Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 何彩莉 | en_US |
dc.contributor.author | Ho, Tsai-Li | en_US |
dc.contributor.author | 鄧清政 | en_US |
dc.contributor.author | Ching-Cheng Teng | en_US |
dc.date.accessioned | 2014-12-12T02:17:07Z | - |
dc.date.available | 2014-12-12T02:17:07Z | - |
dc.date.issued | 1996 | en_US |
dc.identifier.uri | http://140.113.39.130/cdrfb3/record/nctu/#NT850327011 | en_US |
dc.identifier.uri | http://hdl.handle.net/11536/61663 | - |
dc.description.abstract | 本論文,我們研究模糊類神經網路在二維陣列中,估計訊號源位置的應用 。利用模糊類神經網路可以近似任意連續函數的能力,能夠迅速而準確地 追縱高速移動的目標物。另一方面,與以前必須利用規則形狀的陣列估計 水平角、仰角的方法比較起來,我們用少□う熒P測器,也不需限制陣列 的幾何形狀就可以達到判斷訊號源的目地。同時,我們將展示在訊號彼此 相互關聯的環境下,利用模糊類神經網路處理低角度的追縱問題,比起空 間平化的特徵結構法有更好的表現。 In this thesis, we discuss two-dimensional direction estimation by using a fuzzy neural network(FNN). We can track high speed moving target with highresolutions by FNN for its mapping capacity. Furthermore, the array geometry can be quite free, and fewer sensors are necessary comparing to classicazimuth/ elevation direction finding using uniformly regular array. Also, wedemostrate, in a coherent enviornment, how FNN can outperform conventional eigenstructure-based methods with spatial smoothing scheme in dealing with low angle tracking problem. | zh_TW |
dc.language.iso | zh_TW | en_US |
dc.subject | 模糊類神經網路 | zh_TW |
dc.subject | 陣列訊號處理 | zh_TW |
dc.subject | 二維 | zh_TW |
dc.subject | Fuzzy Neural Network | en_US |
dc.subject | Array Signal Processing | en_US |
dc.subject | two-dimensional | en_US |
dc.title | 模糊類神經網路在二維陣列訊號處理上的應用 | zh_TW |
dc.title | Application of Fuzzy Neural Network on 2-D Array Signal Processing | en_US |
dc.type | Thesis | en_US |
dc.contributor.department | 電控工程研究所 | zh_TW |
Appears in Collections: | Thesis |