DriftAnomaly

R2026b

Synthetic drift anomaly model for validating anomaly detection models

Since R2026a

Description

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

The DriftAnomaly object specifies the characteristics of a linear, exponential, or quadratic drift anomaly model that you can inject into a time series using injectAnomaly. Name-value argument specifications in injectAnomaly determine the window location and length during which the anomaly occurs.

You create this model using syntheticAnomaly. DriftAnomaly 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.

Properties

Type — Drift type

"Linear" (default) | "Exponential" | "Quadratic"

Drift type, represented as "Linear","Exponential" or "Quadratic".

• When Type is set to "Linear", injectAnomaly adds a linearly increasing drift over the entire window length that you specify to injectAnomaly, as shown in the following equation.

  $$\widehat{y}(n)=y(n)+\beta n$$

  Here,

  • y(n) is the original time series.

  • β(n) is the drift scaling value.

  • ŷ(n) is the resulting anomalous time series.

• When Type is set to "Quadratic", injectAnomaly incorporates a quadratically increasing drift, as shown in the following equation.

  $$\widehat{y}(n)=y(n)+\beta n^2$$

• When Type is set to "Exponential", injectAnomaly incorporates an exponentially increasing drift, as shown in the following equation.

  $$\widehat{y}(n)=y(n)\beta e^n$$

Scale — Scaling value

1 (default) | numeric scalar

Scaling β for each sample, represented as a numeric scalar.

Object Functions

injectAnomaly

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

Version History

Introduced in R2026a