The Jackson Inequality for the Best L-2-Approximation of Functions on [0,1] with the Weight X
Jian Li,Yongping Liu
2008-01-01
Abstract:Let L 2([0, 11, x) be the space of the real valued, measurable, square summable functions on [0, 1] with weight x, and let L-n, be the subspace of L-2([0, 1], x) defined by a linear combination of J(0)(mu(k)x), where J(0) is the Bessel function of order 0 and {mu(k)} is the strictly increasing sequence of all positive zeros of J(0). For f is an element of L-2([0, 1], x), let E(f, L-n) be the error of the best L-2([0, 1], x), i.e., approximation of f by elements of L-n. The shift operator of f at point X is an element of [0, 1] with step t is an element of [0, 1] is defined byT(t)f (x) = 1/pi integral(0) (pi) f(root (x(2) + t(2) -2xt cos theta)d theta.The differences (I - T(t))(r/2)f = Sigma(infinity)(j=0)(-1)j((r/2)(j))T-j(t)f of order r is an element of (0, infinity) and the L-2([0, 1], x)- modulus of continuity w(r)(f, tau) = sup{||(I - T( t))(r/2)f|| : 0 <= t <= tau} of order r are defined in the standard way, where TO(t) = I is the identity operator. In this paper, we establish the sharp Jackson inequality between E(f,Y,) and cor(f,'r) for some cases of r and r. More precisely, we will find the smallest constant X,,(,r, r) which depends only on n, r, and tau, such that the inequality E(f, L-n) <= k(n)(tau, r)w(r)(f, tau) is valid.