| J. Abrahams, "Hu#man-type codes for infinite source distributions," Journal of the Franklin Institute, vol. 331B, pp. 265--271, 1994. |
....the heap property) While its worst case time complexity is O(log n) for a heap of size n (cf. 12, 6] the average case is O(1) under a suitable uniform probability model (cf. 17, 4] and 2.1) As in Knuth [12] let us call the path formed by the numbers a[n 1] a[# . a[1] the special path. The algorithm analyzed in [17, 4] and Algorithm A in this correspondence) is a bottom up sequential search on the special path. Doberkat [4] derived the probability distribution that a random element falls into the intervals of the special path in a random heap under a ....
....coding tree for the corresponding source coding problem; cf. Figures 1 and 2. In this way, the result in this correspondence becomes an addition to the very few literature on parametric families of source distributions for which the Hu#man codeword lengths can be explicitly characterized. See [1, 5, 7, 8, 9, 10] for further information. In the next section, we state the probability model on which we are developing our arguments, describe the two algorithms mentioned above, and then transform the insertion problem into a coding problem. In Section 3, we prove the main result of this correspondence. ....
[Article contains additional citation context not shown here]
J. Abrahams, "Hu#man-type codes for infinite source distributions," Journal of the Franklin Institute, vol. 331B, pp. 265--271, 1994.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC