Reaction-diffusion CNN chip. I. IC implementation

Arena, Paolo; Branciforte, M.; Bernardo, Giuseppe Di; Lavorgna, Mario; Occhipin, L.

doi:10.1109/iscas.2000.856086

Cited by 8 publications

(10 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In general, the latter utilizes Opamps or Operational Transconductance Amplifiers (OTA) as in [3,12,13]. Furthermore, some of these studies employ a top-down design approach to implement reaction-diffusion systems, namely by considering the commonly used cell functions and structures in CNN theory.…”

Section: Comparisons With Existing Implementationsmentioning

confidence: 99%

Compact MOS Implementation of a Reaction-Diffusion CNN

Karahaliloglu

Balkir

2005

Analog Integr Circ Sig Process

View full text Add to dashboard Cite

A biologically inspired single layer cellular neural network (CNN) with trigger wave formation capability is presented. A novel compact MOS cell circuit is proposed which exhibits a third order I-V characteristic with negative differential resistance (NDR). Certain D.C. characteristics of both the proposed cell and the network are described and corresponding theoretical estimations are presented. It is shown that the CNN formed by resistive coupling of these cells has very low complexity and realizes a reaction-diffusion system. The dynamical network behavior is demonstrated by transient simulations of a 2D cell array at the circuit level.

show abstract

Section: Comparisons With Existing Implementationsmentioning

confidence: 99%

Compact MOS Implementation of a Reaction-Diffusion CNN

Karahaliloglu

Balkir

2005

Analog Integr Circ Sig Process

View full text Add to dashboard Cite

show abstract

“…Moreover, it suggests an interesting time multiplexing technique to overcome the bottleneck that arise when you have to map the target nodes to the network cells. The analogue CMOS circuit presented in [3] is dedicated to the solution of RD-PDE. It is based on the CNN paradigm with the aim to generate autonomous wave suitable to control and drive multi-actuated systems.…”

Section: The Cellular Non-linear Networkmentioning

confidence: 99%

“…For this reason, some dedicated multiprocessor hardware systems are often used to compute the numerical solutions in particular problems. An unconventional way could be the design of a programmable analogue circuit on that the PDE system is mapped [1][2][3]. In this case, such analogue circuit represents a hardware model of the RD equation.…”

Section: Introductionmentioning

confidence: 99%

Programmable CNN Analogue Chip for RD-PDE Multi-Method Simulations

Sargeni

Bonaiuto

2005

Analog Integr Circ Sig Process

View full text Add to dashboard Cite

show abstract

“…In the log-domain circuit technique, state variables { } are currents and are related to node voltages { } through exponential relationships (6) This means that the circuital state variables { } cannot be negative. Consequently, a coordinate transformation is required [21] to map the mathematical state variables { } in (5) to the circuital ones { }…”

Section: Circuit Synthesismentioning

confidence: 99%

“…Applying the log-domain mapping given by (6) to the discretely spaced version of (8) that results after the translation and scaling of the state variables, yields the following relation between circuit voltages:…”

Section: Circuit Synthesismentioning

confidence: 99%

Log-domain implementation of complex dynamics reaction-diffusion neural networks

Serrano-Gotarredona

Linares-Barranco

2003

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Abstract-In this paper, we have identified a second-order reaction-diffusion differential equation able to reproduce through parameter setting different complex spatio-temporal behaviors. We have designed a log-domain hardware that implements the spatially discretized version of the selected reaction-diffusion equation. The logarithmic compression of the state variables allows several decades of variation of these state variables within subthreshold operation of the MOS transistors. Furthermore, as all the equation parameters are implemented as currents, they can be adjusted several decades. As a demonstrator, we have designed a chip containing a linear array of ten second-order dynamics coupled cells. Using this hardware, we have experimentally reproduced two complex spatio-temporal phenomena: the propagation of travelling waves and of trigger waves, as well as isolated oscillatory cells.Index Terms-Analog circuit design, analog very large scale integration (VLSI), current mode, logdomain, neural networks, nonlinear circuits, piecewise-linear circuit, subthreshold, weak inversion.

show abstract

Reaction-diffusion CNN chip. I. IC implementation

Cited by 8 publications

References 5 publications

Compact MOS Implementation of a Reaction-Diffusion CNN

Compact MOS Implementation of a Reaction-Diffusion CNN

Programmable CNN Analogue Chip for RD-PDE Multi-Method Simulations

Log-domain implementation of complex dynamics reaction-diffusion neural networks

Contact Info

Product

Resources

About