Compressão de sequências de eventos com ordem parcial.

Abstract

In distributed systems (DS), there are independent and dependent events. In trace theory, events are represented as symbols of a finite alphabet. Consecutive symbols that represent two independent events may be interchanged, generating a equivalent sequence. The set of equivalent sequences of the same length is defined as interchange class. Interchange entropy is based on interchange class for define information theoretic limits for sequence compression considering partial-order. It is presented two new algorithms aiming compression of sequences of events with partial-order. The performance of algorithms is analyzed in terms of time complexity and compression rate. Furthermore, the algorithms are used to estimate interchange entropy, obtaining values smaller then Shannon entropy for the same finite alphabet.