You are on page 1of 7

TAMKANG JOURNAL OF MATHEMATICS

Volume 42, Number 1, 31-37, Spring 2011


doi:10.5556/j.tkjm.42.2011.31-37
-
-
+
--
-
+ -

Available online at http://journals.math.tku.edu.tw/index.php/TJM

DEGREE OF APPROXIMATION OF A FUNCTION BELONGING


TO WEIGHTED (L r , ξ(t )) CLASS BY (C , 1)(E , q) MEANS

HARE KRISHNA NIGAM

Abstract. In this paper, a¡new¢ theorem on degree of approximation of a function f ∈


W (L r , ξ (t )) class by (C , 1) E , q product summability means of Fourier series has been
proved.

1. Introduction

A good amount of work on degree of approximation of functions belonging to Lipα,


Lip(α, r ), Lip (ξ (t ) , r ) and W (L r , ξ (t )) classes using Cesàro, Nörlund and generalized Nörlund
single summability methods has been done by number of researchers like Alexits [1], Sahney
and Geol [12], Qureshi and Neha [10], Qureshi [7, 8, 9], Chandra [2], Khan [4], Leindler [5] and
Rhoades [11]. But till now nothing seems to have been done in the direction of present work.
Therefore, in this paper, we establish a theorem on degree of approximation of function f be-
¡ ¢
longing to weighted i.e. W (L r , ξ (t ))-class, r ≥ 1, by E , q (C , 1) summability means of Fourier
series.
be a given infinite series with sequence of its n t h partial sum {s n }.
P∞
Let n=0 u n

The (C , 1) transform is defined as the n t h partial sum of (C , 1) summability and is given


by
s0 + s1 + s2 + · · · + sn
tn =
n +1
1 X n
= s n → s as n → ∞
n + 1 k=0
P∞
then the series n=0 u n is summable to the definite number s by (C , 1) method.
If à !
1 n n
q
q n−k s k → s as n → ∞
X
(E , q) = E n =
(1 + q)n k=0 k

Received November 09, 2009.


2000 Mathematics Subject Classification. Primary 42B05, 42B08.
Key
¡ words
¢ and phrases. ¡Degree ¢ of approximation, W (L r , ξ (t )) class of function, (C , 1) summability,
E , q summability, (C , 1) E , q product summability, Fourier series, Lebesgue integral.

31
32 HARE KRISHNA NIGAM

P∞ ¡ ¢
then the infinite series n=0 u n with partial sum s n is said to be summable by E , q method
to a definite number s (Hardy [3]).
A product of (C , 1) transform of (E , q) transform defines (C , 1)(E , q) transform and it can
q
be denoted by C n1 E n .
Thus if n
1 X q q
C n1 E n =
E → s as n → ∞
n + 1 k=0 k
q
where E n denotes the (E , q) transform, then the series ∞
P
n=0 u n is said to be summable by
(C , 1)(E , q) means or summable (C , 1)(E , q) to a definite number s.
Let f (x) be periodic with period 2π and integrable in the sense of Lebesgue. The Fourier
series of f (x) is given by

a0 X∞
f (x) ∽ + (a n cos nx + b n sin nx) (1.1)
2 n=1

with n th partial sum s n ( f ; x).


A function f ∈ Lipα if

f (x + t ) − f (x) = O(|t |α ) for 0 < α ≤ 1,

f ∈ Lip(α, r ) if
³ Z2π ´1
| f (x + t ) − f (x)|r d x = O(|t |α ),
r
0 < α ≤ 1, r ≥ 1
0
(Definition 5.38 of Mc Fadden [6], 1942).

Given a positive increasing function ξ(t ) and an integer r ≥ 1, f ∈ Lip(ξ(t ), r ) if


³ Z2π ´1
| f (x + t ) − f (x)|r d x
r
= O(ξ(t )),
0

and that f (t ) ∈ W (L r , ξ(t )) if


³ Z2π ¯© ¯r ´ 1
¯ f (x + t ) − f (x) sinβ x ¯ d x = O(ξ(t )), β ≥ 0.
¯ ª ¯ r

If β = 0 then W (L r , ξ (t )) coincides with the class Lip (ξ (t ) , r ) and if ξ (t ) = t α then Lip(ξ (t ) , r )


class coincides with the class Lip (α, r ) and if r → ∞ then Lip(α, r ) class reduces to the class
Lipα.
Now we define norm by
³ Z2π ´1
| f (x)|r d t
r
k f kr = , r ≥1 (1.2)
0
DEGREE OF APPROXIMATION OF A FUNCTION BELONGING TO WEIGHTED (L p ,ξ(t)) CLASS 33

and let the degree of approximation E n ( f ) be given by

E n ( f ) = min k f − Tn kr (Zygmund [14]) (1.3)

where Tn (x) is a trigonometric polynomial of degree n.


The following notations are used through out the paper:

φ(t ) = f (x + t ) + f (x − t ) − 2 f (x)
sin(ν + 12 )t o
à !
1 n · 1 k n k ¸
k−ν
X X
K n (t ) = q .
2π(n + 1) k=0 (1 + q)k ν=0 ν sin( 2t )

2. Main theorem

If f is a 2π-periodic function, Lebesgue integrable on [0, 2π] and belongs to W (L r , ξ (t ))


class, then its degree of approximation is given by
h 1
³ 1 ´i
q
kC n1 E n − f kr = O (n + 1)β+ r ξ (2.1)
n +1
provided ξ (t ) satisfies the following conditions:
n ξ(t ) o
be a decreasing sequence, (2.2)
t
½Z 1 ³ t |φ(t )| ´r ¾1 ³ 1 ´
n+1 r
βr
sin tdt =O (2.3)
0 ξ(t ) n +1
and
½Zπ ³ −δ ¾1
t |φ(t )| ´r r n o
dt = O (n + 1)δ (2.4)
1
n+1
ξ(t )
1
where δ is an arbitrary number such that s(1 − δ) − 1 > 0, r + 1s = 1, conditions (2.3) and (2.4)
q
hold uniformly in x and C n1 E n is (C , 1)(E , q) means of the Fourier series (1.1).

3. Lemmas

Following lemmas are required for the proof of our theorem:

Lemma 1.
1
|K n (t )| = O (n + 1) , for 0 ≤ t ≤ .
n +1

1
Proof. For 0 ≤ t ≤ n+1 , sin nt ≤ n sin t

(2ν + 1) sin( 2t ) i¯¯


¯ Ã ! ¯
1 ¯X n h 1 k k
k−ν
X
|K n (t )| ≤ q
¯
2π(n + 1) ¯k=0 (1 + q)k ν=0 ν sin( 2t )
¯ ¯
¯
34 HARE KRISHNA NIGAM

¯ Ã ! ¯
1 ¯Xn h 1 k k i¯
k−ν ¯
X
≤ (2k + 1) q
¯
2π(n + 1) ¯k=0 (1 + q)k
¯ ¯
ν=0 ν
¯
1 Xn
= (2k + 1)
2π(n + 1) k=0
= O(n + 1) 

Lemma 2. ³1´ 1
|K n (t )| = O , for ≤ t ≤ π.
t n +1

1
¡t ¢ t
Proof. For n+1 ≤ t ≤ π, by applying Jordan’s lemma, sin 2 ≥ π and sin nt ≤ 1
¯ Ã ! ¯
¯ n h k k i¯¯
1 1 1
q k−ν ³ ´ ¯¯
¯X X
|K n (t )| ≤ ¯
2π(n + 1) ¯¯k=0 (1 + q)k ν=0 ν t ¯
π
1 n h1 i
(1 + q)k
X

2t (n + 1) k=0 (1 + q)k
1 Xn
= 1
2t (n + 1) k=0
³1´
=O . 
t

4. Proof of main theorem


¡ ¢
Following Titchmarsh [13] and using Riemann-Lebesgue theorem, s n f ; x of the series
(1.1) is given by ³ ´
1
Zπ sin n + 12 t
s n ( f ; x) − f (x) = φ(t ) dt.
2π 0 sin 2t
¡ ¢ ¡ q¢ ¡ ¢
Therefore, using (1.1) the E , q transform E n of s n f ; x is given by
( Ã ! ¶ )
1

φ(t ) n n µ
1
q X n−k
En − f (x) = q sin k + t d t .
2π(1 + q)k sin 2t
¡ ¢
0 k=0 k 2

q
Now denoting (C , 1) E , q transform of s n f ; x as C n1 E n , we write
¡ ¢ ¡ ¢

à !
1 n · 1
Zπ k k
φ(t ) n X ³ 1´ o
¸
q
C n1 E n n−k
X
− f (x) = q sin ν + t d t
2π(n + 1) k=0 (1 + q)k 0 sin( 2t ) ν=0 ν 2
·Z 1 Zπ ¸
n+1
= + φ(t )K n (t )d t
1
0 n+1

= I 1 + I 2 (say). (4.1)
DEGREE OF APPROXIMATION OF A FUNCTION BELONGING TO WEIGHTED (L p ,ξ(t)) CLASS 35

Now Z 1
n+1
|I 1 | ≤ |φ(t )| |K n (t )|d t .
0

We have

|φ(x + t ) − φ(x)| ≤ | f (u + x + t ) − f (u + x)| + | f (u − x − t ) − f (u − x)|.

Hence, by Minkowiski’s inequality,


·Z2π n ¯r ¸ r1
¯φ (x + t ) − φ (x) sinβ x ¯¯ d x
¯ ª
0
·Z2π ¯ ¸ 1r
¯© ª β ¯¯r
≤ ¯ f (u + x + t ) − f (u + x) sin x ¯ d x
0
·Z2π ¯ ¸ 1r
¯© ª β ¯¯r
+ ¯ f (u − x − t ) − f (u − x) sin x ¯ d x
0
= O {ξ (t )} .

Then f ∈ W (L r , ξ (t )) ⇒ φ ∈ W (L r , ξ (t )).
Using Hölder’s inequality and the fact that φ (t ) ∈ W (L r , ξ (t )), (2.3), Lemma 1 and second
mean value theorem for integrals, we have
"Z 1
( ¯ )r # 1 "Z 1
#1
t ¯φ (t )¯ sinβ t
¯ r ½ ¾s s
n+1 n+1 ξ (t )|K n (t )|
|I 1 | ≤ dt dt
0 ξ (t ) 0 t sinβ t
¶ "Z 1 ¾s #1
µ ½ s
1 n+1 (n + 1)ξ (t )
=O dt
n +1 0 t 1+β
¶¾ "Z 1 #1
½ µ s
1 n+1 dt 1
=O ξ for some 0 <∈<
n +1 ∈ t (1+β)s n +1
½ µ ¶¾
1 1 1 1
= O (n + 1)β+ r ξ since + = 1. (4.2)
n +1 r s

Now using Hölder’s inequality, |sin t | < 1, sin t ≥ 2t


¡ ¢
π , Lemma 2, (2.2), (2.4) and mean
value theorem,
"Z ( )r # 1 "Z #1
t −δ ¯φ (t )¯ sinβ t
π
¯ ¯ r π ½ ¾s s
ξ (t )|K n (t )|
|I 2 | ≤ dt dt
1
n+1
ξ (t ) 1
n+1
t −δ sinβ t
"Z ½ # 1
π ¾s s
n o ξ (t )
= O (n + 1)δ dt
t β+1−δ 1
n+1
  ³ ´ s 1
s
n o Zn+1  ξ 1y  d y
= O (n + 1)δ 
 
 y δ−1−β  y 2
1

π
36 HARE KRISHNA NIGAM

¶¾ "Zn+1 #1
½ µ s
δ 1 dy
= O (n + 1) ξ
n +1 1
π
y s (δ−1−β)+2
#1
½ µ ¶¾ " s (1+β−δ)−1 s (δ−1−β)+1 s
1 (n + 1) − π
= O (n + 1)δ ξ ¡ ¢
n +1 s 1+β−δ −1
½ µ ¶¾ h
1 1
i
δ
= O (n + 1) ξ (n + 1)(1+β−δ)− s
n +1
½ µ ¶¾
β+ r1 1 1 1
= O (n + 1) ξ since + = 1. (4.3)
n +1 r s

Now combining (4.1), (4.2) and (4.3), we get


½ µ ¶¾
¯C E − f (x)¯ = O (n + 1)β+ r1 ξ
¯ 1 q ¯ 1
n n
n +1
½Z2π ½ µ ¶¾r ¾ 1r
° 1 q ° β+ 1r 1
°C E − f ° =
n n r O (n + 1) ξ dx
0 n +1
½ µ ¶¾ "½Z2π ¾ r1 #
β+ r1 1
= O (n + 1) ξ dx
n +1 0
½ µ ¶¾
β+ r1 1
= O (n + 1) ξ .
n +1

This completes the proof of the main theorem. 

5. Corollaries

Following corollaries can be derived from our main theorem:

Corollary 1. If β = 0 and ξ (t ) = t α , then the degree of approximation of a function f ∈ Lip (α, r ) , 0 <
α ≤ 1, is given by ( )
° 1 q °
°C E − f (x)° = O 1
n n r 1
.
(n + 1)α− r

Corollary 2. If r → ∞ in Corollary 1, then for 0 < α < 1,


½ ¾
° 1 q °
°C E − f (x)° = O 1
n n .

(n + 1)α

Corollary 3. If β = 0, ξ (t ) = t α and q n = 1 ∀n, then the degree of approximation of a function


f ∈ Lip(α, r ) , 0 < α ≤ 1, is given by
( )
° 1 1 °
°C E − f (x)° = O 1
n n r 1
.
(n + 1)α− r
DEGREE OF APPROXIMATION OF A FUNCTION BELONGING TO WEIGHTED (L p ,ξ(t)) CLASS 37

Corollary 4. If r → ∞ in Corollary 3, then for 0 < α < 1, we have


½ ¾
° 1 1 °
°C E − f (x)° = O 1
n n .

(n + 1)α

Remark. Independent proofs of Corollaries 1 and 3 can be obtained along the same lines of
our theorem.

Acknowledgement

Author is grateful to his beloved parents for their encouragement to his work. Author
is also thankful to the managing editor Prof. Shou-Jen Hu for his kind cooperation during
communication.

References

[1] G. Alexits, Convergence Problems of Orthogonal Series, Pergamon Press, London, 1961.
[2] Prem Chandra, Trigonometric approximation of functions in L p norm, J. Math. Anal. Appl., 275 (2002), 13–26.
[3] G. H. Hardy, Divergent Series, first edition, Oxford University Press, 1949.
[4] H. H. Khan, On degree of approximation of functions belonging to the class Lip(α, p), Indian J. Pure Appl.
Math. 5 (1974), 132–136.
[5] László Leindler, Trigonometric approximation of functions in L p norm, J. Math. Anal. Appl. 302 (2005).
[6] Leonard McFadden, Absolute Nörlund summability, Duke Math. J., 9 (1942), 168–207.
[7] K. Qureshi, On the degree of approximation of a periodic function f by almost Nörlund means, Tamkang J.
Math., 12 (1981), 35–38.
[8] K. Qureshi, On the degree of approximation of a function belonging to the class Lipα, Indian J. pure Appl.
Math., 13 (1982), 898–903.
[9] K. Qureshi, On the degree of approximation of a function belonging to weighted W (L r ,ξ(t)) class, Indian J.
pure Appl. Math., 13 (1982), 471–475.
[10] K. Qureshi and H. K. Neha, A class of functions and their degree of approximation, Ganita., 41 (1990), 37–42.
[11] B. E. Rhaodes, On the degree of approximation of functions belonging to Lipschitz class by Hausdorff means
of its Fourier series, Tamkang J. Math., 34 (2003), 245–247.
[12] B. N. Sahney and D. S. Goel, On the degree of continuous functions, Ranchi University, Math. Jour., 4 (1973),
50–53.
[13] E. C. Titchmarsh, The Theory of functions, Oxford University Press, 1939, 402–403.
[14] A. Zygmund, Trigonometric Series, 2nd rev. ed., Vol. 1, Cambridge Univ. Press, 1959.

Department of Mathematics, Faculty of Engineering and Technology, Mody Institute of Technology and Science
(Deemed University), Laxmangarh-332311, Sikar (Rajasthan), India.

E-mail: harekrishnan@yahoo.com

You might also like