**2. Common architecture designs for IoT irrigation systems in agriculture**

This section will provide an overview of the most popular architectures for these systems. In IoT irrigation management solutions, multi-agent architectures are widespread [43, 44]. These structures provide a distinction between the numerous components that make up its structure. The difference is typically made depending on the architectural strata in which the elements are housed. For example, nodes higher in the hierarchy may act as a broker for nodes lower in the hierarchy [43]. The most of of designs are divided into layers or functional blocks that represent the main tasks that must be done [45]. These blocks or layers are considered generic and are found in the majority of IoT irrigation management system architectural designs. The essential components of these architectures are devices, connectivity, services, administration, applications, and security. IoT systems are made up of devices that are put in a specific location and may perform activities including detection, monitoring, control, and action. In order to convey the essential data, the devices must have interfaces that allow them to communicate with other devices. The information gathered by various sensors will be treated as a whole, and the results will be applied to various actuators. The data collected and the actions taken must then be sent between the devices. The use of communication protocols is required for this task. In the majority of circumstances, different communication protocols are used on the same IoT system in order for it to work together. Services may be required to complete tasks such as device discovery, device control, and data analysis. The user can interact with the system using the programmes. The user will be able to see data acquired through monitoring as well as data extracted once it has been processed utilizing the applications. On numerous occasions, the user can execute actions that he considers relevant to the scenario presented by the data, and the actions can also be performed automatically.

#### *An IoT-based Immersive Approach to Sustainable Farming DOI: http://dx.doi.org/10.5772/intechopen.105449*

Finally, the security of the system may be considered. The three layers of IoT architecture have typically been believed to be perception, network, and application. After several research, an intermediary layer was built between the network and application levels. In cloud and fog computing environments, this layer, also known as the service layer, is used to store and process data. For the past few years, authors such as Ferrández-Pastor [46] have proposed a new architecture based on four layers: objects, edge, communication, and cloud. In their current architectural proposals, the authors use the edge layer to locate critical apps and perform basic control activities. According to [46], cloud (internet/intranet) can also include Web services, data storage, HMI interfaces, or analytic applications. An illustration of the architecture models is shown in **Figure 9**. These designs in Sensors 2020, 20, x 34 of 48, include devices, communications, services, administration, applications, and security. IoT systems are made up of devices that are put in a specific location and may perform activities including detection, monitoring, control, and action. To transfer the essential data, the devices must have interfaces that allow them to communicate with other devices. The information gathered by numerous sensors will be processed in general, and the results will be applied to various actuators. The observed data as well as the response actions must then be sent between the devices. Communication protocols are required for this task. In the majority of circumstances, different communication protocols are used on the same IoT system in order for it to work together. Services may be required to complete tasks like device discovery, device control, or data analysis. The programs enable the user to interact with the system. The user will be able to visualize information collected through monitoring as well as information taken from data after it has been processed using the applications. On numerous occasions, the user can take actions that he considers important to the scenario presented by the data, and these actions can also be taken automatically. Finally, assess the system's security. Traditionally, the three layers of IoT architecture have been thought to be perception, network, and application. Following several research, an intermediary layer between the network and application layers was built. In cloud and fog computing environments, this layer, also called the service layer, is used to store and process data. For the past several years, authors like Ferrández-Pastor [46] have proposed a new architecture that is built on four layers: objects, edge, communication, and cloud. In these current architectural methods, the authors employ the edge layer to locate critical apps and perform basic control operations. According to [46], cloud (internet/intranet) can also include Web services, data storage, HMI interfaces, or analytic applications. **Figure 8** shows a representation of the architecture.

**Figure 8.**

*Evolution of the layered model in IoT architecture [2].*

Both 3-layered [43, 47] and 5-layered [48] designs are accessible in the assessed IoT systems for irrigation. The sensor nodes and actuators are usually found in the lowest layer. The middle layer has a gateway and is concerned with data transport. Finally, the third layer is often responsible for data storage and analysis. Cloud services, databases, and applications are common examples of third layers. The Internet of Underground Things [33] is considering an innovative approach to IoT deployments for precision agriculture. In-situ sensing, wireless communication in underground environments, and the interaction between architectural features like sensors, machinery, and the cloud are all identified as functions by the authors. In the case of IoUT, sensors are implanted underground. Wireless communication between above-ground and beneath devices was examined by the researchers. The route loss link between above ground and subterranean devices achieved 80 dBm over a distance of 50 metres. The distance between underground devices for 80 dBm was roughly 10 m. The authors also explore the impact of soil moisture on route loss.

#### **2.1 Recommendations for putting a smart agriculture irrigation system in place**

In this section, the researcher has presented an architecture suggestion for an IoT irrigation system. To ensure the optimal functioning of the IoT irrigation system for precision agriculture, the architecture should provide interoperability, scalability, security, availability, and robustness. Following a thorough analysis of other researchers' work, we have divided our architecture concept into four tiers, as shown in **Figure 9**, which we refer to as devices, communication, services, and applications. Furthermore, the communication and services levels should solve management and security concerns at the same time.

The first layer is the Device layer, which includes all of the devices that will perform detection, monitoring, control, and action functions. There would be four types of nodes in total. The water quality would be checked at the water monitoring node to verify if it was suitable for crop irrigation. The soil monitoring node would

**Figure 9.** *Architecture proposal for an IoT irrigation system for agriculture [2].*

#### *An IoT-based Immersive Approach to Sustainable Farming DOI: http://dx.doi.org/10.5772/intechopen.105449*

monitor soil moisture, temperature, and other parameters, which would contribute in the irrigation schedule decision-making process. The weather monitoring node would measure air temperature and humidity, precipitation, luminosity, radiation, and wind parameters to facilitate decision-making. Finally, the decision-making process's operations would be carried out by the actuator nodes. The second layer is the communication layer, which has three blocks. The Hop-to-Hop communication block allows for the design of data link layer technologies as well as frame transmission with device layer data. In order to reach far-flung sites, frames will be transmitted from this block to the network communication block. The routing function may be assumed in this block in mesh networks, such as 802.15.4 networks. The end-to-end communication block is responsible for delivering the capabilities of the TCP/IP model's transport and application layers when communication spans various network contexts. Finally, the network communication block is responsible for network communication (routing), hop-to-hop communication at end-to-end blocks using IPv4 and IPv6 addresses, as well as ID resolution. It will also be in charge of overseeing service quality. The following layer is the services layer, which consists of three blocks. The services section includes IoT services as well as the ability to discover and search for them. Users are assigned services by the organization block based on their needs or available resources. Finally, in IoT-related business environments, service block modeling and execution will be triggered by application execution. Management and security are two elements that work on both the communication and service tiers. The management block is built using the fault, configuration, accounting, performance, and security (FCAPS) idea and architecture. This model represents the ISO Telecommunications Management Network [33]. The security block, which consists of four blocks, ensures the security and privacy of the systems. User and service authentication are handled by the authentication block. The authorization block is in charge of access control policies. Furthermore, access control decisions will be made based on access control regulations. To provide secure peer-topeer communication, the key exchange & management block is used. Finally, the trust & reputation block is responsible for scoring the user and evaluating the level of trust in the service. The final layer is the application layer. It allows customers to interact with IoT technologies. This layer allows users to receive alarms, see acquired data in real time, and trigger actuators or actions that have not been configured automatically.
