Related Experiment Video
Updated: Aug 1, 2025
11:54
Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
Published on: March 13, 2017
9.3K
Sensor Management Method of Giving Priority to Confirmed Identified Targets
Chunshan Ding1,2, Chunguo Li1
1School of Information Science and Engineering, Southeast University, Nanjing 210096, China.
Sensors (Basel, Switzerland)
|April 28, 2023
Summary
This study introduces a new sensor management method for faster target identification. By prioritizing identifiable targets, it achieves comparable accuracy while significantly reducing confirmation time.
Area of Science:
- Artificial Intelligence
- Information Theory
- Sensor Systems
Background:
- Traditional sensor management for target identification relies on information theory metrics like information gain.
- These methods reduce overall uncertainty but neglect the speed of target confirmation.
Purpose of the Study:
- To develop a sensor management method that accelerates target identification by prioritizing targets nearing confirmation.
- To improve the accuracy of identification probability prediction in distributed Bayesian frameworks.
Main Methods:
- Proposed an improved identification probability prediction method with global result feedback to local classifiers.
- Developed a sensor management function using information entropy and expected confidence level to optimize identification uncertainty.
- Modeled sensor management as an allocation problem with an objective function prioritizing identifiable targets.
Main Results:
- The proposed method achieves correct identification rates comparable to existing information theory-based approaches.
- Experimental results demonstrate the shortest average time to confirm target identification.
Conclusions:
- The novel sensor management strategy effectively balances identification accuracy and speed.
- Prioritizing targets with higher confidence levels enhances identification efficiency.
