摘要
The problem of distributed detection fusion using multiple sensors for remote underwater target detection is studied. Considering that multiple access channel （MAC） schemes are able to offer high efficiency in bandwidth usage and consume less energy than the parallel access channel （PAC）, the MAC scheme is introduced into the underwater target detection field. The model of underwater distributed detection fusion based on MAC schemes is established. A new method for detection fusion of MAC based on deflection coefficient maximization （DCM） and Neyman-Pearson （NP） rule is proposed. Under the power constraint of local sensors, this paper uses the DCM theory to derive the optimal weight coefficients and offsets. The closed-form expressions of detection probability and false alarm probability for fusion systems are obtained. The optimal detection performance of fusion systems is analyzed and deeply researched. Both the theory analysis and simulation experiments indicate that the proposed method could improve the detection performance and decrease the error probability effectively under power constraints of local sensors and low signal to noise ratio.
The problem of distributed detection fusion using multiple sensors for remote underwater target detection is studied. Considering that multiple access channel （MAC） schemes are able to offer high efficiency in bandwidth usage and consume less energy than the parallel access channel （PAC）, the MAC scheme is introduced into the underwater target detection field. The model of underwater distributed detection fusion based on MAC schemes is established. A new method for detection fusion of MAC based on deflection coefficient maximization （DCM） and Neyman-Pearson （NP） rule is proposed. Under the power constraint of local sensors, this paper uses the DCM theory to derive the optimal weight coefficients and offsets. The closed-form expressions of detection probability and false alarm probability for fusion systems are obtained. The optimal detection performance of fusion systems is analyzed and deeply researched. Both the theory analysis and simulation experiments indicate that the proposed method could improve the detection performance and decrease the error probability effectively under power constraints of local sensors and low signal to noise ratio.
基金
supported by the National Natural Science Foundation of China (60972152)
Northwestern Polytechnical University Foun dations for Fundamental Research (JC201027
JC20100223)
作者简介
Corresponding author:Zhenhua Xu was born in 1985.He is now a Ph.D.candidate student in Northwestern Polytechnical University in information and communication engineering.His research interests include signal processing,distributed detection and estimation,data fusion.E-mail:xzh985@163.com;Jianguo Huang was born in 1945.He is a professor in College of Marine Engineering,Northwestern Polytechnical University.His research areas include array signal processing,underwater communications,and so on.E-mail:jghuang@nwpu.edu.cn;Qunfei Zhang was born in 1968.He is a professor in College of Marine Engineering,Northwestern Polytechnical University.His current research interests include underwater acoustic signal processing,array signal processing and computer simulation.E-mail:zhangqf@nwpu.edu.cn.
