Srinagesh Satyanarayana

A reconfigurable VLSI neural network

Due to the variety of architectures that need be considered while attempting solutions to various problems using neural networks, the implementation of a neural network with programmable topology and programmable weights has been undertaken. A new circuit block, the distributed neuron-synapse, has been used to implement a 1024 synapse reconfigurable network on a VLSI chip. In order to evaluate the performance of the VLSI chip, a complete test setup consisting of hardware for configuring the chip, programming the synaptic weights, presenting analog input vectors to the chip, and recording the outputs of the chip, has been built. Following the performance verification of each circuit block on the chip, various sample problems were solved. In each of the problems the synaptic weights were determined by training the neural network using a gradient-based learning algorithm which is incorporated in the experimental test setup. The results of this work indicate that reconfigurable neural networks built using distributed neuron synapses can be used to solve various problems efficiently. >

Resistive interpolation biasing: a technique for compensating linear variation in an array of MOS current sources

A new technique called resistive interpolation biasing for accurately biasing a large number of analog cells on a VLSI chip is presented. Variations in oxide thickness, mobility, doping concentration, etc., cause inaccuracies in current ratios of two identically biased transistors if they are placed sufficiently far apart on a chip. The proposed technique compensates for these inaccuracies without using any sampling or switching. The technique has been verified using a 2 /spl mu/m n-well CMOS process. Measurements show a factor of 3 improvement in terms of current ratio accuracy when the resistive interpolation technique is used. The circuit can be implemented with a small chip area and low power dissipation. This technique finds applications where extensive current duplication over a large area is required (e.g., analog memories, D/A converters, continuous-time filters, imaging arrays, neural networks, and fuzzy logic systems). >

Video color enhancement using neural networks

Television pictures are not always at their best color saturation settings. We propose an intelligent system that automatically adjusts the color saturation on a field-by-field basis. Incoming pictures are first classified into one of two categories, namely, facial tone and nonfacial tone images, then each category is subcategorized. Depending on the membership of the picture to each subcategory, color saturation changes are computed and used to modify the color saturation of the following field. The categorization is done by using neural network classifiers operating on the color difference signals. We describe the principles of the color correction scheme, its implementation using analog hardware and its interface to a television. Performance of the color adjustment technique implemented using real-time hardware in a television is discussed. Oversaturated and under-saturated pictures in both facial and nonfacial tone categories can be detected and corrected. The technique lends itself to low-cost analog implementations.