How to get a signal without NaN values?

4 vues (au cours des 30 derniers jours)

Haya Ali le 12 Mar 2023

0
Lien

Utiliser le lien direct vers cette question

https://fr.mathworks.com/matlabcentral/answers/1927200-how-to-get-a-signal-without-nan-values

Commenté : Walter Roberson le 14 Mar 2023

I have a signal and I have applied the Shannon's Nyquist Theorem to get the rate and the reconstructed signal by following formulas.

But I am getting NaN values in the Rate Signal. Is there any way that to not get the NaN values? Below is my code.

close all; clear all; clc;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
X=[11477.9
3
4
11413
9
1
9
1
3
9
4
7
4
2
2
6
4
1
4
5
3
1
9
2
3
8
5
8
8
8
9
5
6
11523
11516
1
9
3
9
1
7
5
11463
5
3
1
2
9
8
11456
5
6
1
11454
3
4
11501
8
2
2
8
6
3
5
4
9
6
3
6
2
5
6
6
2
5
11445
6
3
2
3
11456
6
7
8
6
4
11405
6
3
6
8
11405
7
2
5
6
8
4
8
9
11353
8
6
2
2
4
11341
3
8
3
1
7
4
9
6
2
3
5
3
5
7
6
11402
8
3
9
7
5
5
4
5
5
8
4
4
3
6
11409
3
6
3
6
5
6
6
5
6
5
11423
8
4
9
5
7
5
6
8
5
7
11395
8
5
8
3
4
3
11365
8
11404
6
9
5
3
2
1
6
1
3
4
6
5
7
5
3
1
3
6
9
8
9
3
6
9
5
11346
5
2
8
7
2
8
8
6
2
8
9
8
7
3
11295
2
8
1
2
7
5
6
4
3
11327
5
4
7
11367
1
2
7
2
9
3
6
11390
7
11429
1
3
11452
5
6
1
4
4
1
3
9
4
2
5
3
8
8
4
3
1
1
1
11355
8
7
7
9
2
2
5
3
3
8
2
6
9
5
9
8
5
2
7
8
1
6
2
4
7
8
8
6
5
6
7
3
11375
11398
4
2
3
11440
5
11438
1
1
6
1
6
1
8
7
3
7
5
11401
3
4
1
3
5
8
7
4
4
7
9
1
8
9
1
2
11530
11539
9
3
2
7
11490
9
1
9
6
8
9
4
1
11529
2
4
6
9
5
6
7
8
6
4
5
9
5
5
3
5
3
5
5
6
7
2
7
9
9
6
1
3
4
9
4
3
7
6
11355
8
7
9
4
5
11382
11370
6
1
11321
7
11294
5
6
3
9
7
9
2
11341
8
5
7
8
4
2
3
6
3
4
1
11465
5
1
6
7
11491
5
9
8
9
5
11463
3
9
5
3
8
11445
8
9
9
6
1
11402
8
7
11426
7
5
4
1
11423
11450
3
11486
3
1
8
5
5
8
7
2
6
3
7
8
8
11465
5
7
8
2
4
11508
9
9
6
1
4
4
3
8
11385
1
4
5
9
2
6
11323
3
3
5
6
5
4
4
5
11415
9
8
2
9
6
4
11411
2
2
8
7
9
4
9
4
9
4
3
4
11406
9
7
1
4
2
11472
3
7
3
8
1
3
6
9
8
1
5
1
1
6
8
7
5
2
4
5
5
8
4
4
4
9
11340
1
1
6
4
9
4
6
7
1
9
5
2
4
9
4
7
8
2
11746
5
11805
11714
6
9
2
9
11505
9
1
4
6
7
11450
1
1
6
11406
3
2
5
7
2
2
11267
5
4
3
6
1
5
9
8
1
2
2
8
8
1
4
6
8
9
8
11823
9
5
11550
1
9
8
9
11447
3
1
4
9
3
5
2
4
5
3
3
7
3
1
6
8
11395
2
8
3
11299
6
3
4
3
8
5
4
2
6
9
3
6
8
3
5
5
5
6
5
6
3
5
6
6
7
5
1
4
1
4
4
7
9
5
9
2
6
6
11476
11492
11438
4
6
6
6
9
4
1
9
7
11463
8
8
5
8
1
2
9
2
3
6
1
8
1
1
1
8
7
1
9
3
4
8
2
9
9
4
9
2
3
5
9
2
8
6
6
7
5
11439
2
3
11381
2
2
11391
9
8
11402
7
7
9
5
2
7
6
5
4
6
11321
2
6
8
11361
11363
9
1
11394
7
11394
1
4
2
11438
8
7
8
2
3
1
3
8
4
6
7
11317
3
11322
8
2
4
2
3
11325
7
3
11356
7
6
7
3
6
2
8
4
8
5
5
3
3
7
6
6
8
6
3
11380
9
6
4
2
7
11349
3
5
2
6
11385
8
5
5
6
9
3
3
3
2
7
11357
8
1
1
3
3
2
11311
8
1
1
3
4
9
1
1
4
3
5
9
7
5
4
6
3
2
9
6
5
1
5
5
9
2
9
5
4
1
9
5
8
3
3
2
2
7
4
3
8
8
6
2
3
6
5
7
9
2
5
11351
7
5
8
6
5
6
3
1
5
9
9
2
9
4
5
5
1
8
9
4
2
4
5
4
3
11332
9
2
5
11327
4
1
5
3
2
7
1
4
8
8
5
11366
11401
11366
2
6
8
1
3
9
9
9
2
9
1
8
8
8
8
9
6
11390
11363
6
8
5
1
4
5
7
5
2
5
2
8
9
8
3
8
9
2
9
3
7
11311
6
3
2
8
6
5
7
7
3
3
1
4
6
9
3
1
6
9
11360
8
11376
3
2
9
1
9
1
7
8
7
2
3
7
6
7
4
3
6
8
7
8
8
11368
7
6
1
3
3
3
1
4
9
7
2
2
6
2
5
5
7
1
7
6
8
6
7
5
11397
9
3
2
4
1
7
5
6
3
1
5
3
2
2
1
1
5
2
1
11380
3
8
6
3
7
6
7
9
7
11354
8
4
1
3
11370
6
9
5
3
6
11335
7
8
7
11345
7
4
6
6
11346
11325
1
8
5
5
6
2
2
4
4
11372
5
3
4
5
7
11356
2
5
6
7
8
4
3
9
7
8
8
2
2
9
1
11394
5
6
2
5
6
11414
3
4
6
5
6
3
9
2
6
4
7
1
6
8
7
4
5
9
5
6
3
6
1
3
2
4
6
7
5
1
6
2
6
7
1
6
2
7
11370
4
1
6
11336
5
4
5
9
9
5
5
2
5
7
7
4
3
2
2
8
11321
6
9
3
7
4
7
3
3
7
5
7
6
3
11361
9
9
8
];
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Addnum=20;%Number of additional data points in each time interval.
N=Addnum*length(X); %Length of the reconstructed signal(the continuous signal);
Rec=zeros([1,N]);
Rate=zeros([1,N]);
nx=zeros([1,N]);
NX=zeros([1,length(X)]);
zeroline=zeros([1,N]);
index=0;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
for i=1:N
    if mod(i,Addnum)==1
        index=index+1;
        NX(index)=i;
    end
  nx(i)=i;  
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
for i=1:N
Rec(i)=0;
Rate(i)=0;
for j=1:length(X)
    Rec(i)=Rec(i)+X(j)*sinc((1/Addnum)*(i-Addnum*(j-1)));
    Rate(i)=Rate(i)+X(j)*(cos((pi/Addnum)*(i-Addnum*(j-1)))/(i-Addnum*(j-1))-sin((pi/Addnum)*(i-Addnum*(j-1)))/((pi/Addnum)*(i-Addnum*(j-1))^2));
end
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mean_X=mean(X);
X=X-mean_X;
mean_Rec=mean(Rec);
Rec=Rec-mean_Rec;
plot(nx,Rec,'sg');
hold on
plot(nx,zeroline,'k');
plot(nx,Rate,'*b');
plot(NX,X,'or')
axis([1000 10000 -400 600])
legend ('Reconstructed Signal','zeroline','Rate Signal','BOLD Signal X')

4 commentaires
Afficher 2 commentaires plus anciensMasquer 2 commentaires plus anciens

Dyuman Joshi le 12 Mar 2023

What you are trying to do for the variable Rate? Do you have a reference formula?

What value it should be instead of NaN?

Haya Ali le 12 Mar 2023

I have attached a formula for rate signal in the image. The value can be anything other than NaN

Connectez-vous pour commenter.

Connectez-vous pour répondre à cette question.

Réponse acceptée

Image Analyst le 12 Mar 2023

0
Lien

Utiliser le lien direct vers cette réponse

https://fr.mathworks.com/matlabcentral/answers/1927200-how-to-get-a-signal-without-nan-values#answer_1191180

Ouvrir dans MATLAB Online

Try removing the nans or infs

badIndexes = isnan(Rate) | isinf(Rate);
nx2 = nx(~badIndexes);
Rate2 = Rate(~badIndexes);
plot(nx2, Rate2, '*b');

10 commentaires
Afficher 8 commentaires plus anciensMasquer 8 commentaires plus anciens

Image Analyst le 14 Mar 2023

OK. Though it's probably more trouble to avoid/prevent the NaNs than to just fix them afterwards. But whatever you intuitively feel most comfortable with is fine.

Walter Roberson le 14 Mar 2023

The code involves loop summing terms, some of which might be Nan. If you detect the potential for NaN at the term level, then you can omit adding that term to the summation, so you end up with a sum of non-NaN values that might contain useful information. If you wait until afterwards to detect the NaN then you lose the information in the non-NaN terms.

Connectez-vous pour commenter.

Plus de réponses (0)

Connectez-vous pour répondre à cette question.

Catégories

Industries Medical Devices Cardiology

En savoir plus sur Cardiology dans Help Center et File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by