具節能之細胞叢集法及群播排程管理於LTE-A MBSFN 網路

Abstract

3GPP LTE-A (Long Term Evolution-Advanced，長期演進技術) 是目前最受推崇的4G 無線通訊標準之一，其支援高達1Gbps 的傳輸頻寬，及多樣且豐富的媒體串流服務，尤其像Live IPTV、Video over IP 等高頻寬的影音串流服務。然而，隨著串流服務加載的便利，在同一個服務範圍內，不同的行動用戶大量點播相同且高頻寬的影音串流的可能性越來越高，造成頻寬與無線資源的急速耗盡。因此，LTE-A 標準制訂了Multicast Broadcast Single Frequency Network (MBSFN)機制，允許相鄰的細胞叢集成大型服務區域(MBSFN Area)，再透過多細胞協同傳輸方式(coordinated multi-point, CoMP)同步傳輸訊號以降低相鄰細胞間的干擾，增進資源利用率。因此，降低行動用戶接取資料的時間，進而達到節能省電的效果。然而，如何妥善叢集化細胞成為大型服務區域，至今仍是一個開放的問題。有鑑於此，本篇論文主要探討如何有效的叢集細胞以及服務排程問題，並且在確保滿足Quality of Service (QoS)的前提下，最小化UE 接收服務的時 間，即最小化UE 的耗電。我們提出一個高效且低複雜度的方法。該方法共包含兩個階段：第一階段中，我們依序挑選具訊號疊加效果並且不耗費大量無線資源之細胞作叢集，並且確保服務的QoS，如此可降低每個服務所需的無線資源，並且降低UE 的接收時間；在第二階段中，我們利用dynamic programming 最佳化MBSFN Area 中服務的排程序列，如此可最小化用戶資料的接收時間，達成節能省電的效果。根據模擬結果，我們的方法具有優異的省電效益，同時能有效的提高無線資源利用率。

The 3GPP Long Term Evolution-Advanced (LTE-A) is perhaps the most promising technology for next-generation wireless communications. It provides transmission rates up to 1 Gbps and is thus able to support a variety of broadband multimedia services, such as IPTV and Voice/Video-over-IP applications. With the prevalence of multimedia services, the probability that multiple User Equipments (UEs) under the same area subscribe to the same multimedia service may become much higher, thus wasting excessive bandwidths and radio resources bandwidth. Therefore, LTE-A develops the Multicast Broadcast Single Frequency Network Mechanism (MBSFN), which allows multiple cells to be clustered into a big serving area and leverages Coordinated Multi-point Transmission (CoMP) to transmit multicast data to multiple UEs synchronously to reduce inter-cell interference as well as improve system utilization. In this way, the data reception time of UEs can be reduced and thus the receiving power of UEs can also be saved. However, how to cluster cells and schedule multicasts is still an open issue. In this paper, we address the cell clustering and multicast scheduling problem to minimize UEs' energy consumption in terms of data reception time while guaranteeing the Quality of Service (QoS) of multicast services. We propose an efficient scheme, which consists of two phases. The first phase tries to find effective cells as serving areas with less radio resource while satisfying the services' QoS. This potentially reduces UEs' data reception time because UEs can receive the same data with less radio resource (and less access time). The second phase further tries to reduce UEs' energy via dynamic programming to well schedule the multicast services in each serving area. Extensive simulation results show that our scheme can significantly increase the system utilization while saving UEs' energy.