Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

How to measure spectral flatness in Matlab?

Status
Not open for further replies.

omidi_sbu

Banned
Joined
Jul 20, 2005
Messages
56
Helped
1
Reputation
2
Reaction score
0
Trophy points
1,286
Location
iran
Activity points
0
what is the measure of spectral flatness?
how can i implement it in matlab?
 

spectral flatness

You measure using ripple.

Find the freq response and then the variance.
 

spectral flatness matlab

Spectral Flatness Measure is the ratio of the Geometric mean and the Arithmetic mean of the given signal's power spectrum.

Let the given signal be given by \[x[n]\] and let its power spectrum be given by \[S_x(e^{j\omega)}\], then SFM is given by the formula,

\[SFM = \frac{exp\left[\frac{1}{2\pi}\int\limits_{-\pi}^{\pi}\ln \left\{|S_x(e^{j\omega})|\right\}d\omega\right]}{\frac{1}{2\pi}\int\limits_{-\pi}^{\pi}|S_x(e^{j\omega})| d\omega}\]

You can easily see that the numerator term is the GM and the denominator is the AM of the spectrum.

As GM<AM we can say that the maximum value of \[SFM\] can take is \[1\]. If the spectrum is completely flat, then you see that the numerator is \[1\] and also the denominator is \[1\]. Thus \[SFM\] is also one. Thus we say that the above gives the measure of the flatness of the spectrum and hence the name Spectral Flatness Measure
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top