在蜂巢式物聯網群組傳輸下之延遲與能量效益分析

Abstract

本論文旨在研究機器對機器通訊系統下，借由重複使用上行資源的方法，使所有的裝置，能在最有效率的情形下完成傳輸，提升服務品質以及降低能量消耗。 未來物聯網通訊中，群組傳輸系統被廣泛用來管理龐大數量的裝置，我們利用了分群演算法，找出適合的協調者(coordinator)達到較佳的傳輸路徑，以降低能量消耗、減少裝置接入基地台的數目。此種結構式的分群會使得協調者集中在基站附近，協助其他裝置傳送資料。由於觀察到協調者都會聚集在基站周圍，我們提出了一個多區域同時傳輸的方法，重複使用上行的資源，使所有裝置能在更嚴苛的時間內完成傳輸，但在能量上會有部分的損失。此外我們做了能量與延遲的分析，找到了相對有效的資源分配方法，使改善平均延遲與總能量消耗比率達到最大，並使用多區域同時傳輸演算法，以降低總傳輸能量及延遲。

In this dissertation, we reduce the average delay for all devices transmission complete by reusing uplink resources. In this efficient way, we enhance the quality of service (QoS) and also reduce the energy consumption. In Machine to Machine (M2M) network, grouping method was proposed for managing massive devices. We use the grouping algorithm to find some appropriate coordinators to reduce energy consumption and overhead. In this hierarchical grouping method, coordi-nators are concentrated around the base station to help other devices to relay data. We propose a Multi-region concurrent transmission (MRCT) method which reuses the uplink resources to make all devices transmit complete in more stringent time, but the cost is energy consumption. Furthermore, we analyze the tradeoff between delay and energy consumption. We found relative efficient way, which delay improvement to energy de-grade ratio is the maximum. Then apply MRCT algorithm to achieve lower system delay.