NR PDSCH Throughput example SISO and MISO

1 view (last 30 days)
Hello. I am using 5G Toolbox and LTE Toolbox with example of NR PDSCH Throughput to simulate the Throughput and BER. I have an issue that when I assign SISO and MISO with one layer it gives me an error with this Error using nrLayerDemap Expected IN to be finite when you using lteEqualizeZF. I have attached m file of this example here, so could someone help me here how could be fix?

Accepted Answer

Arthi Sathyamurthi
Arthi Sathyamurthi on 23 Mar 2021
The error Expected Input to be Finite” occurs when the input contains non-finite value(s). In this case the equalizer output pdschEq which serves as the input for nrPDSCHDecode function has inf values. This may be introduced if an unexpected value is encountered.
In the case of lteEqualizerZF applies the MIMO zero-forcing equalization to the received data resource grid using the channel information only. When zero-forcing equalizer attempts to invert the channel response matrix it will amplify the noise greatly at frequencies where the channel response matrix has a small magnitude. This would have led to the inf values in the equalization output.
In the case of nrEqualizeMMSE (which was replaced) applies MMSE equalization to the received resource grid by using the estimated channel information and the estimate of the received noise variance which can minimize the total power of the noise and ISI components in the output.
  3 Comments
Arthi Sathyamurthi
Arthi Sathyamurthi on 23 Mar 2021
You as such cannot assign a non-inf value since it is being calculated by the lteEqualizerZF function. But as a workaround/solution, possibly reassign the inf values to a maximum saturated value you think would be right before passing them to the nrPDSCHDecode function for further computation.
Youssef Rawy
Youssef Rawy on 23 Mar 2021
How can I reassign the inf values to a maximum value? could you help me or give me a hit for this please?

Sign in to comment.

More Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!