Skip to main content
Springer logoLink to Springer
. 2018 Apr 20;2018(1):92. doi: 10.1186/s13660-018-1684-0

Pointwise approximation of modified conjugate functions by matrix operators of conjugate Fourier series of 2π/r-periodic functions

Mateusz Kubiak 1, Włodzimierz Łenski 1, Bogdan Szal 1,
PMCID: PMC5910496  PMID: 29706745

Abstract

We extend the results of Xh. Z. Krasniqi (Acta Comment. Univ. Tartu Math. 17:89–101, 2013) and the authors (Acta Comment. Univ. Tartu Math. 13:11–24, 2009; Proc. Est. Acad. Sci. 67:50–60, 2018) to the case when considered function is 2π/r-periodic and the measure of approximation depends on r-differences of the entries of the considered matrices.

Keywords: Rate of approximation, Summability of Fourier series

Introduction

Let L2π/rp(1p<) be the class of all 2π/r-periodic real-valued functions, integrable in the Lebesgue sense with the pth power over Qr= [π/r,π/r] with the norm

fL2π/rp=f()L2π/rp:=(Qr|f(t)|pdt)1/p,

where rN. It is clear that L2π/rpL2π/1p=L2πp and for fL2π/rp

fL2πp=r1/pfL2π/rp.

Taking into account the above relations, we will consider, for fL2π/r1, the trigonometric Fourier series as such a series of fL2π1 in the following form:

Sf(x):=a0(f)2+ν=1(aν(f)cosνx+bν(f)sinνx)

with the partial sums Skf and the conjugate one

S˜f(x):=ν=1(aν(f)sinνxbν(f)cosνx)

with the partial sums S˜kf. We also know that if fL2π1, then

f˜(x):=1π0πψx(t)12cott2dt=limϵ0+f˜(x,ϵ)=limϵ0+f˜r(x,ϵ),

where, for rN,

f˜r(x,ϵ):={1π(m=0[r/2]12mπr+ϵ2(m+1)πrϵ+2[r/2]πr+ϵ(2[r/2]+1)πr)ψx(t)12cott2dtfor an odd r,1πm=0[r/2]12mπr+ϵ2(m+1)πrϵψx(t)12cott2dtfor an even r,

and

f˜(x,ϵ)=f˜1(x,ϵ):=1πϵπψx(t)12cott2dt,

with

ψx(t):=f(x+t)f(xt),

exist for almost all x (cf. [4, Th. (3.1) IV]).

Let A:=(an,k) be an infinite matrix of real numbers such that

an,k0when k,n=0,1,2,,limnan,k=0andk=0an,k=1,

but A:=(an,k)k=0n, where

an,k=0when k>n.

We will use the notations

An,r=k=0|an,kan,k+r|,An,r=k=0n|an,kan,k+r|

for rN and

T˜n,Af(x):=k=0an,kS˜kf(x)(n=0,1,2,)

for the A-transformation of S̃f.

In this paper, we will study the estimate of |T˜n,Af(x)f˜r(x,ϵ)| by the function of modulus of continuity type, i.e. a nondecreasing continuous function ω̃ having the following properties: ω˜(0)=0, ω˜(δ1+δ2)ω˜(δ1)+ω˜(δ2) for any 0δ1δ2δ1+δ22π. We will also consider functions from the subclass L2π/rp(ω˜)β of L2π/rp for rN:

L2π/rp(ω˜)β={fL2π/rp:ω˜β(f,δ)L2π/rp=O(ω˜(δ)) when δ[0,2π] and β0},

where

ω˜βf(δ)L2π/rp=sup0|t|δ{|sinrt2|βψ(t)L2π/rp}.

It is easy to see that ω˜0f()L2π/rp=ω˜f()L2π/rp is the classical modulus of continuity. Moreover, it is clear that for βα0

ω˜βf(δ)L2π/rpω˜αf(δ)L2π/rp

and consequently

L2π/rp(ω˜)αL2π/rp(ω˜)β.

The deviation T˜n,Af(x)f˜r(x,ϵ) was estimated with r=1 in [2] and generalized in [1] as follows:

Theorem A

([1, Theorem 8, p. 95])

If fL2πp(ω˜)β with 1<p< and 0β<11p, where ω̃ satisfies the conditions:

{πn+1π(tγ|ψx(t)|ω˜(t))psinβpt2dt}1/p=Ox((n+1)γ) 1

with 0<γ<β+1p and

{0πn+1(t|ψx(t)|ω˜(t))psinβpt2dt}1/p=Ox((n+1)1), 2

then

|T˜n,Af(x)f˜(x,πn+1)|=Ox((n+1)β+1p+1An,1ω˜(πn+1)).

The next essential generalizations and improvements in [3, Theorem 1] were given. In these results f˜r(x,ϵ) and An,r (with rN) instead of f˜1(x,ϵ)=f˜(x,ϵ) and An,1, respectively, were taken. We can formulate them as follows.

Theorem B

([3, Theorem 1])

If fL2πp, 1<p<, 0β<11p and a function ω̃ of modulus of continuity type satisfies the conditions:

{0πr(n+1)(t|ψx(t)||sinrt2|βω˜(t))pdt}1/p=Ox((n+1)1) 3

for rN,

{2mπr2mπr+πr(n+1)(|ψx(t)||sinrt2|βω˜(t2mπr))pdt}1/p=Ox(1) 4

for a natural r3, where m{1,[r2]} when r is an odd or m{1,[r2]1} when r is an even natural number, and

{2mπr+πr(n+1)2mπr+πr(|ψx(t)||sinrt2|βω˜(t)(t2mπr)γ)pdt}1/p=Ox((n+1)γ), 5

for rN with 0<γ<β+1p, where m{0,[r2]} when r is an odd or m{0,[r2]1} when r is an even natural number. Moreover, let ω̃ satisfy, for a natural r2, the conditions:

{2(m+1)πrπr(n+1)2(m+1)πr(|ψx(t)||sinrt2|βω˜(2(m+1)πrt))pdt}1/p=Ox(1), 6
{2(m+1)πrπr2(m+1)πrπr(n+1)(|ψx(t)||sinrt2|βω˜(t)(2(m+1)πrt)γ)pdt}1/p=Ox((n+1)γ), 7

with 0<γ<β+1p, where m{0,[r2]1}. If a matrix A is such that

k=0(k+1)2an,k=O((n+1)2) 8

and

[l=0nk=lr+l1an,k]1=O(1) 9

with rN are true, then

|T˜n,Af(x)f˜r(x,πr(n+1))|=Ox((n+1)β+1p+1An,rω˜(πn+1)).

Theorem C

([3, Theorem 2])

Let fL2πp, 1<p<, 0β<11p and a function ω̃ of modulus of continuity type satisfy, for rN, the conditions: (4) and (5) with 0<γ<β+1p, where m{0,[r2]} when r is an odd or m{0,[r2]1} when r is an even natural number. Moreover, let ω̃ satisfy, for a natural r2, the conditions (6) and (7) with 0<γ<β+1p, where m{0,[r2]1}. If a matrix A is such that

k=0(k+1)an,k=O(n+1), 10

and (9) with rN are true, then

|T˜n,Af(x)f˜r(x,πr(n+1))|=Ox((n+1)β+1p+1An,rω˜(πn+1)).

In our theorems we generalize the above results considering 2π/r-periodic functions and using simpler assumptions.

In the paper k=ab=0 when a>b.

Statement of the results

To begin with, we will present the estimates of the quantities

|T˜n,Af(x)f˜r(x,πr(n+1))|andT˜n,Af()f˜r(,πr(n+1))L2π/rp.

Finally, we will formulate some remarks and corollaries.

Theorem 1

Suppose that fL2π/rp, 1<p<, rN, 0β<11p and a function ω̃ of the modulus of continuity type satisfies the conditions:

{0πr(n+1)(t|ψx(t)||sinrt2|βω˜(t))pdt}1/p=Ox((n+1)1), 11

when r=1 or

{0πr(n+1)(|ψx(t)||sinrt2|βω˜(t))pdt}1/p=Ox(1), 12

when r2, and

{πr(n+1)πr(|ψx(t)||sinrt2|βω˜(t)tγ)pdt}1/p=Ox((n+1)γ), 13

for rN with 0<γ<β+1p. If a matrix A is such that (8) and (9) are true, then

|T˜n,Af(x)f˜r(x,πr(n+1))|=Ox((n+1)β+1p+1An,rω˜(πn+1)).

Theorem 2

Suppose that fL2π/rp, 1<p<, rN, 0β<11p and a function ω̃ of the modulus of continuity type satisfies the conditions (12) and (13) for rN with 0<γ<β+1p. If a matrix A is such that (10) and (9) are true, then

|T˜n,Af(x)f˜r(x,πr(n+1))|=Ox((n+1)β+1p+1An,rω˜(πn+1)).

Remark 1

The Hölder inequality gives

k=0(k+1)an,k=k=0(k+1)an,k1/2an,k1/2[k=0(k+1)2an,k]1/2[k=0an,k]1/2=[k=0(k+1)2an,k]1/2

and thus the condition (8) implies (10), but the condition (12) implies (11). Therefore Theorems 1 and 2 are not comparable.

Theorem 3

Let fL2π/rp(ω˜)β, 1<p<, rN and 0β<11p. If a matrix A is such that (9) and (8) or (10) are true, then

T˜n,Af()f˜r(,πr(n+1))L2π/rp=Ox((n+1)β+1p+1An,rω˜(πn+1)).

Corollary 1

Taking r=1 the conditions (11) and (13) in Theorem 1 reduce to (1) and (2). Thus we obtain the results from [2] and Theorem A [1, Theorem 8, p. 95], but in the case of [3] (Theorem B and C) we reduce the assumptions.

Next, using more natural conditions when β>0 we can formulate, without proofs, the following theorems.

Theorem 4

Suppose that fL2π/rp, 1<p<, rN, 0<β<11p. Let a function ω̃ of the modulus of continuity type satisfy the conditions:

{πr(n+1)πr(tγ|ψx(t)||sinrt2|βω˜(t))pdt}1/p=Ox((n+1)γ1p) 14

for γ(1p,1p+β) and rN (instead of (13)), and

{0πr(n+1)(t|ψx(t)||sinrt2|βω˜(t))pdt}1/p=Ox((n+1)11p)

when r=1 or

{0πr(n+1)(|ψx(t)||sinrt2|βω˜(t))pdt}1/p=Ox((n+1)1p) 15

when r2 (instead of (11) and (12), respectively). If a matrix A is such that (9) and (8) are true, then

|T˜n,Af(x)fr˜(x,πr(n+1))|=Ox((n+1)β+1An,rω˜(πn+1)). 16

Moreover, if a function ω̃ of the modulus of continuity type and a matrix A satisfy the following conditions: (14) with rN and γ(1p,1p+β), (15) with rN, (9) and (10), then the estimate (16) is also true.

Theorem 5

Let fL2π/rp(ω˜)β with 1<p<, rN and 0<β<11p. If a matrix A is such that (9) and (8) or (10) are true, then

T˜n,Af()f˜r(,πr(n+1))L2π/rp=Ox((n+1)β+1An,rω˜(πn+1)).

Remark 2

We note that our extra conditions (9), (8) and (10) for a lower triangular infinite matrix A always hold.

Corollary 2

Considering the above remarks and the obvious inequality

An,rrAn,1for rN 17

our results also improve and generalize the mentioned result of Krasniqi [1].

Remark 3

We note that instead of L2π/rp(ω˜)β one can consider another subclass of L2π/rp generated by any function of the modulus of continuity type e.g. ω˜x such that

ω˜x(f,δ)=sup|t|δ|ψx(t)|ω˜x(δ)

or

ω˜x(f,δ)=1δ0δ|ψx(t)|dtω˜x(δ).

Auxiliary results

We begin this section by some notations from [5] and [4, Sect. 5 of Chapter II]. Let for r=1,2,

Dk,r(t)=sin(2k+r)t22sinrt2,D˜k,r(t)=cos(2k+r)t22sinrt2

and

D˜k,r(t)=cosrt2cos(2k+r)t22sinrt2=cosrt22sinrt2D˜k,r(t).

It is clear by [4] that

S˜kf(x)=1πππf(x+t)D˜k,1(t)dt

and

T˜n,Af(x)=1πππf(x+t)k=0an,kD˜k,1(t)dt.

Now, we present a very useful property of the modulus of continuity.

Lemma 1

([4])

A function ω̃ of modulus of continuity type on the interval [0,2π] satisfies the following condition:

δ21ω˜(δ2)2δ11ω˜(δ1)for δ2δ1>0.

Next, we present the following well-known estimates.

Lemma 2

([4])

If 0<|t|π then

|D˜k,1(t)|π2|t|,|D˜k,1(t)|π|t|

and, for any real t, we have

|Dk,1(t)|k+12,|D˜k,1(t)|12k(k+1)|t|,|D˜k,1(t)|k+1.

Lemma 3

([5, 6])

Let rN, lZ and (an)C. If t2lπr, then for every mn

k=nmaksinkt=k=nm(akak+r)D˜k,r(t)+k=m+1m+rakD˜k,r(t)k=nn+r1akD˜k,r(t),k=nmakcoskt=k=nm(akak+r)Dk,r(t)k=m+1m+rakDk,r(t)+k=nn+r1akDk,r(t).

We additionally need the following estimate as a consequence of Lemma 3.

Lemma 4

Let rN, lZ and (an,k)R0+ for n.kN0. If t2lπr, then

|12k=0an,kcos(2k+1)t2|12|sinrt2|(An,r+k=0r1an,k)1|sinrt2|An,r.

Proof

By Lemma 3,

12k=0an,kcos(2k+1)t2=12(k=0an,kcosktcost2k=0an,ksinktsint2)=cost22(k=0(an,kan,k+r)Dk,r(t)+k=0r1an,kDk,r(t))sint22(k=0(an,kan,k+r)D˜k,r(t)k=0r1an,kD˜k,r(t))

and our inequalities follow. □

We also need some special conditions which follow from the ones mentioned above.

Lemma 5

Suppose that fL2π/rp, where 1p< and rN. If the condition (12) holds with any function ω̃ of the modulus of continuity type and β0, then

{2(m+1)πrπr(n+1)2(m+1)πr(|ψx(t)|ω˜(2(m+1)πrt))p|sinrt2|βpdt}1p=Ox(1),

where m{0,[r2]1}.

Proof

By the substitution t=2(m+1)πru, we obtain

{2(m+1)πrπr(n+1)2(m+1)πr(|ψx(t)|ω˜(2(m+1)πrt))p|sinrt2|βpdt}1/p={0πr(n+1)(|ψx(2(m+1)πru)|ω˜(u)|sinr2(2(m+1)πru)|β)pdu}1/p={0πr(n+1)(|ψx(u)|ω˜(u)|sinru2|β)pdu}1/p.

Hence, by (12) our estimate follows. □

Lemma 6

Suppose that fL2π/rp, where 1p< and rN. If the condition (12) holds with any function ω̃ of the modulus of continuity type and β0, then

{2mπr2mπr+πr(n+1)(|ψx(t)|ω˜(t2mπr))p|sinrt2|βpdt}1p=Ox(1),

where m{0,[r2]}.

Proof

By the substitution t=2mπr+u, analogously to the above proof, we obtain

{2mπr2mπr+πr(n+1)(|ψx(t)|ω˜(t2mπr))p|sinrt2|βpdt}1/p={0πr(n+1)(|ψx(2mπr+u)|ω˜(u)|sinr2(2mπr+u)|β)pdu}1/p{0πr(n+1)(|ψx(u)|ω˜(u)|sinru2|β)pdt}1/p=Ox(1)

and we have the desired estimate. □

Now, we formulate another two lemmas without proofs. We can prove them in the same way as Lemmas 5 and 6, respectively.

Lemma 7

Suppose that fL2π/rp, where 1p< and rN. If the condition (13) holds with any function ω̃ of the modulus of continuity type and γ,β0, then

{2(m+1)πrπr2(m+1)πrπr(n+1)(|ψx(t)||sinrt2|βω˜(t)(2(m+1)πrt)γ)pdt}1/p=Ox((n+1)γ),

where m{0,[r2]1}.

Lemma 8

Suppose that fL2π/rp, where 1p< and rN. If the condition (13) holds with any function ω̃ of the modulus of continuity type and γ,β0, then

{2mπr+πr(n+1)2mπr+πr(|ψx(t)||sinrt2|βω˜(t)(t2mπr)γ)pdt}1/p=Ox((n+1)γ),

where m{0,[r2]}.

Proofs of theorems

Proof of Theorem 1

It is clear that for odd r

T˜n,Af(x)f˜r(x,πr(n+1))=1π0πψx(t)k=0an,kD˜k,1(t)dt+1π(m=0[r/2]12mπr+πr(n+1)2(m+1)πrπr(n+1)+2[r/2]πr+πr(n+1)(2[r/2]+1)πr)ψx(t)12cott2dt=1π(0πr(n+1)+m=1[r/2]2mπr2mπr+πr(n+1)+m=0[r/2]12(m+1)πrπr(n+1)2(m+1)πr)×ψx(t)k=0an,kD˜k,1(t)dt+1π(m=0[r/2]2mπr+πr(n+1)2(m+1)πr+m=0[r/2]1(2m+1)πr2(m+1)πrπr(n+1))ψx(t)k=0an,kD˜k,1(t)dt=I0(x)+I1(x)+I2(x)+I3(x)+I4(x)

and for even r

T˜n,Af(x)f˜r(x,πr(n+1))=1π0πψx(t)k=0an,kD˜k,1(t)dt+1πm=0[r/2]12mπr+πr(n+1)2(m+1)πrπr(n+1)ψx(t)12cott2dt=1π(0πr(n+1)+m=1[r/2]12mπr2mπr+πr(n+1)+m=0[r/2]12(m+1)πrπr(n+1)2(m+1)πr)×ψx(t)k=0an,kD˜k,1(t)dt+1π(m=0[r/2]12mπr+πr(n+1)(2m+1)πr+m=0[r/2]1(2m+1)πr2(m+1)πrπr(n+1))×ψx(t)k=0an,kD˜k,1(t)dt=I0(x)+I1(x)+I2(x)+I3(x)+I4(x),

whence

|T˜n,Af(x)f˜r(x,πr(n+1))||I0(x)|+|I1(x)|+|I1(x)|+|I2(x)|+|I3(x)|+|I3(x)|+|I4(x)|.

Next, using Lemma 2, (8), the Hölder inequality with p>1 and q=pp1 and (11) when r=1 or (12) when r2 we get

|I0(x)|=O((n+1)2)0πr(n+1)t|ψx(t)|dtO((n+1)2){0πr(n+1)(t|ψx(t)|ω˜(t))psinβprt2dt}1/p{0πr(n+1)(ω˜(t)sinβrt2)qdt}1qO((n+1)2)Ox((n+1)1)ω˜(πr(n+1)){0πr(n+1)(πrt)βqdt}1q=Ox((n+1))ω˜(πr(n+1))(πr(n+1))1qβ=Ox((n+1)β+1p)ω˜(πn+1),

for 0β<11p. We note that applying the condition (9) we have

[(n+1)An,r]1=[l=0nAn,r]1[l=0nk=l|an,kan,k+r|]1[l=0n|k=l(an,kan,k+r)|]1=[l=0nk=lr+l1an,k]1=O(1),

whence

|I0(x)|=Ox((n+1)1+β+1pAn,rω˜(πn+1)).

By Lemma 2

|I1(x)|+|I1(x)|+|I2(x)|1π(m=1[r/2]2mπr2mπr+πr(n+1)+m=0[r/2]12(m+1)πrπr(n+1)2(m+1)πr)|ψx(t)|tdt1π(m=1[r/2]2mπr2mπr+πr(n+1)+m=0[r/2]12(m+1)πrπr(n+1)2(m+1)πr)|ψx(t)|π/rdt

and using the Hölder inequality with p>1 and q=pp1

|I1(x)|+|I1(x)|+|I2(x)|Ox(1)m=1[r/2][2mπr2mπr+πr(n+1)(|ψx(t)|sinβrt2ω˜(t2mπr))pdt]1p×[2mπr2mπr+πr(n+1)(ω˜(t2mπr)sinβrt2)qdt]1q+Ox(1)m=1[r/2]1[2(m+1)πrπr(n+1)2(m+1)πr(|ψx(t)|sinβrt2ω˜(2(m+1)πrt))pdt]1p×[2(m+1)πrπr(n+1)2(m+1)πr(ω˜(2(m+1)πrt)sinβrt2)qdt]1q.

Hence, by Lemmas 5 and 6 with (12) and (9),

|I1(x)|+|I1(x)|+|I2(x)|=Ox(1)ω˜(πr(n+1))[0πr(n+1)(1sinβrt2)qdt]1q=Ox((n+1)β1q)ω˜(πn+1)=Ox((n+1)β+1pAn,rω˜(πn+1)),

for 0β<11p.

In the case of the last integrals, applying Lemma 4 we obtain

|I3(x)|+|I3(x)|+|I4(x)|1π(m=0[r/2]2mπr+πr(n+1)(2m+1)πr+m=0[r/2]1(2m+1)πr2(m+1)πrπr(n+1))|ψx(t)||sint2sinrt2|An,rdt.

Using the estimates |sint2||t|π for t[0,π], |sinrt2|rtπ2m for t[2mπr+πr(n+1),(2m+1)πr], where m{0,,[r/2]} and |sinrt2|2(m+1)rtπ for t[(2m+1)πr,2(m+1)πrπr(n+1)], where m{0,,[r/2]1}, we obtain

|I3(x)|+|I3(x)|+|I4(x)|An,rm=0[r/2]2mπr+πr(n+1)(2m+1)πr|ψx(t)|rtπ(t2mπr)dt+An,rm=0[r/2]1(2m+1)πr2(m+1)πrπr(n+1)|ψx(t)|rtπ[2(m+1)πrt]dt.

By the Hölder inequality with p>1 and q=pp1 we have

|I3(x)|+|I3(x)|+|I4(x)|πrAn,rm=0[r/2][2mπr+πr(n+1)2mπr+πr(|ψx(t)|ω˜(t)(t2mπr)γ|sinrt2|β)pdt]1p×[2mπr+πr(n+1)2mπr+πr(ω˜(t)(t2mπr)γt(t2mπr)|sinrt2|β)qdt]1q+πrAn,rm=0[r/2]1[2(m+1)πrπr2(m+1)πrπr(n+1)(|ψx(t)|ω˜(t)(2(m+1)πrt)γ|sinrt2|β)pdt]1p×[2(m+1)πrπr2(m+1)πrπr(n+1)(ω˜(t)(2(m+1)πrt)γt(2(m+1)πrt)|sinrt2|β)qdt]1q.

Further, using Lemmas 7 and 8 with (13) and Lemma 1 we get

|I3(x)|+|I3(x)|+|I4(x)|Ox(1)An,rm=0[r/2](n+1)γ[2mπr+πr(n+1)2mπr+πr(ω˜(t)(t2mπr)γt(t2mπr)|sinrt2|β)qdt]1q+Ox(1)An,rm=0[r/2]1(n+1)γ[2(m+1)πrπr2(m+1)πrπr(n+1)(ω˜(t)(2(m+1)πrt)γt(2(m+1)πrt)|sinrt2|β)qdt]1q=Ox(1)An,r[m=0[r/2](n+1)γ{πr(n+1)πr(ω˜(t+2mπr)tγ1(t+2mπr)|sinrt2|β)qdt}1q+m=0[r/2]1(n+1)γ{πr(n+1)πr(ω˜(2(m+1)πrt)tγ1(2(m+1)πrt)|sinrt2|β)qdt}1q]=Ox(1)An,r(n+1)γ{πr(n+1)πr(ω˜(t)tγ1t|sinrt2|β)qdt}1q=Ox(1)An,r(n+1)1+γω˜(πr(n+1))(πr(n+1)πrt(γ1β)qdt)1q=Ox(1)An,r(n+1)1+γω˜(πr(n+1))(n+1)1+βγ1q=Ox((n+1)1+β+1pAn,rω˜(π(n+1)))

for 0<γ<β+1p.

Collecting the partial estimates our statement follows.

Proof of Theorem 2

The proof is the same as above, but for estimate of |I0(x)| we only used the inequality |D˜k,1(t)|k+1 from Lemma 2, and the condition (10) instead of (8).

Proof of Theorem 3

We note that for the estimate of T˜n,Af()f˜r(,π(n+1))L2πp we need the conditions on ω̃ from the assumptions of Theorems 1 or 2. These conditions always hold with ψ(t)L2π/rp instead of |ψx(t)| and thus the desired result follows.

Authors’ contributions

MK, WŁ and BS contributed equally in all stages to the writing of the paper. All authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Mateusz Kubiak, Email: M.Kubiak@wmie.uz.zgora.pl.

Włodzimierz Łenski, Email: W.Lenski@wmie.uz.zgora.pl.

Bogdan Szal, Email: B.Szal@wmie.uz.zgora.pl.

References

  • 1.Krasniqi X.Z. Slight extensions of some theorems on the rate of pointwise approximation of functions from some subclasses of Lp. Acta Comment. Univ. Tartu Math. 2013;17:89–101. [Google Scholar]
  • 2.Łenski W., Szal B. Approximation of functions belonging to the class Lp(ω) by linear operators. Acta Comment. Univ. Tartu Math. 2009;13:11–24. [Google Scholar]
  • 3.Łenski W., Szal B. Pointwise approximation of modified conjugate functions by matrix operators of their Fourier series. Proc. Est. Acad. Sci. 2018;67(1):50–60. doi: 10.3176/proc.2018.1.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Zygmund A. Trigonometric Series. Cambridge: Cambridge University Press; 2002. [Google Scholar]
  • 5.Szal B. On L-convergence of trigonometric series. J. Math. Anal. Appl. 2011;373:449–463. doi: 10.1016/j.jmaa.2010.08.003. [DOI] [Google Scholar]
  • 6.Szal B. A new class of numerical sequences and its applications to uniform convergence of sine series. Math. Nachr. 2011;284(14–15):1985–2002. doi: 10.1002/mana.200910061. [DOI] [Google Scholar]

Articles from Journal of Inequalities and Applications are provided here courtesy of Springer

RESOURCES