Ebook: VLSI Artificial Neural Networks Engineering

Engineers have long been fascinated by how efficient and how fast biological neural networks are capable of performing such complex tasks as recognition. Such networks are capable of recognizing input data from any of the five senses with the necessary accuracy and speed to allow living creatures to survive. Machines which perform such complex tasks as recognition, with similar ac­ curacy and speed, were difficult to implement until the technological advances of VLSI circuits and systems in the late 1980's. Since then, the field of VLSI Artificial Neural Networks (ANNs) have witnessed an exponential growth and a new engineering discipline was born. Today, many engineering curriculums have included a course or more on the subject at the graduate or senior under­ graduate levels. Since the pioneering book by Carver Mead; "Analog VLSI and Neural Sys­ tems", Addison-Wesley, 1989; there were a number of excellent text and ref­ erence books on the subject, each dealing with one or two topics. This book attempts to present an integrated approach of a single research team to VLSI ANNs Engineering.

---

VLSI Artificial Neural Networks Engineering offers a unique engineering approach to the design of VLSI Artificial Neural Networks (ANNs). The design of analog, digital and mixed analog/digital VLSI ANNs are represented. A design methodology and a CAD environment are presented to highlight the tradeoff design factors. System applications of ANNs to automatic speech recognition and pattern recognition are included.
Chapter 1 serves as an introduction. Chapters 2, 3, 4 and 5 deal with VLSI circuit design techniques (analog, digital and sampled data) and automated VLSI design environment for ANNs. Chapter 2 reports on a sampled data approach to the implementation of ANNs with application to character recognition. It also contains an overview of the different approaches of VLSI implementation of ANNs; explaining the advantage and disadvantage of each approach. In Chapter 3, the topic of design exploration of mixed analog/digital ANNs at the high level of the design hierarchy is addressed. The need for creating such a design automation environment, with its supporting CAD tools, is a necessary condition for the widespread use of application-specific chips of ANN implementation. In Chapter 4 the same topic of design exploration is discussed, but at the low level of the hierarchy and targeting analog implementation. Chapter 5 reports on all-digital implementation of ANNs using the Neocognitron as the ANN model.
Chapters 6, 7, 8 and 9 deal with the application of ANNs to a number of fields. Chapter 6 addresses the topic of automatic speech recognition using neural predictive hidden Markov models. Chapter 7 deals with the topic of classification using minimum complexity ANNs. Chapter 8 addresses the topic of pattern recognition using a fuzzy clustering ANNs. Chapter 9 deals with speech recognition using pipelined ANNs.
VLSI Artificial Neural Networks Engineering will be useful to researchers and graduated engineers working in the area of VLSI circuit and system design and to the students of upper-undergraduate and graduate level courses on analog circuits, digital circuits, ANNs and VLSI system applications.

---

VLSI Artificial Neural Networks Engineering offers a unique engineering approach to the design of VLSI Artificial Neural Networks (ANNs). The design of analog, digital and mixed analog/digital VLSI ANNs are represented. A design methodology and a CAD environment are presented to highlight the tradeoff design factors. System applications of ANNs to automatic speech recognition and pattern recognition are included.
Chapter 1 serves as an introduction. Chapters 2, 3, 4 and 5 deal with VLSI circuit design techniques (analog, digital and sampled data) and automated VLSI design environment for ANNs. Chapter 2 reports on a sampled data approach to the implementation of ANNs with application to character recognition. It also contains an overview of the different approaches of VLSI implementation of ANNs; explaining the advantage and disadvantage of each approach. In Chapter 3, the topic of design exploration of mixed analog/digital ANNs at the high level of the design hierarchy is addressed. The need for creating such a design automation environment, with its supporting CAD tools, is a necessary condition for the widespread use of application-specific chips of ANN implementation. In Chapter 4 the same topic of design exploration is discussed, but at the low level of the hierarchy and targeting analog implementation. Chapter 5 reports on all-digital implementation of ANNs using the Neocognitron as the ANN model.
Chapters 6, 7, 8 and 9 deal with the application of ANNs to a number of fields. Chapter 6 addresses the topic of automatic speech recognition using neural predictive hidden Markov models. Chapter 7 deals with the topic of classification using minimum complexity ANNs. Chapter 8 addresses the topic of pattern recognition using a fuzzy clustering ANNs. Chapter 9 deals with speech recognition using pipelined ANNs.
VLSI Artificial Neural Networks Engineering will be useful to researchers and graduated engineers working in the area of VLSI circuit and system design and to the students of upper-undergraduate and graduate level courses on analog circuits, digital circuits, ANNs and VLSI system applications.

Content:
Front Matter....Pages i-xv
An Overview....Pages 1-31
A Sampled-Data CMOS VLSI Implementation of a Multi-Character ANN Recognition System....Pages 33-89
A Design Automation Environment for Mixed Analog/Digital ANNs....Pages 91-137
A Compact VLSI Implementation of Neural Networks....Pages 139-156
An All-Digital VLSI ANN....Pages 157-189
A Neural Predictive Hidden Markov Model Architecture for Speech and Speaker Recognition....Pages 191-245
Minimum Complexity Neural Networks for Classification....Pages 247-282
A Parallel ANN Architecture for Fuzzy Clustering....Pages 283-302
A Pipelined ANN Architecture for Speech Recognition....Pages 303-321
Back Matter....Pages 323-329

---

VLSI Artificial Neural Networks Engineering offers a unique engineering approach to the design of VLSI Artificial Neural Networks (ANNs). The design of analog, digital and mixed analog/digital VLSI ANNs are represented. A design methodology and a CAD environment are presented to highlight the tradeoff design factors. System applications of ANNs to automatic speech recognition and pattern recognition are included.
Chapter 1 serves as an introduction. Chapters 2, 3, 4 and 5 deal with VLSI circuit design techniques (analog, digital and sampled data) and automated VLSI design environment for ANNs. Chapter 2 reports on a sampled data approach to the implementation of ANNs with application to character recognition. It also contains an overview of the different approaches of VLSI implementation of ANNs; explaining the advantage and disadvantage of each approach. In Chapter 3, the topic of design exploration of mixed analog/digital ANNs at the high level of the design hierarchy is addressed. The need for creating such a design automation environment, with its supporting CAD tools, is a necessary condition for the widespread use of application-specific chips of ANN implementation. In Chapter 4 the same topic of design exploration is discussed, but at the low level of the hierarchy and targeting analog implementation. Chapter 5 reports on all-digital implementation of ANNs using the Neocognitron as the ANN model.
Chapters 6, 7, 8 and 9 deal with the application of ANNs to a number of fields. Chapter 6 addresses the topic of automatic speech recognition using neural predictive hidden Markov models. Chapter 7 deals with the topic of classification using minimum complexity ANNs. Chapter 8 addresses the topic of pattern recognition using a fuzzy clustering ANNs. Chapter 9 deals with speech recognition using pipelined ANNs.
VLSI Artificial Neural Networks Engineering will be useful to researchers and graduated engineers working in the area of VLSI circuit and system design and to the students of upper-undergraduate and graduate level courses on analog circuits, digital circuits, ANNs and VLSI system applications.

Content:
Front Matter....Pages i-xv
An Overview....Pages 1-31
A Sampled-Data CMOS VLSI Implementation of a Multi-Character ANN Recognition System....Pages 33-89
A Design Automation Environment for Mixed Analog/Digital ANNs....Pages 91-137
A Compact VLSI Implementation of Neural Networks....Pages 139-156
An All-Digital VLSI ANN....Pages 157-189
A Neural Predictive Hidden Markov Model Architecture for Speech and Speaker Recognition....Pages 191-245
Minimum Complexity Neural Networks for Classification....Pages 247-282
A Parallel ANN Architecture for Fuzzy Clustering....Pages 283-302
A Pipelined ANN Architecture for Speech Recognition....Pages 303-321
Back Matter....Pages 323-329
....