Low pass filter VS high pass filter

I have this filter and I want to analyze it. How can I know from the plot that its a highpass filter or a lowpass filter?

Answer 1

[Star Strider]

Calculate the Fourier transform of the filter from the signals.

t = ...;                                                    % Time Vector
s = ...;                                                    % Original Signal Vector
s_filtered = ...;                                           % Filtered Signal Vector
Ts = mean(diff(t));                                         % Sampling Interval (Assumes Constant Intervals)
Fs = 1/Ts;                                                  % Sampling Frequency
Fn = Fs/2;                                                  % Nyquist Frequency
L = numel(t);                                               % Signal Length
FTs = fft(s)/L                                              % Fourier Transform of Original Signal
FTs_filt = fft(s_filtered)/L;                               % Fourier Transform of Filtered Signal
Fv = linspace(0, 1, fix(L/2)+1)*Fn;                         % Frequency Vector (One-Sided Fourier Transform)
Iv = 1:numel(Fv);                                           % Index Vector (One-Sided Fourier Transform)
H = FTs_filt ./ FTs;                                        % Complex Transfer Function

figure
subplot(2,1,1)
plot(Fv, abs(H(Iv)))
ylabel('Amplitude')
grid
subplot(2,1,2)
plot(Fv, angle(H(Iv)))
xlabel('Frequency')
ylabel('Phase')
grid
sgtitle('Bode Plot of Filter')

That should reveal the filter characteristics, as well as reasonably accurate aspects of its design (e.g. passband and stopband frequencies).

Answer 2

[Jan]

On first view you see, thet the high frequencies are removed: the changes between neighboring values are reduced. This means that the filter let low frequencies pass through, so it is a low-pass filter.

Answer 3

[Mathieu NOE]

hello

you can do a bode plot of the transfer function between the output (signal filtered) vs the input (raw signal)

use tfestimate for transfer function estimation from time data and you will be able to fit a filter TF to that