Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Consider $\mathbb R^4$ together with the standard inner product and let $W=\{(x_1, x_2, x_3, x_4) \in \mathbb R^4 |\ x_1+x_2-x_4=0\}$. Find a basis for $W^\perp$.

How do you solve this question? All I can think of is that you have to put the elements in $W$ in terms of others.

share|improve this question
    
No need to download any piece of software. Just write your question in LaTeX. For example, \$x_1\$ to get $x_1$, \$W^\perp\$ to get $W^\perp$. –  Siminore Jun 15 '12 at 14:23
    
Thanks Siminore :) –  Alice Jun 15 '12 at 14:25

5 Answers 5

This is not the most elegant way, but here are some hints.

(1) First note that $W$ is the vector space given by $$W = \{(t, s, u, t+s)\in \mathbb{R}^4 \lvert t,s,u \in \mathbb{R} \}. $$

(2) Then for a vector $(a,b,c,d)$ to be in $W^{\perp}$ we need $$(a,b,c,d)\cdot (t,s,u,t+s) = 0 $$ for all $t,s,u \in \mathbb{R}$.

(3) We get from this that

$$W^{\perp} = \{(a, a, 0, -a) \lvert a \in \mathbb{R}\}. $$

(4) From here you should be able to write down a basis.

share|improve this answer

Since this is a hw, I will only give some hint!

First find a basis for $W$. To do so use the identity $x_4=x_1+x_2$ that defines $W$. You should find that any member $x=(x_1,x_2,x_3,x_4)$ of $W$ is a linear combination of 3 linearly independent vectors $w_1,w_2,w_3$, i.e. $w=\alpha_1 w_1+\alpha_2 w_2+\alpha_3w_3$ for some real numbers $\alpha_1,\alpha_2, \alpha_3$ (e.g. depending on $x_1,x_2,x_3$). Next, find a non-zero vector $v=(v_1,v_2,v_3,v_4)$ which is orthogonal to $w_1, \ w_2$, and $w_3$, meaning that $$ v\cdot w_1=v\cdot w_2= v\cdot w_3=0. $$ Any vector of the form $\alpha.v$, where $\alpha \in \mathbf{R}\setminus\{0\}$ is a basis for $W^\perp$.

share|improve this answer

Consider that $(1,1,0,-1)$ is orthogonal to all $(x_1,x_2,x_3,x_4)\in W$ because by definition $x_1+x_2-x_4=0$. Then $W^{\perp}=\langle(1,1,0,-1)\rangle$.

share|improve this answer
    
I still don't know what to do.. So how would I find the basis using the $W^\perp$? –  Alice Jun 15 '12 at 14:41

Note that $W = \{(1,1,0,-1)\}^\perp =\langle(1,1,0,-1)\rangle^\perp$. So $W^\perp = \big(\langle(1,1,0,-1)\rangle^\perp\big)^\perp= \langle(1,1,0,-1)\rangle$, with basis $\{(1,1,0,-1)\}$.

share|improve this answer

I would first find a basis for the solution space to the homogenous system of equations given by $x_1 + x_2 - x_4 = 0$. This turns out to be a 3 dimensional vector space. To find a basis for the solution space, set $x_2 = 1, x_3 = 0$, and $x_4 = 0$, then solve for $x_1$. This will provide you with your first vector in the basis. For the second vector, repeat this process, again solving for $x_1$, but setting $x_3 = 1$ instead of $x_2$. For the third and final vector, set $x_4 = 1$ and solve for $x_1$. You can then set up a new system of equations where the vector $\langle y_1, y_2, y_3, y_4 \rangle$ inner-producted with each of the basis vectors you just found will be zero. This will give you 3 equations in 4 variables. The solution space to this new system will provide you with a basis for $W^{\perp}$.

share|improve this answer
    
I only found one basis vector for the homogenous system $x_1 +x_2 -x_4 =0$ which is (0, 0, 0, 1). Am I doing this wrong? All I did was put it in row reduced form but this is what I got.. –  Alice Jun 15 '12 at 15:14
    
What happens if you set $x_4 = 0, x_3 = 1,$ $and x_2 = 0$ and then solve for $x_1$? So far, you've found one of the vectors in the basis for the solution set;) Wait, my bad... (0,0,0,1) does not satisfy the equation. You have to change one of the numbers (but not the $x_3$). –  Andrew Jun 15 '12 at 15:15
    
@Alice: Just updated the answer with some more detail. –  Andrew Jun 15 '12 at 15:36

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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