Fog Computing as Solution for IoT-Based Agricultural Applications

Introduction

Before the 21st century, computers were only used as brains to perform predefined tasks that help the users to achieve their goals. Unfortunately, this lead to major restrictions and limitations on applications in which there are billions of information generated by people via other means other than the keyboard Devices like cars, TVs, and cameras, were unable to connect to computer easily or to communicate between each other. The wide spread of Internet and the tremendous jump in the communication field with the introduction of 5G technologies helped in connecting these devices to the Internet through simple built-in or even added separate communication devices. The Internet of Things (IoT) term has become popular for describing situations in which a group of objects, components, sensors, and any possible things are communicating and connected through the Internet connectivity with a computing capability to the data between them (Misra & Simmhan, 2015).

IoT presents a significant means for connecting a massive number of smart and embedded devices, introducing new opportunities to implement new applications without limitations or restrictions by the help of distributed architectures. Building an IoT applications from zero point is a tedious work and can generate various types of errors specially for casual users. Therefore, several open platforms have been introduced such that the users can build their applications easily using these platforms. These platforms greatly make the process simpler and increase the speed of application development for users while providing the application builders with several services such as programming frameworks, data security and possibly storage, data and device administration, and even protocol translation (Misra & Simmhan, 2015), (Vermesan et al's, 2015) and (Giusto et al's, 2011).

IoT applications are being developed to help in making our daily life and work activities easier. Many sectors of life and work are now covered by these applications around the world being commercial, industrial, agricultural and medical. Other examples are smart home, smart city (parking, waste management), utilities (smart grid, smart metering), transportation (connected vehicles), and agriculture. The work in agriculture include crop monitoring, climate, livestock tracking (Giusto et al's, 2011).

One of the challenges known in the agriculture domain is to present the proper information for farmers in timely manner to make better and faster decisions about their investments. Nowadays, it is easy to reach knowledge to assist agriculture branches through websites and applications. However, this is not sufficient in managing the work itself and to take the proper actions in time. It became necessary to introduce new and intelligent solutions to enhance the decision making process in this field. IoT field has aided agriculture domain greatly by introduced the facilities required to manage its applications remotely and with or low human intervention. Most of the previous work depending on the Cloud platform for storage and computing of these applications.

Cloud computing emerged as a platform for providing computing and storage capabilities for limited-resources user devices. This alleviated great burden of the shoulders of such users starting from providing the computational speed required, the large size of storage and the maintenance of those services. They used services on demand and per request. With the advent of IoT, the Cloud computing platforms were directed towards handling this vast area of work. Large Cloud service providers are offering platforms that help the users to build their own IoT applications which target to the IoT as a part of the Cloud business. Different solutions such as Infrastructure as-a-service (IaaS) backend are presented to offer storage area and computing power for both applications and services. Examples for such platforms are: Microsoft Azure IoT (Microsoft, 2018), and IBM Watson IoT platform (Internet of Things Research, 2016).

The growing interest in the Internet of Things (IoT) has led to the creation of IoT data challenges. IoT is identified to be one of the main sources of big data, as it is composed of connected though the Internet large number of smart devices that should send their sensed data of their environments to a higher level or to each other. Big data analytics is the science that recognize and extract important patterns from the data to result in higher levels of visions to be used by the decision makers. These insights are extremely important to the business owners, as they enable them to gain many advantages in the market [5].