AADL based embedded software reliability model building and assessment method

Abstract

The invention discloses an AADL based embedded software reliability model building and assessment method. The method comprises the steps that 1, an AADL reliability model of an embedded software is built; 2, the reliability elements of the AADL reliability model are extracted, a reliability model ZAL is built using a Z language, the model comprises a fault model, a structure model and a behavior model; 3, the discrete time Markov chain DTMC is used to describe the ZAL model, and a PCTL formula of describing the embedded software reliability is provided, the PCTL formula is computed using the proposed reliability assessment algorithm ZARE to perform the calculations, instantaneous availability is obtained, and reliability assessment is completed. The reliability model building and the assessment method expresses the embedded software in layers, clearly and concisely and is easy to understand, and ZAL as a formalized model can use model checking and similar methods to conduct rigorous analyses and assessments for reliability.

Description

technical field [0001] The invention belongs to the field of trusted computing, software engineering, formalized modeling and verification, and specifically proposes an embedded software reliability modeling and evaluation method based on AADL. Background technique [0002] In the field of software reliability research, the study of software reliability model has always occupied an important position, and it is also the field with the most fruitful results. How to establish a software reliability model as early as possible, reduce the complexity of modeling, and enable the model to accurately describe the fault propagation relationship between components is an important content in software reliability research. [0003] Software modeling methods are mainly divided into semi-formal and formal methods. In software engineering, semi-formal specifications are generally used to describe software structure and behavior, which has the advantages of being easy to understand and use...

AI Technical Summary

Problems solved by technology

[0005] However, there are deficiencies in the existing implicit description: Generally, the existing methods assume that the semantics of the transformation are consistent, but the semantic description cannot be consistent and thus not precise enough; the existing model transformation technology is based on the existing semantics of AADL, and Some semantics in AADL are explained in natural language and examples, and the semantics given are not precise enough, which may lead to inaccurate and complete semantic conversion

On the other hand, the existing model conversion methods also have their own shortcomings: although Petri nets can well describe the dynamic characteristics of most asynchronous and concurrent systems, their models tend to become very large; the characteristics of fault trees are It can comprehensively and visually describe the causal relationship of faults, but it is difficult to add the probability of occurrence to the fault events for quantitative analysis

[0007] Since the occurrence and propagation of faults in the ZAL model are based on the characteristics of probability, the model checking method based on the ZAL model should consider probability factors, and there is no research related to probability model checking in the existing research on model checking methods for the Z language. achievement

Images