Description
This book shows you how to develop a hybrid mm-wave chipless Radio Frequency Identification (RFID) system, which includes chipless tag, reader hardware, and detection algorithm that use image processing and machine learning (ML) techniques. It provides the background and information you need to apply the concepts of Artificial Intelligence (AI) into detection and chipless tag signature printable on normal plastic substrates, instead of the conventional peak/nulls in the frequency tags. You’ll learn how to incorporate new AI detection techniques along with cloud computing to lower costs. You’ll also be shown a cost-effective means of image construction, which can lower detection errors. The book focuses on side-looking-aperture-radar (SLAR) with a combination of deep learning to provide a much safer means of chipless detection than the current inverse synthetic-aperture radar (iSAR) techniques. Each chapter includes practical examples of design, and a QA section to answer the possible doubts in simpler phrases. With its emphasis on mm-wave band and the practical side of design and engineering of the chipless tags, reader, and detection algorithms, this is an excellent resource for industry engineers, design engineers and university researchers.
Table Of Contents
Chapter 1 Introduction
1.1 The overview model of RFID
1.2 Different types of RFID
1.3 Different types of chipless RFID
1.4 Market aspects for chipless RFID
1.5 RFID frequency spectrum
1.6 Challenges in implementing chipless RFID in mm-wave spectrum
1.7 Book outlines
Chapter 2 Chipless Tag Design
2.1 Introduction
2.2 Chipless RFID tags
2.2.1 Time-domain tags
2.2.2 Frequency-domain tags
2.2.3 Image-based tags
2.2.4 Letter-based tags
2.2.5 Screen printing for the chipless tags
2.2.6 Screen printing experimental observations
2.3 Letter-based Tag Design
2.3.1 Effect of substrate on backscattered signal
2.3.2 Encoding capacity considerations
2.3.3 Tag design based on Peyote alphabet
2.3.4 Peyote tag frequency response
2.4 Backscattering theory and Radar Cross Section (RCS) calculations
2.5 Tag performance simulations
2.5.1 Tag design improvement
2.5.2 Discussion of the results
2.6 Tag response measurements
2.7 Conclusions
2.8 Tag design, friendly questions and answers
Chapter 3 Chipless Reader Design
3.1 Introduction
3.2 Chipless RFID readers
3.2.1 Frequency-based readers
3.2.2 Image-based readers
3.3 60 GHz System block diagram
3.3.1 Maximum reader power and link budget calculations
3.3.2 Maximum reading distance calculations
3.4 60 GHz Tx/Rx boards
3.5 Designing and Integration: RF, IF, controller and peripheral circuits
3.5.1 60 GHz transmitter/receiver
3.5.2 Local voltage controlled oscillator (LO)
3.5.3 Gain/Phase comparator
3.5.4 Digital control board
3.5.5 Peripheral circuits
3.6 Reader characterization
3.6.1 Scanning time and frequency resolution calculations
3.6.2 RCS calibrations
3.7 Conclusions
3.8 Chipless Reader, friendly questions and answers
Chapter 4 Tag decoding
4.1 Introduction
4.2 Machine learning and pattern recognition
4.2.1 Tag decoding using feed-forward networks and back-propagation
4.2.2 Feed-forward concept
4.2.3 Support Vector Machines (SVM)
4.2.4 KNN as a lazy learner
4.2.5 Decision Trees Ensembles
4.2.6 Deep-learning methods and frameworks
4.2.7 Machine learning in Chipless RFID and the gaps
4.3 Data collection methodology
4.3.1 Data collection in the simulations
4.3.2 Data collection in the experiments
4.4 Using feed-forward networks
4.4.1 Feed-forward network results
4.5 Using pattern recognition methods
4.5.1 Pattern recognizers results
4.6 Using CW-SLAR imaging
4.6.1 1-port VNA
4.6.2 2-port reader
4.6.3 Computational costs
4.6.4 Tag imaging and experimental results
4.7 A reliable tag decoder architecture
4.8 Conclusions
4.9 Chipless Tag Decoding, friendly questions and answers
Chapter 5 Cloud-based Deep Learning
5.1 Introduction
5.2 Cloud computing considerations
5.2.1 Cloud computing challenges
5.3 Cloud-computing hardware architecture
5.3.1 IaaS model
5.3.2 SaaS model
5.4 Deep-learner in action
5.4.1 2D image representation of 1D frequency data
5.4.2 Data augmentation
5.4.3 Deep-learner structure
5.4.4 Deep-learning results
5.5 A reliable reader based on cloud deep-learning
5.6 Conclusions
5.7 Cloud-based deep learning, friendly questions and answers
Chapter 6 Conclusions
6.1 Conclusions
6.2 Fulfilling research goals
6.3 Recommendations for future work
Appendices
Chapter A Code Listing
Chapter B PCB Layout
Author
-
Nemai Chandra Karmakar
, a current senior member of the IEEE, obtained his B.Sc (EEE) and M.Sc (EEE) from Bangladesh University of Engineering and Technology, Dhaka, his M.Sc. degree in Electrical Engineering from the University of Saskatchewan, Saskatoon, Canada, his PhD degree from the University of Queensland, Brisbane, Australia in 1999, Postgraduate Diploma in Teaching in Higher Education from The National Institute of Education, Nanyang Technological University, Singapore and Master in Higher Education from Griffith University, Australia. Previously, he worked as an Assistant Engineer in Electronics Institute, Atomic Energy Research Establishment, Dhaka, Bangladesh, a Research Assistant at the Communications Research Group, University of Saskatchewan, Canada, as a Microwave Design Engineer at Mitec Ltd., Brisbane, Australia where he contributed to the development of land mobile satellite antennas for the Australian Mobilesat. Dr. Karmakar currently works in the Faculty of Engineering at Monash University as an Associate Professor within the department of Electrical and Computer Systems Engineering and he is the director of Monash Microwave, Antennas, RFID and Sensors (MMARS) Lab.
-
Larry M. Arjomandi
, a current senior member of the IEEE, received his B.Sc. and M.Sc. degrees in electronics/telecommunications from Ferdowsi University of Mashhad, Iran, and a Ph.D. degree in electrical and computer science from Monash University, Australia in 2019. Previously he worked as a Research Engineer with the Amir-Kabir University Microwave Lab and Niroo Research Center, as an Information analyzer with Tehran Stock Exchange, and as a Radio Optimization manager with Ericsson. Currently he is a sessional Senior Research Engineer in embedded systems at Swinburne University, and a Senior System Engineer at Transport Certification Australia. Dr. Arjomandi received the PARC US Research Associate award, Palo-Alto, California in 2018 and has two translated books, a few national and international papers, and two international patents. His current research interests include chipless RFID, image-processing techniques, and real-life applications of machine learning.