Abstract
In this paper, we propose an energy-efficient data gathering scheme in wireless sensor networks utilizing spatial correlation of sensor readings to achieve energy saving. First, we utilize a mechanism to group sensor nodes into clusters by exploiting spatial correlation and the tolerable threshold so that the members in each cluster can be represented by their cluster head for data reporting. Next, to further reduce energy consumption, we merge the neighboring clusters with similar readings into fewer, larger clusters while providing user-acceptable reported data. The problem of cluster merging is modeled as an infer-graph set problem and a greedy heuristic algorithm is designed to solve it. Finally, we propose a concept of safe region to increase the probability of cluster members and the cluster head staying in the same cluster, thereby avoiding the cost of new cluster creation. Experimental results show that the proposed scheme outperforms the representative approach in terms of message size and number of cluster heads.
Original language | English |
---|---|
Pages (from-to) | 5007-5018 |
Number of pages | 12 |
Journal | Information (Japan) |
Volume | 16 |
Issue number | 7 B |
State | Published - Jul 2013 |
Keywords
- Data gathering
- Spatial correlation
- Wireless sensor networks