記憶型多重存取萊斯衰減通道之衰減數

Abstract

在本篇論文中，我們分析記憶型萊斯衰減多重存取通道的總通道容量。在此通道中，衰減程序為高斯分佈並且有一個可目視的路徑成分，而且，有一個以上的使用者在同一時間裡傳送資料。為了簡化我們的分析，我們只考慮單傳送天線單接收天線的情況，也就是說，所有傳送端的使用者和接收端都僅使用單一天線。 在衰減通道容量的分析中，我們還不知道通道容量的精準表示式。我們使用一種稱作漸進分析的方法，在極限當可用的功率趨近無限大時得到通道容量。它顯示通道總容量在高訊號與雜訊比時會以雙指數成長達到無限大。而在高訊號與雜訊比的展開式中，第二項是一個叫做衰減數的常數。 在我們的研究中，我們找到一個單傳送天線單接收天線，m個使用者，一般記憶型萊斯衰減多重存取通道衰減數的上界。和其自然的下界－單使用者，單傳送天線單接收天線通道的衰減數結合後，我們得到精確的單傳送天線單接收天線，m個使用者，一般記憶型萊斯衰減多重存取通道衰減數。為了達到這個衰減數，我們必須停止比較差的使用者們傳送，並且讓最好的使用者們用分享時間的方式傳送。

In this thesis we analyze the sum-rate capacity of the Rician fading multiple-access channel (MAC) with memory. The fading process of the channel is Gaussian in addition to a line-of-sight component. Moreover, there are more than one user sending data at the same time. To simplify our analysis, we consider the single-input single-output (SISO) case, i.e., all the transmitters and the receiver use one antenna. In the analysis of the fading channel capacity, the exact expression of the capacity is not yet known. A way called asymptotic analysis is used to derive the channel capacity in the limit when the available power tends to infinity. It is shown that at high signal-to-noise ratio (SNR), the sum-rate capacity grows to infinity doublelogarithmically. The second term in the high-SNR expansion is a constant called fading number. In our work, we derive an upper bound on the fading number of the general m-user SISO Rician fading MAC with memory. Combining the natural lower bound on the fading number of the single-user SISO channel, we then obtain the exact fading number of the general m-user SISO Rician fading MAC with memory. To achieve the fading number, we have to switch off the worse users and allow the best users communicate by time-sharing.