D2D通訊在LTE蜂巢式網路中之步進式位置推薦演算法

Abstract

在長程演進技術(Long Term Evolution, LTE)中有一項重要的技術，也就是設備對設備(Device-to-Device, D2D)的通訊，D2D通訊可以不透過基地台做設備與設備間的直接傳輸，這個做法可以減輕基地台的負擔以及增加整體系統的網路容量。 本論文提出一種基於LTE蜂巢式網路中D2D的步進式位置推薦法，希望能透過演算法規劃出一條推薦路徑，使D2D移動至推薦的路徑，並藉此能改善網路整體的系統容量及效能。 實驗模擬在一個含有傳統大型基地台及其使用者與多組D2D的環境來驗證本論文提出的步進式位置推薦演算法，並使用LTE-Sim開源工具來量測出模擬的結果，實驗結果顯示整體系統傳輸效能在多種不同的情況下可達明顯的提升效果。

Device-to-Device(D2D) communication is an important technique in Long Term Evolution (LTE) system, which can communicate with each other without connecting to the base station. Therefore, the loading of the base station could be reduced and the network capacity of the whole system could be increased. In this thesis, we purpose a step by step location recommendation algorithm for D2D communication underlying LTE cellular networks. With this algorithm, a recommendation path could be designed where D2D pairs move to. System capacity and efficiency of the whole network could be improved by this way. The experiments simulate an environment containing main base stations and several pairs of D2D to verify the step-by-step location recommendation algorithm proposed in this paper for D2D communications, and the LTE-Sim open source tool is utilized here to measure the simulation results. Experimental results show that the transmission performances of overall system can achieve significant improvement in various circumstances.