College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China

---

Abstract—Currently, two main approaches to data dependence of EFSM(Extended Finite State Machine) haven’t refined intra-transition data dependence, instead they consider that every definition variable in a transition depends on all the use variables (including condition variables). For data dependence of a specific definition variable, not only the relevant use variables but also the irrelevant use variables (including condition variables) are considered, which obviously causes redundancy. Without a doubt, further analysis based on this brings hidden danger to the dependent analysis of the entire system and practical application. With the idea of introducing program dependence graph into to EFSM, this paper studies intratransition data dependence, and describes the data dependence between every intra-transition definition variable and the use and condition variables which influence or are influenced by it. Thus irrelevant dependence variables are removed to reduce redundancies and errors. Also, theoretical and experimental analyses are conducted.

Index Terms—Extended Finite State Machine(EFSM); Dependence Analysis; Intra-Transition Data Dependence(IaTDD)

[PDF]

Cite: Shenghui Shi, Qunxiong Zhu, Zhiqiang Geng, Wenxing Xu, "Intra-Transition Data Dependence," Journal of Software vol. 7, no. 12, pp. 2663-2670, 2012.