Название: Smart Embedded Systems: Advances and Applications
Автор: Arun Kumar Sinha, Abhishek Sharma, Luiz Alberto Pasini Melek
Издательство: CRC Press
Год: 2024
Страниц: 300
Язык: английский
Формат: pdf (true)
Размер: 67.2 MB

"Smart Embedded Systems: Advances and Applications" is a comprehensive guide that demystifies the complex world of embedded technology. The book journeys through a wide range of topics from healthcare to energy management, autonomous robotics, and wireless communication, showcasing the transformative potential of intelligent embedded systems in these fields. This concise volume introduces readers to innovative techniques and their practical applications, offers a comparative analysis of wireless protocols, and provides efficient resource allocation strategies in IoT-based ecosystems. With real-world examples and in-depth case studies, it serves as an invaluable resource for students and professionals seeking to harness the power of embedded technology to shape our digital future.

The use of Machine Learning (ML) and Deep Learning (DL) techniques in healthcare has been widespread over recent decades. ML algorithms include multi-layer perceptron (MLP), artificial neural networks (ANN), support vector machine (SVM), multi-class SVM, random forest (RF), XGBoost, K-nearest neighbor (KNN), naïve Bayes (NB), and extreme learning machine (ELM). These, and DL algorithms like convolutional neural network (CNN), deep CNN, long-short-term memory (LSTM) and bi-directional LSTM (Bi-LSTM), have been employed in the detection of epileptic seizures. In recent years, many researchers have worked on field programmable gate array (FPGA)-based ML algorithms for the detection of epileptic seizures, which have advantages of reliability, flexibility and lower costs. Lichen Feng et al. implemented feature extraction (FE) and SVM modules using Verilog HDL. The FE module has discrete wavelet transform (DWT), mean absolute value (MAV) and variance sub-modules.

An embedded system, as the name suggests, is the combination of a processor (single/multi-core), memory and several peripheral devices that define/execute a task within a larger entity/system, often electronic or mechanical. Recent advances in the area of neuromorphic embedded systems have introduced many parallel core architectures. A recently reported research example is the spiking neural network architecture, often referred to as SpiNNaker, designed with 57,600 processing nodes, each with 18 ARM968 processors. Total cores were 1,036,800 with 7 terabytes of memory. Technische Universität München developed a robotic platform called SpOmnibot that hosts the integrated SpiNNaker neuromorphic computing board and enables nearly one million simple neurons to be simulated in real time. It also hosts embedded dynamic vision sensors (eDVSs) and has a 9 DOF (degrees of freedom) inertial measurement unit. One of its use cases is autonomous simultaneous localization mapping.

Salient Features:

1. The book provides a comprehensive coverage of various aspects of smart embedded systems, exploring their design, implementation, optimization, and a range of applications. This is further enhanced by in-depth discussions on hardware and software optimizations aimed at improving overall system performance.

2. A detailed examination of Machine Learning techniques specifically tailored for data analysis and prediction within embedded systems. This complements the exploration of cutting-edge research on the use of AI to enhance wireless communications.

3. Real-world applications of these technologies are extensively discussed, with a focus on areas such as seizure detection, noise reduction, health monitoring, diabetic care, autonomous vehicles, and communication systems. This includes a deep-dive into different wireless protocols utilized for data transfer in IoT systems.

4. This book highlights key IoT technologies and their myriad applications, extending from environmental data collection to health monitoring. This is underscored by case studies on the integration of AI and IoT in healthcare, spanning topics from anomaly detection to informed clinical decision-making. Also featured is a detailed evaluation and comparison of different system implementations and methodologies.

This book is an essential read for anyone interested in the field of embedded systems. Whether you're a student looking to broaden your knowledge base, researchers looking in-depth insights, or professionals planning to use this cutting-edge technology in real-world applications, this book offers a thorough grounding in the subject.