Signal Processing in the Time Domain

Signal Processing in the Time Domain

• Correlation Analysis
  • Autocorrelation: similarity a variable with its past, for finding repeating patterns
  • Cross-correlation: similarity of two variables as a function of time-lag applied
• Covariance Analysis
  • Autocovariance: variability a variable with its past
  • Cross-covariance:
• Statistical Features in the Time Domain
  • Mean: mathematical average
  • Variance: population variance represents a measure of dispersion
  • Standard deviation; a measure of variability
  • Mean absolute deviation
  • Root amplitude
  • Root mean square (RMS): a measure of signal magnitude
  • Square mean root (SMR): a measure of signal magnitude
  • Peak: maximum value
  • Peak to Peak
  • Skewness: normalized third central moment represents a measure of symmetry
  • Kurtosis: normalized fourth central moment represents a measure of distribution spread
  • Crest factor: ratio of the peak level to the RMS level
  • Latitude factor: ratio of the peak level to the SMR level
  • Clearance factor
  • Shape factor (form factor): ratio of RMS to mean of absolute
  • Impulse factor: indication of shock in a waveform
  • THIKAT
  • TALAF
  • Slope Sign Change
  • Willison Amplitude
  • Zero crossing
• Others
  • Derivative
  • Integral
  • Convolution

Advantages

• Simple implementation
• Low computational cost
• General trend information

Weakness

• Limited dynamic information

References

• leiIntelligentFaultDiagnosis2017 link DOI
• delgadoprietoContributionsElectromechanicalSystems2012 link DOI