為了縮短響應時間與兼顧私有資料的保密性，霧計算或邊緣計算成為一個新興的研究課題，在後雲時代，傳統上以雲計算為核心的物聯網將難以滿足因有限頻寬的即時終端服務。本研究設計一個三MCU核心的物聯網閘道器，架構為一主從式的微控制器系統，其三個核心分別為通訊協定管理引擎，其運行方式是使用MQTT Client與雲端服務層進行橋接傳輸。裝置管理引擎，其運行方式是使用藍牙裝置與感知層進行橋接傳輸。資料庫管理引擎，負責資料排序管理功能，讓資料庫管理上具有高效率的資料查詢功能。藉此將物聯網終端裝置所需的資料通訊、資料管理、分析工作從雲端轉移至閘道器端，得以解決裝置通訊的即時性和隱私性問題。實驗結果中也顯示本論文的三MCU核心系統在傳輸效能與穩定度相對於單核心閘道器具有明顯的優勢。;In order to shorten response time without compromising the confidentiality of private data, fog calculation or marginal computing has become a new research topic, in the post-cloud era, the traditional network of objects with cloud computing being the core will face difficulties in meeting the standard of the real-time terminal services with limited bandwidth. This study designed a network of objects gateway with three MCU cores, structured as a master–slave microcontroller system. The three cores are the communication management engines, and uses a MQTT Client and cloud service layer for bridging transmission. The device management engine uses a blue tooth device to bridge the transmission with the sensing layer. The database management engine is responsible for the management of data sorting to allow database management to be equipped with efficient data searching ability. As a result, the required information communication and data management and analysis can be transferred from the cloud to the gateway end. In this way, the problem of immediacy and privacy of communication using the device can be solved. The experimental results have also shown that our three-MCU cores system has obvious advantages in the transmission efficiency and stability compared with the single core gate.
