How to Implement Box Function in better way in Matlab? - matlab

Question

This function in Matlab presents Box_Function:
output = 1 while sample n belongs to the range (-a, +a).
otherwise the output is 0 outside that range.
So how can I implement this function in Matlab in a better way to shift the plot in case of negative values of time, without assigning negative values to the array.
Thanks in Advance
function B_X = Box_Func(N,K,a)
if(N <= 0)||(K+a > N)
warning('Please Enter Valid Positive Integer !');
else
B_X = zeros([-N N]);
for i = -N : 1 : N
if (i >= K-a) && (i <= K+a)
B_X(i)=1;
end
end
end
end

Answer 1

Your question is unclear since it does not really explain what you want to do, and in one comment you state that you know where the error comes from. I suggest to read the docs (also in a comment), but here I'll show you my problems with your code, provide some simple ways of testing your code and I hope this helps to solve your problem and to understand how to ask better questions.
First, one remark to the lines
if(N <= 0)||(K+a > N) % if samples Number wrong, or shifting exceeds limit
% of Samples Print a warning.
warning('Please Enter Valid Positive Integer !');
I suggest to throw an error instead of a warning if the input parameters are wrong and will lead to an error anyway. Otherwise you could omit the test and let Matlab throw the respective error.
The next misunderstanding is
B_X = zeros([-N N])
What do you expect B_X to be after this line if, let's say, N=2? Test if the result is what you expect by simply entering this command in the command line directly:
>> zeros([-2 2])
ans =
0×2 empty double matrix
I guess that's not what you expect. As the docs state, zeros(N) will yield a square matrix with N rows and N columns; zeros(M,N) will yield a matrix with M rows and N columns. Look:
>> zeros(2)
ans =
0 0
0 0
>> zeros(2,1)
ans =
0
0
I do not know what you expect from zeros([-2 2]), but I guess that you are looking for one of the following:
>> N = 2;
>> zeros(2*N+1,1)
ans =
0
0
0
0
0
>> zeros(1,2*N+1)
ans =
0 0 0 0 0
>> zeros(2*N+1)
ans =
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
My guess is that somehow you expect the function zeros to operate on some range of indices you provide. Your misunderstanding might be that you expect zeros([-2 2]) to provide a vector of zeros into which you can index using -2:2 (that is, one of -2,-1,0,+1,+2). If you assume this, your assumption is wrong.
I guess this from the line
for i = -N : 1 : N
in your code. Due to this line, I'll first thought that
>> B_X = zeros(1, 2*N+1)
B_X =
0 0 0 0 0
is what you expect. However, from the comment
% if samples Number wrong, or shifting exceeds limit
I guessed that N might be just the number of data points in the result. This would mean
>> B_X = zeros(1, N)
B_X =
0 0
(which would not make much sense for N=2). So, for the next question you ask (or an edit to this question): Clearly explain the meaning of the function inputs!
Since later, you set the limits of your x-axes to [-N N], I'll keep my first assumption, thus the number of data points (and therefore the argument to zeros) should be 2*N+1.
The next argument to your function is K, you call it the shift. k only occurs in combination with the third input of the function, a. You do not provide any inforamtion about a.
From this line I guess that a is something that specifies a width:
if (i >= K-a) && (i <= K+a)
Now, slowly, if one also considers
B_X(i)=1;
and the usage of the word box (and heaviside, which is still in the comments), one can get a clue of what you want to do. Together with your comment that you want your function
to appear on the plotting shifted on X Axis, so it looks like that the center of the Box Function is in the negative area of X axis
Might this be your goal: I want to plot a vector from -N to N (in steps of 1) with zero values except for the region of -K±a, where I want it to be one?
If this is the case, one attempt would be as follows (it remains to you to put it into a function):
>> N=15;
>> K=7;
>> a = 3;
Get the x-values:
>> x = -N:N;
(-15, -14, ... 14, 15). Next, allocate B_X:
>> B_X = zeros(1, 2*N+1);
Last, use logical indexing (this might help to understand this) to set the values around -K±a to one:
>> B_X(x>(-K-a) & x<(-K+a)) = 1;
Eventually, plot the result:
plot(x,B_X);
and adjust the axis limits:
>> ax=gca;
>> ax.YLim = [-.2 1.2];
Result is:

Answer 2

function B_X = Box_Func(N,K,a)
% Box_Func This Function takes Number of Samples N,Count of shift K, and
% half of Box width a then it stem Them ,
% Note it works only for positive shifting
% that means K should be positive or Zero
if(N <= 0)||(K+a > N) % if samples Number wrong, or shifting exceeds limit
% of Samples Print a warning.
warning('Please Enter Valid Positive Integer !');
else % if The inputs are fine , then :
B_X = zeros([1 2*N+1]);
x = -N:N;
B_X(x>=(K-a) & x<=(K+a)) = 1;
end
%===========================================
% Plotting the Results
%===========================================
figure('Name','Box Function','NumberTitle','off');
stem(x,B_X)
hold on
xlabel('Samples')
ylabel('Box Shifeted Function')
xlim([-N N]) ;
ylim([-1 2]);
grid on
hold off
end