search

brianmc95 / OpenCV2X

OMNeT++ V2X simulation framework for ETSI ITS-G5
GNU General Public License v2.0
37 stars 15 forks source link

The number of decoded packets at the physical layer is inconsistent with the number of packets received by the application layer #25

Open PeiZhonghui opened 3 years ago

PeiZhonghui commented 3 years ago

Hi @brianmc95，

During the simulation, I found that the number of packets received by the application layer was far more than the number of packets successfully decoded by the physical layer. image In line 1475 of "LtePhyVUeMode4", the decoded result sent to the MAC layer only considers "interference_result". Other factors that cause TB reception failure are not considered, such as "tbFailedDueToNoSCI" and "TbFailedDuetoProp". Should the "setDeciderResult" sent to the upper layer here take all the factors that cause the decoding failure into consideration?

In fact, the results show that the number of TB decoding failures due to no corresponding SCI is extremely high, with the probability even exceeding 50%. This is because the probabilistic model used on lines 1292-1301 results in a large number of "sciUnsensed_". image image However, even though the probability of "TBFailedDueToNoSci" is high, the packet delivery ratio calculated by "TBDecoded" is still very high (obviously over 80%). Do you know what caused such contradictory results? Or is it just because of the 500m distance limit used when calculating PDR? image Thank you.

Best regards, Zhonghui

PeiZhonghui commented 3 years ago

I have just debugged the program, and the conflict result between TBFailedDueToNoSci and PDR is indeed caused by the statistical distance. This result is no problem, sorry for that.