一種基于神經(jīng)網(wǎng)絡(luò)分形模型的一維信號(hào)表示方法
USING ITERATED FUNCTION SYSTEM BASED ON NEURAL NETWORK TO MODEL TIME SEQUENCES
-
摘要: 本文提出了一種基于神經(jīng)網(wǎng)絡(luò)的分形模型,討論了映射收縮條件,并對(duì)湖底回波進(jìn)行了實(shí)驗(yàn)。實(shí)驗(yàn)結(jié)果表明,在此基礎(chǔ)上求解分形逆問(wèn)題,得到的吸引子能很好地逼近指定信號(hào)。
-
關(guān)鍵詞:
- 分形; 分形逆問(wèn)題; 神經(jīng)網(wǎng)絡(luò); 非線性
Abstract: A new method to resolve fractal inverse problem based on neural network was presented in this paper which can be employed to model a time sequences.The precondition to assure the model was also provided. A piece of echo from a lake was taken to test the algorithm. The result is satisfying. -
Barnsley M F, Demko S. Iterated function system and the global construction of fractals. Proc. Roy. Soc. London, 1985, A-399(1817): 243-275.[2] Mazel D S, Hayes M H. Using iterated function system to model discrete sequences. IEEE Trans. on SP, 1992, SP-40(7): 1724-1734.[2]趙耀,袁保宗. 一種基于新仿射變換的分形序列圖像編碼方法. 電子學(xué)報(bào),1997, 25(7): 28-31.[3]謝和平, 薛秀謙. 分形應(yīng)用中的數(shù)學(xué)基礎(chǔ)與方法. 北京: 科學(xué)出版社,1997, 27-40, 99-106, 170-175.[4]Hecht-Nielsen R. Theory of the backpropagation neural network. IJCNN, San Diego: 1989, 1, 593-605.[5]羅發(fā)龍, 李衍達(dá). 神經(jīng)網(wǎng)絡(luò)信號(hào)處理. 北京: 電子工業(yè)出版社, 1993, 7-7. -
計(jì)量
- 文章訪問(wèn)數(shù): 2321
- HTML全文瀏覽量: 185
- PDF下載量: 415
- 被引次數(shù): 0
下載:
