00 Journal of Inequalities and Applications
a SpringerOpen Journal
RESEARCH Open Access
Weighted analogues of Bernstein-type inequalities on several intervals
Mehmet Ali Akturk1,2 and Alexey Lukashov1,3*
"Correspondence: alukashov@fatih.edu.tr 1 Department of Mathematics, Fatih University, Istanbul, 34500,Turkey 3Department of Mechanics and Mathematics, Saratov State University, Saratov, 410012, Russia Full list of author information is available at the end of the article
Abstract
We give weighted analogues of Bernstein-type inequalities for trigonometric polynomials and rational functions on several intervals. MSC: 41A17; 42A05; 41A20
Keywords: weighted polynomial inequalities; inequalities for derivatives of rational functions
ft Spri
1 Introduction
Inequalities for polynomials have been a classical object of studies for more than one century. Modern expositions can be found in books and surveys [1-3] and [4]. Recently weighted analogues of classical polynomial inequalities were considered (see, for instance, [5,6] and [7]). Other ways of generalizations are in replacing the domain of polynomials by more complicated (disconnected) sets and (or) in considering polynomials in more general Chebyshev systems. The main goal of the paper is to give simple proofs of weighted analogues of Bernstein-type inequalities on several intervals. They are inspired by weighted Bernstein-type inequalities from Section 5.2, E.4 in [1]. It turns out that for disconnected sets similar ideas allow to write down weighted versions with an explicit constant.
Throughout the paper, we use the notations
PC = p: p(x) = £ akxk, ak e R(C) (1)
I k-o J
for the set of algebraic polynomials and
TC = 11: t(x) = y + (ak cos kx + bk sin kx), ak, bk e R(C) J (2)
for the set of trigonometric polynomials with real (complex) coefficients; as a weight w, we consider an arbitrary continuous positive function on a suitable set, || • || is the uniform norm on this set.
The first theorem is a weighted analogue of the Bernstein-type inequality on several intervals.
Theorem 1 Let E be a set consisting of a finite number l > 2 of disjoint intervals, E =
UL [aj, bj] c [0,1], a1 < b1 < a2 < ••• < b2l, then there exists no depending on w and E such
ringer
©2013 Akturk and Lukashov; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
p'n(x)w(x)
ni(x - aj)(x - bj)\
< n\\pnw\\E, x e E
for every polynomial pn e Pf, n > no.
Next result is a weighted version of the Bernstein-type inequality for trigonometric polynomials on several intervals.
Theorem 2 Let w be any function which is continuous and positive on
E = Utj 02j] £ 2n ], < 02 < ••• < hl < 02l+l = 01 + 2n j=1
5(0) = J~[ sin
0 - Qj
Then there exists n° depending on w and E such that \t'n(0)w(0|5(0)\\ < n\\tnw\\£, 0 e E
for every polynomial tn e Tf, n > n°. Inequality (6) is sharp in the sense that it is not possible to replace n in (6) by n(1 - e) for arbitrary e > 0.
Now let 0 < a < n, and let
K = {ei0 | 0 e [-a,a]}
be the circular arc on the unit circle of central angle 2a and with a midpoint at 1. Next result is a weighted version of the inequality from [8].
Theorem 3 With the above notations and for any continuous positive function w, there exists n° depending on w and a such that
\p'n(ei0) w{eie) I
< n\\pnw\\Ka, 0 e [-a,a]
for every polynomial pn e Pf, n > n°.
Next we recall the definition of the harmonic measure a)(z, G, D) of a set G c dD at a point z e D relative to the domain D,
w(z, G, D) =
if 9 , - 7-gD (Z, n Jg dn
z)|dZ |,
where gD (Z, z) is the Green function of the domain D and n is the exterior normal at Z (see, for example, [9]).
Our last result is an extension of Rusak's inequality [10, p.57] to the case of several intervals.
Theorem 4 Let rn be a complex-valued algebraic fraction
rn(x) = xn + bn, (0)
V IPv (x)|
where b1,...,bn e C, pv (x) = nj=1(x - Xj )vj is a real polynomial of degree v which is positive on E = Ujl=1 [a2;_i, a2j], a1 < a2 < • •• < an satisfying the condition
\rn(x) \ < 1, x e E, (1)
and y (x) is a differentiablefunction on E. Then the estimate
\(rn(x)y(x))'\ {tp'n(x))2y2(x) + y/2(x), x e int(E) (12)
is valid. Here
pn(x) = — ( (2n - v)mE(to,x) + ^^ Vj^E(xj,x) J, (13)
2 \ j=1 '
mE(z, x) = dx («(z,[a1, x] n E, C\E)). Ifx is not a multiple root of y (x), then the equality sign is valid only for algebraic fractions
rn(x) = e cos pn (x), | e | =1 (14)
at the points x satisfying
(y (x) sin pn(x))' = 0 (15)
in the case when
(2n - v)«(TO,[a2^-1,a2k],C\E) + ^ vj«(x;,[a2k-1,a2k],C\E) = qk, (16)
where qk e N, k = 1,..., l.
Remark 1 A Markov-type inequality, which is obtained by a similar method, was announced in the conference [11].
In the following we use several auxiliary results.
Lemma 1 [12] Consider any algebraic fraction
rn(x)=xn+b1x:^ ••+bn, ()
V pv (x)
where bi,...,bn e R, and pv (x) = nV=i(x - xj)vj is a real polynomial of degree v which is
positive onE = [a1, a2] U ••• U [a2l-1, a2l] c R, a1 < ••• < a2l. Then
+ ^ <........ (
where yn(x) is given by (13).
For further reference, it is convenient to give a particular case of a version of Lemma 1 from [13].
Lemma 2 The following inequality holds for any trigonometric polynomial tn e Tn and
0 e int(E), E is a real compact subset of [0,2n]:
+t2(01 <■«■'& • (
rn£(z,x) = d^fc n \ei0: inf E < 0 < x},, (20)
and T£ = {ei0: 0 e E}.
Lemma 3 [12,14] The following assertions are equivalent. 1. The trigonometric polynomial tn deviates least from zero on
E = [01,02] U ••• U [02l-1,02i], 01< 02 < ••• < 02i with respect to the sup-norm among all trigonometric polynomials of degree N/2 with leading coefficients cos ty and sin ty, i.e.,
max I tn (0 )l = inf max
0eE 1 ci,dieR 0eE
NN cos ty cos — 0 + sin ty sin — 0 r 2 r 2
^ N - 2j J N - 2j + y / c, cos —2— 0 + d;- sin —2— 0
has the maximal possible number ofextremum points on E.
2. For every j = 1,..., l, the equilibrium measures of the arcs rj- = {ei0: 0 e [02j-1,02j]} are positive rational numbers. More precisely,
Nw(<x, r,, C\ rE ) = qN, q(N) e N, j = 1,..., l. (22)
3. There is a real trigonometric polynomial aN- i of order N -2 such that for a constant An >0,
4(0)-S(0)al_ i (0)= AN, (3)
where S(0) is given by (5).
If any of those assertions is valid, then
(a) the numbers qjN) are equal to the number ofzeros ofrN(Q) on Ej = [ify-i, &2j], j = 1,..., l;
(b) the polynomial tn may also be written in terms of mE (z, x) as xn(Q)=Ans cos ( W Nm£ (to, Z) dZ), Q e E, (24)
J£n[Q1,t
where e e {-1,1}.
Lemma 4 [15] The density of the equilibrium measure from (20), E = [0^ 02] U ••• U [02H, 02l], 01 < 02 < ••• < 02l is given by
^ (to, 0 ) = ± , (5)
where Q(0) = fj sin( ^), and fy e [02j, 02j+1], j = 1, ...,l, 02j+1 := 01 + 2n , are uniquely determined by
f 02j+1 Q(0)
-j=== dZ =0, j =1,...,l. (6)
Proof We want to present here a different proof of the lemma which uses the representations of extremal polynomials in (24).
(1) Suppose firstly «(to, Vj, C\Ve) = , pj e N, j = 1,..., l. Then by Lemma 3 the function
tn(Q) = cos I tf/ 2Nrns (to, Z) dZ , Q e E (27)
is a real trigonometric polynomial of order N. If we take a derivative, we get
2N-l 0 _ „
Tn(0)=NQ(0)[] sin—^, (28)
where fy, j = 1, ...,2N -1, are zeros of aN _ (0) and there is a real trigonometric polynomial l of order N - l such that
4(0) - S(0)o2n_l (0) = 1. (29) Hence
ia, o' • 0 - Pj sin(n fen[01,0]2N(to, z) dZ)
on ; (0) = c sin —-— =- .-. (30)
N-j( ) Ji 2 v '
Moreover, tn (0) has a maximal number of deviation points, and inner zeros of its derivative coincide with zeros of aN_ (0), and tn has one zero fy at each gap (02j, 02j+1), j = 1,..., l.
sin(n /Vn[„ 0] 2nmE(to, Z) dZ) Tjn (0) = ±n jen[01,0 ] -—— q(0)
= - sin n
/ 2NmE(to,Z)dZ )n2NmE(to,0),
Je n[01,0 ]
so we have
me (to, 0) =
!Q(0)l
yiS(0)T'
Now equality (26) follows from the representation (24). Uniqueness of §j's follows from the uniqueness of extremal trigonometric polynomials in Lemma 3.
(2) Using density of the systems of l arcs satisfying «(to, r, C\rE) e Q, j = 1,..., l, among all systems of l arcs (see, for instance, [16, 17] and references therein), we obtain the lemma. □
2 Proofs
Proof of Theorem 2 First consider tn e Tn. By the Weierstrass approximation theorem, for any n >0, there is qk e Tk such that
w(0) <
qk (0)
n!=1! sin( ^)!
— < (1 + n)w(0), 0 e E,
where are given by (26) in Lemma 4. Hence
Itn(0)w(0)|S(0)l1/2| <
tn(0)qk(0) l |S(0^ nj=1! sin(^)!
|S(0)|1/2
(tnqk )'(0)
tn(0)qk (0)
nl=1! sin(0-j)! !S(0)!1/2
nl=1! sin^)!
and, using Lemmas 2 and 4, we have
|tn(0)w(0)|S(0)|1/2| < (n + k)"tnqk'' e + "tn "e k"qk "e < (n + k)(1 + n)"tnw"E
0 ( 0-
1 I smf
+ —(1 + n)"tnW'E "w"e m
.70- §
< n"tnw"E
k, , 1 ,
1 + n + -(1 + n) +-(1 + n)"w"e
n f 0-
| | sin f
where m := min{w(0) : 0 e E}. Now, for every tn e Tn and e > 0, provided n > 0 is suffi-
n \ m 1
ciently small, n > n0 such that e > n + + mm II w||f ), we get
4(0)w(0)|S(0)|1/2| < n(1 + e)||tnW|E
n (0 - %
I I sinf
and because of || ]~[|=i sin() IIf < 1, we obtain, for sufficiently small e >0,
|i«(0)w(0)^S(0)j| < «UnwUs.
Thecaseof tn e T^ is proved then similarly to the proof of [1, Corollary 5.1.5]. The theorem is sharp even for the case w = 1. Namely, we cannot replace the multiplier n by n(1 - e) with any e > 0 in the right-hand side of (6). Take E = [-a, a], 0 < a < n. Then we have S(0) = sin(0—a) sin(^), fy1 = n and
• ( 0 - % sin1
Consider
tn(0) =cos( 2narccosl —
Take 0 = 0n = 2 arcsin(sin a sin n), then
K (0n)| = n
cos f J\S(0n)\
|S(0n)|| = n cos 2 > n(1 — e)
for sufficiently large n such that
sin — <
4n sin2 2 '
(2) □
Proofs of Theorems 1 and 3 are quite analogous and they use related inequalities from [18,19].
Proof of Theorem 4 Firstly we consider the case when the numerator pn(x) has real coefficients. Put rn(x) = cos w = cos(arccos rn(x)); using Lemma 1, we obtain
^rn(x)y (x)) =
sin wr'n (x)y (x)
1 —r2(x)
+ cos WY '(x)
< sin2 w + cos2 w
(r!n (x)) Y (x) ,2
1 — r2(x) < V (<p'n (x))2Y 2(x) + Y '2(x).
+ Y ' (x)
The validity of the estimate for complex-valued algebraic fractions is proved by the same trick as in [1, Corollary 5.1.5]. Equality sign in the last inequality in (43) is valid only for the function rn(x) = ecos<n(x), !e! = 1 if (16) holds [13, 20]. Equality sign in the second inequality in (43) then holds only for the same function at those points where
- sin ^2n(x) = cos <?2n(x) (4)
v'i-(x)y (x) Y '(x) ' Equality (44) is equivalent to (y(x) sin<2n(x))' = 0. □
Competing interests
The authors declare that they have no competing Interests. Authors' contributions
Allauthors jointly worked on the results and they read and approved the finalmanuscript. Author details
1 Department of Mathematics, Fatih University, Istanbul, 34500, Turkey. 2Department of Engineering Sciences, Istanbul University, Istanbul, 34320, Turkey. 3 Department of Mechanics and Mathematics, Saratov State University, Saratov, 410012, Russia.
Acknowledgements
The first author would like to thank the Scientific and TechnologicalResearch Councilof Turkey (TUBlTAK) for the financial support. The authors are deeply gratefulto reviewers for theircareful reading of the manuscript and remarks which helped to improve the presentation.
Received: 7 February 2013 Accepted:11 October 2013 Published: 07 Nov 2013 References
1. Borwein, P, Erdelyi, T: Polynomials and Polynomial Inequalities. Springer, New York (1995)
2. Dubinin, VN: Methods of geometric function theory in classicaland modern problems for polynomials. Russ. Math. Surv. 67(4), 599-684(2012)
3. Milovanovic, GV, Mitrinovic, DS, Rassias, TM: Topics in Polynomials: ExtremalProblems, Inequalities, Zeros. World Scientific, Singapore (1994)
4. Rahman, QI, Schmeisser, G: Analytic Theory of Polynomials. London MathematicalSociety Monographs. New Series. Oxford University Press, Oxford (2002)
5. Andrievskii, VV: Weighted Lp Bernstein-type inequalities on a quasismooth curve in the complex plane (English summary). Acta Math. Hung. 135(1), 8-23 (2012)
6. Erdelyi, T: Markov-Bernstein-type inequality for trigonometric polynomials with respect to doubling weights on [-w, w]. Constr. Approx. 19(3), 329-338 (2003)
7. Mastroianni, G, Totik, V: Weighted polynomial inequalities with doubling and ,4to weights. Constr. Approx. 16(1), 37-71 (2000)
8. Nagy, B, Totik, V: Bernstein's inequality for algebraic polynomials on circular arcs. Constr. Approx. 37(2), 223-232 (2013)
9. Ransford, T: PotentialTheory in the Complex Plane. Cambridge University Press, Cambridge (1995)
10. Rusak, VN: RationalFunctions as an Apparatus of Approximation. Belorus. State Univ. Press, Minsk (1979)
11. Akturk, MA, Lukashov, AL: Weighted Markov inequality on severalintervals. AIP Conf. Proc. 1479(1), 568-569 (2012)
12. Lukashov, AL: Inequalities for derivatives of rationalfunctions on severalintervals. Izv. Math. 68(3), 543-565 (2004)
13. Dubinin, VN, Kalmykov, SI: A majoration principle for meromorphic functions. Mat. Sb. 198(12), 37-46 (2007)
14. Peherstorfer, F, Steinbauer, R: Orthogonalpolynomials on arcs of the unit circle: II. Orthogonalpolynomials with periodic reflection coefficients. J. Approx. Theory 87(1), 60-102 (1996)
15. Peherstorfer, F, Steinbauer, R: Strong asymptotics of orthonormalpolynomials with the aid of Green's function. SIAM J. Math. Anal. 32(2), 385-402 (2000)
16. Khrushchev, S: Periodic Schur functions and slit discs. J. Approx. Theory 159(2), 293-307 (2009)
17. Totik, V: The norm of minimalpolynomials on severalintervals. J. Approx. Theory 163(6), 738-746 (2011)
18. Totik, V: Polynomialinverse images and polynomialinequalities. Acta Math. 187(1), 139-160 (2001)
19. Totik, V: Bernstein-type inequalities. J. Approx. Theory 164(10), 1390-1401 (2012)
20. Kalmykov, SI: Majoration principles and some inequalities for polynomials and rationalfunctions with prescribed poles. J. Math. Sci. 157(4), 623-631 (2009)
10.1186/1029-242X-2013-487
Cite this article as: Akturk and Lukashov: Weighted analogues of Bernstein-type inequalities on several intervals.
Journal of Inequalities and Applications 2013, 2013:487
