首页> 外文会议>International conference on formal engineering methods >Fast Translation from LTL to Biichi Automata via Non-transition-based Automata
【24h】

Fast Translation from LTL to Biichi Automata via Non-transition-based Automata

机译:通过基于非转换的自动机从LTL到Biichi自动机的快速翻译

获取原文

摘要

In model checking, properties are typically defined in linear temporal logic (LTL) and are translated into non-deterministic Biichi automata (NBA). In this paper, we propose a new, efficient translation method that is different from those used in LTL2BA, Spot and LTL3BA. Our method produces non-transition-based generalised Biichi automata (GBA) as an intermediate object, whereas LTL2BA, Spot, and LTL3BA use transition-based generalised Biichi automata (TGBA). Our method enables fast conversion because the data structure representing the object is simpler than that used in conversions via TGBA. Furthermore, we have developed techniques to reduce the number of states, similar to techniques that have heretofore only been available for conversions via TGBA. We also propose a technique to suppress the increase in the number of states that normally occurs while GBA is converted into NBA, using characteristics of strongly connected components of the GBA. We implemented our method with these techniques and experimentally compared our method with LTL2BA, Spot, and LTL3BA, which are the fastest translators to date. Our conversion method was much faster than LTL2BA and Spot, and was competitive with LTL3BA. In addition, the number of states in the NBA resulting from our method was comparable to that produced by LTL2BA, Spot, and LTL3BA.
机译:在模型检查中,属性通常在线性时间逻辑(LTL)中定义,并被转换为非确定性Biichi自动机(NBA)。在本文中,我们提出了一种与LTL2BA,SPOT和LTL3BA中使用的新的高效翻译方法。我们的方法产生基于非转化的广义Biichi自动机(GBA)作为中间物体,而LTL2BA,斑点和LTL3BA使用基于转变的广义Biichi自动机(TGBA)。我们的方法可以快速转换,因为表示对象的数据结构比通过TGBA的转换中使用的数据结构更简单。此外,我们开发了减少州数量的技术,类似于通过TGBA仅可用于转换的技术。我们还提出了一种技术来抑制通常发生的状态的状态增加,而GBA转换为NBA,则使用GBA的强连接组件的特性。我们利用这些技术实施了我们的方法,并通过实验与LTL2BA,SPOT和LTL3BA进行了实验,这些方法是迄今为止最快的翻译人员。我们的转换方法比LTL2BA和斑点更快,并且与LTL3BA竞争。此外,由我们的方法产生的NBA中的状态的数量与LTL2BA,斑点和LTL3BA产生的态度相当。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号