A framework to support dependability evaluation of WSNs from AADL models

Abstract

Wireless Sensor Networks (WSNs) have been widely considered as a promising solution to support different types of applications on industrial environments. Many of these applications impose strict dependability requirements, since a system failure may result in economic losses, or damage for human life or to the environment. The absence of an effective approach enabling the dependability evaluation of WSNs prevents system designers to forecast these type of scenarios or to optimize decisions regarding the criticality of the devices, network topology, levels of redundancy and network robustness that minimize the occurrence of faults. To bridge the gap between research achievements and industrial development, we present in this paper a framework to support the dependability evaluation of WSNs based on the automated generation of analytical dependability models from high level AADL (Architecture Analysis and Description Language) architecture models. The main objective of this framework is to relieve the end user from a deep knowledge of dependability modeling techniques and evaluation methods, focusing on their knowledge of the behavior and structure of the system.

A Comprehensive Dependability Model for QoM-Aware Industrial WSN When Performing Visual Area Coverage in Occluded Scenarios

“…The flowchart of the proposed evaluation methodology is presented in Figure 5 and depicts the methodology in four stages: coverage analysis , connectivity analysis , Fault Tree generation and dependability analysis . Backwards, the dependability analysis (the last stage) is performed in terms of its quantitative attributes (reliability and availability), as it is commonly addressed in the literature [ 23 , 24 , 25 , 26 ]. For this task, the SHARPE (Symbolic Hierarchical Automated Reliability and Performance Evaluator) tool [ 27 ] is used to numerically compute a hierarchical model of a fault tree embedded with Markov chains.…”

Automated Methodology for Dependability Evaluation of Wireless Visual Sensor Networks

“…Since analytical modeling is a time-consuming task, demanding much effort for complex topologies, some authors propose automated model generation approaches [ 6 , 8 , 12 , 40 , 41 ]. Cinque et al.…”

“…Martins et al. [ 8 ] propose a toolset to support the evaluation of the dependability of WSNs in industrial environments, focusing on the automatic generation of analytical dependability models from AADL (Architecture Analysis and Design Language) architecture models. The proposed framework is integrated with the SHARPE tool and decides which modeling technique is the most appropriate for each case according to the system structure and dependability metrics.…”

“…Actually, dependability is a generic concept including attributes as availability, reliability, safety, integrity and maintainability [ 5 ], being commonly addressed by its quantitative attributes (reliability and availability) [ 6 , 7 , 8 , 9 ]. According to Avizienis et al.…”

“…Moreover, the use of simulation for dependability evaluation brings an important concern: for an accurate estimation of dependability measures, frequent observations of the system-failure event are necessary, which, by definition are rare events. This results in a substantial increase of the simulation time, which could lead to impractical values [ 8 , 10 , 11 , 12 ]. Therefore, considering this complex scenario for dependability evaluation, in this paper, we will focus on analytical modeling, since it is generally cheaper, faster and can provide an acceptable level of accuracy.…”