Download Cellular Neural Networks: Analysis, Design and Optimization by Martin Hänggi, George S. Moschytz PDF

By Martin Hänggi, George S. Moschytz

Cellular Neural Networks (CNNs) represent a category of nonlinear, recurrent and in the community coupled arrays of exact dynamical cells that function in parallel. ANALOG chips are being built to be used in functions the place refined sign processing at low strength intake is needed.
sign processing through CNNs merely turns into effective if the community is carried out in analog undefined. In view of the actual obstacles that analog implementations entail, strong operation of a CNN chip with appreciate to parameter diversifications should be insured. by means of a long way now not all mathematically attainable CNN initiatives could be performed reliably on an analog chip; a few of them are inherently too delicate. This publication defines a robustness degree to quantify the measure of robustness and proposes an actual and direct analytical layout procedure for the synthesis of optimally strong community parameters. the tactic relies on a layout centering method that is in general appropriate the place linear constraints must be happy in an optimal manner.
Processing pace is often an important whilst discussing signal-processing units. relating to the CNN, it really is proven that the environment time might be laid out in closed analytical expressions, which allows, at the one hand, parameter optimization with recognize to hurry and, nevertheless, effective numerical integration of CNNs. Interdependence among robustness and velocity matters also are addressed. one other target pursued is the unification of the speculation of continuous-time and discrete-time structures. via a delta-operator strategy, it truly is confirmed that an analogous community parameters can be utilized for either one of those periods, whether their nonlinear output services vary.
extra complicated CNN optimization difficulties that can't be solved analytically necessitate resorting to numerical tools. between those, stochastic optimization suggestions akin to genetic algorithms end up their usefulness, for instance in photo class difficulties. because the inception of the CNN, the matter of discovering the community parameters for a wanted activity has been considered as a studying or education challenge, and computationally pricey equipment derived from typical neural networks were utilized. additionally, various worthwhile parameter units were derived via instinct.
during this publication, a right away and special analytical layout process for the community parameters is gifted. The strategy yields suggestions that are optimal with admire to robustness, a side that is an important for winning implementation of the analog CNN that has usually been ignored.
`This superbly rounded paintings offers many attention-grabbing and priceless effects, for either CNN theorists and circuit designers.'
Leon O. Chua