Prev: system of partial differential and differential equations
Next: Is there a MATLAB function similiar to TableCurve or FindGraph?
From: Steven_Lord on 9 Aug 2010 12:54
"Shamma Raghib" <critico.nino(a)gmail.com> wrote in message
news:i3p9vc$kj7$1(a)fred.mathworks.com...
> "Andy " <myfakeemailaddress(a)gmail.com> wrote in message
> <i3p8gg$g6r$1(a)fred.mathworks.com>...
>> Well, since you haven't posted your distance function, the only error I'm
>> getting is that the distance function is undefined. Could you post your
>> error message, rather than forcing us to guess at it?
>
>
> ERROR:
> ??? Attempted to access RSX(2); index out of bounds because numel(RSX)=1.
> Error in ==> polytry2 at 29
> d=distance(MSX,MSY,RSX(j),RSY(j))
>
> Here is my distance function:
> Like I said its plain and simple :)
> %%%%%%%%
> function d=distance(a, b, x, y)
> d=sqrt((a-x).^2+(b-y).^2);
>
> end

Looking at a piece of your original code:


d_tot=0;
for j=1:6
[RSX,RSY]=pol2cart(relay_theta(j),relay_r(j)); %transform RS
position to the cartesian coordinate system


hold on;
plot(RSX,RSY,'b*');
end
for j=1:6
d=distance(MSX,MSY,RSX(j),RSY(j));
end


Since you overwrite RSX and RSY during each iteration of the code I quoted,
when you reach your distance call RSX and RSY don't have 2nd elements --
they are 1-by-1 scalars. POL2CART is vectorized; it can accept a vector of
theta and r values, so use that instead of the first loop I quoted. Then
RSX and RSY will be vectors and your distance call will work. But actually,
your distance function will work with vectors x and y (since you're using
the elementwise .^ operator instead of the matrix ^ operator so you don't
need _that_ loop either.

--
Steve Lord
slord(a)mathworks.com
comp.soft-sys.matlab (CSSM) FAQ: http://matlabwiki.mathworks.com/MATLAB_FAQ
To contact Technical Support use the Contact Us link on
http://www.mathworks.com

From: Shamma Raghib on 9 Aug 2010 13:34
Hey thanks Andy, Jan, Steve,

Yea you were right about the vector thing. I changed the second for loop code to:

for j=1:6
d=distance(user_theta,user_r,relay_theta(j),relay_r(j));
end

but now a new problem occurs I am assuming it is something similar....in the code:
%%%%%line after the second distance for loop
d_tot=d_tot+d(j)^-gamma;

error: ??? Attempted to access d(6); index out of bounds because numel(d)=1.
Error in ==> polytry2 at 32
d_tot=d_tot+d(j)^-gamma;

Does it mean that since d_tot is scalar, the loop is not progressing? :S
This is where the debugger stops. I just want a summation of all the 6 vector distances that I calculates in the loop. Is this the wrong approach?

Thanks!
From: Shamma Raghib on 9 Aug 2010 13:44
I forgot to mention something. I think my last code (above messege of mine) was wrong since I wanted to calculate the distance in terms of polar coordinates.

(user_r, user_theta ) are the coordinates for the random MS points
(relay_r, relay_theta) are the coordinates for the fixed 6 points

Now after the loop runs, I wanted to find the distance of one MS as it keeps on adding a point at each iteration. So at first I converted it to normal cartesian coordinate and then tried calculating the distance. but the first error occured.

Then i tried changing the vector value as Steve suggested, but i realized that both the points are in polar coordinates so the distance function won't be logical.

So I have to do it in terms of polar distance calculation...
From: Shamma Raghib on 9 Aug 2010 13:54
Hey guys! I solved it :)

Thanks a lot. Now I understand the vector error issue. Thank Jan for the debug command. Thanks Andy for your help. Thanks Steve for pointing it out to me!

Cheers!
First  |  Prev  | 
Pages: 1 2
Prev: system of partial differential and differential equations
Next: Is there a MATLAB function similiar to TableCurve or FindGraph?