Question with respect to NPUSCH BLER Example

Hello All,

I am trying to simulate the NPUSCH BLER example and would like to understand the hidden assumptions when we disable perfect channel estimation for this example? The simulation results in BLER = 0 for SNR = -25dB when SCS=3.75kHz for 1RU and 1 subcarrier. I would like to use a custom channel (say Rician) for example. I have made the changes accordingly, but it does not seem to work. Please could you help me figure out what is happening in the background?

Update: I realised that noiseEst is always 0 when 'perfectChannelEstimate=false' and this is giving me wrong results for lower SNR.

Thanks in advance

Answer 1

[Broy]

Hi Kavyashree,

I understand your issue, when you make perfectChannelEstimate as “false” in the NPUSCH BLER example, the results appear incorrect, and you would like to know what is happening in the background when perfectChannelEstimate = false.

When perfectChannelEstimate = false, the simulation no longer uses the ideal channel response. Instead, it calls the practical estimator lteULChannelEstimateNPUSCH, which performs:

• Channel estimation (hEst) using the NPUSCH Demodulation Reference Signal (DRS).
• Noise power estimation (noiseEst) by averaging least-squares estimates of the DRS across time and frequency.

In cases like very small allocations e.g., 1 RU, 1 subcarrier, SCS = 3.75 kHz, the number of DRS symbols is very limited. If the estimator cannot extract or average enough valid pilots, noiseEst can become zero or very small.

Helpful documentation links:

https://www.mathworks.com/help/releases/R2023b/lte/ug/nb-iot-npusch-block-error-rate-simulation.html

https://www.mathworks.com/help/releases/R2023b/lte/ref/lteulchannelestimatenpusch.html

Hope this helps.

[Kavyashree]

Hello Broy,

Thanks for the clarification.