TY - GEN
T1 - Model-based validation of QoS properties of Biomedical Sensor networks
AU - Tschirner, Simon
AU - Liang, Xuedong
AU - Wang, Yi
PY - 2008
Y1 - 2008
N2 - A Biomedical Sensor Network (BSN) is a small-size sensor network for medical applications, that may contain tens of sensor nodes. In this paper, we present a formal model for BSNs using timed automata, where the sensor nodes communicate using the Chipcon CC2420 transceiver (developed by Texas Instruments) according to the IEEE 802.15.4 standard. Based on the model, we have used UPPAAL to validate and tune the temporal configuration parameters of a BSN in order to meet desired QoS requirements on network connectivity, packet delivery ratio and end-to-end delay. The network studied allows dynamic reconfigurations of the network topology due to the temporally switching of sensor nodes to power-down mode for energy-saving or their physical movements. Both the simulator and model-checker of UPPAAL are used to analyze the average-case and worst-case behaviors. To enhance the scalability of the tool, we have implemented a (new text-based) version of the UPPAAL simulator optimized for exploring symbolic traces of automata containing large data structures such as matrices. Our experiments show that even though the main feature of the tool is model checking, it is also a promising and competitive tool for efficient simulation and parameter tuning. The simulator scales well; it can easily handle up to 50 nodes in our experiments. The model checker installed on a notebook can also deal with networks with 5 up to 16 nodes within minutes depending on the properties checked; these are BSNs of reasonable size for medical applications. Finally, to study the accuracy of our model and analysis results, we compare simulation results by UPPAAL for two medical scenarios with traditional simulation techniques using OMNeT++, one of the most used simulation tools for wireless sensor networks. The comparison shows that our analysis results coincide with the simulation results by OMNeT++ in most cases although there are some differences caused the simplified wireless channel model in UPPAAL.
AB - A Biomedical Sensor Network (BSN) is a small-size sensor network for medical applications, that may contain tens of sensor nodes. In this paper, we present a formal model for BSNs using timed automata, where the sensor nodes communicate using the Chipcon CC2420 transceiver (developed by Texas Instruments) according to the IEEE 802.15.4 standard. Based on the model, we have used UPPAAL to validate and tune the temporal configuration parameters of a BSN in order to meet desired QoS requirements on network connectivity, packet delivery ratio and end-to-end delay. The network studied allows dynamic reconfigurations of the network topology due to the temporally switching of sensor nodes to power-down mode for energy-saving or their physical movements. Both the simulator and model-checker of UPPAAL are used to analyze the average-case and worst-case behaviors. To enhance the scalability of the tool, we have implemented a (new text-based) version of the UPPAAL simulator optimized for exploring symbolic traces of automata containing large data structures such as matrices. Our experiments show that even though the main feature of the tool is model checking, it is also a promising and competitive tool for efficient simulation and parameter tuning. The simulator scales well; it can easily handle up to 50 nodes in our experiments. The model checker installed on a notebook can also deal with networks with 5 up to 16 nodes within minutes depending on the properties checked; these are BSNs of reasonable size for medical applications. Finally, to study the accuracy of our model and analysis results, we compare simulation results by UPPAAL for two medical scenarios with traditional simulation techniques using OMNeT++, one of the most used simulation tools for wireless sensor networks. The comparison shows that our analysis results coincide with the simulation results by OMNeT++ in most cases although there are some differences caused the simplified wireless channel model in UPPAAL.
KW - Modelling and verification
KW - Wireless sensor networks
UR - https://www.scopus.com/pages/publications/70349240607
U2 - 10.1145/1450058.1450069
DO - 10.1145/1450058.1450069
M3 - 会议稿件
AN - SCOPUS:70349240607
SN - 9781605584683
T3 - Proceedings of the 8th ACM International Conference on Embedded Software, EMSOFT'08
SP - 69
EP - 78
BT - Proceedings of the 8th ACM International Conference on Embedded Software, EMSOFT'08
PB - Association for Computing Machinery (ACM)
T2 - 8th ACM International Conference on Embedded Software, EMSOFT 2008
Y2 - 19 October 2008 through 24 October 2008
ER -