How to smooth a FFT signal?

I managed to plot the FFT spectrum using the below code. But I couldn't plot the smoothed spectrum. Can anyone suggest me a way to go through it? and How to choose order and frame length for the sgolayfilt code.

clc
data = xlsread('Reading 5(10000).xlsx') ; %Loading Sensor data from Excel file
t = data (1:1024,1);                %Selecting Time vector
s = data (1:1024,4);                %Selecting Z axis vibrations
L = numel(t);                       %Signal length
Ts = mean(diff(t));                 %Sampling interval
Fs = 1/Ts;                          %Sampling frequency
Fn = Fs/2;                          %Nyquist frequency
FTs = fft(s)/L;                     %Fast fourier transform (s- data)
Fv = linspace(0,1, fix(L/2)+1)*Fn;  %Frequency vector
Iv = 1:numel(Fv);                   %Index vector
IV = Iv-mean(Iv);
FTsiga = double(abs(FTs(Iv))*2);    %Magnitude,Convert to double
S_smth = sgolayfilt(FTsiga,20,501);  %Savitzky-Golay filtering of FFT

h = figure(1);
subplot(4, 1, 1);                   %plotting top pane
plot(t,s);                          %Acceleration vs time
set(gca,'xlim',[1 51.2]);             %Scale to fit
grid;                               %Grids on
title ({'Graph 5','Acceleration vs time'});     
xlabel('time(s)');
ylabel('Acceleration (m/s^2)');

subplot(4, 1, 2);                   %Plotting bottom pane
plot(Fv, abs(FTs(Iv))*2,'red');     %FFT - Amplitude vs Frequency
grid
title ('Fast fourier transform'); 
xlabel('Frequency (Hz)');
ylabel ('Amplitude (m)');

subplot(4, 1, 3);                     %FFT
% plot(Fv, abs(FTs(Iv))*2,'red');     %FFT - Amplitude vs Frequency
% grid
% title ('Fast fourier transform'); 
% xlabel('Frequency (kHz)');
% ylabel ('Amplitude (m)');
% hold on
Y = fft(s-mean(s));                     %Subtracting the mean before FFT to remove DC
P2 = abs(Y/L);                          
P1 = P2(1:L/2+1);
plot(Fv,P1,'Green');
title ('Fast Fourier Transform - "DC removed"');
xlabel('Frequency (Hz)');
ylabel ('Amplitude (m)');

subplot(4,1,4);                         %Plotting filtered FFT
plot (Fv,S_smth,'-k');
title ('Fast Fourier Transform - "Filtered"');
xlabel('Frequency (kHz)');
ylabel ('Amplitude (m)');
% fig = figure(1);
% print(fig, 'myFileName.pdf', '-dpdf','-r0')
% saveas (h,'Graph5.png')
% saveas (h,'Graph5.eps')
% saveas (h,'Graph5.jpg')

[Star Strider]

I recognise something that appears to have begun as code I posted here!

The sgolayfilt parameters (as with much else in signal processing) are arrived at by trial-and-error until you get the result you want. This is obviously highly subjective, so with respect to sgolayfilt, I generally begin with a low frame length and order, and then increase them as necessary.

So start with a low order (perhaps 3) and frame length (perhaps 9) and increase first the frame length. If that doesn’t do what you want, change the order as well, however higher orders with the same frame length will provide a less smooth result, so only change one parameter at a time.

[Wakeel Mohammed]

Yes ofcourse! I get the knowledge from on of your post couple of years ago! and thank you very much for your answer above, you are great man!

[Star Strider]

Thank you!