Preliminary results from an analog implementation of first-order TDCNN dynamics

摘要: This work falls into the category of linear cellular neural network (CNN) implementations. We detail first investigative attempt on CMOS analog VLSI implementation a recently proposed formalism, which introduces time-derivative ‘diffusion’ between CNN cells for nonseparable spatiotemporal filtering applications—the temporal-derivative CNNs (TDCNNs). The reported circuit consists an array Gm-C filters arranged in regular pattern across space. show that state–space coupling Gm-C-based elements realizes stable and first-order (temporal) TDCNN dynamics. is based linearized operational transconductance amplifiers Class-AB current mirrors. Measured results from prototype chip confirms stability linearity realized are provided. has been built AMS 0.35 µm technology occupies total area 12.6 mm sq, while consuming 1.2 µW per processing cell. Copyright © 2010 John Wiley & Sons, Ltd.