dsp.MovingAverage

Moving average

Description

The dsp.MovingAverage System object™ computes the moving average of the input signal along each channel independently over time. The object uses either the sliding window method or the exponential weighting method to compute the moving average. In the sliding window method, a window of specified length moves over the data sample by sample, and the block computes the average over the data in the window. To change the length of the sliding window during simulation, pass the window length as an input to the object along with the data input. For an example, see Tune Window Length During Simulation. (since R2026a) In the exponential weighting method, the object multiplies the data samples with a set of weighting factors. The average is computed by summing the weighted data. For more details on these methods, see Algorithms.

The dsp.MovingAverage object and the movmean function both compute the moving average of the input signal. However, the object can process large streams of real-time data and handle system states automatically. The function performs one-time computations on data that is readily available and cannot handle system states. For a comparison between the two, see System Objects vs MATLAB Functions.

To compute the moving average of the input:

1. Create the dsp.MovingAverage object and set its properties.

2. Call the object with arguments, as if it were a function.

To learn more about how System objects work, see What Are System Objects?

Creation

Syntax

movAvg = dsp.MovingAverage

movAvg = dsp.MovingAverage(Len)

movAvg = dsp.MovingAverage(Len,Overlap)

movAvg = dsp.MovingAverage(PropertyName=Value)

Description

movAvg = dsp.MovingAverage returns a moving average object, movAvg, using the default properties.

movAvg = dsp.MovingAverage(Len) sets the WindowLength property to Len.

movAvg = dsp.MovingAverage(Len,Overlap) sets the WindowLength property to Len and the OverlapLength property to Overlap.

movAvg = dsp.MovingAverage(PropertyName=Value) specifies additional properties using Name,Value pairs. Unspecified properties have default values.

Example: movAvg = dsp.MovingAverage(Method="Exponential weighting",ForgettingFactor=0.9);

Properties

Unless otherwise indicated, properties are nontunable, which means you cannot change their values after calling the object. Objects lock when you call them, and the release function unlocks them.

If a property is tunable, you can change its value at any time.

For more information on changing property values, see System Design in MATLAB Using System Objects.

Method — Averaging method

"Sliding window" (default) | "Exponential weighting"

Averaging method, specified as "Sliding window" or "Exponential weighting".

• "Sliding window" — A window of length specified by SpecifyWindowLength is moved over the input data along each channel. For every sample the window moves by, the object computes the average over the data in the window.

• "Exponential weighting" — The object multiplies the samples with a set of weighting factors. The magnitude of the weighting factors decreases exponentially as the age of the data increases, never reaching zero. To compute the average, the algorithm sums the weighted data.

For more details on these methods, see Algorithms.

SpecifyWindowLength — Specify window length

true (default) | false

Flag to specify a window length, specified as a scalar boolean.

• true — The length of the sliding window is equal to the value you specify in the WindowLength property.

• false — The length of the sliding window is infinite. In this mode, the average is computed using the current sample and all the past samples.

Dependencies

This property applies when you set Method to "Sliding window".

WindowLengthSource — Source of window length

"Property" (default) | "Input port"

Since R2026a

Source of the window length, specified as one of these:

• "Property" –– Specify the window length using the WindowLength property.

• "Input port" –– Specify the window length as an input argument.

Dependencies

To enable this property, set:

• Method to "Sliding window".

• SpecifyWindowLength to true.

Data Types: char | string

WindowLength — Length of sliding window

4 (default) | positive integer

Length of the sliding window in samples, specified as a positive integer that is less than or equal to the value you specify in the MaxWindowLength property.

Dependencies

To enable this property, set:

• Method to "Sliding window".

• SpecifyWindowLength to true.

• WindowLengthSource to "Property". (since R2026a)

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

MaxWindowLength — Maximum length of sliding window

20 (default) | positive integer

Since R2026a

Maximum length of the sliding window in samples, specified as a positive integer.

Dependencies

To enable this property, set:

• Method to "Sliding window".

• SpecifyWindowLength to true.

• WindowLengthSource to "Input port".

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

OverlapLength — Overlap length between windows

WindowLength − 1 (default) | nonnegative integer

Overlap length between sliding windows, specified as a nonnegative integer. The value of overlap length varies in the range [0, WindowLength − 1]. If not specified, the overlap length is WindowLength − 1.

When you set the WindowLengthSource property to "Input port", the object assumes an overlap length of WindowLength − 1 and provides one sample of output for every sample of input. (since R2026a)

Dependencies

To enable this property, set:

• Method to "Sliding window".

• SpecifyWindowLength to true.

• WindowLengthSource to "Property". (since R2026a)

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

ForgettingFactor — Exponential weighting factor

0.9 (default) | non-negative real scalar in the range [0,1]

Exponential weighting factor, specified as a non-negative real scalar in the range [0,1]. A forgetting factor of 0.9 gives more weight to the older data than does a forgetting factor of 0.1. A forgetting factor of 1.0 indicates infinite memory and all the past samples are given an equal weight. A forgetting factor of 0 indicates no memory and the past samples have no weight on the current computation.

Since this property is tunable, you can change its value even when the object is locked.

Tunable: Yes

Dependencies

To enable this property, set Method to "Exponential weighting".

Data Types: single | double

Usage

Syntax

y = movAvg(x)

y = movAvg(x,WL)

Description

y = movAvg(x) computes the moving average of the input signal, x, using either the sliding window method or exponential weighting method.

y = movAvg(x,WL) computes the moving average of the input signal using a sliding window of length WL. (since R2026a)

Input Arguments

x — Data input

vector | matrix

Data input, specified as a vector or a matrix. If x is a matrix, each column is treated as an independent channel. The moving average is computed along each channel.

The object accepts variable-size inputs. Once the object is locked, you can change the size of each input channel, but you cannot change the number of channels.

Data Types: single | double
Complex Number Support: Yes

WL — Length of sliding window

positive integer

Since R2026a

Length of the sliding window in samples, specified as a positive integer that is less than or equal to the value you specify in the MaxWindowLength property.

To enable this input argument, set:

• Method to "Sliding window".

• SpecifyWindowLength to true.

• WindowLengthSource to "Input port".

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

Output Arguments

y — Moving average

vector | matrix

Moving average of the input signal, returned as a vector or a matrix.

When you input a signal of size m-by-n to the object, and if you set Method to "Sliding window" and SpecifyWindowLength to true, the output has an upper bound size of ceil(m/hop size)-by-n. Hop size is window length − overlap length. In other cases, the output has a size of m-by-n.

When you generate code from this object, the variable-size behavior of the output in the generated code depends on the input frame length and whether the size of the input signal is fixed or variable. For more details, see Code Generation.

Data Types: single | double
Complex Number Support: Yes

Object Functions

To use an object function, specify the System object as the first input argument. For example, to release system resources of a System object named obj, use this syntax:

release(obj)

Common to All System Objects

step	Run System object algorithm
release	Release resources and allow changes to System object property values and input characteristics
reset	Reset internal states of System object

Examples

Moving Average of Noisy Ramp Signal

Compute the moving average of a noisy ramp signal using the dsp.MovingAverage object.

Initialization

Set up movavgWindow, movavgWindow_overlap, and movavgExp objects. movavgWindow uses the sliding window method with a window length of 50 samples and a default overlap length of 49 samples, which is one sample less than the specified window length. movavgWindow_overlap uses a window length of 50 samples and an overlap length of 45 samples. movavgExp uses the exponentially weighting method with a forgetting factor of 0.95.

Create a time scope for viewing the output.

FrameLength = 1001;
Fs = 1000;
movavgWindow = dsp.MovingAverage(50);
movavgWindow_overlap = dsp.MovingAverage(50,45);
movavgExp = dsp.MovingAverage(Method='Exponential weighting',...
    ForgettingFactor=0.95);
scope  = timescope(SampleRate=[Fs, Fs, Fs/(50-45), Fs],...
    TimeSpanOverrunAction='Scroll',...
    ShowGrid=true,...
    YLimits=[-0.5 1.5]);
title = 'Moving Average';
scope.Title = title;
scope.ChannelNames = {'Original Signal',...
    'Sliding window of 50 samples with default overlap',...
    'Sliding window of 50 samples with an overlap of 45 samples',...
    'Exponential weighting with forgetting factor of 0.95'};

Compute the Average

Generate a ramp signal with an amplitude of 1 and a time span of 2 seconds. Apply the sliding window average and exponentially weighted average to the ramp. View the output in the time scope.

for i = 1:500
    t = (0:0.001:1)';
    unitstep = t>=0;
    ramp = t.*unitstep;
    x = ramp + 0.1 * randn(FrameLength,1);
    y1 = movavgWindow(x);
    y2 = movavgWindow_overlap(x);
    y3 = movavgExp(x);
    scope(x,y1,y2,y3);
end

Tune Window Length During Simulation

Since R2026a

Create two dsp.MovingAverage objects, one with a fixed window length of 50, and the other whose window length varies with the input noise power. Compute the moving average of a noisy square wave signal using these objects. Initialize a time scope to view and compare the moving average outputs.

FrameLength = 100;
Fs = 100;
movAvgFixedWL = dsp.MovingAverage(WindowLength=50);
movavgWL = dsp.MovingAverage(WindowLengthSource="Input port",MaxWindowLength=1000);
scope  = timescope(SampleRate=Fs,...
    TimeSpanOverrunAction='Scroll',...
    TimeSpanSource='Property',...
    TimeSpan=1000,...
    ShowGrid=true,...
    BufferLength=1e7,...
    YLimits=[-30,40]);
title = 'Moving Average with Tunable Window Length';
scope.Title = title;
scope.ChannelNames = {'Noisy signal',...
    'Moving average with fixed window length (50)',...
    'Moving average with varying window length'};

Create a noisy square wave signal. The variance of the noise changes every time the signal amplitude changes. Vary the window length of one of the dsp.MovingAverage objects such that it is directly proportional to the noise power. Higher the noise power, greater is the window length.

Compute the moving average of the noisy square wave signal with a fixed window length of 50 and with the varying window length. Compare the two outputs in the time scope.

When the window length is large, the moving average output is smooth even though there is more noise.

count = 1;
noisepower =  [1 7 10 1];
signalBase = 10*[1 2 -1 3];
for index = 1:length(noisepower)
    np = noisepower(index);
    yexp = signalBase(index)*ones(FrameLength,1);
    WL = 50*np;
    for i = 1:250
        x = yexp+sqrt(np)*randn(FrameLength,1);
        y1 = movAvgFixedWL(x);
        y2 = movavgWL(x,WL);
        scope(x,y1,y2)
    end
end

Algorithms

Sliding Window Method

In the sliding window method, the output for each input sample is the average of the current sample and Len – 1 previous samples. Len is the length of the window in samples. To compute the first output sample, the algorithm waits until it receives the hop size number of input samples. Hop size is defined as window length – overlap length. Remaining samples in the window are considered to be zero. As an example, if the window length is 5 and the overlap length is 2, then the algorithm waits until it receives 3 samples of input to compute the first sample of the output. After generating the first output, it generates the subsequent output samples for every hop size number of input samples.

When you do not specify the window length, the algorithm chooses an infinite window length. In this mode, the output is the moving average of the current sample and all the previous samples in the channel.

For an example, see Sliding Window Method and Exponential Weighting Method.

Exponential Weighting Method

In the exponential weighting method, the moving average is computed recursively using these formulas:

$$\begin{array}{l}{w}_{N,\lambda }=\lambda w_{N-1,\lambda }+1\\ {\overline{x}}_{N,\lambda }=\left(1-\frac{1}{w_{N,\lambda }}\right){\overline{x}}_{N-1,\lambda }+\left(\frac{1}{w_{N,\lambda }}\right)x_N\end{array}$$

• $${\overline{x}}_{N,\lambda }$$ — Moving average at the current sample

• $$x_N$$ — Current data input sample

• $${\overline{x}}_{N-1,\lambda }$$ — Moving average at the previous sample

• λ — Forgetting factor

• $$w_{N,\lambda }$$ — Weighting factor applied to the current data sample

• $$\left(1-\frac{1}{w_{N,\lambda }}\right){\overline{x}}_{N-1,\lambda }$$ — Effect of the previous data on the average

For the first sample, where N = 1, the algorithm chooses $$w_{N,\lambda }$$ = 1. For the next sample, the weighting factor is updated and used to compute the average, as per the recursive equation. As the age of the data increases, the magnitude of the weighting factor decreases exponentially and never reaches zero. In other words, the recent data has more influence on the current average than the older data.

The value of the forgetting factor determines the rate of change of the weighting factors. A forgetting factor of 0.9 gives more weight to the older data than does a forgetting factor of 0.1. A forgetting factor of 1.0 indicates infinite memory. All the previous samples are given an equal weight.

For an example, see Sliding Window Method and Exponential Weighting Method.

References

[1] Bodenham, Dean. “Adaptive Filtering and Change Detection for Streaming Data.” PH.D. Thesis. Imperial College, London, 2012.

Extended Capabilities

C/C++ Code Generation

Generate C and C++ code using MATLAB® Coder™.

Usage notes and limitations:

• See System Objects in MATLAB Code Generation (MATLAB Coder).

• When you set Method to "Sliding window" and SpecifyWindowLength to true, and generate code from this object, the variable-size behavior of the output depends on the input frame length and whether the size of the input signal is fixed or variable.

  See this table for more details.

  Input signal	Input size	Output signal
  Fixed-size	m-by-n	When the input frame length is a multiple of the hop size, the output signal has a fixed-size of (m/hop size)-by-n. When input frame length is not a multiple of the hop size, the output signal is variable-sized and has an upper bound of ceil(m/hop size)-by-n.
  Variable-size	m-by-n	Output is a variable-size signal. Output has an upper bound size of ceil(m/hop size)-by-n.

Version History

Introduced in R2016b

R2026a: Input window length and tune value during simulation

When you set the averaging method of the dsp.MovingAverage object to sliding window, you can specify the window length as an input to the object, and tune its value during simulation. This functionality enables you to vary the window length based on the characteristics of the input signal. For an example, see Tune Window Length During Simulation.

The window length input must be a positive integer that is less than or equal to the value of the MaxWindowLength property.

R2024b: Specify a forgetting factor of 0

You can now specify a forgetting factor of 0 in the dsp.MovingAverage object.

R2022b: Change in variable-size behavior for output signal in generated code

Starting in R2022b, if you generate code from this object with the Method property set to "Sliding window" and the SpecifyWindowLength property set to true, and if you input a signal with a frame length that is a multiple of the hop size (window length − overlap length), this object generates a fixed-size output signal in the generated code. For more details, see Code Generation.