WarpingAnomaly

R2026b

Synthetic magnitude-warping anomaly model for validating anomaly detection models

Since R2026a

expand all in page

Description

Add-On Required: This feature requires the Time Series Anomaly Detection for MATLAB add-on.

The warpingAnomaly object specifies the characteristics of an anomaly model that you can inject into a time series using injectAnomaly. This anomaly causes the time series to experience magnitude warping for the duration of the anomaly.

Magnitude warping causes a time series to be multiplied by a second term, as shown in the following equation.

y^(n)=y(n)*f(θ,n)

In this equation,

  • ŷ(n) is the warped time series.

  • y(n) is the original time series

  • f(θ,n) is the warping curve, specified as either a Gaussian process regression (GPR) or a cubic spline.

The name-value argument specifications in injectAnomaly determine the window location and length during which the anomaly occurs.

You create this model using syntheticAnomaly. WarpingAnomaly is one type of anomaly model in a set of anomaly objects that you can use to perturb a time series in multiple ways. You can then use this perturbed time series to help validate anomaly detection models against different anomaly types.

Plot of a magnitude-warping anomaly within a ramp signal, as shown by a drop followed by constant rise for a number of samples near the middle of the plot.

Properties

expand all

Algorithm to use for warping curve, represented as "GPR", or "CubicSpline".

Algorithm to use for warping curve, represented as "GPR", or "CubicSpline".

Object Functions

injectAnomaly Inject anomalies defined by one or more anomaly models into a univariate time series

Version History

Introduced in R2026a

See Also

Topics