An Internet of Things gateway serves as a key intermediary between numerous smart things and their corresponding cloud networking servers. A typical conventional gateway system uses a high-level embedded microcontroller (MCU) as its core; that MCU performs low-level perception-layer device network management, upper-level cloud server functions, and remote mobile computation services. However, in edge computing, many factors need to be considered when designing an IoT gateway, such as minimizing the response time, the power consumption, and the bandwidth cost. Regarding system scalability, computational efficiency, and communication efficiency, solutions that use a single MCU cannot deliver IoT functionality such as big data collection, management, real-time communication, expandable peripherals, and various other services. Therefore, this article proposes an innovative multi-MCU system framework combining a field-programmable- gate-array-based hardware bridge and multiple scalable MCUs to realize an edge gateway of a smart sensor fieldbus network. Through distributed and collaborative computing, the multi-MCU edge gateway can efficiently perform fieldbus network management, embedded data collection, and networking communication, thereby considerably reducing the real-time power consumption and improving scalability compared to the existing industrial IoT solutions.

中文翻译：

---

使用多个协作微控制器的工业物联网边缘计算网关

物联网网关是众多智能事物及其相应的云网络服务器之间的关键中介。典型的常规网关系统以高级嵌入式微控制器（MCU）为核心。MCU执行低层感知层设备网络管理，高层云服务器功能和远程移动计算服务。但是，在边缘计算中，设计IoT网关时需要考虑许多因素，例如最小化响应时间，功耗和带宽成本。关于系统可扩展性，计算效率和通信效率，使用单个MCU的解决方案无法提供IoT功能，例如大数据收集，管理，实时通信，可扩展的外围设备以及各种其他服务。所以，本文提出了一种创新的多MCU系统框架，该框架结合了基于现场可编程门阵列的硬件桥和多个可扩展MCU，以实现智能传感器现场总线网络的边缘网关。通过分布式和协作计算，多MCU边缘网关可以有效地执行现场总线网络管理，嵌入式数据收集和网络通信，从而与现有的工业IoT解决方案相比，大大降低了实时功耗并提高了可扩展性。