The use of machine learning in the Internet of Things - overview and taxonomy of ML algorithms

Aug 5, 2026

Upload by Himbo

The paper examines how machine learning enables Internet of Things and wireless sensor networks to manage large-scale connected devices and the data volumes they generate. It addresses key operational challenges including topology changes, link failures, memory limits, interoperability, scalability, network traffic, security, and sequestration. It surveys core ML paradigms and algorithms, then introduces a new taxonomy to organize ML approaches for WSN and IoT. It also highlights major operational and research challenges and outlines future research directions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Machine Learning in WSN and IoT

Core ML concepts and iterative learning

The use of machine learning in the Internet of Things Wafaa Ennabigha1*, Ahmed Moutabir2, and Abderraoouf Aboudou 1 1Hassan 2 University, High School of Technology, Laboratory in Computer Networks, Telecommunications and Multimedia, Casablanca, Morocco. 2Hassan 2 University, Aïn Chok Faculty of Science (FSAC), Laboratory of Electrical and Industrial Engineering, Information Processing, IT and Logistics, Casablanca, Morocco. Abstract. In the Internet of Things and wireless sensor networks period, a large number of connected objects and seeing bias are devoted to collecting, transferring, and inducing a huge quantum of data for a wide variety of fields and operations. To effectively run these complex networks of connected objects, there are several challenges like topology changes, link failures, memory constraints, interoperability, network traffic, content, scalability, network operation, security, and sequestration to name many. therefore, overcoming these challenges and exploiting them to support this technological outbreak would be one of the most pivotal tasks of the ultramodern world. Recently, the development of Artificial Intelligence(AI) led to the emergence of Machine Learning (ML), which has become the crucial enabler to figure out results and literacy models in an attempt to enhance the quality of service parameters of Internet of Things and wireless sensor networks. By learning from one gest, ML ways aim to resolve issues in the Internet of Things and wireless sensor networks and fields by erecting algorithmic models. In this paper, we're going to punctuate the most abecedarian generalities of ML orders and Algorithms. We start by furnishing a thorough overview of the Internet of Things and wireless sensor network technologies. We also bandy the vital part of ML ways in driving up the elaboration of these technologies. also, as the crucial donation of this paper, a new taxonomy of ML algorithms is handed. We also epitomize the major operations and exploration challenges that abused ML ways in the WSN and IoT. ultimately, we dissect the critical issues and list some unborn exploration directions 1. Introduction In recent times, there has been a worldwide interest in WSN and IoT. It'll not be a magnification to consider WSN and IoT as two of the most delved areas in the last decade. are many overviews of exploration literature on IoT and WSNs which can be set up at[1][2][3][4][ 5][ 6][ 7][8] . A WSN can be distinct as a network of small bias, called detector * [Corresponding author:](Corresponding author: wafaa.nabigha@gmail.com)[ ](Corresponding author: wafaa.nabigha@gmail.com)[wafaa.nabigha@gmail.com](Corresponding author: wafaa.nabigha@gmail.com) © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/) . bumps, which are spatially dispersed and uniting to transfer information from sources to cesspools wirelessly. also, IoT can be defined as a global network structure composed of colorful connected bias that calculates on communication, sensitivity, information processing technologies, and networking and IoT technologies offer multitudinous advantages over conventional networking results, similar as trustability, delicacy, lower costs, inflexibility, and ease of deployment that enable their use in a wide range of different fields and operations. use in a wide range of diverse fields and applications. By 2020, statistical results show that the number of connected biases is anticipated to reach 50 billion[9] . The increase in the number of connected bias will enhance network content buton the other hand, it'll also increase the size of collected data as well as computational complexity at the centralized base station. The cooperative nature of WSN and IoT brings several advantages, including tone-association, inflexibility, rapid-fire deployment, and processi

FAQs

What problem does the paper address in IoT and wireless sensor networks?

It focuses on enabling IoT and WSN to effectively collect, transfer, and process large volumes of data despite challenges such as topology changes, link failures, scalability, and security.

How does machine learning contribute to IoT and WSN according to the paper?

Machine learning is described as a crucial enabler that builds algorithmic models from data, helping improve quality of service and tackle complex tasks in IoT and WSN.

What is included as a key contribution of the paper?

The paper claims a new taxonomy of machine learning algorithms, along with an outline of major operational and research challenges addressed by ML in WSN and IoT.

Aug 5, 2026

Upload by Himbo