LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

waveform recognition

Hi Guys:

I am doing a chime signal recognition project,I have a Midea player,which will play different chime signals,I use a PXI-4462 to acquire the signal,and write it to .wav file,and I have to judge which chime signal is playing.Here I post four different wave files.My question is how to reccognize the signal I get?

Download All
0 Kudos
Message 1 of 8
(3,765 Views)

What is a "Midea" player?  Do you mean "media"?

 

It looks like the chimes are different frequencies.  Perform an FFT on the incoming waveform, or use Extract Single Tone Information, and you should be able to find the fundamental frequency and identify which waveform is playing.

 

 

0 Kudos
Message 2 of 8
(3,726 Views)

Yes,it is a Media player,it is my spelling mistake.

Question1:

There is a fact that the wavform I acquire of the chime signal is not the same every time.so I doubt is it right to perform an FFT,the following picture shows my doubt:

5.png

Note:Graph2 is part of Graph1,Graph1 is a waveform file of 441000 points.Actually,maybe I will acquire Graph2,or another graph which is also part of Graph1.Although Graph1 and Graph2 have the same Freq,but there FFT spectrum is different.

Question2:

Beep(750Hz) and Gong(750Hz) have the same Freq ,so it's diffcult for me to use fundamental frequency to identify them:

Beep(750Hz) has Freq:272

1.png

 

Gong(750Hz) also has Freq:272

3.png

 

And it's same for the 2000Hz chime signals.

 

 

 

0 Kudos
Message 3 of 8
(3,719 Views)

You are taking a portion of the last waveform into your FFT in your Graph2 examples.  Graph1 is complete and could be looped creating a continuous repeating pattern where that could not happen in Graph2.  Try to look either at the periodic sound pulses or the periodic sound pulses including its off time (downtime / quiet time) as the complete periodic waveform to perform the FFT.  You need to be consistent. 

Help the Community (and future reviewers) by marking posts as follows:
If it helped - KUDOS
If it answers the issue - SOLUTION
0 Kudos
Message 4 of 8
(3,705 Views)

Okay.  So FFT is not enough to determine which is which.  It can help.  So your problem is to figure out what is different about the Beep vs. Gong and determine some other parameters you can analyze and compare.

 

Note, you will never have something exactly equal to something else.  You will have to put some tolerances in your evaluation.

 

Look at the signal analysis palette.  There are functions called convolution and deconvolution that might help.

0 Kudos
Message 5 of 8
(3,700 Views)

We had already discussed how taking the amplitude would be capable of exhibiting the difference in another post. https://forums.ni.com/t5/Automotive-and-Embedded-Networks/automative-priority-testing/td-p/1088553

 

Again, as this is sound, the issue is dependent on any number of environmental variables.  It will be hard to pick a hard and fast value for what amplitude separates a Beep from a Gong.

Help the Community (and future reviewers) by marking posts as follows:
If it helped - KUDOS
If it answers the issue - SOLUTION
0 Kudos
Message 6 of 8
(3,697 Views)

Hi guys:

I am back.Sorry for absence these days,because I am working on other project.I will update my status for this topic tomorrow,because it is late in china.I have to go home now,byebye!

 

0 Kudos
Message 7 of 8
(3,657 Views)

Hi Minions:

I have use another method to judge which chime signal is playing,beep or gong.I use IMAQ Match Pattern,I save the graph to .bmp file and use Match pattern to detect the number of matches,if it is bigger than 0,the chime signal is the type of Template.2017-06-12_130643.png

 

You can find the vi in the attachment.

Now I have make it clear that I have the following parameters to measure:

2017-05-26_201247.png

1:Sound duty cycle;

2:Sound cadence Period;

3:Number of repetitions;

Number of repetitions can be measured in the chime check.vi,it is the "Number of Matches ".

My problem is to measure "Sound duty cycle" and "Sound cadence Period".

 

 

Download All
0 Kudos
Message 8 of 8
(3,624 Views)