Why do I obtain edge effects or oscillations when using the...?

Illustration
willison_ad - 2021-01-04T11:27:28+00:00
Question: Why do I obtain edge effects or oscillations when using the...?

Why do I obtain edge effects or oscillations when using the RESAMPLE function to perform non-integer resampling on my signal in the Signal Processing Toolbox 6.7 (R2007a)? I am using the RESAMPLE function on my signal with Signal Processing Toolbox 6.7 (R2007a) and I see that the resampled signal suffers from edge effects, i.e. oscillations at the edges. This can be seen from the following example.     fs1 = 10; % Original sampling frequency in Hz t1 = 0:1/fs1:1; % Time vector x = t1 + 100; % Define a linear sequence y = resample(x,3,2); % Now resample it t2 = (0:(length(y)-1))*2/(3*fs1); % New time vector plot(t1,x,'*',t2,y,'o',(-0.5:0.01:1.5),(-0.5:0.01:1.5)+100,':') legend('original','resampled'); xlabel('Time') I know that this problem could be prevented if I were able to specify initial and final conditions for my signal in the RESAMPLE function. However, the function does not allow that. I want to compensate for this and obtain correct results.  

Expert Answer

Profile picture of Prashant Kumar Prashant Kumar answered . 2025-11-20

The ability to specify initial and final values for the signal is not available in the RESAMPLE function in Signal Processing Toolbox 6.7 (R2007a).
The RESAMPLE function assumes that the signal values before and after the specified signal are zeros. However, this may not be true for your particular biased signal as in the example above. This assumption is what causes the edge effects (or oscillations) to occur.
To work around this issue, you should pad your input signal at the beginning and end with the appropriate values, and then use RESAMPLE. The relevant signal can then be extracted from the results of resampling.
fs1 = 10;             % Original sampling frequency in Hz

t1 = 0:1/fs1:1;       % Time vector

x = t1 + 100;         % Define a linear sequence

xpad = [repmat(x(1), 1, 10), x, repmat(x(end), 1, 10)];

tpad = [-1/fs1*10 : 1/fs1: 0-1/fs1, t1, 1+1/fs1:1/fs1:1+1/fs1*10];

ypad = resample(xpad,3,2);  % Now resample it

t2 = (0:(length(ypad)-1))*2/(3*fs1) - 1;  % New time vector

plot(t1,x,'*',t2,ypad,'o',(-0.5:0.01:1.5),(-0.5:0.01:1.5)+100,':')

legend('original','resampled'); xlabel('Time')


Not satisfied with the answer ?? ASK NOW

Get a Free Consultation or a Sample Assignment Review!