Advances in Pervasive Computing and Networking

Gilbert Chen, Joel Branch, Michael Pflug, Lijuan Zhu and Boleslaw Szymanski
Department of Computer Science
Rensselaer Polytechnic Institute, Troy, NY 12180
{cheng3, brancj, pflugm, zhul4, szymansk}@cs.rpi.edu
| Abstract | A new network simulator, called SENSE, has been developed for simulating wireless sensor networks. The primary design goal is to address such factors as extensibility, reusability, and scalability, and to take into account the needs of different users. The recent progresses in component-based simulation, namely the component-port model and the simulation component classification, provided a sound theoretical foundation for the simulator. Practical issues, such as efficient memory usage, sensor network specific models, were also considered. Consequently, SENSE becomes an ease-of-use and efficient simulator for sensor network research. |
| Keywords: | Wireless sensor networks, network simulation, component-based simulation. |
The emergence of wireless sensor networks created many open issues in network design [1]. The three main traditional techniques for analyzing the performance of wired and wireless networks were analytical methods, computer simulation, and physical measurement. However, many constraints imposed on sensor networks, such as energy limitation, decentralized collaboration, and fault tolerance necessitate the use of complex algorithms for sensor networks that usually defy 'analytical methods. Furthermore, few sensor networks have come into existence, for there are still many unresolved research, design and implementation problems, so measurements are virtually impossible. It appears that simulation is currently the primary feasible approach to the quantitative analysis of sensor networks.
ns2 (http://www.isi.edu/nsnam/ns/), perhaps the most widely used network simulator for research, has been extended to include some basic facilities...