利用小波-類神經網路於近場光碟機的尋軌控制

Abstract

為了提升儲存容量和密度，先進光碟機的讀寫機制朝著近場光學的方向發展。近場光碟機的讀寫頭比現今習見的光碟機更接近硬碟機的飛行讀寫頭的架構。小波轉換為自1980年代開始發展的新興數學解析工具，相較於短時間傅利葉轉換，能在頻率-時間平面上將訊號局部的特徵表現出來。本研究以現行硬碟機的讀寫裝置模擬近場光碟機，結合小波分析的方法與類神經網路發展控制器，進行近場光碟機的尋軌控制，從中獲得小波轉換在控制應用上的初步成果。本研究並且分別利用線性電阻和光學尺作量測，驗證控制方法的成效。

The flying optical pickup head in near-field optical disk drives applies the near-field optics theory to overcome the limit of conventional optical disk techniques, thereby substantially increasing data storage capacity. The flying pickup head and disk constitute an air bearing similar to that in hard disk drives. Wavelet transform has been an emerging analysis tool since 1980’s. As a time-frequency transform, the wavelet transform is an alternative to the short-time Fourier transform. Based on both wavelet transform and neural network, this study aims to develop wavelet network controllers to achieve track-seeking for near-field optical disk drives. To validate the proposed method, this work conducts experiments, in which a resistance device and an optical grating respectively read the position feedback signal of the pickup head.