Razvoj algoritama za izračunavanje autokorelacije prekidačkih funkcija preko dijagrama odlučivanja

Abstract

The autocorrelation is a mathematical operation with important applications in computer science and engineering. The space and time complexity of algorithms for computing the autocorrelation is exponential in the number of variables in the switching function. Most of existing algorithms focus on obtaining the autocorrelation coeffcients of singleoutput switching function. However, in practical applications are usually required to work with multi-output switching functions. With this motivation, this doctoral thesis describes new algorithms for the effcient computation of the complete total autocorrelation for multiple-output switching functions with large number of inputs and outputs over various types of binary decision diagrams. Experimental results over benchmarks conrm the effciency of the proposed algorithms.