Escolar Documentos
Profissional Documentos
Cultura Documentos
mtanveer8689@gmail.com
0316-7017457
Lecture # 1
Metric Space:
Let X be a non-empty set. Defined a function
f: X X R s.t
: d(x,y) 0 x,y X
: d(x,y) = 0 x = y
: d(x,y) = d(y,x)
: d(x,z) d(x,y) + d(y,z) x,y,z
Then d is called Metric in X and (X,d) is called Metric Space.
d: × R s.t d(x,y) = ∑ | − |
x={ } , y={ }
s.t ∑ | | < ∞ then ( , ) is Metric Space.
: d(x,y) 0 ∑ | − | 0
: d(x,y) = 0 ∑ | − | =0
∑ | − | =0
− =0 k =1, 2……… ∞
= , ,…………. = , ,……….
x=y
: d(x,y) = ∑ | − | = ∑ | − | = d(y,x)
: d(x,z) d(x,y) + d(y,z)
∑ | − | = d(x,z) = ∑ | − + − |
∑ | − | + ∑ | − |
= d(x,y) + d(y,z)
| + | | | + | | (Minkowski Inequality)
∑ | + | ∑ | | + ∑ | |
= d(y,x) d is Metric
( , )
: (x,z) =
( , )
( , ) ( , )
( , ) ( , )
( , ) ( , )
( , ) ( , )
+ ( , ) ( , )
d(x,y) = ∑ | − | is metric on X.
: d(x,y) 0 ∑ | − | 0
: d(x,y) = 0 ∑ | − | =0
∑ | − | =0
− =0 i =1, 2……… ∞
= , ,…………. = , ,……….
x=y
: d(x,y) = ∑ | − | = ∑ | − | = d(y,x)
: d(x,z) d(x,y) + d(y,z)
∑ | − | = d(x,z) = ∑ | − + − |
∑ | − | + ∑ | − |
= d(x,y) + d(y,z)
| + | | | + | | (Minkowski Inequality)
∑ | + | ∑ | | + ∑ | |
Question:
d(x, y) − d( ′ , ′ ) d(x, ′ ) + d(y, ′ )
By : d(x, y) d(x, ′ ) + d( ′ , ′ )+ d(y, ′ )
d(x, y) − d( ′ , ′ ) d(x, ′ ) +d(y, ′ ) (1)
′ ′
Interchanging x by & y by
d( ′ , ′ ) − d(x, y) d( ′ , )+ d( ′ , )
Multiply by -1
-[ d(x, y)+ d( ′ , ′ )] d(x, y)- d( ′ , ) (2)
-r x − r x
− x + x
x
[ − , + ]
Sphere:
S( )= { x X : d(x, ) = }
For real line
S( )= { x R : |x − |= } x
x
|x − |= x
x− =r
x x
x = r
{ − , + } Set
8
f(n) = 2n n
f(n) = 5n n
Floor Brackets
f(n) = n
Ceiling Brackets
f(n) = ⌈ ⌉ n
Convergent Sequence:
A sequence { } in metric space (X,d) is said to be converges x X. If given
any a natural number.
= (
s.t n;n d( ,x) <
or n;n lim →¥ =x
or n;n → x when →¥
Question:
Prove that a sequence in a (X,d) converges to one and only one limit.
Solution:
Let (X,d) be a metric space { } be a convergent sequence converges to two
distinct points x and of X.
Let r = d(x, ) : r > 0
Since → x , for s.t
n;n d( ,x) <
10
Take = Max( , )
n;n d( ,x) <
d( , ) <
Now r = d(x, ) ( , ) + d( , )
r = d(x, ) < +
r<r =r
Which is contradiction.
So x=
Note: Limit and limit points for a convergent sequence is same.
Cauchy Sequence:
Let (X,d) be a metric space a { } is said to a Cauchy sequence.
If = ( Depending
on
m,n ; m,n d( , ) <
Theorem:
Prove that every convergent sequence is a Cauchy sequence.
Proof: Suppose that { } is a convergent sequence in (X,d) metric space. And
converges to a point x X.
Then for = (
s.t n d( ,x) <
Take = Max( ,
n,n d( , ) d( , ) + d( , )
n,n d( , ) < +
n,n d( , ) <
→xX
Note: Every sequence itself is subsequence of it.
Complete Space:
Let { } be a Cauchy sequence in (X,d) if → x X then (X,d) is said to
be complete space. e.g. R & C are complete spaces.
Dense Subset: If A X s.t = X then A is dense in X.
Somewhere & Nowhere Dense Subset:
If A X , ( ) then it is somewhere dense subset.
If ( ) = then it is nowhere dense subset. Super Set
Nested Sequence:
Let , … … , … .. be a sequence of non-empty set in (X,d) s.t
(i) , n = 1,2,…. (ii) ( ) ® 0 as n ®
{ } is nested sequence of set. 12
: ‖a ‖ = |a|‖ ‖ , F
: ‖ + ‖ ‖ ‖+‖ ‖ x,y N
‖.‖ is Norm and (N, ‖.‖) is Normed space.
Example:
Prove that space consisting of all sequence x }, F under
‖ .‖ : R such that
: ‖ ‖ 0 ∑ | | 0
: ‖ ‖= 0 ∑ | | =0
‖ ‖ = 0 { , ,……. , ….} = 0
‖ ‖= 0 x=0
: ‖a ‖ = ∑ |a |
= |a| ∑ | | = |a|‖ ‖
: ‖ + ‖ ∑ | + |
By Minkoswki Inequality
‖ + ‖ ∑ | | + ∑ | |
‖ + ‖ ‖ ‖‖ ‖Hence ( , ‖.‖) is normed space.
13
Solution:
n
: ‖ ‖ 0 Sup | | 0
i 1
n
: ‖ ‖= 0 Sup | | =0
i 1
| | = 0 { , ,……. } = (0,0,0….0)
‖ ‖= 0 x=0
n n
: ‖ + ‖ Sup | + |
i 1
n
‖ + ‖ Sup | |+ | |
i 1
‖ + ‖ ‖ ‖‖ ‖
Hence ( , ‖.‖) is normed space.
Question: Show that a normed space is a metric space.
Proof:
Let d: N × N R such that
d(x,y) = ‖ − ‖ x,y N
: d(x,y) 0 ‖ − ‖0 By
: d(x,y) = 0 ‖ − ‖ = 0 x = y by
: d(x,y) = ‖ − ‖
= ‖−1( − )‖ = |−1|‖ − ‖ ‖a ‖ = |a|‖ ‖
14
: d(x,z) = ‖ − ‖
=‖ − + − ‖
‖ − ‖ + ‖ − ‖ By
‖ ‖ − ‖ ‖ ‖ − ‖ (1)
‖ ‖ =‖ − + ‖
‖ − ‖+‖ ‖+‖ ‖ By
‖ ‖ ‖ − ‖ + ‖ ‖
−‖ − ‖ ‖ ‖ − ‖ ‖ (2)
By (1) and (2)
−‖ − ‖ ‖ ‖ − ‖ ‖ ‖ − ‖ Proved
|‖ ‖ − ‖ ‖| ‖ − ‖
Question:
x, y, z N prove that
(i) d(x+y ,y+z) = d (x,z)
(ii) d(x,y) = |a|d(x,y)
Solution:
We know that
d: N × N R such that
d(x,y) = ‖ − ‖
(i) d(x+y ,y+z) = ‖ − + − ‖
15
16
17
18
20
:‖ ‖=0 ∈[ , ]
| ( )|= 0
| ( )|= 0
f=0
: ‖a ‖ = ∈[ , ]
|a ( )| |a|
∈[ , ]
| ( )|
= |a| ‖ ‖
: ‖ + ‖= ∈[ , ]
| ( ) + ( )|
| ( )| + | ( )|
∈[ , ] ∈[ , ]
‖ + ‖ ‖ ‖ + ‖ ‖
Now let ‖ ‖ = ∈[ , ]
| ( )|
f(t) = sint
and g(t) = cost where ∈ [0, ]
‖ ‖= |sint| = 1 , ‖ ‖= |cost| = 1
∈[ , ] ∈[ , ]
‖ + ‖= |sint + cost|
∈[ , ]
‖ + ‖ = + = = √2
√ √ √
‖ − ‖= |sint − cost|
∈[ , ]
Question #6:
How can you prove that inner product space is a normed space and hence a
metric space also show that its converse may not be true [ normed space is not
an inner product space V]
Solution:
In an inner product space V
‖. ‖ : V such that
‖ ‖ = √< , > xV
:‖ ‖0 since < x , x > 0 By
: ‖ ‖ = 0 √< , > =0
<x,x> =0
x=0 By
Let ‖ ‖ = ∈[ , ]
| ( )|
‖ ‖= |sint| = 1 , ‖ ‖= |cost| = 1
∈[ , ] ∈[ , ]
‖ + ‖= |sint + cost|
∈[ , ]
23
‖ − ‖= |sint − cost|
∈[ , ]
‖ − ‖=1
By //gram identity
‖ + ‖ +‖ − ‖ = 2‖ ‖ + 2‖ ‖
i.e. if P = 2
+ =1
q=2
24
0 = (0,0,0,0,…….0)
is a Linear space over F (R or C).
‖. ‖ : Such that
‖ ‖= ∑ | |
: ‖ ‖0 ∑ | | 0
: ‖ ‖=0 ∑ | | =0
= 0
( , , ,…… ) = (0,0,0,……0)
=0
: ‖a ‖ = ∑ |a |
= |a| ∑ | |
= |a|‖ ‖
: ‖ + ‖= ∑ | + |
∑ | | + ∑ | |
‖ + ‖ ‖ ‖+ ‖ ‖
Hence ( , ‖. ‖) is a normed space.
(ii). is a Banach Space.
( )
Let { } be a cauchy in
Then for > 0 , = ()
( ) ( ) ( ) ( )
s.t p,q ; p,q − = ∑ −
25
( ) ( )
p,q ; p,q −
( )
{ } is a cauchy sequence in R.
Since R is complete so
( )
R as P i = 1,2,3,…….,n
For > 0 ( Already choosen) s.t
( )
P; P − <
√
( )
Take x = lim → then x
Let = ( , ,….. )
( ) ( )
P;P − = ∑ | − |
( )
P;P − = + + … … . … . +
( )
P;P − <
x where P
is complete and Hence a Banach space.
Question: Prove that is a normed space and hence a Banach Space.
Proof:
The space is Euclidian space where
=( , ,……. )
0 = (0,0,0,……0)
is a Linear space over F (R or C)
‖. ‖ : Such that
‖ ‖= ∑ | |
26
: ‖a ‖ = ∑ |a |
= |a| ∑ | |
= |a|‖ ‖
: ‖ + ‖= ∑ | + |
By Minkowski inequality
∑ | | + ∑ | |
‖ + ‖ ‖ ‖+ ‖ ‖
Hence ( , ‖. ‖) is a normed space.
(ii). is a Banach space
( )
Let { } be a cauchy in
Then for > 0 = ()
( ) ( ) ( ) ( )
s.t p,q ; p,q − = ∑ −
( ) ( ) ( ) ( )
p,q ; p,q − ∑ −
( ) ( )
p,q ; p,q −
( )
{ } is a cauchy sequence in R.
Since R is complete so
( )
R as P i = 1,2,3,…….,n
For > 0 ( Already choosen) s.t
( )
P; P − <
√
27
Let = ( , ,….. )
( ) ( )
P;P − = ∑ | − |
( )
P;P − = + + … … . … . +
( )
P;P − <
z where P
is complete and Hence a Banach space.
28
xi 0
( x1 , x2 , x3 ,..........xn ,....) (0,0,.........0,......)
x 0
N3 : a x Sup a xi
i 1
a Sup xi
i 1
a xi
N 4 : x y Sup xi yi
i 1
Sup xi Sup yi
i 1 i 1
29
S .t p, q ; p, q no x ( p ) x ( q ) xi( p ) xi( q )
p, q ; p, q no x ( p ) x ( q ) x ( p ) x ( q )
p, q ; p, q no x ( p ) x( q )
( )
Hence { } is a Cauchy sequence in R or C. Since R or C is complete so
( )
i = 1,2,…….
( )
Take x = { } we show that x and lim → =x
( )
Since So, N, = () s.t
( )
p; p − < , i = 1,2,…..
( ) ( )
That is p ; p − = − <
x ( p ) x Also from
xi xi xi( p ) xi( p )
xi xi( p ) xi( p )
xi k p ( finite number )
2
Hence x {xi }l
l is Banach space.
30
N1 : x 0 Sup xi 0
i 1
N2 : x 0 Sup xi 0
i 1
xi 0
( x1, x2 , x3 ,..........xn ,....) (0,0,.........0,......)
x 0
N 3 : a x Sup a xi
i 1
a Sup xi
i 1
a xi
N 4 : x y Sup xi yi
i 1
Sup xi Sup yi
i 1 i 1
p ; p no xi( p ) xi x ( p ) x
3
p ; p no x ( p ) xi i 1, 2,.........
3
( ) ( )
Now consider the sequence ={ }
( )
Then C and so is convergent.
But a convergent sequence will be a Cauchy sequence.
So, for > 0 N s.t
( ) ( )
i,j ; i,j − < 3
Take = max ( , ) then
( ) ( ) ( ) ( )
i,j ; i,j − = − + − + −
( ) ( ) ( ) ( )
− + − + −
3 + 3+ 3
i,j ; i,j − <
{ } is a Cauchy sequence in R or C .
Since R or C is complete. So x = { } is convergent and hence x C
Thus, c is closed subspace of
Thus, c is complete and hence Banach space.
“ A subspace Y of complete metric space X is complete iff Y is close in X”
32
p ; p po xi( p ) xi x( p ) xi
3
p ; p po x( p ) xi i 1,2,.........
3
( ) ( )
Now consider the sequence ={ }
( )
Then and so is convergent.
But a convergent sequence will be a Cauchy sequence.
So, for > 0 N s.t
( ) ( )
i,j ; i,j − < 3
Take = max( , ) then
( ) ( ) ( ) ( )
i,j ; i,j − = − + − + −
33
3 + 3+ 3
i,j ; i,j − <
{ } is a Cauchy sequence in R or C.
Since R or C is complete. So x = { } is convergent and hence x
Thus, is closed subspace of
Thus, is complete and hence Banach space.
Question:
Give an example of space which is not a Banach space.
Solution:
The space Q of rational number is a subset of the Banach space R
We Know
=RQ
Q is not closed in R =
34
36
x + (1 − ) ( ,r)
( ,r) is convex in N
Theorem:
Let C be a convex set in Linear space N then for 0 , 0 . Prove that
(+)C = C + C
Proof: Case-I
If = 0 or = 0
Then (+)C = C+ C
Holds trivially
Case-II
Let > 0 , > 0
And z (+)C then cC
s.t z = (+)c
z = c+c C+C
(+)C C + C (1)
Conversely, Let u C+ C
38
= (+) (+)C
C+C (+) C …..(2)
a a
where =( + (1 − )d) C
a b a b
is convex in N
39
It should be noted that N/S (quotient space) is a linear space under addition and
scalar multiplication defined by
x + S + y + S = x+y+ S ; x,y N
(x+S) = x + S ;xN,F
x S 1 Inf x s
sS
Question:
Show that a quotient space N/S is a norm space under the norm
x S 1 Inf x s
sS
Solution:
N1 : x S 1 0 Inf x s 0
sS
N2 : x S 0 Inf x s 0
sS
40
Inf a x a s
sS
a Inf xs
sS
a xS
Lim x xn x S 1 , Lim y yn y S 1
n n
x xn y yn
Taking limit as n
x s y S 1 x y S 1 Lim x xn Lim y yn
n n
41
x S 1 Inf x s
sS
Solution:
Let { + }; N Cauchy sequence in N/S
Then for > 0 N, n1 n1 ( ) s.t
m, n ; m, n n1 ( xm S ) ( xn S ) 1 xm xn S .......(1)
x S Inf x s
sS
1
Take , m n1 and n n1 1
2
1
m, n ; m, n n1 ( xn1 S ) ( xn11 S ) xn1 xn11 S
2
If we choose = N, n2 n2 ( ) s.t
1
m, n ; m, n n2 ( xn2 S ) ( xn2 1 S ) xn 2 xn 2 1 S
4
Continuing in this way
. . . .
. . . .
. . . .
. . . .
42
yk y y k y k 1 y k 1 y k 2 y k 2 ............ y k 1 y
k' k'
y k y k 1 y k 1 y k 2 .................... y k 1 y
k'
1 1 1 1
yk y .....
k' 2 k
2 k 1
2 k 2
2 k 1
1
k a
2 S
1 1 r
1
2
1
yk y as k
k' 2 k 1
Thus { } is convergent sequence in N.
Since N is complete so yN
xnk S y S N / S
N / S is complete
We use this theorem in above proof
“A Cauchy sequence converges iff it has convergent subsequence”
Equivalent Norms:
Let N be a Norm space ‖∙‖ and ‖∙‖ be norms define on N then ‖∙‖ is said to
be equivalent to ‖∙‖ (‖∙‖ ‖∙‖ ) if a > 0 , b > 0 be real numbers such that
a‖ ‖ ‖ ‖ b‖ ‖ xN
Question:
Show that the relation being equivalent to among the norms that can be defined
on a linear space N is an equivalence relation.
Solution:
43
a‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖
and b‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖
‖ ‖ ‖ ‖ ‖ ‖
‖∙‖ ‖∙‖
i.e. Relation is symmetric
Step-III Transitive Relation:
Let ‖∙‖‖∙‖ and ‖∙‖ ‖∙‖ then a > 0 , b > 0 , >0, >0 s.t
a‖ ‖ ‖ ‖ b‖ ‖ (1)
‖ ‖ ‖ ‖ ‖ ‖ (2) xN
From (2)
‖ ‖ ‖ ‖ and ‖ ‖ ‖ ‖
‖ ‖ ‖ ‖ …..(3) and ‖ ‖ ‖ ‖ …..(4)
Using (1) ,(3) and (4)
‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖
‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ ‖ = , =
‖∙‖‖∙‖ i.e. Relation is transitive
Hence Relation is Equivalence
44
‖ ‖ ‖ ‖ ‖ ‖ …(1) xN
We shall prove that a Basic open set in (N, ‖×‖) and conversely.
For this let x N and B(x,r) be an open ball in (N, ‖×‖). We show that B(x,r)
will open in (N, ‖×‖)
Let y B(x,r) and ‖ − ‖ = <r
‖ − ‖<r By def.
∗
Conversely ( , ) is an open ball in (N, ‖×‖)
Where = >0
∗
Then for any z ( , )
‖ − ‖ <
And ‖ − ‖ = ‖ − + − ‖
‖ − ‖+‖ − ‖
b‖ − ‖ + ‖ − ‖ by (1) ‖ ‖ ‖ ‖
<b +
‖ − ‖<r
Hence z B(x,r)
∗
y ( , ) B(x,r)
B(x,r) is open ball in (N, ‖. ‖ )
Similarly, every open ball in (N, ‖. ‖ ) will be open in (N, ‖×‖)
Hence topologies induced by ‖×‖ and ‖. ‖ are same.
45
‖ ‖ ‖ ‖ ‖ ‖ …(1) xN
Let { } be a Cauchy sequence in (N, ‖×‖) Then for
Then for = (
m,n ; m,n ‖ − ‖<
‖ ‖ ‖ ‖ ‖ ‖ …(1) xN
Let { } be a convergent sequence in (N, ‖×‖)
Then for = (
n; n ‖ − ‖<
n;n ‖ − ‖ ‖ − ‖<
by (1)
n;n ‖ − ‖ <
n;n ‖ − ‖ < =
46
x+y
(iii) The addition and scalar multiplication as function
+:VVV s.t
(x,y) x+y x,y V v v
And ‘.’ : F V V s.t Fv
(,x) x F,xV
x x
Linear Operator: F v
Let N and M are topological linear spaces then function T : N M is said to be
linear operator if for every
a,b F , x,y N
T(ax+by) = aT(x) + bT(y) Or
(i) T(x+y) = T(x) + T(y)
(ii) T(x) = T(x) , F and x N
Example:
I: N N s.t
T(x) = x xN
For , F , , N
I( + ) = I( ) + I( )
= I( ) + I( )
Example:
O linear operator
O : NN s.t
, F , , N
O( + ) = O( ) + O( )
= .O( ) + .O( )
47
48
n
T (ai ei ) T (0) T (0) 0
i 1
n
a e 0
i 1
i i
so y T ( x)
n
T (ai ei )
i 1
n
y a T (e )
i 1
i i
Theorem:
Let T : NM be a linear operator then prove that T is continuous on N iff T is
bounded .
Proof:
Suppose that T is continuous on N then it is continuous N so for
s.t x N; x x0 T ( x) T ( x0 ) ...(1)
‖ − ‖ ‖ ( ) − ( )‖ = ‖ ( − )‖
K‖ − ‖
< K.
<K
‖ ( ) − ( )‖
T : NM is continuous
50
Since T is linear
n n
T ( x) T ( xi ei ) T ( x) x T (e )
i i
i 1 i 1
n n
T ( x) xi T (ei ) xi T (ei )
i 1 i 1
n n
b xi .......(1) b Sup T (ei )
i 1 i 1
a x
i 1
i i c ai "
i 1
n n
In this case x xi ei c xi by lemma
i 1 i 1
n
1
x c
i 1
i x put in (1)
n
b
x
i 1
i
c
x
b
K x K 0
c
51
Zero Norm:
n
x 0
Sup ai
i
Question:
Suppose that ‖×‖ ‖. ‖
n
x 0 Sup ai on a Norm space N1
i 1
52
1
Since . 0 . by (1)
a
1 ( p)
p, q : p, q n0 y ( p ) y ( q ) y y(q)
0 a a
n n
p , q : p , q n0 Sup aip aiq x 0
Sup ai
i 1 a i 1
p, q : p, q n0 ai( p ) ai( q )
a
( ) ( )
is a cauchy sequence in F (R or C) Since F is complete so as
p
i.e. a i( p ) a i( q ) 0 as p .......(2)
n
put y a x
i i
i i
n
Then y N and y p y a i
p
xi ai xi
i 1
n
p
(a
i 1
i ai )xi
n
a ip a i xi
i 1
n n
y p y k aip ai 0 as p by (2) k Sup xi
i 1 in
yp y N
N is complete
53
n
x cs S ai
i 1
1
s x ........(2)
c
n
Also x 1 a x
i 1
i i
1
n
a i xi 1
i 1
n
x 1 kS k Sup xi 1
i 1
k
x 1 x
c
c
x x
k 1
c
a x 1 x ...(3) a 0
k
54
x 1 c' S
1
S x …(4)
c' 1
n
So, x a x
i 1
i i
n
ai xi
i 1
n n
K ' ai K ' Sup xi
i 1 i 1
x K 'S
K'
x x1
c'
K'
x b x1 ...(5) b
c'
By (3) and (5)
a x1 x b x1 x N
‖. ‖ ~‖×‖
Hence any two norms on a finite dimensional Linear space are equivalent
55
Solution:
n
= xi y i
i 1
n
xi y i
i 1
n n
Sup xi yi
i 1 i 1
‖ ‖ n‖ ‖ ….(1)
n n
Also ‖ ‖ = Sup xi x i
i 1 i 1
‖ ‖ ‖ ‖ …(2)
From (1) and (2)
1. ‖ ‖ ‖ ‖ n‖ ‖ x
‖. ‖ ‖. ‖
Question:
n n
2
Show that ‖. ‖ ‖. ‖ where ‖ ‖ = x i
,‖ ‖ x i , xRn
i 1 i1
56
n 2
2 n
x
i 1
i xi
i1
n n
2
xi
i 1
xi
i 1
‖ ‖ ‖ ‖ …(1)
n
Also ‖ ‖ = x i
i 1
n
= xi y i y (1,1,......1)
i 1
n
xi y i
i 1
n n
2 2
‖ ‖ xi
i 1
. y
i 1
i
‖ ‖ √ ‖ ‖
‖ ‖ ‖ ‖ …(2)
√
‖ ‖ ‖ ‖ ‖ ‖
√
‖. ‖ ‖. ‖
57
Solution:
n
1 xi yi
: d(x,y) 0 2 i
0
i 1 1 xi yi
n
1 xi yi
: d(x,y) = 0 2
i 1
i
1 xi yi
0
1 x1 y1 1 x2 y2 1 xn yn
1
2 ..... n 0
2 1 x1 y1 2 1 x2 y2 2 1 xn yn
1 x1 y1 1 x2 y2 1 xn yn
0, 0 ,....., 0
21 1 x1 y1 22 1 x2 y2 2n 1 xn yn
x1 y1 0, x2 y2 0,......., xn yn 0
x1 y1 0, x2 y2 0,......., xn yn 0
x1 y1 , x2 y2 ,......., xn yn
d(x,y) = 0 x = y
n
1 xi yi
: d(x,y) = 2
i 1
i
1 xi yi
n
1 yi xi
= 2
i 1
i
1 yi xi
= d(y,x)
n
1 xi zi n
1 xi yi yi zi
: d(x,z) = i
i 1 2 1 xi zi
i 1 2i 1 xi yi yi zi
n
1 xi yi yi zi
i 1 2i 1 xi yi yi zi
58
Solution:
n
xi yi
: d(x,y) 0 0
i 1 1 xi yi
n
xi yi
: d(x,y) = 0 1 x y
i 1
0
i i
x1 y1 x2 y2 xn yn
..... 0
1 x1 y1 1 x2 y2 1 xn yn
x1 y1 x2 y2 x yn
0, 0 ,....., n 0
1 x1 y1 1 x2 y2 1 xn yn
x1 y1 0, x2 y2 0,......., xn yn 0
x1 y1 0, x2 y2 0,......., xn yn 0
x1 y1 , x2 y2 ,......., xn yn
(x1, x2 ,....., xn ) ( y1, y2 ,....., yn )
d(x,y) = 0 x = y
n
xi yi
: d(x,y) = 1 x y
i 1 i i
59
n
xi zi n
xi yi yi zi
: d(x,z) = i 1 1 xi zi
i 1 1 xi yi yi zi
n
xi yi yi zi
i 1 1 xi yi yi zi
n xi yi y i zi
i 1 1 xi yi yi zi 1 xi yi yi zi
n
xi yi n
yi zi
d(x,z)
i 1 1 xi yi
i 1 1 yi zi
Solution:
2 2
: d(x,y) 0 x y 0
2 2
: d(x,y) = 0 x y 0
2 2
x y 0
2 2
x 0 , y 0
x 0 , y 0
d(x,y) = 0 x y
2 2
: d(x,y) = x y
60
2 2 2 2 2
: d(x,z) = x z = x yy z
2 2 2 2 2 2 2 2
x y y z x y + y z
61
Proof:
Case-I If x = 0 , y = 0 then
x, y x y holds trivially
0 a b b c
b b b b
a b b . c
c c c c
2
b
a
c
2
ac b
c
0 ac −| |
| | ac
2 2 2
x, y x, x . y, y x, x x , y, y y
2 2 2
x, y x . y
x, y x y
62
Question:
(a) For any sequence { } and { } in inner product space V. x ,
y then prove < , > <x ,y>
(b) If { } and { } are Cauchy sequence in V then < , > is
convergent sequence in F (R or C)
Solution (a):
Consider
xn , yn x, y xn , yn x, yn x, yn x, y
xn x, yn x, yn y
xn x, yn x, yn y ab a b
xn x yn x yn y x, y x y
x ‖ − ‖ 0
Given ….(1)
y ‖ − ‖ 0
xn , yn x, y 0 by (1)
63
xn , yn ym xn xm , ym
xn , yn ym xn xm , ym ab a b
xn yn ym xn xm ym x, y x y
xn , yn xm , ym 0 as m , n
64
n 2
L.H.S = x i 1
i
n n
2
= xi , x j x x, x
i 1 j 1
n n
= x , x
i 1 j 1
i j
n
= x , x i i xi , x j 0 i j
i 1
n
2
= x
i 1
i = R.H.S
Question:
(a) Define orthonormal system in an inner product space.
(b) xa xb 2
(c) Consider the set S = Sn (t ), c0 (t ), cn (t ); n 1,2,.........
Where (t) = , n = 1,2,….. , (t) =
√ √
(t) = cosnt , n = 1,2,………
√
In the real space c[o, 2 ] with inner product space V defined by
<x,y> = ∫ ( ) ( ) . Then show that S is an orthonormal system.
Solution: (a)
A set A = { a ; } of non-zero vectors in an inner product space V is said
to be orthonormal system if < a , b > = 0 , , , and < a , a > =1
2
xa 1 xa 1 ,a
2
Solution: (b) L. H . S xa xb
65
= xa , xa xb , xa xa , xb xb , xb
xa , xb xb , xa 0 , a b
= xa , xa xb , xb
2 2
= xa xb
=1+1=2
2
xa xb 2
xa xb 2
Solution (c) :
Step-I : To prove Sn , Cm 0 m, n 1,2,....... for m n
1 1
Sn , Cm sin nt , cos nt
2
1 1
Sn , Cm sin nt ,cos nt sin nt cos ntdt
0
2 2
1 sin nt 1
= = (0)
2n 0 2 n
S n , Cm = 0
Step-II For n m
1 1
Sn , Cm sin nt , cos mt
2
1 1
Sn , Cm sin nt ,cos mt sin nt cos mtdt
0
2 2
1 1
= 0 2sin nt cos mtdt 2 (sin(n m) sin(n m))tdt
2 0
66
=0+0=0
1 1 1
Also Sn , Sn sin nt , sin nt sin nt ,sin nt
2
1 2
sin tdt
=
0
2
1 1 cos nt
tdt
=
0
2
2
1 sin 2nt
t
2 2n 0
1
= (2 ) 1
2
Similarly, Cn , Cn 1 , C0 , C0 1
S is orthonormal system
67
Theorem:
A finite dimensional normed or Linear Space N is isomorphic to its second dual
space ∗∗ i.e. N ∗∗
∗∗
Proof: Let N be a finite dimensional normed or linear space of dim = n and
be its second dual space. Define
68
70