本論文針對異質網路之干擾通道下，考慮使系統功率最小化的問題。異質網路即包含高功率基地台與低功率節點的網路，在異質網路中主要問題是由於所有基地台使用相同的頻段，會產生細胞間的干擾。為了要解決細胞間的干擾，我們提出遞迴式位元承載演算法來使系統功率最小化，在此演算法中，使用者在讀取位元時會考慮到干擾通道，我們從訊號與干擾加噪音比公式中推導出一個和使用者訊號與干擾加噪音比、通道資訊、背景噪音和使用者的功率的關係式，使用者可以藉由關係式來計算增加位元的功率需求，並且以功率需求來創造一個功率矩陣來幫助使用者選擇最小功率需求的位元，而我們提出的遞迴是指以當前的干擾再分配一次位元到每一個使用者，而且定義重新分配的總功率為新功率總和，而模擬數據結果顯示我們提出的遞迴式位元承載演算法和現有的論文相比有較低的功率和複雜度。;In this thesis, the problem for minimizing the power in the heterogeneous networks is considered. Heterogeneous networks is the mix of Macro base stations (BS) and some low power nodes. We consider Pico BS as low-power nodes. The major problem in heterogeneous networks is the co-channel inter-cell interference because the spectrum is shared between all BSs. In order to resolve the interference problem, we proposed the recursive bit loading algorithm to minimize the total transmit power, which included Marco BS power and Pico BSs power. In this recursive algorithm, users are loaded bits with considering of interference channels. We derive the equations from an SINR expression, which gives a relationship with SINR of user, channel state information, background noise, and power of user. Users can use the relation to calculate the power cost for adding each bit. Based on the power cost of adding each bit, it can create a power matrix to help users to choose the bit with the minimum cost. The proposed recursion is defined as doing allocation for all users again with current interference and we set the new allocated bit’s power as new total power. Numerical results show the proposed algorithm achieves lower transmit power with a lower computational complexity.
