Applications of IoT through Machines' Integration to Provide Smart Environment

Abstract

Internet of Things (IoT) is considered a very large-scale platform where billions of smart devices are integrated together to provide an easy and smart way of life. In IoT, machines can communicate with each other with less human intervention. One of the major challenges that will be faced due to the increase in IoT data is its processing and analysis, which includes processing time of data and response time of an application. Normally data processing, analysis, and knowledge extraction will be done in cloud infrastructure, which will increase the burden of cloud and delay response time that can't be acceptable for real-time stream processing applications. Some of the challenges that IoT faces is faster analytics in small-scale platforms, that is, fast decisions and accurate results to be made, one example of this is autonomous cars and streaming data to be processed that are coming from multiple sources. Machine interactions can be done with exchange of Information and data via IoT devices in either edge or cloud network. In this chapter, we will discuss machine level interaction in edge computing structure and cloud computing through machine learning and deep learning (ML/DL) methodologies and some of the applications of IoT in this smart world like automatic traffic control management, environmental monitoring, healthcare management, analysis and prediction of diseases from medical images, smart agriculture, etc. ML and DL individually or together play an important function in IoT data processing and analysis that can be performed in edge or fog infrastructure. Overall discussions will be based on ML/DL tools that are good for data exploration, knowledge extraction of IoT data. After discussing various DL/ML models such as classification, regression, clustering, convolutional neural networks (CNNs), and recurrent neural networks (RNNs), you will have a clear picture of the importance of IoT and machine communication to this smart world scenarios. ML/DL also plays a very important role in IoT security challenges such as encryption, access control, network, and application security. Through intelligent IoT gateway in edge server, we can process the data faster and can solve any complex analytical tasks. Some of the main constraints are energy management, power consumption of devices, decentralized process, and dynamic allocation of resources. Thus, ML and DL provide a way to effectively handle real-time machine interactions and intelligent processing over the data.