-
The point value maximization problem for positive definite functions supported in a given subset of a locally compact group
Authors:
Sándor Krenedits,
Szilárd Gy. Révész
Abstract:
The century old extremal problem, solved by Carathéodory and Fejér, concerns a nonnegative trigonometric polynomial normalized by a0 = 1, and the quantity to be maximized is the coefficient a1. In the complex exponential form, the coefficient sequence (ck) will be supported in [-n; n] and normalized by c0 =1. Reformulating, nonnegativity of T translates to positive definiteness of the sequence (ck…
▽ More
The century old extremal problem, solved by Carathéodory and Fejér, concerns a nonnegative trigonometric polynomial normalized by a0 = 1, and the quantity to be maximized is the coefficient a1. In the complex exponential form, the coefficient sequence (ck) will be supported in [-n; n] and normalized by c0 =1. Reformulating, nonnegativity of T translates to positive definiteness of the sequence (ck), and the extremal problem becomes a maximization problem for the value at 1 of a normalized positive definite function c : Z --> C, supported in [-n; n]. Boas and Katz, Arestov, Berdysheva and Berens, Kolountzakis and Révész and recently Krenedits and Révész investigated the problem in increasing generality, reaching analogous results for all locally compact Abelian groups. We prove an extension to all the known results in not necessarily commutative locally compact groups.
△ Less
Submitted 2 May, 2015; v1 submitted 15 April, 2015;
originally announced April 2015.
-
The Caratheodory-Fejér type extremal problem on locally compact Abelian groups
Authors:
Sándor Krenedits,
Szilárd Gy. Révész
Abstract:
We consider the extremal problem of maximizing a point value jf(z)j at a given point z 2 G by some positive definite and continuous function f on an Abelian group G, where for a given symmetric open set 3 z, f vanishes outside and is normalized by f(0) = 1. Denote the extremal value as CG(; z). This extremal problem was investigated in R and Rd and for a 0-symmetric convex body in a paper of Boas…
▽ More
We consider the extremal problem of maximizing a point value jf(z)j at a given point z 2 G by some positive definite and continuous function f on an Abelian group G, where for a given symmetric open set 3 z, f vanishes outside and is normalized by f(0) = 1. Denote the extremal value as CG(; z). This extremal problem was investigated in R and Rd and for a 0-symmetric convex body in a paper of Boas and Kac in 1943. Arestov and Berdysheva extended the investigation to Td, where T := R=Z. Kolountzakis and R?ev?esz gave a more general setting, considering arbitrary open sets, in all the classical groups above. Also they observed, that such extremal problems occurred in certain special cases and in a different, but equivalent formulation already a century ago in the work of Carath?eodory and Fej?er. Moreover, following observations of Boas and Kac, Kolountzakis and R?ev?esz showed how the general problem can be reduced to equivalent discrete problems of "Carath?eodory-Fej?er type" on Z or Zm := Z=mZ. We extend the results of Kolountzakis and R?ev?esz to locally compact Abelian groups.
△ Less
Submitted 11 December, 2013; v1 submitted 30 March, 2013;
originally announced April 2013.
-
Special quadrature error estimates and their application in the hardy-littlewood majorant problem
Authors:
Sándor Krenedits
Abstract:
The Hardy-Littlewood majorant problem has a positive answer only for expo- nents p which are even integers, while there are counterexamples for all p =2 2N. Montgomery conjectured that there exist counterexamples even among idempotent polynomials. This was proved recently by Mockenhaupt and Schlag with some four-term idempotents. However, Mockenhaupt conjectured that even the classical 1 + e^{2\pi…
▽ More
The Hardy-Littlewood majorant problem has a positive answer only for expo- nents p which are even integers, while there are counterexamples for all p =2 2N. Montgomery conjectured that there exist counterexamples even among idempotent polynomials. This was proved recently by Mockenhaupt and Schlag with some four-term idempotents. However, Mockenhaupt conjectured that even the classical 1 + e^{2\piix} \pm e^{2\pii(k+2)x} three- term character sums, should work for all 2k < p < 2k+2 and for all k \in N. In two previous papers we proved this conjecture for k = 0; 1; 2; 3; 4, i.e. in the range 0 < p < 10, p \notin 2N. Here we demonstrate that even the k = 5 case holds true. Refinements in the technical features of our approach include use of total variation and integral mean estimates in error bounds for a certain fourth order quadrature. Our estimates make good use of the special forms of functions we encounter: linear combinations of powers and powers of logarithms of absolute value squares of trigonometric polynomials of given degree. Thus the quadrature error estimates are less general, but we can find better constants which are of practical use for us.
△ Less
Submitted 27 July, 2012;
originally announced July 2012.
-
On Mockenhoupt's Conjecture in the Hardy-Littlewood Majorant Problem
Authors:
Sándor Krenedits
Abstract:
The Hardy-Littlewood majorant problem has a positive answer only for expo- nents p which are even integers, while there are counterexamples for all p =2 2N. Montgomery conjectured that even among the idempotent polynomials there must exist some counterex- amples, i.e. there exist some finite set of characters and some ? signs with which the signed character sum has larger pth norm than the idempot…
▽ More
The Hardy-Littlewood majorant problem has a positive answer only for expo- nents p which are even integers, while there are counterexamples for all p =2 2N. Montgomery conjectured that even among the idempotent polynomials there must exist some counterex- amples, i.e. there exist some finite set of characters and some ? signs with which the signed character sum has larger pth norm than the idempotent obtained with all the signs chosen + in the character sum. That conjecture was proved recently by Mockenhaupt and Schlag. However, Mockenhaupt conjectured that even the classical 1 + e2?ix ? e2?i(k+2)x three- term character sums, used for p = 3 and k = 1 already by Hardy and Littlewood, should work in this respect. That remained unproved, as the construction of Mockenhaupt and Schlag works with four-term idempotents. In our previous work we proved this conjecture for k = 0; 1; 2, i.e. in the range 0 < p < 6, p =2 2N. Continuing this work here we demonstrate that even the k = 3; 4 cases hold true. Several refinement in the technical features of our approach include improved fourth order quadra- ture formulae, finite estimation of G02=G (with G being the absolute value square function of an idempotent), valid even at a zero of G, and detailed error estimates of approximations of various derivatives in subintervals, chosen to have accelerated convergence due to smaller radius of the Taylor approximation.
△ Less
Submitted 11 March, 2012;
originally announced March 2012.
-
Three-term idempotent counterexamples in the Hardy-Littlewood majorant problem
Authors:
Sándor Krenedits
Abstract:
The Hardy-Littlewood majorant problem was raised in the 30's and it can be formulated as the question whether $\int |f|^p\ge \int|g|^p$ whenever $\hat{f}\ge|\hat g|$. It has a positive answer only for exponents $p$ which are even integers. Montgomery conjectured that even among the idempotent polynomials there must exist some counterexamples, i.e. there exists some finite set of exponentials and s…
▽ More
The Hardy-Littlewood majorant problem was raised in the 30's and it can be formulated as the question whether $\int |f|^p\ge \int|g|^p$ whenever $\hat{f}\ge|\hat g|$. It has a positive answer only for exponents $p$ which are even integers. Montgomery conjectured that even among the idempotent polynomials there must exist some counterexamples, i.e. there exists some finite set of exponentials and some $\pm$ signs with which the signed exponential sum has larger $p^{\rm th}$ norm than the idempotent obtained with all the signs chosen + in the exponential sum. That conjecture was proved recently by Mockenhaupt and Schlag. \comment{Their construction was used by Bonami and Révész to find analogous examples among bivariate idempotents, which were in turn used to show integral concentration properties of univariate idempotents.}However, a natural question is if even the classical $1+e^{2πi x} \pm e^{2πi (k+2)x}$ three-term exponential sums, used for $p=3$ and $k=1$ already by Hardy and Littlewood, should work in this respect. That remained unproved, as the construction of Mockenhaupt and Schlag works with four-term idempotents. We investigate the sharpened question and show that at least in certain cases there indeed exist three-term idempotent counterexamples in the Hardy-Littlewood majorant problem; that is we have for $0<p<6, p \notin 2\NN$ $\int_0^{\frac12}|1+e^{2πix}-e^{2πi([\frac p2]+2)x}|^p > \int_0^{\frac12}|1+e^{2πix}+e^{2πi([\frac p2]+2)x}|^p$. The proof combines delicate calculus with numerical integration and precise error estimates.
△ Less
Submitted 20 September, 2011; v1 submitted 2 June, 2010;
originally announced June 2010.
