Main Content

dspdata

DSP data parameter information

Syntax

Hs = dspdata.dataobj(input1,...)

Description

Note

The use of dspdata.dataobj is not recommended. Use the appropriate function interface instead.

Hs = dspdata.dataobj(input1,...) returns a dspdata object Hs of type dataobj. This object contains all the parameter information needed for the specified type of dataobj. Each dataobj takes one or more inputs, which are described on the individual reference pages. If you do not specify any input values, the returned object has default property values appropriate for the particular dataobj type.

Note

You must use a dataobj with dspdata.

Data Objects

A data object, dataobj, for dspdata specifies the type of data stored in the object. Available dataobj types for dspdata are shown below.

dspdata.dataobj

Description

Corresponding Functions

dspdata.msspectrum

Mean-square spectrum data (power)

periodogram

pwelch

dspdata.psd

Power spectral density data (power/frequency)

pburg

pcov

periodogram

pmcov

pmtm

pwelch

pyulear

dspdata.pseudospectrum

Pseudospectrum data (power)

peig

pmusic

For more information on each dataobj type, use the syntax help dspdata.dataobj at the MATLAB® prompt or refer to its reference page.

Methods

Methods provide ways of performing functions directly on your dspdata object. You can apply these methods directly on the variable you assigned to your dspdata object.

Method

Description

avgpower

This method applies only to dspdata.psd objects.

avgpower(Hs) computes the average power of a signal, Hs, in a given frequency band. The technique uses a rectangle approximation of the integral of the signal's power spectral density (PSD). If the signal is a matrix, the computation is done on each column. The average power is the total signal power. The SpectrumType property determines whether the total average power is contained in the one-sided or the two-sided spectrum. For a one-sided spectrum, the range is [0,pi] if the number of frequency points is even and [0,pi) if it is odd. For a two-sided spectrum, the range is [0,2pi).

avgpower(Hs,freqrange) specifies the frequency range over which to calculate the average power. freqrange is a two-element vector containing the lower and upper bounds of the frequency range. If a frequency value does not match exactly the frequency in Hs, the next closest value is used. The first frequency value in freqrange is included in the calculation and the second value is excluded.

centerdc

centerdc(Hs) or centerdc(Hs,true) shifts the data and frequency values so that the DC component is at the center of the spectrum. If the SpectrumType property is 'onesided', it is changed to 'twosided' and then the DC component is centered.

centerdc(Hs,'false') shifts the data and frequency values so that the DC component is at the left edge of the spectrum.

findpeaks

findpeaks(Hs) finds local maxima or peaks. If no peaks are found, findpeaks returns an empty vector.

[pks,frqs] = findpeaks(x) returns the peaks’ values, pks, and the frequencies, frqs, at which they occur.

findpeaks(x,'minpeakheight',mph) returns only peaks greater than the minimum peak height mph, where mph is a real scalar. The default is -Inf.

findpeaks(x,'minpeakdistance',mpd) returns only peaks separated by the minimum frequency units distance mpd, which is a positive integer. Setting the minimum peak distance ignores smaller peaks that may occur close to larger local peaks. The default is 1.

findpeaks(x,'threshold',th) returns only peaks greater than their neighbors by at least the threshold, th, which is a real, scalar value greater than or equal to 0. The default is 0.

findpeaks(x,'npeaks',np) returns a maximum of np number of peaks. When np peaks are found, the search stops. The default is to return all peaks.

findpeaks(x,'sortstr',str) specifies the sorting order, where str is 'ascend', 'descend', or 'none'. When str is set to 'ascend', the peaks are sorted from smallest to largest. When str is set to 'descend' the peaks are sorted in descending order. When str is set to 'none', the peaks are returned in the order in which they occur.

halfrange

halfrange(Hs) converts the spectrum of Hs to a spectrum calculated over half the Nyquist interval. All associated properties affected by the new frequency range are adjusted automatically. This method is used for dspdata.pseudospectrum objects.

The spectrum is assumed to be from a real signal. That is, halfrange uses half the data points regardless of whether the data is symmetric.

normalizefreq

normalizefreq(Hs) or normalizefreq(Hs,true) normalizes the frequency specifications in the Hs object to Fs so the frequencies are between 0 and 1. It also sets the NormalizedFrequency property to true.

normalizefreq(Hs,false) converts the frequencies to linear frequencies.

normalizefreq(Hs,false,Fs) sets a new sampling frequency, Fs. This can be used only with false.

onesided

onesided(Hs) converts the spectrum of Hs to a spectrum calculated over half the Nyquist interval and containing the total signal power. All associated properties affected by the new frequency range are adjusted automatically. This method is used for dspdata.psd and dspdata.msspectrum objects.

The spectrum is assumed to be from a real signal. That is, onesided uses half the data points regardless of whether the data is symmetric.

plot

Displays the data graphically in the current figure window.

For a dspdata.psd object, it displays the power spectral density in dB/Hz.

For a dspdata.msspectrum object, it displays the mean–square in dB.

For a dspdata.pseudospectrum object, it displays the pseudospectrum in dB.

sfdr

This method applies only to dspdata.msspectrum objects.

sfdr(Hs) computes the spurious-free dynamic range (SFDR) in dB of a mean square spectrum object Hs. SFDR is the usable range before spurious noise interferes with the signal.

[sfd,spur,frq] = sfdr(Hs) returns the magnitude of the highest spur and the frequency frq at which it occurs.

sfdr(Hs,'minspurlevel',msl) ignores spurs below the minimum spur level msl, which is a real scalar in dB.

sfdr(Hs,'minspurdistance',msd) includes spurs only if they are separated by at least the minimum spur distance msd, which is a real, positive scalar in frequency units.

twosided

twosided(Hs) converts the Hs spectrum to a spectrum calculated over the whole Nyquist interval. All associated properties affected by the new frequency range are adjusted automatically. This method is used for dspdata.psd and dspdata.msspectrum objects.

If your data is nonuniformly sampled, converting from onesided to twosided may produce incorrect results.

wholerange

wholerange(Hs) converts the Hs spectrum to a spectrum calculated over the whole Nyquist interval. All associated properties affected by the new frequency range are adjusted automatically. This method is used for dspdata.pseudospectrum objects.

If your data is nonuniformly sampled, converting from half to wholerange may produce incorrect results.

For more information on each method, use the syntax help dspdata/method at the MATLAB prompt.

Plotting a dspdata Object

The plot method displays the dspdata object spectrum in a separate figure window.

Modifying a dspdata Object

After you create a dspdata object, you can use any of the methods in the table above to modify the object properties. For example, to change an object, Hs, from two-sided to one-sided, use onesided(Hs).

Examples

See the dspdata.msspectrum, dspdata.psd, and dspdata.pseudospectrum reference pages for specific examples.

Version History

Introduced before R2006a