Welch method in Signal Analyis toolbox and using the matlab function

1 次查看（过去 30 天）

Javier Abanades 2017-8-1

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/351101-welch-method-in-signal-analyis-toolbox-and-using-the-matlab-function

编辑： Prasanth Sunkara 2017-8-9

Dear folks,

I have done a signal analysis using the welch function in matlab and using the toolbox. I have found quite funny that using the matlab function takes quite long; however using the signal analysis toolbox takes nothing.

What is the difference to find these differences?

It would be great to be capable of generating the code from the toolbox in order to develope a code that would allow to analyse multiple series without the necessity of imporing every variable (i.e. automatise the process).

Thank you very much for your time

All the best, Javier

3 个评论
显示 1更早的评论隐藏 1更早的评论

Javier Abanades 2017-8-7

编辑：Javier Abanades 2017-8-9

Dear Prasanth,

Thank you very much for your answer.

When I meant the toolbox, I mean the SPTool, which allows importing variables and create spectra by means of different methods: FFT, Welch,...

This operation is carried out in terms of seconds.

However when I program my own code using for instance pwelch as follows:

[ppx,f]=pwelch(variable,1024,0,0.3:0.0001:0.75,5);

%pwelch(variable, window,overlap,frequencies,fs);

the computational costs are huge.

I hope you can clarify this.

Thank you

All the best, Javier

Edit my answer: profiling the code, it seems that goertzel function is consuming most part of the time (90%). Is there any way to optimise this? I have managed to reduce the computational costs by means of reducing the precision of the results, but using spTool, the precision is even larger than the one used in my code.

Prasanth Sunkara 2017-8-9