作者: N. A. Kant , M. R. Dar , F. A. Khanday , C. Psychalinos
DOI: 10.1109/RTEICT.2016.7807904
关键词: Block (data storage) 、 Domain (software engineering) 、 Companding 、 Artificial neural network 、 Computer architecture 、 Simulation 、 Computer science 、 CMOS 、 Cellular neural network 、 Realization (systems) 、 Image processing
摘要: Temporal Derivative Cellular Neural Network (TDCNN) is an important class of neural networks. These networks find a lot application in real life mostly the real-time image processing. However, main challenge to implement this network hardware. Therefore, paper, sinh-domain realization single cell architecture TDCNN which forms only building block complex introduced. The design offers advantages of; a) low-power operation, b) electronic tunability, c) grounded components, and, d) Class AB nature. functioning has been verified by simulation results achieved through HSPICE tool employing CMOS 0.35μm process.