一種快速全局優(yōu)化的神經(jīng)網(wǎng)絡及其在數(shù)據(jù)融合中的應用
A FAST GLOBAL OPTIMIZATION NEURAL NETWORK AND ITS APPLICATION TO DATA FUSION
-
摘要: 本文將遺傳算法的全局性和EM算法的快速性相結合,提出了一種快速全局優(yōu)化神經(jīng)網(wǎng)絡,并將其應用于數(shù)據(jù)融合中。理論與實驗結果表明該算法在數(shù)據(jù)融合中具有很強的魯棒性。Abstract: This paper presents a fast global optimization neural network and applies it to the data fusion. This neural network is based on the global property of genetic algorithm and the high speed property of expectation maximization (EM) algorithm. The simulation results show that this neural network is robust in the data fusion.
-
Moon T K. The expectation maximization algorithm[J].IEEE Signal Processing Magazine.1996, 13(6):47-60[2]Jamshidian M, Jennrich R. Acceleration of the EM algorithm飾using quasi-Newton methods[J].J. R. Statist. Soc. B.1997, 59(3):569-587[3]陳國良, 等. 遺傳算法及其應用. 北京:人民郵電出版社,1996, 69-77.[4]Long M W. Airborne Early Warning System Concepts. Boston, London: Artech House, 1992, 293-321.[5]Sarma V V S. Multisensor data fusion and decision support for airborne target identification. IEEE Trans.on SMC., 1991, SMC-21(5): 969-978.[6]王功伯. 噴氣式飛機的紅外輻射特性. 紅外與激光技術,1989, 2(1): 26-29.[7]楊新星. 模糊神經(jīng)網(wǎng)絡數(shù)據(jù)融合算法的研究: [碩士論文]. 西安:西安電子科技大學,1997. -
計量
- 文章訪問數(shù): 2090
- HTML全文瀏覽量: 156
- PDF下載量: 650
- 被引次數(shù): 0
下載:
