D2D通訊於LTE蜂巢式網路中使用粒子群最佳化演算法之位置推薦

Abstract

在此篇論文中，我們提出了D2D (Device-to-Device, D2D) 通訊於LTE蜂巢式網路中使用粒子群最佳化演算法之位置推薦。我們所提出的方法是讓D2D使用者移動到我們所推薦的位置以達到更好的系統容量以及資源利用率，利用我們設計的成本函數來決定我們的資源分配機制，避免與相鄰的大型蜂窩式網路基地台或其他的D2D用戶產生干擾並且平衡MUE (Marcocell User Equipment, MUE)用戶及D2D用戶的體驗品質(Quality of Experience, QoE)，希望兩者都可以達到最佳的體驗效果。 最後經由模擬結果證明，粒子群演算法不同的迭代結果與傳統暴力法的位置推薦方法比較之後，我們的方法可以有效的降低位置推薦所產生的推薦成本及時間。

In this paper, we present a Particle swarm optimization algorithm based location recommendation for Device-to-Device (D2D) Underlying Cellular Networks. The proposed scheme enables D2D users to transmit to the best position we recommended for better system capacity and resource reuse utilization ratio. Moreover, with our cost function to decide how to distribute resource, the interference between neighboring Cellular Networks base station and D2D users could be avoided. Also, it can balance Quality of Experience(QoE) between MUE (Marcocell User Equipment, MUE) users and D2D users, hoping for the best user experience on both side. As simulation results shows, PSO LR scheme with different iterations effectively reduce both the cost and time of location recommendation in comparison with brute-force LR scheme.