From: A novel approach to extracting useful information from noisy TFDs using 2D local entropy measures
α | Entropy order |
β | Constant |
χ | Entropy mask |
δ | Window size |
δ+ | Optimal window size |
Γ | Threshold parameter |
ω(t) | Smoothing window |
\(\overline {L}\) | The largest lower limit of D |
σ | Estimation variance |
τ | Continuous lag |
τ+ | Optimal threshold |
\(\underline {U}\) | The smallest upper limit of D |
C(t,f) | TFD of the signal |
D | Confidence interval |
E | Energy |
f | Continuous frequency |
FN | False negatives |
FP | False positives |
g | Time smoothing window |
H | Entropy |
h | Frequency smoothing window |
H(s) | Shannon entropy |
\(H^{\delta }_{\rho }\) | Rényi entropy for window size |
\(H^{\text {RICI}}_{\rho (t,f)}\) | Optimal Rényi entropy |
L | Lower limit of D |
M | Frequency bins |
Mp | Distribution |
N | Number of samples |
O | Overlapping confidence intervals |
pi | Probability value |
R(n,δ) | Ratio of finite interval and confidence interval |
Rc | Threshold |
t | Continuous time |
TN | True negatives |
TP | True positives |
U | Upper limit of D |
x(t) | Signal |