Optimising Data Transmission in Heterogeneous Sensor Networks

The transfer rate of messages in distributed sensor network applications is a critical factor in a system's performance. The Sensor Abstraction Layer (SAL) is one such system. SAL is a middleware integration platform for abstracting sensor specific technology in order to integrate heterogeneous types of sensors in a network. SAL uses Java Remote Method Invocation (RMI) as its connection method, which has unsatisfying transfer rates, especially for streaming data.  This paper analyses different connection methods to optimize data transmission in SAL by replacing RMI.  Our results show that the most promising Java-based connections were frameworks for Java New Input/Output (NIO) including Apache MINA, JBoss Netty, and xSocket. A test environment was implemented to evaluate each respective framework based on transfer rate, resource usage, and scalability. Test results showed the most suitable connection method to improve data transmission in SAL JBoss Netty as it provides a performance enhancement of 68%.





References:
[1] M. Stonebraker, U. Çetintemel, and S. Zdonik, "The 8 requirements of
real-time stream processing," ACM SIGMOD Rec., vol. 34, pp. 42-47,
2005.
[2] G. Gigan and I. Atkinson, "Sensor Abstraction Layer: a unique software
interface to effectively manage sensor networks," in Third International
Conference Intelligent Sensors, Sensor Networks and Information
Processing (ISSNIP 07), Melbourne, Australia, 2007.
[3] C. Huddlestone-Holmes, G. Gigan, and I. Atkinson, "Infrastructure for a
sensor network on Davies Reef, Great Barrier Reef," in 3rd International
Conference Intelligent Sensors, Sensor Networks and Information
Processing (ISSNIP 07) Melbourne, Australia: IEEE Xplore, 2007.
[4] J. Trevathan, H. Ghodosi, and T. Myers, "Efficient Batch Authentication
for Hierarchical Wireless Sensor Networks," in Proceedings of the
Seventh International Conference on Intelligent Sensors, Sensor
Networks and Information Processing (ISSNIP 2011), Adelaide,
Australia, 2011, pp. 217-222.
[5] J. Trevathan, I. Atkinson, W. Read, N. Bajema, Y. J. Lee, R. Johnstone,
and A. Scarr, "Developing Low-Cost Intelligent Wireless Sensor
Networks for Aquatic Environments," in Proceedings of the 6th
International Conference on Intelligent Sensors, Sensor Networks and
Information Processing (ISSNIP'10) Brisbane, Australia, 2010, pp. 13-
18.
[6] S. P. Ahuja and R. Quintao, "Performance evaluation of Java RMI: a
distributed object architecture for Internet based applications," in
Modeling, Analysis and Simulation of Computer and
Telecommunication Systems, 2000. Proceedings. 8th International
Symposium on, 2000, pp. 565-569.
[7] W. R. Cook and J. Barfield, "Web Services versus Distributed Objects:
A Case Study of Performance and Interface Design," in Proceedings of
the IEEE International Conference on Web Services: IEEE Computer
Society, 2006.
[8] C. Demarey, G. Harbonnier, R. Rouvoy, and P. Merle, "Benchmarking
the round-trip latency of various java-based middleware platforms,"
Studia Informatica Universalis Regular Issue, vol. 4, pp. 7-24, 2005.
[9] M. Juric, I. Rozman, B. Brumen, M. Colnaric, and M. Hericko,
"Comparison of performance of Web services, WS-Security, RMI, and
RMI SSL," Journal of Systems and Software, vol. 79, pp. 689-700,
2006.
[10] G. Aloisio, D. Conte, C. Elefante, G. P. Marra, G. Mastrantonio, and G.
Quarta, "Globus Monitoring and Discovery Service and SensorML for
Grid Sensor Networks," in Proceedings of the 15th IEEE International
Workshops on Enabling Technologies: Infrastructure for Collaborative
Enterprises, Manchester, United Kingdom 2006, pp. 201-206.
[11] S. Heinzl and M. Mathes, Middleware in Java: Leitfaden Zum Entwurf
Verteilter Anwendungen: Vieweg+ teubner Verlag, 2005.
[12] T. Lee. (2011) Rapid Network Application Development with Apache
MINA(Online).Available: http://developers.sun.com/learning/j
avaoneonlin e/2008/pdf/TS-4814.pdf