
226 Practical Biomedical Signal Analysis Using MATLAB
R
ysis of matching pursuit [Jedrzejczak et al., 2004]. Matching pursuit decomposes sig-
nals into components of specific amplitudes, frequencies, latencies, and time spans,
so these parameters are given explicitly. A comparison of different methods used
for OAE analysis is shown in Figure 4.61. The simulated signal used for testing the
methods was constructed from the so-called gammatones—the functions resembling
the shape of click evoked OAE at single resonant frequency. The gammatone is ex-
pressed by a function, whose envelope rises as t
3
and decays exponentially with a
constant Γ:
γ(t)=γ
0
t
3
e
−2