{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

ECE210a_lecture2_small

# ECE210a_lecture2_small - Vector spaces Vector spaces A...

This preview shows pages 1–9. Sign up to view the full content.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Vector spaces Vector spaces A vector is a set of elements taken from a ﬁeld, .77. Our most common choices for the ﬁeld are 7?, (real numbers) or C (complex numbers). Addition: 3: + y (where as and y are vectors) Scalar multiplication: am, where oz 6 .7 and a: is a vector. Deﬁning properties A vector space, V, satisﬁes the following properties: ForallaEJ-‘andallmandyEV, Al 3: —— y E V (closure) A2 (:1; + y) + z = x + (y + z) (distributive) A3 a: —— y = y —|— a: (commutative) A.4 There exists a unique 0 E V such that a: + 0 = x A.5 There exists —x e V such that a: + (—12) = 0 Roy Smith, ECE 210a, 2: 1 Vector spaces Vector spaces For all a and ﬂ 6 fand all a: and y E V, M.1 as: E V M2 (amm = a(ﬁx) M.3 a(x + y) = ax + ay M.4 (a + my: 2 am + ﬁzz: M.5 There exists 1 E 7-" such that 1x = 1: Some examples 1. \$1 Hill “More, v=1zn<orc> an ' aln, A: s s , m;eR(orC), V=RmX”(orCmX”) Roy Smith, ECE 210a, 2: 2 Vector spaces More examples 3. Functions, f : ’R,[0, 1] —> R, mapping [0, 1] on ’R, to R. 4. Real-valued polynomials of order n. Subspaces A subspace, S Q V (with the same ﬁeld, addition and multiplication deﬁnitions), is also a vector space. Therefore, 1. x,yeS => w+ye\$ 2.016.7-‘andz68 2) axes The key property of subspaces is closure. Other properties are inherited from the vector space. S = {0} is the trivial subspace. Roy Smith, ECE 210a, 2: 3 Vector spaces Span Given a set of vectors, 5: {111,U2,...,v,,}, 12,61}, the span of S is the set of all linear combinations. span(S) = {011111 + 02112 + - - - + anvn I 011‘ E «7} Properties Clearly a: E spanS => at E V. Furthermore, the spanS is also closed and is therefore a subspace. Proof: Roy Smith, ECE 210a, 2: 4 Vector spaces Question: What is the span of the following set of vectors? 1001 S: 2,0,0,2 3014 Sum of subspaces Given subspaces, X Q V and 32 Q V, X+y= {I+y|\$€?€,y€y} If X = span(Sx) and 37 = span(Sy), then X + 3} = span(SX U 3y). Roy Smith, ECE 210a, 2: 5 Vector spaces Linear independence A set of vectors, S = {01, . . . ,v,, } is linearly independent if, Tl Emu-=0 2 a;=0, i=1,...,n. [:1 In other words, zero is the only solution. This is a property of a set of vectors7 not an individual vector. Example 1 n 1 0 0 01 Z (171}; = a1 0 + 052 1 + 053 0 = 012 = 0, {:1 0 0 1 (13 which clearly implies that a1 = a2 = 05;; = 0. Roy Smith, ECE 210a, 2: 6 Vector spaces Example 2. Consider a square matrix as a. collection of column vectors, 0*2 A 6 RM” = “*1 aw Column vector notation: an; := Now consider the set of column vectors in A, Show that S is a linearly independent set if and only if det(A) aé 0. Basic idea of the proof: If det(A) = 0 then there exists an eigenvector, a? 75 0, such that, As there exists at least one 2.,- 75 0, this equation has a non-trivial solution, implying linear dependence of the vectors. Roy Smith, ECE 210a, 2: 7 Vector spaces Example 3. Vandermonde matrices. 1 x1 xf - -~ 261171 1 x2 x? - -~ 267171 2 2 . . . l/mxn :2 2 : 2 : , Wlth 1;} 3A {IIJ' for all 2 7é j. 2 71 1 CB," 1' m I ~ . \$le Show that if n = m, det(Van) 7é 0. (In the case Where n < m an analogous result can be stated and proven identically). Proof: Roy Smith, ECE 210a, 2: 8 Vector spaces Bases A basis of a vector space is a set of linearly independent vectors which span the vector space. Every nontrivial subspace has a basis. The number of vectors in a basis set is the dimension of the vector space. Example 1. The “standard” basis for R". 1 0 0 0 1 : . 7 : 7 7 0 == {517527---aen}- 0 0 1 Example 2. M2“: space of 2 X 2 real-valued matrices. {léﬁllﬁélil‘iﬁllﬁﬁl} °r {Elllﬁﬂllﬂéll‘ffil} Example 3. 5'2”: space of symmetric 2 )< 2 real-valued matrices. What is the dimension of this space? Roy Smith, ECE 210a, 2: 9 Vector spaces Example 4 The space of solutions, x(t) E C [07 1] (continuous functions on [0,1]), that satisfy, 2 (1 E1501) +x(t) = 0. What is the dimension of this space? Find a basis. Unique representations Givenabasis, B = {b1,...,bn} E V. For every x E V, there exists ai, i = 1, . . . ,n, such that7 I? x = E 01in and the a. are unique. [71 This choice of basis for any given space is not unique. Roy Smith, ECE 210a, 2: 10 Vector spaces Representation in terms of a basis Our standard vector notation is actually a representation in terms of a particular basis. For example, 1 1 0 0 5 = 1 0 + 5 1 + (—2) 0 2 161+ 562 + (—2)e3. —2 0 O 1 The “vector” is really just a list of the (1,- factors. The standard notation may confuse this. More precise notation: If we have a basis, B = {b1,...,b,,} G V, then the v E V, expressed with respect to the basis, B, is denoted by [v] 5‘ ’01 72 [’U]b, = = Zia-bi. i:l vIL This is important when we express the same vector in different bases. Roy Smith, ECE 210a, 2: 11 Vector spaces Example: Consider, B = {sint, cost }, deﬁned on t E ’R[0, 1]. This is actually a basis for all continuous solutions of on the interval, R[O, 1]. Any solution can be expressed as, x(t) = a1 sint + a2 cost. Therefore, [x0013 = [3:]- All solutions are therefore associated with a vector in R2 and it is often easier to manipulate them in R2 than in their original form. Roy Smith, ECE 210a, 2: 12 Linear transformations Vector spaces Given two vector spaces, U and V, over a ﬁeld, .73, a linear trwusfownatim is a mapping, T, from u to V (written T : u —> V) satisfying, 1. T(ar+ y) = T(x) + T(y) for all :1: and y G Ll. 2. T(am) = 047(x) for all as E L! and a E .7-'. In the case Where T : u —> U (i.e. mapping back into the same space), T is called a linear operator. The zero transformation is: 0(x) = 0, or more precisely 0([zr]u) = My. The idcntity operator is: I(a:) = 1:. Example 1. Matrix multiplication: T : R” —> Rm. T(:c) = Ax, Where A E Rm“. This is obviously linear; A(om: + y) = ozAm + Ag. Example 2. Consider L1 as the space of continuous functions, f (t) : ’R,[0, 00) —> R[0, oo). Deﬁne the operator: T( f) := /\ f (t)dt. (more precisely written as: T( f) (11)) 0 This linear because, Adriaf(t)+g(t)]dt = 0‘ lJilif(1t)dt+ / 1:9(t)dt, f0ra11f(t)and9(t) eu. 0 Example 3. Deﬁne Q6 : R2 —> R2 as a rotation of a vector in R2 by a ﬁxed angle of 0. mcosO—ysiné] _ [cos0 —sin0] [ac] ac, then 99%) .— [xsin0+y0050 sin6 0050 y . Ifu=[ y This is linear: Q9(ozu + v) = 019901) + 96(7))- Note the correspondence with matrix multiplication. Roy Smith, ECE 210a, 2: 13 Vector spaces Roy Smith, ECE 210a, 2: 14 Vector spaces Space of linear transformations Given vector spaces U and V over .7, the set of (1,1 l linear transformations, [.(Ll,V) :2 {T:Z/l—)V | Tis linear} is itself a vector space over .7. Basis foru: Bu = {u1,...,un} (dim(Ll) = n) Basis for V: BV = {111, . . .,vm} (dim(V) = m) We can deﬁne a linear transformation T : U —> V by its action on each of the basis vectors in U. T will map u; (the jth basis vector in U) to some vector in V, Now any a: E U is expressed in terms of the basis Bu, 931 TI MB“ 22%” = j:l 3777 Roy Smith, ECE 210a, 2: 15 Vector spaces Coordinate matrix representations So now, [mama = [7 (quﬂ m = Z[T(\$juj)le = .2ng [7%)le = 2961' 206007: Bv We now express a vector y E V, the result of a linear transformation, y = ’1'(rc)7 x E u, in terms of the basis By, m [lev = 2.712% = {:1 From above7 we can identify each of the y values, 71 7'! yr = E 95,7060 = E 017:ij- j:1 1:1 This has an obvious matrix interpretation: 3/] yin 3/1 (111 ‘ ' ' aln \$1 \$1 =: 3V [T] Bu 5 . (Meyer’s notation: [’1'] n] 4 13V) ym aml ' ’ ' arm! {1}” x71 Roy Smith, ECE 210a, 2: 16 Vector spaces Key facts about linear transformations 1. Every linear transformation can be represented by a matrix. 2. The action of a linear transformation T : u —> V, on x E u, can be represented by matrix multiplication between the coordinates of the bases. l7(\$)le = Bi; [7] Ba lxlBu' 3. dim([,(u, 12)) = (dim(u)) (dim(V)). Example Differentiation for polynomials of degree 3. we» == did—f) Select the basis:Bu = B]; = { 1,t,t2,t3 } We can show that, 13v [T] Bu 2 COCO OOOH COMO OOOOO Roy Smith, ECE 210a, 2: 17 Vector spaces More about notation Consider operators, ’P : X —> 3} and Q : W —> X. y=’Px and x=Qw. Note that this is taken to mean that 7’ maps a: E X to y E 37. .7163] :vEX wEW ‘ In matrix form these equations are: [y] M = By [’P] EX [:5] By and [it] b’l- = Bx [Q] 3W [wlifw This is equivalent to, [ley = By [73le Bar [Q] 13W [w] W. And we usually write the associated matrix/vector mapping as y = PQw. Roy Smith, ECE 210a, 2: 18 ...
View Full Document

{[ snackBarMessage ]}