원문정보
초록
영어
Application of sensor network which is working on the critical condition such as fire and radiation leak must be fast, reliable and fault tolerant. If sensors cannot satisfy those requirements, they rapidly lose their own functionality and bring untold losses ultimately. To satisfy these requirements and prolong the lifetime of the sensor network, an efficient routing protocol based on clustering approach called HPEQ was proposed. However, the protocol takes somewhat much time for electing clustering head due to notification message for next cluster head. It means that at least the time consumption of cluster head election is required for fault recovery. So we propose our scheme for reducing the clustering head election time by using candidate and existing message instead of resending the messages. Our proposed scheme has 2 key ideas. One is management fields added to event messages and the other is delayed propagation of management messages. Detailed description of ideas is given in the section 3. The improvement is showed by the NS2 simulation in section 4.
목차
1. Introduction
2. Related work
3. Cluster Head Candidate
3.1 Key Ideas
3.2 Candidate Selection
3.3 Cluster Head Advertisement
3.4 A fault in the cluster head
3.5 A fault in the candidate head
4. Performance Analysis
4.1 Effect of extended message
4.2 Numerical analysis
4.3 Simulation experiments
5. Conclusion
References
