Wrong answer to straightforward problem:

* Write a function called triangle_wave that computes the sum

for each of 1001 values of t uniformly spaced from 0 to 4π inclusive. The input argument is a scalar non-negative integer n, and the output argument is a row vector of 1001 such sums—one sum for each value of t. You can test your function by calling it with n == 20 or greater and plotting the result and you will see why the function is called “triangle_wave”.

My code is:

function [ out ] = triangle_wave( n )
%TRIANGLE_WAVE Summary of this function goes here
%   Detailed explanation goes here
t = 4*pi/1000;
tvector = 0:t:4*pi;
length(tvector)
out = 0;
for count = 1:n+1
    A= -1^(count-1);
    B = sin((2*(count-1)+1)*tvector);
    C = (2*(count-1)+1)^2;
    triangle = A*B/C;
    out = out + triangle;
end
end

This seems like it should be fairly straightforward.

Answer 1

[Steven Lord]

One problem is in this line.

A= -1^(count-1);

The power operator ^ is higher in operator precedence than the unary minus operator - so this is the equivalent of:

A= -(1^(count-1));

Wrap -1 in parentheses so you're raising minus 1 to a power rather than raising 1 to a power then negating it.

A= (-1)^(count-1);

You'll notice that this looks much more like the formula in your image. You should change the limit in your for loop, though, from 1:n+1 into 0:n to match the summation. [If you were storing each term individually in a vector or matrix you'd need to add 1 to the index into which to store the term, to account for the 1-based indexing of MATLAB, but you're not storing the terms just the sum of the terms.]

Answer 2

[DJ V]

Thanks a lot, that was the problem.