how to design a lowpass filter for RF signal

2 views (last 30 days)
Mohamed Jamal on 17 Jul 2020
Answered: Kiran Felix Robert on 23 Jul 2020
Hi , Im trying to do my own LPF that aren't moving my DC signal after filtering, this means the DC of my filetered signal must be the same as the input signal before filter, I used designfit simulink LPF, it's already designed LPF, and after filtering my input signal it leads to move my DC signal from my DC of the input signal, the DC of my filtered signal is moved in compare to the DC of my input signal.
Im trying to do in matlab LPF that aren't moving my DC signal, my input signal is like what Im attaching here:
this signal is actualy sampled signal from original signal (from transmitted signal), signal transmitted on those parameters:
f=868MHZ, symbol rate(symbol rate)=100KBaud, deviation: 100Khz.
The frequency of sampling is 2048Khz.
what Im trying to do is to look my signal is sinusoidal but without changing the DC of my input signal, exactly I need my LPF filter to is to implicitly to make my filtered signal to look as sinusoidal and smearing/plastering the peaks(where there's on the peaks something not sinusoidal or not smearing to make it looks like smearing/sinusoidl)
attaching here what Im trying to do like this: (NOTE: NOT LETTING MY DC of my filtered signal to move from the DC of my input signal)
regarding to the design of LPF does't matter for me how manny bits it , doesn't matter the length of my LPF.
if needed: in my cases the DC of my input signal is zero or 0.000005 (approximate to zero)
thanks for helpers!

Kiran Felix Robert on 23 Jul 2020
Hi Mohamed Jamal,
It is my understanding that your have a problem with the DC offsets, that is the DC offset of the input signal is different from the DC offset of the filtered signal. DC offset for an AC signal is basically the mean value over a time period. So, a possible work-around for this would be to adjust the bias point of the filtered signal, this can be done as shown,
Input_signal = Y;
DC_input = mean(y);
filtered_signal = Z;
DC_Filt = mean(Z);
req_signal = Z - DC_Filt + DC_input;
This will reset the DC Bias of the filtered signal back to that of the input signal
Kiran