Using bsxfun to plot

%% I have this following code: 

n=10000;
g = exp(-linspace(-2,2,round(n/30)).^2);
[origsig, signal] = deal(conv(randn(n,1),g,'same'));
propnoise = 0.02;
noisepnts = randperm(n);
noisepnts = noisepnts(1:round(n*propnoise));

for ti = 1:length(noisepnts)
    tidx = noisepnts(ti):noisepnts(ti)+ceil(rand*10);
    signal(tidx) = NaN;
end

%%
islands = bwconncomp (~isfinite(signal));
islands.N = cellfun(@length,islands.PixelIdxList);
filtsig = signal;
for ii = 1:islands.NumObjects
        bndpoints = [islands.PixelIdxList{ii}(1)-1 islands.PixelIdxList{ii}(end)+1];
       
        if bndpoints(1)<1 || bndpoints(2)>n
            continue
        end

        linterp = linspace(0,1,islands.N(ii)+2)*diff(signal(bndpoints)) + signal(bndpoints(1));
        filtsig(bndpoints(1):bndpoints(2)) = linterp;
end

[DGM]

What is your question?

[Walter Roberson]

I do not see a question? You showed your code but I do not see where you ask for information or advice?

[Jae Won Lee]

Hi,

I have made the time series with using a faulty device that could eliminate the certain data points and using linear interpolation to fill in the missing values. I am wondering how I could plot these plots using bsxfun.

bsxfun(@time,[signal filtsig],???)

Answer 1

[Matt J]

Now idea why you think bsxfun should be involved. Why not just,

t=linspace(-2,2,n);

plot(t,signal,t,filtsig)