Real-Time Adaptive Smoothing with a 1-D Nonlinear Relaxation Network in Analogue VLSI Technology
Reconstruction of given noisy data is an ill-posed problem and a computationally intensive task. Non-linear regularisation techniques are used to find a unique solution under certain constraints. In our contribution we present a parallel mixed-signal architecture which solves this non-linear problem within microseconds. By connecting all parallel cells in a circular manner it is possible to process noisy data vectors of infinite length. This is achieved by virtually shifting the non-linear adaptive filter kernel over the noisy data vector. A 1-D experimental chip has been fabricated using 0.8μxm CMOS technology. On-chip measurements are shown to agree with results from numerical simulations. Results from applying the 1-D chip to nonlinear smoothing of image data will also be given correspondence.
Unable to display preview. Download preview PDF.
- 1.Jose Pineda de Gyvez, Lei Wang, and E. Sanchez-Sinencio. Large-image CNN hardware processing using a time multiplexing scheme. In IEEE International Workshop on Cellular Neural Networks and their Applications, pages 405–410, 1996.Google Scholar
- 2.Peter Kinget and Michiel Steyaert. Analog VLSI integration of massive parallel signal processing systems. Kluwer Academic Publishers, 1997.Google Scholar
- 5.Bing J. Sheu and Joongho Choi. Neural Information Processing and VLSI, chapter 14, pages 397–466. Kluwer Academic Publishers, 1995.Google Scholar
- 6.Klaus Wiehler, Ralph Lembcke, Rolf-Rainer Grigat, Josef Heers, Christoph Schnörr, and H.-Siegfried Stiehl. Dynamic circular cellular networks for adaptive smoothing of multi-dimensional signals. In IEEE International Workshop on Cellular Neural Networks and their Applications, London, 1998.Google Scholar
- 7.Ralf Wöbke. Störungsanalyse eines eindimensionalen nichtlinearen Relaxationsnetzwerks zur Signalverarbeitung. Diplomarbeit, Universität Hamburg, Fachbereich Informatik, Arbeitsbereich Kognitive Systeme, 1997.Google Scholar