How do I apply a Gaussian pulse shaping filter on a random bit stream using Communications Toolbox?

22 views (last 30 days)

Accepted Answer

MathWorks Support Team
MathWorks Support Team on 27 Jun 2009
The ability to apply a Gaussian pulse shaping filter on a signal is not available in Communications Toolbox.
As a workaround, you may use the GAUSSFIR function in the Signal Processing Toolbox. The following example demonstrates how this can be done.
%%Signal Source
% Build a set of test data.
x = randint(5,1);
x = x*2 -1;
rxsig = upsample(x,33); % Upsampled signal
rxsig = [zeros(16,1); rxsig];
stem(rxsig); hold on
%%Setup
% Define parameters.
bt = 0.3;
o=8;
n = 2;
%%Filter Definition
% Define filter-related parameters.
h = gaussfir(bt,n,o) ;
%%Filter the Signal
% y = filter(h,1,rxsig)/(h*h'); %Scaling by energy
y = filter(h,1,rxsig)/(max(h)); %Scaling by magnitude
plot(y(17:end, :), 'r');hold off;
legend('Upsampled Binary Signal','After Pulse Shaping')
%%Plot the impulse response of the Gaussian Filter
figure;impz(h);

More Answers (1)

Tasos Giannoulis
Tasos Giannoulis on 25 Jan 2017
Alternatively, you can construct an FIR Gaussian filter using dsp.FIRFilter (from DSP System Toolbox) and gaussdesign (from Signal Processing Toolbox):
bt = 0.3;
span = 4;
sps = 4;
num = gaussdesign(bt, span, sps);
gaussianFilter = dsp.FIRFilter('Numerator', num);
filtered = gaussianFilter(yourSignal);

Products


Release

R2006a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!