降低在LTE網路中對於M2M服務所產生的MME負載

Abstract

由於通訊技術在這近20年來的突飛猛進，從家用的寬頻網路到手機無線網路的普及，使得無線網路逐漸趨於成熟，在Human-to-Human (H2H) 的應用更加多元化，同時也漸漸飽和，促使另一種新興的應用開始發展，也就是Machine-to-Machine (M2M)。M2M networks是由一群Machine-Type Communication (MTC) devices 相互連接而所形成的網路，彼此之間進行資料的交換、處理、維護或是做一些簡單的修復動作，其過程中不需要人為的介入操作也可以運行正常，而 3rd Generation Partnership Project (3GPP) LTE也針對這類的應用制定標準。 本篇論文針對標準中所提及的高負載議題來研究，在一段允許時間 (Grant Time Interval) 中，如何讓大量的MTC devices連接到3GPP網路時降低MME的負載，我們提出了三種方式來處理所連接上來的MTC devices，盡量地降低MME負載，分別是Heuristic-based、Binary-based和Fibonacci-based三種方法。最後在模擬結果中，我們分析了三種方式，實驗結果顯示，我們所提出的方法可以改善MME的overload問題。

In the past decades, communication technologies have been advanced from broadband networks to the popularity of mobile wireless networks. Wireless networks have become more and more mature. Because applications in Human-to-Human (H2H) are gradually saturated, prompting another emerging applications such as Machine-to-Machine (M2M) communications has begun to develop. M2M networks are networks formed by a group of Machine-Type Communication (MTC) devices which are interconnected with each other. The process does not require human intervention. The operation can also be operated normally. The 3rd Generation Partnership Project (3GPP) LTE is also developing standards for such applications. In this thesis, we investigate how to make many MTC devices connected to a 3GPP network in the Grant Time Interval (GTI). To reduce the MME load, we propose four techniques to handle the connection of MTC devices, namely Heuristic, Binary exponential backoff, Fibonacci backoff, and random backoff. Finally, we analyze the proposed four techniques by using simulation. The experimental results show that our proposed techniques can reduce the overload of the MME effectively.