Accurate analog computer generation of Bessel functions for large ranges

Abstract

The generation of Bessel functions J_n(x) is required when simulating many systems on an analog computer. Because of an indeterminate expression at the origin, however, they have not been amenable to an accurate generation for the range 0 ≤ x ≤ X_max for large X_max.

This paper extends an idea of Van Remoortere to use an approximation for 0 ≤ x ≤ X₁ and solve a differential equation for X₁ ≤ x ≤ X_max, combining both phases by switching. The technique described here uses Chebyshev polynomials to minimize equipment in the approximation phase and generates the function 1/x by an integration process in the differential-equation phase to extend the range. The examples given for J ₀ and J ₉ indicate excellent accuracy for at least 0 ≤ x ≤ 100.

Get full access to this article

View all access options for this article.

References

A.S. Jackson Analog computation New York

McGraw-Hill

1960 p 152

A.E. Rogers T.W. Connolly Analog computation in engineering design New York

McGraw-Hill

1960 pp 135-136

S. Fifer Analogue computation vol 2 New York

McGraw-Hill

1961 pp 368-371

S.P. Bingulac E.A. Humo

Analog computer generation of Bessel functions of arbitrary order

IEEE Transactions on Electronic Computers vol EC-14 pp 886-889 December 1965

P. Van Remoortere

Etude analogique des equations differentielles lineaires a coefficients variables, singulieres a l'origine

Annales de 1'Association internationale pour le calcul analogique pp 124-128 April 1965

R.W. Hamming Numerical methods for scientists and engineers New York

McGraw-Hill

1962 pp 249-262

M Abramowitz I A Stegun editors Handbook of mathematical functions NBS Applied Mathematics Series 55 Washington DC National Bureau of Standards June 1964

A. Hausner Generating Bessel functions with an analog computer Tech Memo TM-66-14 Harry Diamond Labs Washington DC 20438 September 1966