BP Neural Network Algorithm for Multi-sensor Trajectory Separation Based on Maximum Membership Degree
Aiming at the problem of trajectory separation which belongs to the data fusion technology, the multi-sensor trajectory separation algorithm of BP neural network based on the maximum membership degree is presented in this paper. The trajectory points can be predicted by establishing the trajectory prediction model which is based on the BP neural network, and the new radar data can be judged whether they belong to the prescriptive trajectory; Based on the prediction of the BP neural network, the multi-trajectory separation algorithm of fuzzy clustering of maximum membership degree is added to improve the effectiveness and accuracy of the trajectory separation. The experimental tests show that the algorithm which is presented in this paper effectively improves the efficiency and accuracy of the trajectory separation, and has a better application value.
Keywords: BP neural network, maximum membership degree, fuzzy clustering, mean square error.
KeywordsMembership Function Fuzzy Cluster Membership Degree Fusion Center Wavelet Neural Network
Unable to display preview. Download preview PDF.
- 2.Wang, Q., Nie, W.: Distributed Multi-sensor data fusion. Sensor Technology 16(5), 8–10 (1997)Google Scholar
- 3.Haykin, S., McMaster: Neural Networks and Learning Machines. China Machine Press, Beijing (2009)Google Scholar
- 5.Wang, Z., Li, H.: The fracture parameter prediction based on BP neural network algorithm. Chinese Journal Of Applied Mechanics 26(2), 379–382 (2009)Google Scholar
- 6.Yu, Q., Zhao, H.: Nenural network prediction model based on differential evolution algorithm and its application. Computer Engineering add Applications 44(14), 246–248 (2008)Google Scholar
- 8.Liang, B., Cao, D.: Fuzzy Mathematics and Its Applications. Establishing of Science Press, Beijing (2007)Google Scholar
- 9.Fang, C., Hang, X., Wang, X.: Target Threat Evaluating and Sequencing Method Based on the Maximum Degree of Membership. Systems Engineering and Electronics 25(1), 50–52 (2003)Google Scholar
- 11.Yu, Q., Zhao, H.: Track Fusion with NN Compensated in a Multi-sensor Environment. Journal of Shanghai Jiaotong University 44(14), 246–248 (2008)Google Scholar
- 15.Pei, L.: Fuzzy comprehensive evaluation on radar track fusion effectiveness. Electronics Optics and Control 13(6), 18–20 (2006)Google Scholar