Calculation of PER based on BER with FEC in InterferenceHelper in wifi module

[Bora Karaoglu]

Dear all,

I am trying to understand the rationale behind the packet error rate calculation in the interference helper with error models supporting forward-error calculation. In other words, I am trying to understand the assumptions with which the given method makes sense. Please do not consider this as criticism since a network level simulator is not expected to reflect PHY layer performance very accurately. 

It seems like ChunkSuccessRate's for each interval is calculated using the empirical formulation with the given FEC and puncturing algorithm. Chunks are defined as the intervals during which SINR stays constant. This tells me that the algorithm is trying to calculate the performance under varying SINR. 

However, the chunk success rates are combined using psr *= CalculateChunkSuccessRate ( ... ). In other words, if there is a high power interfering signal even for a very short duration, there is very little chance for the packet to survive no matter how strong of a FEC algorithm is used. 

I would appreciate any comments or if you could direct me any alternative model that would allow packets to survive in a similar scenario.

All the best,

[Sebastien Deronne]

I do not really get what your question is.
Note that any more accurate PHY layer model is always more than welcome :-)

[Bora Karaoglu]

Thank you Sebastien. I guess mine was a general inquiry about whether there are groups working on such PHY layer models. I'd like to discuss it with other people working on the subject of how a reasonable PHY model should be. I appreciate any contributions in deciding what should be implemented. 

Browsing through this forum, I came accross PhySim-Wifi that seem to do symbol level simulation. However, some users point out it that it requires lots of processing power and is not suitable for large scale simulations. 

I think I reasonable trade-off could be assuming perfect scrambling and hard-decision decoding. One can calculate the number of bit errors in each chunk(using a random number generator) and determine whether the packet is received correctly based on the error correction capabilities of the FEC. If soft-decision decoding is used, maybe SNIR could be adjusted (2dB for AWGN channel). Does this sound like a reasonable trade-off?