Algorithms for Piecewise Polynomials and Splines with Free Knots
نویسندگان
چکیده
We describe an algorithm for computing points o = io < ii < • • • < xjt < Zfc+i = b which solve certain nonlinear systems d(z,_i,xt) = d(i,,x,+i), i = 1,... ,fc. In contrast to Newton-type methods, the algorithm converges when starting with arbitrary points. The method is applied to compute best piecewise polynomial approximations with free knots. The advantage is that in the starting phase only simple expressions have to be evaluated instead of computing best polynomial approximations. We finally discuss the relation to the computation of good spline approximations with free knots. 0. Introduction. Let [a, b] be an interval of the real line and k be a natural number. Moreover, let D = {(x,y) e R2: a < x < y < b} and d: D —> R be a function with the following properties: (0.1) d is continuous, (0.2) d(x, x) = 0 for all (x, x) e D, (0.3) d(x,y) < d(x,y) if [x,y] C [x,y] c [a,b]. A partition a = xo < xi < ■ ■ ■ < Xk < Xk+i = b is called a leveled set if (0.4) d(xi-i,Xi) = d(xi,xi+i), i=l,...,k, and it is called an optimal set if (0.5) max d(zt,xt+1) < max d(yi,yi+i) 0<i<k 0<i<fc for all knots a = yo < yi < ■ ■ ■ < yk < 2/fc+i = b. It is easy to verify that every leveled set is optimal. Using the idea of an algorithm for segment approximation in [10], in Section 1 we give an algorithm to compute a sequence of knot sets converging to a leveled (and therefore to an optimal) set. Simultaneously, a sequence converging to the optimal value m*; = min max d(yi,yi+i) {yi,-,yk)o<i<k is determined. In contrast to Newton-type methods, the algorithm converges for arbitrarily chosen (e.g., equidistant) knots. In Section 2 we apply the above algorithm to best uniform approximation of a given function / e C[a, b] by piecewise polynomials with free knots. This approximation problem will be solved in two phases. Received November 20, 1987. 1980 Mathematics Subject Classification (1985 Revision). Primary 41A15, 41A10, 65D07; Secondary 41A50, 65D15, 65H10. ©1989 American Mathematical Society 0025-5718/89 $1.00 + $.25 per page
منابع مشابه
On monotone and convex approximation by splines with free knots
We prove that the degree of shape preserving free knot spline approximation in L p a; b], 0 < p 1 is essentially the same as that of the non-constrained case. This is in sharp contrast to the well known phenomenon we have in shape preserving approximation by splines with equidistant knots and by polynomials. The results obtained are valid both for piecewise polynomials and for smooth splines wi...
متن کاملOn Monotone and Convex Spline Interpolation*
This paper is concerned with the problem of existence of monotone and/or convex splines, having degree n and order of continuity k, which interpolate to a set of data at the knots. The interpolating splines are obtained by using Bernstein polynomials of suitable continuous piecewise linear functions; they satisfy the inequality k < n — k. The theorems presented here are useful in developing alg...
متن کاملDiscrete GB-Splines and Their Properties
Discrete generalized splines are continuous piecewise defined functions which meet some smoothness conditions for the first and second divided differences at the knots. Direct algorithms and recurrence relations are proposed for constructing discrete generalized B-splines (discrete GB-splines for short). Properties of discrete GB-splines and their series are studied. It is shown that discrete G...
متن کاملFinite Element Methods with Hierarchical WEB-splines
Piecewise polynomial approximations are fundamental to geometric modeling, computer graphics, and finite element methods. The classical finite element method uses low order piecewise polynomials defined on polygonal domains. The domains are discretized into simple polygons called the mesh. These polygons might be triangles, quadrilaterals, etc., for two-dimensional domains, and tetrahedra, hexa...
متن کاملDetection of Singularities Using Segment Approximation
We discuss best segment approximation (with free knots) by polynomials to piecewise analytic functions on a real interval. It is shown that, if the degree of the polynomials tends to infinity and the number of knots is the same as the number of singularities of the function, then the optimal knots converge geometrically fast to the singularities. When the degree is held fixed and the number of ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010