We give an O(log n) algorithm to compute the nth convergent of a periodic continued fraction. The algorithm is based on matrix representation of continued fractions, due to Milne-Thomson. This approach also allows for the computation of first n convergents of a general continued fraction in O(log n) time using O( n log n) processors. © 1991.
|Journal||Computers and Mathematics with Applications|
|Publication status||Published - 1 Jan 1991|