Abstract
Constant linear velocity (CLV) and constant angular velocity (CAV) methods are frequently employed in the data reading of a CD-ROM device. By CLV method, data reconstruction is easier; instead, CAV method provides higher data rate and shorter access time. However, CLV method is utilized to record the datum on a compact disk for increasing the data density of a compact disk. Therefore, by CAV method for data reading, the linear velocity of the outer track is 2.4 times faster than that of the inner track. This thesis presents a novel scheme capable of efficiently reconstructing the datum from a compact disk in CD-ROM device in which CAV technology is used. Characteristic of the 3T ~ 11T digital datum is analyzed in detail and then reconstructed using the notion of real-time pattern recognition realized by competition learning networks (modified from Winner-Take-All algorithm). The design is also implemented by MATLAB simulation with the measured RF signals from a compact disk to demonstrate the effectiveness of the proposed algorithm and theoretical discussion.