基于随机时间自动机和统计模型检验技术的无线传感网络协议建模与分析

Abstract

Recently, sensor technology has seen significant achievement. Meanwhile, Wireless Sensor Network (WSN) technique has attracted lots of attention because of its open and dynamic behaviors, and has become a very important component of the Internet-based computing. The behavior of WSN system is very complex and may encounter lots of stochastic uncertainties and disturbances like message loss and node dynamics. Furthermore, WSN system is difficult to change and maintain once it is deployed. Thus, it is critical to ensure the quality of the low level protocols of WSN system in design phase. Designers should ensure the logical correctness of a protocol, as well as evaluate the performance of the protocol under certain target environments. To handle these issues, in this paper, we propose a framework to analyze and evaluate WSN protocols based on stochastic timed automata and statistical model checking. We propose to address the modeling of the uncertainties in the realistic behaviors by weighted stochastic transitions in the timed automata model of the protocol. Furthermore, we propose that the performance of the protocol under realistic environments should be evaluated by statistical model checking which is much cheaper and more scalable. To illustrate the feasibility and detail of the modeling and verification approach presented in this paper, two well-known WSN protocols, Timing-sync Protocol for Sensor Networks (TPSN) and Flooding Time Synchronization Protocol (FTSP), are studied thoroughly throughout the paper.