The internet of things (IoT) technology, which connects internet-connected devices, continues to extend the current internet by allowing communication and interactions across the physical and cyber worlds. The IoT generates big data that is characterized by its velocity in terms of time and place dependency, with a diversity of diverse modalities and fluctuating data quality, in addition to increased volume. The key to designing smart IoT applications is intelligent data processing and analysis. This chapter first describes the details of how cloud services and edge technology work and support the internet of things with many challenges and limitations in the overall internet services. Second, it describes the support of different machine learning algorithms (MLA) in the different fields of internet of things applications. Finally, there is a description of the future research scopes and open issues in the field of the internet of things with machine learning algorithms for further research work initiation.
TopIntroduction
A plenty of infamy in recent years has gained by the Internet of Things components. It is a collection of different hardware devices, and they are operated by various software programming, those compounds are involved to connect real world environment with the internet(Merenda,2020). There is a dramatic increment in the count IoT device because of extraordinary development in IoT environment. The year wise increment of devices from the year 2016 to 2021 with respect to the Exabyte per month according to all the devices as M2M(Machine to Machine), Smartphones, Non-smartphones, TVs, PCs, Tablets and others are comparatively twice.
Usually the IoT devices have restricted computing efficiency, limited storage and will generate huge volume of data. A real world environment of homes, vehicles and industries has many sensors with low efficiency and connected with IoT. Here there is a requirement of cloud computing concept to provide efficient computing and large storage services to the connected devices in the internet(Statista, 2020). The processing data in computing framework is an important part of IoT, those are named as edge and cloud computing(M. Aazam, 2015). The edge computing is a technology to support cloud computing services in the IoT for end user devices(Joakar 2016).
Cloud computing is a technology to provide services to the end user devices through internet[5]. The cloud computing provides services in various types, they are
- •
Infrastructure as a Service(IaaS): This type of cloud computing provides hardware and servers utilization to the end user or industrial users through internet.
- •
Platform as a Service(PaaS): In this type of cloud computing support to use operating systems and software environments to develop applications without installing it.
- •
Software as a Service(SaaS): Here there are many software applications provided as a service to the end users.
Unfortunately, there are many limitations and complexities to provide those services for different type of end user devices. Here there are different edge computing implemented with the support of Machine Learning (ML) algorithms for efficient and reliable services to the IoT devices(Liu, 2019).
This paper provides the detailed view of Internet of Things with different kind of challenges in the next section, then the different types of Machine learning algorithms which are supporting to the Internet of Things, and the research trends and open issues are discussed, finally there is a conclusion regarding this paper.
TopCloud And Edge Computing In The Internet Of Things
The concept of Internet of Things(IoT) used to achieve a smarter environment and comfortable life with low cost, efficient in the sense of time and energy. As a result massive investments and numerous studies are increased in field IoT technology in recent years(C. Cortes, 1995) . The technology named as internet of things includes the different kind of service provider and service receivers. The cloud computing technology act as a service provider to provide different kind of internet services through internet. The end user devices act as service receivers to receive the requirements through the internet. In between the service provider and the end receiver there must be a intermediate layer named as intimidator or edge which includes routers, gateways, base station etc. The overall IoT environment can be categorized into different layer according to the efficiency of processing, those are shown in the following figure 1. with three layers(Statista, 2020). The name cloud refers the collection of large storage devices, high resource processors and virtual application platforms, these all are accesses and used by the end user devices through edge middle ware. The name edge refers with the collection of gateways, routers and base stations for providing support of communication between service provider to the end user devices. The name end devices refers the collection of all kind of devices used by the end users.
Figure 1. Illustration of edge computing