Hi friends.
In the evaluation section of some WSN routing protocols (e.g. CBR-mobile), the authors set the percentage of CHs as 5% of all nodes, and the transmission range as 19-20 meters (in a network with 100 nodes and 100*100m coor. isn't something wrong here ?
I think 5 cluster head with 20 meter for radio range is not enough ! so, what is wrong ?
please help.
thanks in advance
The above question should be analyzed with specific network scenarios. Cluster heads are selected by the centralized or distributed algorithms. The purpose is to cover the whole network nodes, thus reducing network management complexity. From this perspective, if a node is not covered by any cluster head, then it will automatically become a cluster head.
As said by Prof. Hang Shen, CH are selected based on algorithms such that it covers maximum node within its range. It depends upon the node positions in the network and the node which act as CH. Please refer to heterogeneous networks also. Also refer,
Xiaojiang Du, Moshen Guizani, Yang Xiao and Hsiao-Hwa Chen, “Two Tier Secure Routing Protocol for Heterogeneous Sensor Networks,” IEEE transactions on Wireless Communications, Vol. 6, No. 9, September 2007, pp. 3395-3401.
Depending on the transmission range, you need to choose the number of cluster heads. Otherwise, it may happen that a node cannot communicate with other nodes (cluster head). Therefore, there should have some justification among the number of nodes, cluster heads, and nodes' communication range.
thank to all. yeah dear Samaresh. thats right. but in leach (and other cluster based methods) the authors said that the best range for number of cluster heads is between 3 and 5 (with out considering network size) heinzelman. 2001 LEACH protocol and CBR mobile by samer awwad
- Badges
- Science topic