You are not logged in.
We have the following questions on the PHY model:
a) Could you please provide algorithmic constructs of the TxControl and TrainingData modules? We are trying to modify these two block for higher FFT size. For example how control signals like FFT_En are generated in the TxControl block. How data has been read from the BRAM in the TariningData block. What is the algorithm used for slicer module in the TariningData block.
b) For higher FFT size, the packet gets longer in duration. What kind of changes are expected in these two blocks?
thanks in advance.
Offline
a) We don't have formal documentation for those state machines. Their basic design is based on a master counter, which starts when the transmitter is triggered. At various values of this counter, Tx subsystems are triggered. Additional counters are used for individual subsystems- like counting the number of samples in each FFT frame to trigger each transform.
b) Most of the control logic in the PHY assumes a 64-point FFT. The basic structure of the control logic would work for larger FFTs, but the various constants which are used as comparisons with the counter values would need to be changed.
Offline