Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94

May 30 1994-June 2 1994

Go

Filter Results

Displaying Results 1 - 25 of 126
  • Chaotic signals generated by digital filter overflow

    Publication Year: 1994, Page(s):17 - 20 vol.6
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (363 KB)

    The paper surveys the chaotic behaviour of digital filter overflow oscillations. The behaviour patterns of 1st, 2nd and 3rd order filters are described and the available stochastic characteristics of oscillations with chaotic features are illustrated by means of probability density functions and power density spectra for a range of filter parameters and initial states. The analysis and simulation ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Influence of the drive-signal resonant overtone on subharmonic injection-locking characteristics

    Publication Year: 1994, Page(s):209 - 212 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (334 KB)

    The influence of a small but nonzero resonant overtone of the drive-signal on the static and dynamic characteristics of subharmonic injection-locked oscillators is investigated by means of a perturbation based phasor-domain approach. Both theoretical and computer simulation results show that, if certain conditions on the circuit parameters occur, even a very small amplitude overtone (e.g., -60 dBc... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural networks for quaternion-valued function approximation

    Publication Year: 1994, Page(s):307 - 310 vol.6
    Cited by:  Papers (19)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (305 KB)

    In the paper a new structure of a Multi-Layer Perceptron, able to deal with quaternion-valued signals, is proposed. A learning algorithm for the proposed Quaternion MLP (QMLP) is also derived. Such a neural network allows one to interpolate functions of a quaternion variable with a smaller number of connections with respect to the corresponding real valued MLP.<> View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Time-multiplexing CNN simulator

    Publication Year: 1994, Page(s):407 - 410 vol.6
    Cited by:  Papers (9)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (296 KB)

    A novel approach to simulate Cellular Neural Networks (CNN) is presented in this paper. The approach, time-multiplexing simulation, is prompted by the need to simulate hardware models and test hardware implementations of CNN. For practical size applications, due to hardware limitations, it is impossible to have a one-on-one mapping between the CNN hardware processors and all the pixels of the imag... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Minimal training set size estimation for neural network-based function approximation

    Publication Year: 1994, Page(s):403 - 406 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (372 KB)

    A new approach to the problem of n-dimensional continuous and sampled-data function approximation using a two-layer neural network is presented. The generalized Nyquist theorem is introduced to solve for the optimum number of training examples in n-dimensional input space. Choosing the smallest but still sufficient set of training vectors results in a reduced learning time for the network. Analyti... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Arbitrarily shaped cell placement by three-layer self-organizing neural networks

    Publication Year: 1994, Page(s):399 - 402 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (376 KB)

    In this paper, a three-layer self-organizing neural network is designed to resolve the cell placement problem with arbitrarily-shaped rectilinear cells. The proposed model with additional hidden layer can easily model the rectilinear cells by a set of hidden neurons which correspond to the partitioned rectangles of cells, called “molecule model”. With the collective computing property ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance analysis of Hopfield neural networks for DOA estimation

    Publication Year: 1994, Page(s):395 - 398 vol.6
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (264 KB)

    In this paper, we give a simple analysis of the DOA estimation method based on Hopfield's optimization neural network. In this case, the problem is mapped onto the quadratic energy function of the network as an optimization problem. Although the method can avoid the eigendecomposition of the data autocorrelation matrix and the orthogonality search of parameter space, theoretical analysis and compu... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Character recognition by neural networks with single-layer training and rejection mechanism

    Publication Year: 1994, Page(s):327 - 330 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (316 KB)

    For many real applications of pattern classification problems, it is more important to reduce the misclassification rate than to increase the rate of successful classification. In this paper, we propose a single-layer neural network with two rejection mechanisms for character recognition problems, which guarantees a very low misclassification rate. The proposed architecture is a cascaded connectio... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural modeling and identification of nonlinear systems

    Publication Year: 1994, Page(s):391 - 394 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (236 KB)

    This paper provides a brief overview of a rigorous framework, developed by the author, for the modeling and identification of nonlinear dynamical systems by artificial neural networks. The system model is obtained as a best approximation of the operator(s) representing the system in a “neural space”, under interpolating or smoothing constraints imposed by the input-output training data... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Handwritten numeral recognition with multiple features and multistage classifiers

    Publication Year: 1994, Page(s):323 - 326 vol.6
    Cited by:  Papers (11)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (316 KB)

    Multiple expert system is shown to be a promising strategy for handwritten numeral recognition. This paper presents a multiple expert system using neural networks. In the proposed system, the authors have developed (1) an incremental clustering neural network algorithm with merging and canceling process, (2) a modified directional histogram feature extraction method and (3) a subclass method with ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Cellular neural networks: the analogic microprocessor?

    Publication Year: 1994, Page(s):289 - 294 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (452 KB)

    The various special features of cellular neural networks are presented. It is stated what has been achieved so far and what future work is still needed in order to obtain a truly universal building block for information processing systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A class of order statistics learning vector quantizers

    Publication Year: 1994, Page(s):387 - 390 vol.6
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (308 KB)

    A novel class of Learning Vector Quantizers (LVQs) based on multivariate order statistics is proposed in order to overcome the drawback that the estimators for obtaining the reference vectors in LVQ do not have robustness either against erroneous choices for the winner vector or against the outliers that may exist in vector-valued observations. The performance of the proposed variants of LVQ is de... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A recurrent neural network for solving the shortest path problem

    Publication Year: 1994, Page(s):319 - 322 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (324 KB)

    The shortest path problem is the classical combinatorial optimization problem arising in numerous planning and designing contexts. In this paper, a recurrent neural network for solving the shortest path problem is presented. The proposed recurrent neural network is able to generate optimal solutions to the shortest path problem. The performance and operating characteristics of the recurrent neural... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extraction of depth information by cellular neural networks

    Publication Year: 1994, Page(s):281 - 284 vol.6
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (264 KB)

    This paper describes dynamic depth extraction for binocular stereo visual information by CNN (cellular neural network). The quantization for the funneling information is done by parallel neurons. And, the correspondence problem can be solved by pattern recognition for analog images reconstructed from the transmitted funneling halftoning images. The competitive CNN is used. The computer simulation ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Gaussian synapse circuit for analog VLSI neural networks

    Publication Year: 1994, Page(s):483 - 486 vol.6
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (316 KB)

    Back-propagation neural networks with Gaussian function synapses have a better convergence property over those with linear-multiplying synapses. A compact analog Gaussian synapse cell which is not biased in the subthreshold region has been designed for fully-parallel operation. This cell can approximate a Gaussian function with accuracy around 98% in the ideal case. Device mismatch induced by fabr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A study on the stochastic computation using the ratio of one pulses and zero pulses

    Publication Year: 1994, Page(s):471 - 474 vol.6
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (268 KB)

    Stochastic computation uses pulse streams to represent numbers. In this paper, we have studied the novel method to implement the number system which uses the ratio of the number of one (high) pulses and the number of zero (low) pulses in a pulse stream. With this number system, if we let P be the probability that the pulse is one in a pulse stream, then the number Y we want to deal with is defined... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analogue CMOS VLSI implementation of cellular neural networks with continuously programmable templates

    Publication Year: 1994, Page(s):367 - 370 vol.6
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (312 KB)

    This paper presents the CMOS realisation of a programmable cellular neural network. An analogue programmable synapse circuit was designed that allows a large dynamic range for the template weights. Tuning circuits transform the digital values specified by the user to the corresponding analogue tuning voltages. A chip containing a 4x4 programmable CNN has been fabricated in a 2.4 μm CMOS process View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Sensitivity to errors in artificial neural networks: a behavioral approach

    Publication Year: 1994, Page(s):459 - 462 vol.6
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (352 KB)

    A behavioral approach to the impact of errors due to faults in neural computation is analyzed. Starting from a geometrical description of errors affecting neural values, we derive the probability of error detection at the neuron's output and at the network's outputs View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • HD and IMD prediction techniques for active filters

    Publication Year: 1994, Page(s):169 - 172 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (224 KB)

    Two single-domain techniques for evaluating the output distortion of nonlinear systems are described as a means of circumventing the time-frequency transformations of standard methods. These are compared with respect to accuracy, computational efficiency and flexibility for a bandpass active filter containing a saturation-type nonlinearity View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Perturbations of CNNs

    Publication Year: 1994, Page(s):221 - 224 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (232 KB)

    If a CNN template, through the application of one of the sign-changing transformations T defined by Chua and Wu (see Int. Journal of Circuit Theory and Applications, vol. 20, p.497-517, 1992), is positive and cell-linking and has isolated equilibria then the convergence of the original CNN follows; this is theorem 4 of the aforementioned paper. The main results of this paper use the condition of i... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • CMOS current-mode chaotic neurons

    Publication Year: 1994, Page(s):499 - 502 vol.6
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (300 KB)

    This paper presents two nonlinear CMOS current-mode circuits that implement neuron soma equations for chaotic neural networks, and another circuit to realize programmable current-mode synapse using CMOS-compatible BJT's. They have been fabricated in a double-metal, single-poly 1.6 μm CMOS technology and their measured performance reached the expected function and specifications. The neuron soma... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A simple neural learning algorithm for total least-squares adaptive filtering

    Publication Year: 1994, Page(s):383 - 386 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (316 KB)

    A Hebbian-type learning algorithm for the total least-squares parameter estimation is presented. An asymptotic analysis is carried out to show that the algorithm allows the weight vector of a linear neuron unit to converge to the eigenvector associated with the smallest eigenvalue of the correlation matrix of the input signal. When the algorithm is applied to solve parameter estimation problems, t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fast convergence with low precision weights in ART1 networks

    Publication Year: 1994, Page(s):237 - 240 vol.6
    Cited by:  Papers (2)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (396 KB)

    A new learning law, the Direct Coding Rule, is proposed for bottom-up long term memory learning in Adaptive Resonance Theory (ART) networks. This law requires less computational precision than the traditional Weber Law Rule and modifies the search dynamics of the network to accelerate convergence. Following a brief mathematical analysis of the new learning law, an ART1 network based on this law is... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modelling sound with chaos

    Publication Year: 1994, Page(s):93 - 96 vol.6
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (328 KB)

    This paper is about exploiting chaos theory to provide a means for modelling sound on a computer. A new technique is described for the analysis and resynthesis of a sound using a chaotic system. Preliminary results are promising: a synthetic version of `air noise' is generated which sounds very much like the original View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural network approach for 3-D object recognition

    Publication Year: 1994, Page(s):315 - 318 vol.6
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (308 KB)

    In this paper we propose a new algorithm for recognizing 3-D objects from 2-D image. The algorithm takes the multiple view approach in which each 3-D object is modeled by a collection of 2-D projections from various viewing angles where each 2-D projection is called an object model. To select the candidates for the object model that has the best match with the input image, the proposed algorithm c... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.