大規模M2M行動網路之群體位置管理

Abstract

Machine-to-Machine (M2M) 通訊是一種支援Internet of Things (IoT) 應用程式的新通訊架構。為了使機器(Machine)能夠傳送與接收資料，網路端運行位置管理(Location Management)追蹤機器的位置。然而數以百萬計的機器使得位置管理對Mobile Communication Network (MCN) 產生了大量的信令流量。在這樣大規模的M2M行動網路中，本論文研究3GPP Machine Type Communications (MTC) 所定義的機器移動的群體性。首先，我們定義何謂信令傳輸中的``correlated mobility'，當移動中的機器執行位置更新時。接著我們提出Group Location Management (GLM) 機制解決信令壅塞的問題。在GLM機制中，我們將移動路徑相似的機器們群聚在一起。最後，我們模擬成果顯示GLM機制可以降低機器的註冊信令並且增加佈署M2M通訊在大規模的M2M環境的可行性。

Machine-to-Machine (M2M) communications have emerged as a new communication paradigm to support Internet of Things (IoT) applications. To deliver data to and from machines, the network performs Location Management to track machine locations (i.e., Location Area Identity; LAI). Yet this process incurs large signaling traffic to the Mobile Communication Networks (MCN), which is composed of millions to trillions of machines. To address such large-scale M2M mobile networking, this paper investigates the grouping characteristics of machine movement identified in the 3GPP Machine Type Communications (MTC). We first define the ``correlated mobility' on the signaling transmissions of moving machines when performing location update. Based on the definition, we propose a Group Location Management (GLM) mechanism to mitigate the signaling congestion problem. In GLM, we group machines based on the similarity of their mobility patterns. Through our performance study, we show how the GLM mechanism can reduce registration signaling from machines and increase the feasibility to deploy M2M communications at a large scale M2M environment.