Currently, I am having a hard time trying to understand what exactly are basis vectors and how to properly apply them in different mathematical scenarios.

Allow me to give an example: Find an orthonormal basis in which the vector $C = \begin{pmatrix} 4 \\ 5 \end{pmatrix}$ is colinear with one of the basis vectors.

With the above prompt, I could then state that:

$$x_1 ' = \begin{pmatrix} \frac{4}{\sqrt{41}} \\ \frac{5}{\sqrt{41}} \end{pmatrix}$$ $$x_2 ' = \begin{pmatrix} -\frac{5}{\sqrt{41}} \\ \frac{4}{\sqrt{41}} \end{pmatrix}$$

However, why is that? How did the above "solution" even appear? Why is is "correct"?

What happens when I start extending my vector $C$ to a $3 \times 1$ matrix? How do I find - for example - $x_2 '$ and $x_3 '$ basis vectors?

Any assistance in helping me understand the definition and method of use of basis vectors would be greatly appreciated. Thank you for reading through my question!

  • $\begingroup$ How computing an orthonormal basis, e.g. in the case of $C$ a vector in 3 dimensions, look at the Gram-Schmidt process. $\endgroup$ Apr 24 '20 at 23:17
  • $\begingroup$ You know one of the basic vectors has to be parallel to $C$, so that means it will be of the form $u_1 = \alpha \begin{bmatrix} 4 \\ 5 \end{bmatrix}$ for some scalar $\alpha$. The scalar $\alpha$ should be chosen to make $u_1$ be a unit vector. Then the next basis vector has to be orthogonal to $u_1$, and it also has to be a unit vector. $\endgroup$
    – littleO
    Apr 24 '20 at 23:18
  • $\begingroup$ @littleO, Ah, that makes perfect sense. Thank you for the help! $\endgroup$
    – Athenian
    Apr 25 '20 at 0:38

A basis is a set of vectors from which you can construct any vector in your space by taking a linear combination of the basis elements. We say that the basis spans the space.

To find an orthonormal basis so that $(4,5)$ is collinear with one of the vectors in the basis, say let the basis vectors be $(a,b),\,(c,d).$ Then we have that $(a,b)\cdot(c,d)=ac+bd=0,$ and $a^2+b^2=c^2+d^2=1.$ That means the vectors are orthogonal, and of unit norm. Finally, we must have that $(a,b)=\lambda(4,5)$ for some nonzero real $\lambda.$ It suffices to take $\lambda$ so that the given vector is of unit norm. That is, let $$\lambda=\frac{1}{\sqrt{4^2+5^2}}.$$

Since you now know what $a,\,b$ are, substitute in the previous set of three equations to get a quadratic system in $c,\,d,$ which you should be able to solve.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.