I`m designing the fir filter coefficients of AD9361, and found that the output waveform of tx is abnormal:

Is this because the design of the filter is unreasonable? Or are there other reasons? Thanks

I`m designing the fir filter coefficients of AD9361, and found that the output waveform of tx is abnormal:

Is this because the design of the filter is unreasonable? Or are there other reasons? Thanks

I tried the following two groups：

1

{1280000000, 160000000 ,80000000 ,40000000, 20000000, 20000000},// rx_path_clock_frequencies[6] *** adi,rx-path-clock-frequencies

{1280000000 ,160000000 ,80000000 ,40000000, 20000000 ,20000000},// tx_path_clock_frequencies[6] *** adi,tx-path-clock-frequencies2

{960000000,480000000,240000000,120000000,60000000,60000000},// rx_path_clock_frequencies[6] *** adi,rx-path-clock-frequencies

{960000000,240000000,120000000,60000000,60000000,60000000},// tx_path_clock_frequencies[6] *** adi,tx-path-clock-frequenciesthe picture is the output of the first one, the second one also has this problem.

Hi Michael:

I used:

{1280000000, 160000000 ,80000000 ,40000000, 20000000, 20000000},// rx_path_clock_frequencies[6] *** adi,rx-path-clock-frequencies

{1280000000 ,160000000 ,80000000 ,40000000, 20000000 ,20000000},// tx_path_clock_frequencies[6] *** adi,tx-path-clock-frequenciesto read register 0x002 and 0x003, both of them are 0x2E.

thank you for your reply

Can you try to design a filter that uses interpolation or decimation > 1, for example 2?

Please also try to read register 0x02 and 0x03 after you have enabled the filter.

-Michael