can see the distribution fits, but the curve is shifted a bit. Implementation-wise we coudn’t really find the reason for this small shift.

    Thank you for your feedback and please let us know if you have any further question regarding this topic or about the simulator in general.

    Best regards,


    Thanks for your reply!
    If I have a question, I will ask again.
    Thank you for your kind reply again.

    Best regards,


    How close were the calibration results? I have tried to calibrate, but got results not similar to the document given with simulator.

    Also, in “Plot_singular_values_calibration.m”, (line 49) H = sum(H_all,3)/6; Why are the 600 sub-carriers added and divided only by 6? I think the Eigen values should be found out per sub-carrier or you can average it with number of sub-carriers added, in this case it should be /600.

Viewing 4 posts - 1 through 4 (of 4 total)
  • You must be logged in to reply to this topic.