MATH 304 Linear Algebra Lecture 28: Orthogonal bases. The Gram-Schmidt orthogonalization process.
Orthogonal sets Let V be an inner product space with an inner v, v . product , and the induced norm v = Denition. Nonzero vectors v1 , v2 , . . . , vk V form an orthogonal set if they are orthogonal to each other: vi , vj = 0 for i = j . If, in addition, all vectors are of unit norm, vi = 1, then v1 , v2 , . . . , vk is called an orthonormal set. Theorem Any orthogonal set is linearly independent.
Orthonormal bases Let v1 , v2 , . . . , vn be an orthonormal basis for an inner product space V . Theorem Let x = x1 v1 + x2 v2 + + xn vn and y = y1 v1 + y2 v2 + + yn vn , where xi , yj R. Then (i) x, y = x1 y1 + x2 y2 + + xn yn , 2 + x 2 + + x 2. (ii) x = x1 n 2
Proof: (ii) follows from (i) when y = x.
n n n n
x, y =
i =1
xi vi ,
j =1 n n
yj vj
=
i =1
xi
n
vi ,
j =1
yj vj
=
i =1 j =1
xi yj vi , vj =
i =1
xi yi .
Orthogonal projection Theorem Let V be an inner product space and V0 be a nite-dimensional subspace of V . Then any vector x V is uniquely represented as x = p + o, where p V0 and o V0 . The component p is the orthogonal projection of the vector x onto the subspace V0 . We have o = x p = min x v .
vV0
That is, the distance from x to the subspace V0 is o .
x o
V0
Let V be an inner product space. Let p be the orthogonal projection of a vector x V onto a nite-dimensional subspace V0 . If V0 is a one-dimensional subspace spanned by a x, v vector v then p = v. v, v If v1 , v2 , . . . , vn is an orthogonal basis for V0 then x, v2 x, vn x, v1 v1 + v2 + + vn . p= v1 , v1 v2 , v2 vn , vn
n
Indeed, p, vi =
j =1
x, vi x, vj vj , vi = vi , vi = x, vi vj , vj vi , vi
xp, vi = 0 = xp vi = xp V0 .
The Gram-Schmidt orthogonalization process Let V be a vector space with an inner product. Suppose x1 , x2 , . . . , xn is a basis for V . Let v1 = x1 , x2 , v1 v1 , v1 , v1 x3 , v1 x3 , v2 v3 = x3 v1 v2 , v1 , v1 v2 , v2 ................................................. xn , v1 xn , vn1 vn = xn v1 vn1 . v1 , v1 vn1 , vn1 v2 = x2 Then v1 , v2 , . . . , vn is an orthogonal basis for V .
x3 v3
p3 Span(v1 , v2 ) = Span(x1 , x2 )
Any basis x1 , x2 , . . . , xn
Orthogonal basis v1 , v2 , . . . , vn
Properties of the Gram-Schmidt process: vk = xk (1 x1 + + k 1 xk 1 ), 1 k n; the span of v1 , . . . , vk is the same as the span of x1 , . . . , xk ; vk is orthogonal to x1 , . . . , xk 1 ; vk = xk pk , where pk is the orthogonal projection of the vector xk on the subspace spanned by x1 , . . . , xk 1 ; vk is the distance from xk to the subspace spanned by x1 , . . . , xk 1 .
Normalization Let V be a vector space with an inner product. Suppose v1 , v2 , . . . , vn is an orthogonal basis for V . v1 v2 vn Let w1 = , w2 = ,. . . , wn = . v1 v2 vn Then w1 , w2 , . . . , wn is an orthonormal basis for V . Theorem Any nite-dimensional vector space with an inner product has an orthonormal basis. Remark. An innite-dimensional vector space with an inner product may or may not have an orthonormal basis.
Orthogonalization / Normalization
An alternative form of the Gram-Schmidt process combines orthogonalization with normalization.
Suppose x1 , x2 , . . . , xn is a basis for an inner product space V . Let v1 = x1 , w1 =
v1 v1
,
v2 v2
v2 = x2 x2 , w1 w1 , w2 =
,
v3 v3
v3 = x3 x3 , w1 w1 x3 , w2 w2 , w3 =
................................................. vn = xn xn , w1 w1 xn , wn1 wn1 , vn wn = v . n Then w1 , w2 , . . . , wn is an orthonormal basis for V .
Problem. Let V0 be a subspace of dimension k in Rn . Let x1 , x2 , . . . , xk be a basis for V0 . (i) Find an orthogonal basis for V0 . (ii) Extend it to an orthogonal basis for Rn .
Approach 1. Extend x1 , . . . , xk to a basis x1 , x2 , . . . , xn for Rn . Then apply the Gram-Schmidt process to the extended basis. We shall obtain an orthogonal basis v1 , . . . , vn for Rn . By construction, Span(v1 , . . . , vk ) = Span(x1 , . . . , xk ) = V0 . It follows that v1 , . . . , vk is a basis for V0 . Clearly, it is orthogonal. Approach 2. First apply the Gram-Schmidt process to x1 , . . . , xk and obtain an orthogonal basis v1 , . . . , vk for V0 . Secondly, nd a basis y1 , . . . , ym for the orthogonal complement V0 and apply the Gram-Schmidt process to it obtaining an orthogonal basis u1 , . . . , um for V0 . Then v1 , . . . , vk , u1 , . . . , um is an orthogonal basis for Rn .
Problem. Let be the plane in R3 spanned by vectors x1 = (1, 2, 2) and x2 = (1, 0, 2). (i) Find an orthonormal basis for . (ii) Extend it to an orthonormal basis for R3 . x1 , x2 is a basis for the plane . We can extend it to a basis for R3 by adding one vector from the standard basis. For instance, vectors x1 , x2 , and x3 = (0, 0, 1) form a basis for R3 because 1 2 2 1 2 1 0 2 = = 2 = 0. 1 0 0 0 1
Using the Gram-Schmidt process, we orthogonalize the basis x1 = (1, 2, 2), x2 = (1, 0, 2), x3 = (0, 0, 1): v1 = x1 = (1, 2, 2), v2 = x2 3 x2 , v1 v1 = (1, 0, 2) (1, 2, 2) v1 , v1 9 = (4/3, 2/3, 4/3), v3 = x3
x3 , v1 x3 , v2 v1 v2 v1 , v1 v2 , v2 4/3 2 (4/3, 2/3, 4/3) = (0, 0, 1) (1, 2, 2) 9 4 = (2/9, 2/9, 1/9).
Now v1 = (1, 2, 2), v2 = (4/3, 2/3, 4/3), v3 = (2/9, 2/9, 1/9) is an orthogonal basis for R3 while v1 , v2 is an orthogonal basis for . It remains to normalize these vectors. v1 , v1 = 9 = v1 = 3 v2 , v2 = 4 = v2 = 2 v3 , v3 = 1/9 = v3 = 1/3 w1 = v1 / v1 = (1/3, 2/3, 2/3) = 1 3 (1, 2, 2), 1 (2, 1, 2), w2 = v2 / v2 = (2/3, 1/3, 2/3) = 3 1 w3 = v3 / v3 = (2/3, 2/3, 1/3) = 3 (2, 2, 1). w1 , w2 is an orthonormal basis for . w1 , w2 , w3 is an orthonormal basis for R3 .
