91Ó°ÊÓ

When it comes to calculating the inner product between two functions, a common tool is integration by parts. This technique is particularly helpful when the product of two functions is not straightforward to integrate directly. It is based on the product rule of differentiation and is formally expressed as the integral \[\int u dv = uv - \int v du\], where \(u\) and \(dv\) are chosen from the product of the functions.

For instance, when we have a case like in the exercise Part c, where the product \(x^2 \sin(x)\) doesn't have an elementary antiderivative, we resort to integration by parts. A suitable choice is to let \(u = x^2\) and \(dv = \sin(x)dx\), then differentiate \(u\) and integrate \(dv\) to obtain \(du\) and \(v\), respectively. Ultimately, the process simplifies the integral into more manageable terms, although it may sometimes require multiple applications to reach a solution or even another method to conclude that the integral evaluates to zero.

The concept of a symmetric interval, like \([-1,1]\) in Part b of the exercise, is crucial in understanding the properties of the functions being integrated. A symmetric interval is one that is centered around zero, meaning that every positive distance 'a' from zero on the number line, there is a corresponding negative distance '-a'.

In terms of integration, functions that are either even or odd have specific symmetries that affect their integrals over symmetric intervals. An even function, such as \(g(x) = x^2\), has the property \(g(-x) = g(x)\), while an odd function, such as \(f(x) = x\), satisfies \(f(-x) = -f(x)\). When multiplying an even function by an odd function, the result is an odd function. And the integral of an odd function over a symmetric interval is always zero because the areas on either side of the y-axis cancel each other out. This was precisely the reason the inner product in Part b evaluated to zero.

In our context, orthogonal functions are functions that, when taken the standard inner product over a particular interval, result in zero. This orthogonality implies that the functions are perpendicular or independent of each other in terms of the function space they occupy.

This concept is analogous to orthogonality in geometry, where two vectors are orthogonal if their dot product is zero. In the function scenario, the 'dot product' is replaced with the inner product. From the exercise, we see various pairs that turn out to be orthogonal. For example, in Part d, \(f(x) = 1\) and \(g(x) = \cos(x)\) are orthogonal over the interval \([-pi,pi]\) because their inner product also gives zero, reflecting the orthogonality of the constant function and the cosine function over one period of the cosine.

The inner product requires the functions to be continuous over the interval of interest because integration assumes the function can be drawn without lifting the pen from the paper. That is, there are no gaps, jumps, or vertical asymptotes. Continuity of functions is a crucial aspect since the inner product involves integration covering the entire interval.

All the functions considered in the exercise are continuous on their respective intervals. The continuity ensures that the properties of the inner product, such as orthogonality, can be appropriately assessed. If the functions were not continuous, we might encounter undefined behavior in the integrals or the need for more sophisticated mathematical tools. Consequently, the computations would become more complex or even impossible with standard calculus techniques.