Why cant I get the same results IFFT block and HDL Optimized IFFT block.

2 views (last 30 days)
I'd like to perform OFDM on FPGA's and for that I had to use IFFT block. To perform the IFFT I used the example "OFDM Receiver with 512-Point Streaming I/O FFT". In the example OFDM transmitter part there is an IFFT block and I insert HDL optimized IFFT block and compare them with the spectrum analyzer. However the resultts are different. How can I öconstruct the OFDM transmitter? I put the screen shots of spectrum analyzers and the model.
<<
>>

Accepted Answer

Bharath Venkataraman
Bharath Venkataraman on 30 Sep 2014
The HDL Optimized FFT and IFFT don’t support natural order output in R2014a. The output of the HDL Optimized IFFT in your model is in Bit Reversed order. To be able to compare the output of the HDL Optimized IFFT with IFFT, you need to put the output in natural order. Connect a deserializer and Bit Reverse Order block after your HDL Optimized IFFT and then connect it to spectrum analyzer. You can find these two blocks at the top level of the same example (OFDM Receiver with 512-Point Streaming I/O FFT).
  3 Comments
DN
DN on 2 Aug 2016
This example from Mathworks is not very useful in the sense that half of the HDL part uses blocks that is not HDL compatible. For example in the "serializer" and "deserializer" blocks there are the standard buffer and un buffer blocks. These blocks cannot be converted to HDL code, so even if you do use the optimized HDL FFT technique you are still stuck with useless blocks without using FIFO's and doesn't demonstrate a good comparison with the FFT that is non-HDL.
DN
DN on 2 Aug 2016
Actually forget my comment, I think this example was demo'ed before 2014b release which seems to have brought out "Serializer" and "Deserializer" HD blocks, which will allow you to ditch the standard buffer and unbuffer Simulink blocks that was used in this example on the HDL side yet not HDL ready. Thanks.

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!