Hierarchical classifier design in high-dimensional numerous classcases

In applying pattern recognition methods in remote sensing problems, an inherent limitation is that there is almost always only a small number of training samples with which to design the classifier. A hybrid decision tree classifier design procedure that produces efficient and accurate classifiers for this situation is proposed. In doing so, several key questions are addressed, among them the question of the feature extraction techniques to be used and the mathematical relationship between sample size, dimensionality, and risk value. Empirical tests comparing the hybrid design classifier with a conventional single layered one are presented. They suggest that the hybrid design produces higher accuracy with fewer features. The need for fewer features is an important advantage, because it reflects favorably on both the size of the training set needed and the amount of computation time that will be needed in analysis     

Spectral predictions for pulse interval and width modulation

An expression for the frequency spectrum of pulse interval and width modulation (PIWM) is presented. Experimental results obtained have shown close agreement with spectral predictions from the formula. Performance figures are also included from a practical PIWM optical-fibre transmission system intended for use with TV and video signals.<>     

The digi-neocognitron: a digital neocognitron neural network modelfor VLSI

One of the most complicated ANN models, the neocognitron (NC), is adapted to an efficient all-digital implementation for VLSI. The new model, the digi-neocognitron (DNC), has the same pattern recognition performance as the NC. The DNC model is derived from the NC model by a combination of preprocessing approximation and the definition of new model functions, e.g., multiplication and division are eliminated by conversion of factors to powers of 2, requiring only shift operations. The NC model is reviewed, the DNC model is presented, a methodology to convert NC models to DNC models is discussed, and the performances of the two models are compared on a character recognition example. The DNC model has substantial advantages over the NC model for VLSI implementation. The area-delay product is improved by two to three orders of magnitude, and I/O and memory requirements are reduced by representation of weights with 3 bits or less and neuron outputs with 4 bits or 7 bits.     

Microstructural developments in Mg-Ti-PSZ systems

The microstructure of titania-added Mg-partially-stabilized zirconia (PSZ) is dramatically influenced by thermal treatments. Effects of various sintering, heat-treatment and thermal shock cycling parameters on the microstructure of the Mg-Ti-PSZ system are described. Conditions favourable for the growth of needle-like Ti-rich reinforcements in highly thermalshock-resistant Mg-Ti-PSZ ceramics are identified. TiO₂ seems to play a catalytic role in the formation of Zr-rich networks during high-temperature (1700°C) sintering of the Mg-Ti-PSZ system, quite similar to those found in Mg-PSZ, heat-treated above 1300 °C.     

A parallel systems approach to universal receivers

The problem of communication over a channel with unknown characteristics is addressed. The true channel is from a known set of channels, but the transmitter and receiver do not know which of these channels is actually in effect. The goal of a universal receiver is to provide nearly optimal demodulation regardless of the channel that is actually in effect. A parallel receiver implementation is proposed for a universal scheme to cope with such uncertainty. The parallel system consists of a finite number of receivers with the property that, for each channel in the set, the performance of at least one of the receivers will be within a specified performance range. Data verification is accomplished by an appropriate coding system. Sufficient conditions for the existence of such a universal receiver for a prescribed set of channels are established, procedures are outlined for the receiver design, and an example is given to illustrate the applicability of the theory. For M-ary signaling it is shown that, from an information-theoretic viewpoint, the data verification can be achieved at no extra cost by use of the intrinsic side information that is provided by an appropriate coding scheme that also provides error correction     

An orthogonal descent algorithm to find the zero of a convex function,unsolvability test,and rate of convergence

The paper is the first in a series of articles on the orthogonal descent method. The unsolvability test is considered and the rate of convergence is determined for the algorithm that finds a point with a prescribed value of a convex function.Translated from Kibernetika i Sistemnyi Analiz, No. 5, pp. 87–96, September–October, 1992.