How to remove pwd residue with auto-correlation

I have a sum of sine waves with transmission loss applied in a row.

And pwd is 10e-3. Because of this, the residue of the previous pwd remains in the part where the signal is not generated,

blurring the entire signal. My mentor said that I can remove it using auto-correlation.

However, he said to remove it using the filter format instead of using functions such as xcorr.

I don't understand this part. If there is an expert in this field, please let me know what the intended part is along with the method.

I've attached a picture plotting one row of those signals.

If you need any further information, please let me know and I will get back to you as soon as possible.

%sampling F
Fs = 100e3; % 

time_step = 1 / Fs; 

distance_step = 2 * time_step * soundSpeed;

max_time = min(L_values(:));

pwd = 10e-3;

tt = 0 : time_step : pwd;

F = 1000; %sine wave F

input_sig = sin(2 * pi * F * tt);

round_distance = soundSpeed * dist_time; %soundspeed = 1500(m/s)
TL = 40 * log10(round_distance);

TL_signal = input_sig * 10^(-TL/20);

Answer 1

[Rahul]

Hi @taetae,

According to the description given by you, I understand that you are trying remove the residue of the previous pwd.

You might want to consider using a high pass filter to remove the low-frequency components that may correspond to the residue.

You can use the functions 'designfilt' and 'filter' to achieve this. Then you can pass the TL_signal to this filter.

hp_filter = designfilt('highpassiir', 'FilterOrder', 8, ...
                       'HalfPowerFrequency', cutoff_freq, ...
                       'SampleRate', Fs);

filtered_signal = filter(hp_filter, TL_signal);

% You can set the parameters of the filter according to your requiremenets.

You can refer to the following documentations to know more about these functions:

'designfilt': https://www.mathworks.com/help/releases/R2024a/dsp/ref/designfilt.html?searchHighlight=designfilt%5C&s_tid=doc_srchtitle

'filter': https://www.mathworks.com/help/releases/R2024a/matlab/ref/filter.html?searchHighlight=filter&s_tid=doc_srchtitle

Hope this helps! Thanks!

[taetae]

Ah, something seems right, but I'm not getting the results I want. I used FFT to specify the cutoff frequency.

I tried it with various cutoff frequencies, but it was far from the results I wanted.

It's too difficult ..