高速光學交換網路可以至少1Tb/s 的速度進行光訊息的交換，然而當任意二個光訊息同時通過一交換元件(switch element)時，會相互干擾(crosstalk)，這是光學網路一直以來會遇到的問題。為了建立更健全的交換網路系統，訊息干擾的問題勢必需要被解決。在此計劃中，我們將從二個截然不同的角度切入，對此干擾問題做更深入的研究。這二個向度分別是空間擴張法（space dilation approach）和波長擴張法（wavelength dilation approach）。就空間擴張的角度來說，此計劃預計改進已被提出俱備無干擾問題(crosstalk free)的共軛網路（conjugate network） [1]，藉由新的演算法的提出來減少共軛網路的長度，而網路長度的縮短可避免訊息傳遞時的衰減，假若此新網路被提出，將實際切合現實環境的需要。就波長擴張的角度而言，此計劃預計在Benes 網路[2]上提出新的演算法，使當任一訊息有二種波長選擇做為傳輸時，仍可保證此網路沒有干擾的問題。另外，對於log2N 網路[3]，此計劃將研究在二波長分配(2-wavelength assignment)下，此網路仍保證無干擾問題時，所需的複製網路 (plane networks)個數。 ; High-speed photonic switching networks can switch optical signals at the rate of several terabits per second. However, they suffer from an intrinsic crosstalk problem when two optical signals cross at the same switch element. The crosstalk problem must be overcome in building a robust switching system. In this project, we will do further research on crosstalk problems in two different approaches: space dilation approach and wavelength dilation approach. As for space dilation approach, a new topology based on the conjugate network [1] is expected to be proposed. The new design is not only crosstalk-free but also has shorter length than the general conjugate network. These two advantages exactly meet the demands of reality. As for wavelength dilation approach, an algorithm will be studied in a Benes network [2] in which each input has two wavelengths to choose. In addition, for log2N network [3], the total number of plane networks under 2-wavelength assignment will be studied as well. ; 研究期間 9808 ~ 9907
