Scheduled System Maintenance:
On May 6th, system maintenance will take place from 8:00 AM - 12:00 PM ET (12:00 - 16:00 UTC). During this time, there may be intermittent impact on performance. We apologize for the inconvenience.
By Topic

Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on

Issue 11 • Date Nov. 1997

Filter Results

Displaying Results 1 - 13 of 13
  • In Memoriam - Mac Elwyn Van Valkenburg (October 5, 1921-March 19, 1997)

    Publication Year: 1997 , Page(s): 1041 - 1044
    Save to Project icon | Request Permissions | PDF file iconPDF (24 KB)  
    Freely Available from IEEE
  • Comments on "New conditions for global stability of neural networks with application to linear and quadratic programming problems"

    Publication Year: 1997 , Page(s): 1099 - 1101
    Cited by:  Papers (21)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (139 KB)  

    For original paper, see M. Forti and A. Tesi, ibid., vol. 42, pp. 354-66 (1995). This letter makes the following comments: 1) the assumption of all neuron activation functions to vanish at the origin, which is utilized in the proof of the result implying the existence and uniqueness of the network equilibrium point, can be actually omitted; 2) in the infinite sector case, the result of global asymptotic stability (GAS) remains true with respect to the class of increasing (not necessarily strictly) activations, as in the finite sector case. Consequently, a result about absolute stability (ABST) of neural networks, which can represent a generalization of the existing related ones, is also obtained. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new sufficient condition for nonsymmetric CNNs to have a stable equilibrium point

    Publication Year: 1997 , Page(s): 1092 - 1095
    Cited by:  Papers (10)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (132 KB)  

    This letter gives a new sufficient condition for nonsymmetric CNNs to have at least one stable equilibrium point. Existence of a stable equilibrium point is important for nonsymmetric CNNs because it is a necessary condition for complete stability. It is shown that our sufficient condition is a generalization of a previous result concerning the existence of a stable equilibrium point, and that it can easily be applied to space invariant CNNs with a 3×3 neighborhood View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Two new temperature compensated current sources

    Publication Year: 1997 , Page(s): 1071 - 1074
    Cited by:  Papers (3)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (156 KB)  

    Two new temperature compensated current sources implemented from bipolar transistors have been inspected. They are both deduced from the schema of the cascode current mirror and necessitate a total of six or seven transistors. Their performance have been compared to an earlier introduced current source with four transistors. Simulation and experimental results are reported. They indicate that for 5 V supply the new current sources are both less temperature sensitive than the preceding one over the range -55°C to +125°C. One of them was also found particularly suited to design current sources with a low supply voltage approaching 2 VBE View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An accelerated recurrent network training algorithm using IIR filter model and recursive least squares method

    Publication Year: 1997 , Page(s): 1082 - 1086
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (144 KB)  

    A new approach for the training algorithm of a fully connected recurrent neural network based upon the digital filter theory is proposed. Each recurrent neuron is modeled by an infinite impulse response (IIR) filter. The weights of each layers in the network are updated by optimizing IIR filter coefficients and optimization is based on the recursive least squares (RLS) method. Our results indicate that the proposed algorithm is capable of providing an extremely fast convergence rate. In this letter, the algorithm is validated by applying to sunspots time series, Mackey-Glass time series and nonlinear function approximation problems. The convergence speed of the RLS based algorithm are compared with other fast algorithms. In the obtained results, they show that the proposed algorithm could be up to 200 times faster than that of the conventional backpropagation algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A novel neural network-based approach for designing 2-D FIR filters

    Publication Year: 1997 , Page(s): 1095 - 1099
    Cited by:  Papers (10)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (252 KB)  

    A novel 2-D FIR filter design approach based on neural network optimization (NNO) technique is proposed in the present letter. To demonstrate the feasibility of the NNO design approach, a Tank-Hopfield neural network (THNN) model is chosen and the relation between the MSE (mean square error) criterion and the Lyapunov energy function is also established. The implementation of the approach is described together with some design guidelines. Two 2-D FIR filter design examples are given, and the advantages of the NNO approach over conventional methods are illustrated View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discrete interval system reduction using Pade approximation to allow retention of dominant poles

    Publication Year: 1997 , Page(s): 1075 - 1078
    Cited by:  Papers (10)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (144 KB)  

    This brief presents a method of reduction for discrete interval systems using Pade approximation to allow the retention of dominant poles. For a given system, the denominator of the reduced model is formed by retaining the dominant poles of the original system, while the numerator is obtained by matching interval time moments. Two numerical examples illustrate the procedure View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Terminal sliding mode control of MIMO linear systems

    Publication Year: 1997 , Page(s): 1065 - 1070
    Cited by:  Papers (75)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (204 KB)  

    A new terminal sliding mode control of MIMO linear systems is proposed in this paper. It is shown that, by inserting a nonlinear term of the system state in the MIMO linear sliding mode, a new terminal sliding mode is developed for MIMO linear systems. A terminal sliding mode controller can then be designed to drive the system state variables to reach and retain in the terminal sliding mode. By suitably designing the parameter matrices of the terminal sliding mode, the system state variables reach the system origin in finite time and the closed loop system is infinite stable in the terminal sliding mode View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Error estimation of recurrent neural network models trained on a finite set of initial values

    Publication Year: 1997 , Page(s): 1086 - 1089
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (164 KB)  

    This letter addresses the problem of estimating training error bounds of state and output trajectories for a class of recurrent neural networks as models of nonlinear dynamic systems. The bounds are obtained provided that the models have been trained on N trajectories with N independent random initial values which are uniformly distributed over [a,b]m ∈ Rm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Electrical network theory of countable graphs

    Publication Year: 1997 , Page(s): 1045 - 1055
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (600 KB)  

    The purpose of this dissertation is to derive the equations governing electrical networks of countable graphs and to give conditions assuring that these are gradient dynamical systems on semi-Riemannian Hilbert manifolds. Furthermore, we show how symmetries in the graph give rise to a G-action on the manifold of states. Finally, we present some mathematical results about gradient dynamical systems on semi-Riemannian Banach manifolds View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Generalized state-space averaging approach for a class of periodically switched networks

    Publication Year: 1997 , Page(s): 1078 - 1081
    Cited by:  Papers (12)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (136 KB)  

    A generalized state-space averaging (GSSA) technique for the analysis of a class of periodically switched networks is proposed in the present brief. The application of GSSA is illustrated via the example of buck zero-current switching quasi-resonant converter (ZCS-QRC). Its accuracy is verified by exact numerical simulation results View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The 2-D σ-reciprocal system for model reduction

    Publication Year: 1997 , Page(s): 1056 - 1064
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (472 KB)  

    In this paper we define the σ-reciprocal system of a two-dimensional (2-D) linear time-invariant system represented by the Roesser model. We introduce the 2-D σ-reciprocal system and show that the 2-D generalized singular perturbational approximation (2-D GSPA) is related to direct truncation of its σ-reciprocal system. The stability analysis and the H-norm of the error for the GSPA reduction of 2-D denominator separable systems are presented. A model reduction algorithm based on the analysis of the σ-reciprocal system is given for 2-D systems. Simulation results show that the H-norm of the error of the 2-D GSPA reduction is less or equal to the H-norm of the error if the 2-D singular perturbational approximation (2-D SPA) or the direct truncation method are used View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nonlinear H synchronization of Lur'e systems: dynamic output feedback case

    Publication Year: 1997 , Page(s): 1089 - 1092
    Cited by:  Papers (25)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (172 KB)  

    We introduce a new master-slave synchronization scheme for Lur'e systems, which makes use of vector field modulation and dynamic output feedback in order to recover a message signal. The synchronization scheme is represented in standard plant form according to modern control theory. I/O properties of the scheme are analyzed using a dissipativity approach with a quadratic storage function and a supply rate with finite L2-gain. The method avoids transmission of the full state vector. The controller design is based on a matrix inequality, corresponding to nonlinear H synchronization. The new scheme is illustrated on Chua's circuit View full abstract»

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