TY - GEN
T1 - DyST
T2 - 24th International Conference on Web Engineering, ICWE 2024
AU - Korkan, Ege
AU - Ramirez, Silvia Oliva
AU - Steinhorst, Sebastian
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024
Y1 - 2024
N2 - The comprehension of a distributed system and its verification is one of the most challenging problems in today’s software engineering, commonly referred to as observability. The complexity increases when one cannot control all the components, like in IoT systems composed of third-party devices. The Web of Things standards by the W3C help with this by describing what one can do with an IoT device via network messages. However, no work has leveraged these standards to offer an observability solution that works with any set of IoT devices. This work addresses this gap by proposing a method to verify the correctness of the system by mining its specification from device interactions. Our approach can reverse engineer complex application logic in the form of UML Sequence Diagrams from the analysis of network messages of any protocol between the devices during system runtime, which can be used to programmatically assert the correctness of the mined specification. We have evaluated our approach with three case studies to assess our mining technique, the performance of our algorithms, and the applicability of our contributions to system verification in the IoT. Our results show that our approach can produce accurate Sequence Diagrams that help understand and verify the behavior of IoT systems.
AB - The comprehension of a distributed system and its verification is one of the most challenging problems in today’s software engineering, commonly referred to as observability. The complexity increases when one cannot control all the components, like in IoT systems composed of third-party devices. The Web of Things standards by the W3C help with this by describing what one can do with an IoT device via network messages. However, no work has leveraged these standards to offer an observability solution that works with any set of IoT devices. This work addresses this gap by proposing a method to verify the correctness of the system by mining its specification from device interactions. Our approach can reverse engineer complex application logic in the form of UML Sequence Diagrams from the analysis of network messages of any protocol between the devices during system runtime, which can be used to programmatically assert the correctness of the mined specification. We have evaluated our approach with three case studies to assess our mining technique, the performance of our algorithms, and the applicability of our contributions to system verification in the IoT. Our results show that our approach can produce accurate Sequence Diagrams that help understand and verify the behavior of IoT systems.
KW - Internet of Things
KW - Process Mining
KW - Web of Things
UR - http://www.scopus.com/inward/record.url?scp=85197763109&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-62362-2_11
DO - 10.1007/978-3-031-62362-2_11
M3 - Conference contribution
AN - SCOPUS:85197763109
SN - 9783031623615
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 147
EP - 162
BT - Web Engineering - 24th International Conference, ICWE 2024, Proceedings
A2 - Stefanidis, Kostas
A2 - Systä, Kari
A2 - Matera, Maristella
A2 - Heil, Sebastian
A2 - Kondylakis, Haridimos
A2 - Quintarelli, Elisa
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 17 June 2024 through 20 June 2024
ER -