article · Technologies
Billions of Internet of Things devices generate massive volumes of sensor data, making conventional cloud processing inefficient and overburdening network infrastructure. Deploying machine learning directly on resource-constrained devices at the network edge allows local data interpretation, accurate predictions, and swift decision-making. Hardware platforms such as Raspberry Pi, NVIDIA Jetson, Arduino Nano, and Google Coral Dev Board run custom artificial intelligence frameworks like TensorFlow Lite and Core ML to execute tasks such as object detection and gesture recognition. An analysis of one thousand recent publications reveals key application domains across industry, healthcare, agriculture, transportation, smart cities, and assisted living. However, practical deployment faces major hurdles, notably encrypting sensitive user data locally, managing edge node resources through distributed learning, and balancing the strict energy constraints of small devices against the high power demands of machine learning models.
Transmitting huge volumes of device data to central cloud servers creates network bottlenecks and delays. Running machine learning models directly on low-power devices helps everyday systems in healthcare, farming, and transport make fast, autonomous decisions locally. Understanding the current technology landscape and existing limitations helps engineers and researchers build more energy-efficient, secure, and responsive smart devices.
This work outlines technology options for developers and organisations building smart systems across industrial automation, healthcare, agriculture, and intelligent transport. Solutions rely on accessible hardware such as Raspberry Pi and Jetson boards paired with lightweight frameworks like TensorFlow Lite. As a review and literature analysis of current techniques and bottlenecks, it represents foundational landscape analysis rather than a standalone market-ready product, highlighting that commercial deployment requires solving local encryption and energy consumption challenges.
AI-generated from the published abstract. Always read the original work before citing.
Internet of Things (IoT) devices often operate with limited resources while interacting with users and their environment, generating a wealth of data. Machine learning models interpret such sensor data, enabling accurate predictions and informed decisions. However, the sheer volume of data from billions of devices can overwhelm networks, making traditional cloud data processing inefficient for IoT applications. This paper presents a comprehensive survey of recent advances in models, architectures, hardware, and design requirements for deploying machine learning on low-resource devices at the edge and in cloud networks. Prominent IoT devices tailored to integrate edge intelligence include Raspberry Pi, NVIDIA’s Jetson, Arduino Nano 33 BLE Sense, STM32 Microcontrollers, SparkFun Edge, Google Coral Dev Board, and Beaglebone AI. These devices are boosted with custom AI frameworks, such as TensorFlow Lite, OpenEI, Core ML, Caffe2, and MXNet, to empower ML and DL tasks (e.g., object detection and gesture recognition). Both traditional machine learning (e.g., random forest, logistic regression) and deep learning methods (e.g., ResNet-50, YOLOv4, LSTM) are deployed on devices, distributed edge, and distributed cloud computing. Moreover, we analyzed 1000 recent publications on “ML in IoT” from IEEE Xplore using support vector machine, random forest, and decision tree classifiers to identify emerging topics and application domains. Hot topics included big data, cloud, edge, multimedia, security, privacy, QoS, and activity recognition, while critical domains included industry, healthcare, agriculture, transportation, smart homes and cities, and assisted living. The major challenges hindering the implementation of edge machine learning include encrypting sensitive user data for security and privacy on edge devices, efficiently managing resources of edge nodes through distributed learning architectures, and balancing the energy limitations of edge devices and the energy demands of machine learning.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/technologies12060081
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
