Issue link: https://resources.mouser.com/i/1442808
25 Recent Advancements B efore the advent of modern 32-bit microcontrollers and miniature Wi-Fi connectivity chipsets, many sensors and actuators were not capable of directly connecting to the Internet and communicating with backend systems. However, in recent years, microcontrollers and networking chipsets have advanced to the point where they can instantly and securely connect to the Internet, obtain an Internet Protocol (IP) address, and maintain bidirectional communications with backend systems. Microcontrollers and networking chipsets generally have a lower bill of materials (BOM) cost than an IoT gateway. However, at the same time, edge computing principles may necessitate placing more data processing power close to the physical sensors and actuators. These and other factors represent engineering and cost trade-offs when deciding whether to include an IoT gateway in a solution design. Basic Architectural Patterns There are two basic architectural patterns to consider (note that in this section for brevity the term sensor is used to refer to both sensors and actuators): • Sensor-to-cloud: The sensor connects directly to an IP network through a wired or wireless network connection and can directly communicate with a backend system. • Sensor-to-gateway-to-cloud: The sensor connects to an IoT gateway through a wired or wireless connection, and the gateway connects to an IP network. Communications between the sensor and cloud go through the gateway, which acts as an intermediary. The cloud side in these patterns could be a public or private cloud computing environment, a set of backend web services within your enterprise, or a combination of the two. Functions of an IoT Gateway An IoT gateway can provide several different functions, and some of the most common include: • Sensor aggregation: Combining data from multiple sensors into one information packet that gets sent to the cloud • Local data processing/edge computing: Processing data near the sensors for reasons of reduced latency, sensor fusion, data reduction (e.g., filtering, averaging, thresholding, and compression), and being able to continue operating during transient network outages • Protocol conversion/interface bridging: Converting data formats or electronic interfaces from one form to another (e.g., converting RS-232 or Modbus data messages to JavaScript Object Notation (JSON) data objects or connecting Bluetooth sensors to a Long-Term Evolution (LTE) wireless network) • Authentication and security: Establishing secure communication channels with cloud backend systems, encrypting and decrypting data, managing digital certificates, maintaining audit trails, and providing firewall functions • Utility functions: Managing remote login access for diagnostics, system monitoring, network time synchronization, network routing, remote software updates, and Wi-Fi hotspot services