*2.2. IoT Architectures*

Wireless Sensor Networks (WSNs) have extensively been adopted in agriculture [47] as well as in livestock farming [48] due to installation flexibility especially when wireless transmission introduces a significant reduction and simplification in wiring and harness [30,49]. In addition, greenhouse technology profits from this technology through automation and informatization. In [46] an intelligent system, controlling and monitoring greenhouse temperature has been described, aiming at reducing consumed energy while maintaining good conditions that improve productivity. A review of the common wireless nodes and sensors capturing environmental parameters related to crops in the agriculture domain is reported in [29]. Agriculture in the Internet era is quickly becoming a data-intensive industry. Farmers need to gather and evaluate a massive amount of information from meteorological and physical sensors to increase production efficiency [50]. In [51] a description of a modular IoT architecture for several applications including but not limited to healthcare, health monitoring, and PA is reported. All the proposed subsystem choices used in that research are cheap off-the-shelf components with open-source software libraries.

#### *2.3. Radio and Wireless Protocols in PA*

The goal of optimizing water use for crops leads also to the development of automated irrigation systems. In [52] wireless sensors are linked by ZigBee radio transceivers, implementing a WSN where soil water content and temperature data are transferred. The wireless information unit also features a GPRS module that connects to a web server via the public mobile network. An online graphical application through Internet access devices allows operators to remotely monitor the information data. The feasibility of the implemented automated irrigation system was demonstrated. However, the total cost was high for some applications. The cost of each wireless sensor unit was ∼USD 100, whereas the wireless information unit cost was ∼USD 1800.

In addition to ZigBee, the IoT world is pushing new technologies. The Long-Range (LoRa) technology, originally developed for IoT, is investigated in [10] to demonstrate its use for implementing Distributed Measurement Systems. The LoRa wireless technology is designed for sending small packets at a low data rate (0.3–5.5 kbps) at relatively long distances. The protocol can be used in IoT nodes where energy efficiency is considered the most critical parameter.

The LoRaWAN™ protocol exploits the unlicensed radio spectrum in the Industrial, Scientific, and Medical band. Operating frequencies (433 MHz, 868 MHz, or 915 MHz) depend on the particular geographical region. Formally, LoRaWAN™ is a member of the low power LPWAN family, i.e., WAN wireless communications that are designed to minimize the power consumption while covering large areas but offering a relatively small bit rate. The specification defines the device-to-infrastructure of LoRa physical layer parameters and the LoRaWAN™ protocol. The LoRa physical layer or PHY exploits a Chirp Spread Spectrum (CSS) modulation. Fundamental keywords are low power transmission, low throughput, and optimum coverage. The LoRaWAN™ network architecture is deployed in a star-of-stars topology in which gateways relay messages between end-devices and a central network server. The gateways are connected to the network server via standard IP connections and act as a transparent bridge, simply converting RF packets to IP packets and vice versa [53].

LoRa is having success, as confirmed by the high number of papers adopting it (see e.g., [54–58] and references therein). The LoRa alliance sponsors the integration of LoRaWAN™ into the IoT, and some open implementations of network servers are available helping the constant growth of the LoRaWAN™ ecosystem. A fairly complete analysis of the scalability of networks based on LoRaWAN™ is reported in [58]. Employing analytic and simulation-based approaches, the authors explore the dimensions of the LoRa network configuration. The chosen spreading factor, a parameter directly related to the bitrate of the LoRa message, significantly depends on the number of sensors deployed in the field and on the transmission rate, given in packets/day.
