Digital Signal Processing Reference
In-Depth Information
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
0
500
1000
1500
2000
2500
3000
3500
4000
Frequency, Hz
Figure 3.60:
A plot of first order differences of bin magnitudes between adjacent frames. The first order
differences are all normalized by the respective mean bin magnitudes over all frames in the matrix, which
in this case includes all frames of the test signal.
function LVxDF TEqualization(tstSig,p,k,SR,xplotlim)
% Creates a test signal; tstSig=1 gives 32 periods of a
% length-16 square wave followed by 16 samples equal to zero;
% tstSig=2 yields 64 periods of a length-16 square wave;
% tstSig=3yields a length-1024 chirp from 0 to 512 Hz.
% p is a single real pole to be used to generate a decaying
% magnitude profile which weights SR/4 samples of random noise
% of standard deviation k to generate the test channel
% impulse response. SR is the length of FF T to be used to
% model the channel impulse response and create the inverse
% filter.
% xplotlim is the number of samples of each of the relevant signal
% to plot; the relevant signals are the test signal, the distorted
% test signal, and two equalized versions created, respectively, by
% frequency domain (FD) deconvolution and time domain(TD)
% deconvolution.
% Since the impulse response is random noise with a decaying
%
magnitude,every call to the function generates a completely
%
different impulse response for the simulated channel, and the
%
distorted test signal will have a different appearance.
% LVxDF TEqualization(1,0.9,0.5,2048,250)
