Raised cosine filter in ad9361 FIR


I know that this topic has been already discussed many times on this forum, however, I could not find answers to all of my questions.

I've been able to implement a 128-tap filter with x4 interpolation in matlab and uploaded it to the board, this is clear for me.

However, the FIR filter is not the only one in the Tx/Rx signal chain and the Nyqiust criterion for zero-ISI is not met for the filter cascade of FIR + HB.

When the filter wizard app computes the weights of a low-pass filter it can compensate for all other filters in the cascade, so that there is a constant gain in the passband. Are there any similar formulas to compensate for the HB filters in the case of zero-ISI?

Or could all HB filters be disabled if FIR is set in x4 interpolation mode?

Thank you in advance.



  • 0
    •  Analog Employees 
    on Jul 19, 2019 6:56 PM

    The cascade of PFIR+HBs on the Tx side, channel, and HBS+PFIR on the Rx side can be thought of as the overall channel. As long as the bandwidth of the signal is within 56 MHz, this cascade channel satisfies the condition for nearly ISI-free response. Nearly because, the PFIRs and HBs will contribute some ISI since they are practical filters. Subject to this condition, RRCs at Tx and Rx should help satisfy Nyquist criterion.

    >> Or could all HB filters be disabled if FIR is set in x4 interpolation mode?

    The ADC should be run as fast possible so that with all HB filters applied, the ADC output waveform experiences bitgrowth. 

    - Srikanth