Related Experiment Videos
A Multi-Objective Partition Method for Marine Sensor Networks Based on Degree of Event Correlation
Dongmei Huang1, Chenyixuan Xu2, Danfeng Zhao3
1College of Information Technology, Shanghai Ocean University, Shanghai 201306, China. dmhuang@shou.edu.cn.
Sensors (Basel, Switzerland)
|September 22, 2017
Summary
This study introduces a graph-based approach to optimize marine sensor networks for faster disaster data retrieval. By partitioning sensor data in the cloud, it enhances real-time decision-making for marine events.
Area of Science:
- Marine Science
- Computer Science
- Data Management
Background:
- Current marine sensor networks are geographically fragmented, hindering real-time data access for cross-area events.
- Independent data storage in separate centers leads to retrieval delays, impacting critical decision-making.
Purpose of the Study:
- To develop a fast data retrieval service for marine sensor networks.
- To improve the efficiency of accessing data during marine events across multiple regions.
Main Methods:
- Abstracting the marine sensor network as a graph, with sensors as vertices and marine events as edges.
- Constructing a multi-objective balanced partition method for distributed cloud storage.
- Designing an incremental optimization strategy for dynamic network updates.
Main Results:
- The proposed method achieves optimal distributed storage layout for disaster data retrieval in the China Sea.
- Effectively optimizes data partitions when new sensors (buoys) are added.
- Demonstrates increased sensor correlation and decreased data retrieval costs.
Conclusions:
- The novel approach provides an efficient data access service for marine events.
- Enhances the performance of marine sensor networks for disaster monitoring and response.