How make a high pass filter and give input to it simultaneously?

I want to implement a line coding scheme with the length of bit sequence = 30. After plotting the bit sequence wrt time and its fourier transform, I need to pass it through a High pass filter with a given cutoff frequency (lets say 100Hz). There are functions on matlab like Highpass but it doesn't pass an input signal into it. Hence How can i make a filter which can have a given cutoff frequency and also passes a custom made input (in the code).

len = 30; %length of waveform

waveform = randi([0,1],1,len); %radnom bit generator of length 30

Tb = 1*10^-6; %bit period

freq = 1/Tb; %bits per second

figure;

[t,s] = uni_nrz(waveform, freq);

subplot(1,2,1);

plot(t,s);

xlabel('time');

ylabel('Voltage');

grid on;

title('Unipolar not return to Zero');

function[t,out] = uni_nrz(waveform, freq)

T = length(waveform)/freq; %time of bit sequence

n = 200; %sampling of waveform

N = n*length(waveform);

delta = T/N;

t = 0:delta:T;

out = zeros(1,length(t)); %initialize output signal as 0

for i = 0:length(waveform)-1 %using i+1 to perform operation on the next bit sequence when

%pervious is already operated upon

if waveform(i+1) == 1

out(i*n+1:(i+1)*n) = 5 * 1; %voltage Vo is ±5V

else

out(i*n+1:(i+1)*n) = 5 * 0;

end

end

subplot(1,2,2);

plot(abs(fft(out)));

xlabel('Frequency');

ylabel('Amplitude');

end

Answer 1

[Paul]

"How can i make a filter which can have a given cutoff frequency and also passes a custom made input (in the code)."

Check out highpass. Use it by itself if you want the feature of "compensates for the delay introduced by the filter." Otherwise use the second output argument to return the digitalFilter object and use that to filter the input.

Answer 2

[Star Strider]

‘There are functions on matlab like Highpass but it doesn't pass an input signal into it.’

Yes, it does!

The input signal is the first argument to the highpass function (and related functions). You can then use the second argument (the digital filter object) with filtfilt to filter other signals with the same filter, so it is not necessary to create the same filter each time.

Try something like this —

Fs = 1E+6;
s = randn(1, 30*Fs);
t = linspace(0, numel(s)-1, numel(s))/Fs;
[s_filt1, df] = highpass(s, 100, Fs, 'ImpulseResponse','iir');

figure
plot(t, s)
hold on
plot(t, s_filt1)
hold off
grid
xlim([0 1/100])

s_filt2 = filtfilt(df, s);

figure
plot(t, s)
hold on
plot(t, s_filt2)
hold off
grid
xlim([0 1/100])

.