This book presents a comprehensive description of the emerging technology of cellular neural networks (CNNs), the first general purpose analog microprocessors with applications including real-time image and audio processing, image recognition, and the solution of partial differential equations. It discusses some realistic industrial applications of CNNs (including automatic fruit classification, nuclear magnetic resonance spectra image processing, environmental modeling and simulation for pollution distribution forecast). Particular attention is paid to the study of CNNs in the context of nonlinear circuit theory. Emphasis is also given to chaotic oscillators and their application in secure communication and spread-spectrum systems. Discussed in addition is the subject of spatio-temporal dynamic phenomena in two-dimensional CNNs. It is shown how traveling wavefronts, spirals, and Turing patterns can develop in a regular and topologically simple array. The book is completed by the description of a real CMOS discrete-time switched-current chip implementation of a CNN. The book offers thorough discussions that range from issues at the system-level, which are characterized by a rigorous analytic approach, to the technological and IC design aspects. Examples, simulation studies and experimental results complement the theoretical results throughout.
The book is about new applications of cellular neural networks, new theoretical results and implementation issues. The reader will find thorough discussion on the subject, ranging from the highest-level aspects treated in rigorous analytic way, to the technological challenges and the circuit design aspects.
This book covers the whole spectrum of cellular neural networks. Examples, simulation results, and experimental results are complemented by theoretical discussions throughout the whole book. Includes supplementary material: sn.pub/extras
Gabriele Manganaro
CMOS VLSI circuit circuit design communication complexity dynamical systems neural networks physiology robotics systems theory
From the reviews:
"The book is divided into eight chapters and each of them guides us through one area where the CNNs can be used … . I recommend the publication to everybody, who is interested in the CNN and its application and implementation, but also to those who face some of the technologies described there, such as nonlinear dynamics, synchronization, signal processing or motion control, because the CNN can show a quite novel and beneficial point of view." (Václav Dekanovský, Neural Network World, Vol. 15 (5), 2005)