TinyML for Edge Intelligence in IoT and LPWAN Networks

1st Edition - May 29, 2024
Editors: Bharat S Chaudhari, Sheetal N Ghorpade, Marco Zennaro, Rytis Paškauskas
Language: English
Paperback ISBN:
9 7 8 - 0 - 4 4 3 - 2 2 2 0 2 - 3
eBook ISBN:
9 7 8 - 0 - 4 4 3 - 2 2 2 0 3 - 0

TinyML for Edge Intelligence in IoT and LPWAN Networks presents the evolution, developments, and advances in TinyML as applied to the Internet of Things (IoT) and low-power… Read more

Purchase options

LIMITED OFFER

Save 50% on book bundles

Immediately download your ebook while waiting for your print delivery. No promo code needed.

Institutional subscription on ScienceDirect

Request a sales quote

TinyML for Edge Intelligence in IoT and LPWAN Networks presents the evolution, developments, and advances in TinyML as applied to the Internet of Things (IoT) and low-power wide area networks (LPWANs). It starts by providing the foundations of IoT/LPWANs, low-power embedded systems and hardware, the role of AI and machine learning in communication networks in general, and cloud/edge intelligence. It then presents the concepts, methods, algorithms, and tools of TinyML. Practical applications of TinyML are given from the healthcare and industrial sectors, providing practical guidance on the design of applications and the selection of appropriate technologies.

Cover image
Title page
Table of Contents
Copyright
List of contributors
Preface
Acknowledgments
Chapter 1: TinyML for low-power Internet of Things
Abstract
1.1. Introduction
1.2. IoT/LPWANs
1.3. LPWAN applications/nodes constraints and requirements
1.4. Edge AI and TinyML
1.5. Challenges for edge AI and TinyML
1.6. Hardware and software considerations
References
Chapter 2: Embedded systems for low-power applications
Abstract
2.1. Introduction
2.2. Key considerations and requirements for microcontroller units
2.3. Component selection
2.4. Embedded software tools for MCUs
2.5. Important hardware platforms
References
Chapter 3: Cloud and edge intelligence
Abstract
3.1. Introduction
3.2. Edge intelligence paradigms
3.3. Edge intelligence: advantages and applications
3.4. Challenges and opportunities for edge intelligence
3.5. AI solutions for optimizing computation of edge intelligence
3.6. Architectural layers in the roadmap for edge intelligence
3.7. Conclusions
References
Chapter 4: TinyML: principles and algorithms
Abstract
4.1. Introduction
4.2. Overview of TinyML
4.3. Workflow of TinyML
4.4. AI and edge computing
4.5. Edge computing optimization using AI solutions
References
Chapter 5: TinyML using neural networks for resource-constrained devices
Abstract
5.1. Introduction
5.2. Background
5.3. Neural networks for TinyML
5.4. Concerns/issues in TinyML
5.5. Applications
5.6. Conclusion
References
Chapter 6: Reinforcement learning for LoRaWANs
Abstract
6.1. Introduction
6.2. Reinforcement learning algorithms
6.3. Related work
6.4. Interference avoidance and energy efficiency for LoRa
6.5. Applications
6.6. Conclusion
References
Chapter 7: Software frameworks for TinyML
Abstract
7.1. Introduction
7.2. Overview of TinyML
7.3. Challenges
7.4. Software and libraries
7.5. Advantages of TinyML
References
Chapter 8: Extensive energy modeling for LoRaWANs
Abstract
8.1. Introduction
8.2. LoRa/LoRaWAN components and elements
8.3. LoRa energy modeling
8.4. Methodology
8.5. Case study and experimental setup
8.6. Conclusion
References
Chapter 9: TinyML for 5G networks
Abstract
9.1. Introduction
9.2. 5G network architecture
9.3. Requirements for 5G use case
9.4. Security in 5G wide scope
9.5. 5G mobile network and virtualization systems
9.6. Multi-access edge computing for 5G
9.7. Open5GS, srsRAN/srsLTE, and OpenAirInterface
9.8. Low-power communication, massive machine type communication (mMTC), and IoT
9.9. 5G edge intelligence features
9.10. Machine learning modeling for 5G
9.11. TinyML for edge intelligence in 5G
9.12. TinyML for 5G security and data anomaly
9.13. Design and modeling for LPWAN-based 5G
9.14. 5G MEC latency standards
9.15. TinyML for 5G computation complexity
References
Chapter 10: Non-static TinyML for ad hoc networked devices
Abstract
10.1. Introduction
10.2. Backbones and TinyML boards
10.3. Neural network model reduction
10.4. Transfer learning
10.5. Purely distributed federated learning
10.6. Challenges
10.7. Conclusions
References
Chapter 11: Bayesian-driven optimizations of TinyML for efficient edge intelligence in LPWANs
Abstract
11.1. Introduction
11.2. Background and related work
11.3. Methodology
11.4. Proposed Bayesian optimization algorithms of TinyML in LPWANs
11.5. Experimental results
11.6. Conclusion and future directions
References
Chapter 12: 6TiSCH adaptive scheduling for Industrial Internet of Things
Abstract
12.1. Introduction
12.2. Benefits of TinyML in IIoT
12.3. IEEE 802.15.4e TSCH
12.4. IPv6 over the TSCH mode of IEEE 802.15.4e (6TiSCH)
12.5. 6TiSCH architecture
12.6. 6TiSCH protocol stack
12.7. 6TiSCH scheduling strategies
12.8. Deterministic scheduling for IIoT networks
12.9. Traffic-aware and reliable 6TiSCH scheduling
12.10. Energy efficiency and reliability considerations
12.11. 6TiSCH simulation software
12.12. Challenges and limitations
12.13. Conclusion and future directions
References
Chapter 13: Securing TinyML in a connected world
Abstract
13.1. Introduction
13.2. Related work
13.3. Issues and challenges in securing TinyML
13.4. Existing solutions and approaches for TinyML security
13.5. Threat models in TinyML
13.6. Applications of TinyML
13.7. Implications for future research in TinyML security
References
Chapter 14: TinyML applications and use cases for healthcare
Abstract
14.1. Introduction
14.2. Wearable devices
14.3. Remote patient monitoring
14.4. Role in diagnosis and treatment by healthcare providers
14.5. Drug discovery
14.6. Ethics and TinyML
14.7. Conclusion
References
Chapter 15: Machine learning techniques for indoor localization on edge devices
Abstract
15.1. Introduction
15.2. The challenges of indoor localization
15.3. State-of-the-art techniques for indoor localization
15.4. A tiny indoor localization implementation
15.5. Conclusions
References
Chapter 16: Embedded intelligence in Internet of Things scenarios: TinyML meets eBPF
Abstract
Acknowledgements
16.1. Introduction
16.2. State-of-the-art
16.3. Design and implementation
16.4. Case of study
References
Chapter 17: A real-time price recognition system using lightweight deep neural networks on mobile devices
Abstract
17.1. Introduction
17.2. Related works
17.3. Machine learning-based techniques for image processing
17.4. Material and methods
17.5. Experimental results and discussion
17.6. Conclusion
References
Chapter 18: TinyML network applications for smart cities
Abstract
18.1. Introduction
18.2. TinyML applications in smart cities
18.3. Analysis and discussion
18.4. Conclusion
References
Chapter 19: Emerging application use cases and future directions
Abstract
19.1. Introduction
19.2. Overview of TinyML
19.3. Workflow steps, challenges, and proposed solutions
19.4. Applications of TinyML and future directions
References
Index

Bharat S Chaudhari

Dr. Bharat S. Chaudhari is working as a Professor (HAG) in Electrical/Electronics and Communication Engineering at MIT World Peace University, Pune, India. He graduated in Industrial Electronics Engineering from Amravati University in 1989 and received M. E. Telecom. E. and PhD (Engineering) from Jadavpur University, Kolkata, India in 1993 and 2000, respectively. He has previously held positions like Principal, Dean, and Professor at various Institutes in Pune. He is a recipient of the 2020 IETE N V Gadadhar Memorial Award, the MAEER Pune’s Ideal Teacher Award 2015, and the Young Scientist Research Grant from the Department of Science and Technology, Government of India, in 2003. He was appointed as a Visiting Scientist (Simons Associate/Regular Associate) to Wireless Network Research Group, ICTP, Trieste, Italy from January 2007 to December 2020. He is a Senior Member of IEEE, Fellow of IETE, Fellow of IE (I), and a Founder Chairman of the IEEE Pune Section (2010 – 2011). He also chaired the IEEE Communications Society, Pune Chapter. He serves as a Program Evaluator of the Engineering Accreditation Commission (EAC) of ABET, United States, for accreditation of computer, communications, and similar engineering programs. His current research interest includes LPWANs, AIoT, TinyML, and Si-Photonics.

Sheetal N Ghorpade

Dr. Sheetal Ghorpade is working as Director (Data Sciences) at Rubiscape Pvt. Limited, Pune, Maharashtra, India, where she is actively involved in the research and development of data science products. She received her PhD in Applied Mathematics and MSc in Mathematics from the University of Pune, India. She is a Regular Associate (Visiting Scientist) at the Abdus Salam International Centre for Theoretical Physics in Trieste, Italy. Her career represents a great mix of data analytics and high-quality internationally collaborative research on algorithms and optimization for Industry problems. Her research interests are Optimization Techniques, Artificial Intelligence, Tiny ML, etc. She is instrumental in bridging the institute-industry gap and is passionate about democratizing data science. She is on the advisory board of various engineering and management institutes.

Marco Zennaro

Dr. Marco Zennaro is a Research Scientist at the Abdus Salam International Centre for Theoretical Physics in Trieste, Italy, where he coordinates the Science, Technology, and Innovation Unit. He received his PhD from KTH-Royal Institute of Technology, Stockholm, and his MSc in Electronic Engineering from the University of Trieste. He is a visiting Professor at KIC-Kobe Institute of Computing, Japan, and a Senior Member of IEEE and ACM. His research interest is in ICT4D, the use of ICT for Development, and in particular, he investigates the use of IoT in Developing Countries. He is TinyML4D Chair and Academic Network Co-Chair in the framework of the TinyML4D initiative. He has given lectures on wireless technologies in more than 30 different countries. More info here: https://www.ictp.it/member/marco-zennaro#biography

Rytis Paškauskas

Dr. Rytis Paškauskas is a Postdoctoral Fellow at the Abdus Salam International Centre for Theoretical Physics in Trieste, Italy, where he works on embedded machine-learning applications. He received his PhD in Physics from Georgia Institute of Technology and has worked at ELETTRA Sincrotrone Trieste, CNR Pisa, CNRS Lyon, and the National Institute for Theoretical Physics (NiTheP) in Stellenbosch, South Africa.

Life Sciences

Physical Sciences & Engineering

Social Sciences & Humanities

Health