A tale of two graph models: a case study in wireless sensor networks

Archibald, B. , Kulcsár, G. and Sevegnani, M. (2021) A tale of two graph models: a case study in wireless sensor networks. Formal Aspects of Computing, 33(6), pp. 1249-1277. (doi: 10.1007/s00165-021-00558-z)

[img] Text
247006.pdf - Published Version
Available under License Creative Commons Attribution.


Publisher's URL: https://link.springer.com/article/10.1007/s00165-021-00558-z


Designing and reasoning about complex systems such as wireless sensor networks is hard due to highly dynamic environments: sensors are heterogeneous, battery-powered, and mobile. While formal modelling can provide rigorous mechanisms for design/reasoning, they are often viewed as difficult to use. Graph rewrite-based modelling techniques increase usability by providing an intuitive, flexible, and diagrammatic form of modelling in which graph-like structures express relationships between entities while rewriting mechanisms allow model evolution. Two major graph-based formalisms are Graph Transformation Systems (GTS) and Bigraphical Reactive Systems (BRS). While both use similar underlying structures, how they are employed in modelling is quite different. To gain a deeper understanding of GTS and BRS, and to guide future modelling, theory, and tool development, in this experience report we compare the practical modelling abilities and style of GTS and BRS when applied to topology control in WSNs. To show the value of the models, we describe how analysis may be performed in both formalisms. A comparison of the approaches shows that although the two formalisms are different, from both a theoretical and practical modelling standpoint, they are each successful in modelling topology control in WSNs. We found that GTS, while featuring a small set of entities and transformation rules, relied on entity attributes, rule application based on attribute/variable side-conditions, and imperative control flow units. BRS on the other hand, required a larger number of entities in order to both encode attributes directly in the model (via nesting) and provide tagging functionality that, when coupled with rule priorities, implements control flow. There remains promising research mapping techniques between the formalisms to further enable flexible and expressive modelling.

Item Type:Articles
Additional Information:This work is supported by the Engineering and Physical Sciences Research Council, under grants EP/N007565/1 (S4: Science of Sensor Systems Software); PETRAS SRF grant MAGIC (EP/S035362/1), and by the German Research Foundation (DFG) as part of project A1 within CRC 1053-MAKI.
Glasgow Author(s) Enlighten ID:Archibald, Dr Blair and Sevegnani, Dr Michele
Authors: Archibald, B., Kulcsár, G., and Sevegnani, M.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:Formal Aspects of Computing
ISSN (Online):1433-299X
Published Online:17 August 2021
Copyright Holders:Copyright © 2021 The Authors
First Published:First published in Formal Aspects of Computing 33(6): 1249-1277
Publisher Policy:Reproduced under a Creative Commons License

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
172422Science of Sensor System Software (SSSS)Muffy CalderEngineering and Physical Sciences Research Council (EPSRC)EP/N007565/1Computing Science