NonUniformPoissonModel

Nonuniform Poisson clutter model

Since R2025a

Description

A NonUniformPoissonModel object contains a clutter model that assumes a Poisson distribution of clutter with a nonuniform clutter density. A Poisson distribution of clutter is a common assumption in many multi-object tracking algorithms. In this model, you can specify the clutter density as a function of measurement, which enables you to define regions in the measurement space where clutter is more likely to appear for the first time. You can use this model to define the clutter model for a CustomSensor object.

Creation

To create a NonUniformPoissonModel object, use the sensorClutterModel function with the "nonuniform-poisson" input argument. For example:

clutterModel = sensorClutterModel("nonuniform-poisson")

Properties

ClutterDensityFcn — Function that updates clutter density

function handle

Function that updates the clutter density as a function of measurement and other clutter model data, specified as a function handle.

ClutterDensityFcn must have one of these signatures:

• density = ClutterDensityFcn(z);

• density = ClutterDensityFcn(z,data);

z is the measurement reported by the sensor and data is the property ModelData.

The clutter density is defined as the number of clutter measurements appearing per unit measurement space volume. The volume of measurement space is connected to the measurement model of the sensor. For example, if a radar reports azimuth and range measurements, its clutter density can be calculated as a function of measurement:

$$\text{ClutterDensity}\left(azimuth,range\right)=\frac{\text{ClutterRate}\left(azimuth,range\right)}{\text{AzimuthResolution}\cdot \text{RangeResolution}}$$

The denominator represents the volume of one resolution cell of the radar. The ClutterRate (azimuth,range) specifies the average number of clutter measurements appearing in one resolution cell located at (azimuth,range) per update.

ModelData — Additional data to compute clutter density

[] (default) | data input for ClutterDensityFcn

Additional data to compute the clutter density. ModelData can be any MATLAB^® data type, including numeric types, character vectors, cell arrays, structures, and objects.

You can specify this data to define any additional information you require to compute the clutter density. To specify additional information, use the function signature density = ClutterDensityFcn(z,data) of the ClutterDensityFcn property. The ModelData property is sent to the ClutterDensityFcn function as its data input argument. For an example using additional data to compute the clutter density, see Maritime Target Tracking Using Radar Under Rough Sea Conditions. In this example, ModelData is a structure.

Version History

Introduced in R2025a