Implementing Clenshaw-Curtis quadrature, II computing the cosine transformation

ABSTRACT

In a companion paper to this, “I Methodology and Experiences,” the automatic Clenshaw-Curtis quadrature scheme was described and how each quadrature formula used in the scheme requires a cosine transformation of the integrand values was shown. The high cost of these cosine transformations has been a serious drawback in using Clenshaw-Curtis quadrature. Two other problems related to the cosine transformation have also been troublesome. First, the conventional computation of the cosine transformation by recurrence relation is numerically unstable, particularly at the low frequencies which have the largest effect upon the integral. Second, in case the automatic scheme should require refinement of the sampling, storage is required to save the integrand values after the cosine transformation is computed. This second part of the paper shows how the cosine transformation can be computed by a modification of the fast Fourier transform and all three problems overcome. The modification is also applicable in other circumstances requiring cosine or sine transformations, such as polynomial interpolation through the Chebyshev points.

REFERENCES

• 1: Glenn D. Bergland, Numerical Analysis: A fast fourier transform algorithm for real-valued series, Communications of the ACM, v.11 n.10, p.703-710, Oct. 1968 [doi>10.1145/364096.364118]
• 2: Clenshaw, C.W., and Curtis, A.R. A method for numerical integration on an automatic computer. Numer. Math. 2 (1960), 197-205.
• 3: Gentleman, W.M. An error analysis of Goertzel's (Watt's) method for computing Fourier coefficients. Comput. J. 12 (1969), 160-165.
• 4: Gentleman, W.M., and Sande, G. Fast Fourier transformsfor fun and profit. Proc. AFIPS 1966 FJCC, Vol. 29, Spartan Books, New York, pp. 563-578.
• 5: W. Morven Gentleman, Implementing Clenshaw-Curtis quadrature, I methodology and experience, Communications of the ACM, v.15 n.5, p.337-342, May 1972 [doi>10.1145/355602.361310]
• 6: W. Morven Gentleman, Algorithm 424: Clenshaw-Curtis quadrature [D-1], Communications of the ACM, v.15 n.5, p.353-355, May 1972 [doi>10.1145/355602.355603]
• 7: Havie, T. On a modification of the Clenshaw-Curtis quadrature formula. BIT 9 (1969), 338-350.
• 8: Kenneth Hillstrom, Certification of Algorithm 279: Chebyshev quadrature, Communications of the ACM, v.10 n.5, p.294, May 1967 [doi>10.1145/363282.363309]
• 9: F. R. A. Hopgood , C. Litherland, Algorithm 279: Chebyshev quadrature, Communications of the ACM, v.9 n.4, p.270, April 1966 [doi>10.1145/365278.365508]
• 10: Toyohisa Kaneko , Bede Liu, Accumulation of Round-Off Error in Fast Fourier Transforms, Journal of the ACM (JACM), v.17 n.4, p.637-654, Oct. 1970 [doi>10.1145/321607.321613]
• 11: O'Hara, H., and Smith, F.J. Error estimation in the Clenshaw- Curtis quadrature formula. Comput. J. 11 (1968), 213-219.

INDEX TERMS

Keywords:
Chebyshev series, Clenshaw-Curtis quadrature, cosine transformation, fast Fourier transformation