Here is the code:
close all;
%Define number of samples to take
fs = 100;
f = 400; %Hz
%N =length(signal);
%Define signal
t = 0:1/fs:1-1/fs;
%signal = sin(2*pi*f*t);
signal = xlsread('testdata.xlsx');
%Plot to illustrate that it is a sine wave
plot(t, signal)
title('Time-Domain signal');
%Take fourier transform
fftSignal = fft(signal);
%apply fftshift to put it in the form we are used to (see documentation)
fftSignal2 = fftshift(fftSignal);
%xdft = xdft(1:length(s)/2+1);
%Next, calculate the frequency axis, which is defined by the sampling rate
f = fs/2*linspace(-1,1,fs);
%Since the signal is complex, we need to plot the magnitude to get it to
%look right, so we use abs (absolute value)
figure;
plot(f, abs(fftSignal2));
title('magnitude FFT of sine');
xlabel('Frequency (Hz)');
ylabel('magnitude');