A New Algorithm for Crossbar Switch Problem Using Hopfield Neural Network with Continuous Hysteresis Neurons

In this paper, we propose a continuous hysteresis neurons Hopfield neural network architecture for efficiently solving crossbar switch problems. A Hopfield neural network architecture with continuous hysteresis and its collective computational properties are studied. It is proved theoretically and confirmed by simulating the randomly generated Hopfield neural network with continuous hysteresis neurons. The network architecture is applied to a crossbar switch problem and results of computer simulations are presented and used to illustrate the computation power of the network architecture. The simulation results show that the Hopfield neural network architecture with continuous hysteresis neurons is much better than the previous works including the original Hopfield neural network architecture, maximum neural network and Hopfield neural network with hysteresis binary neurons for crossbar switch problem in terms of both the computation time and the solution quality.

[1]  Yoshiyasu Takefuji,et al.  An artificial maximum neural network: a winner-take-all neuron model forcing the state of the system in a solution domain , 2004, Biological Cybernetics.

[2]  J. J. Hopfield,et al.  “Neural” computation of decisions in optimization problems , 1985, Biological Cybernetics.

[3]  Lipo Wang.  Discrete-time convergence theory and updating rules for neural networks with energy functions , 1997, IEEE Trans. Neural Networks.

[4]  J. Hopfield,et al.  Computing with neural circuits: a model. , 1986, Science.

[5]  武藤 佳恭.  Neural network parallel computing , 1992 .

[6]  A. Marrakchi,et al.  A neural net arbitrator for large crossbar packet-switches , 1989 .

[7]  Yoichi Takenaka,et al.  Maximum Neural Network Algorithms for N-Queen Problems , 1996 .

[8]  S. Tamura,et al.  Comments on "Artificial neural networks for four-coloring map problems and K-colorability problems" , 1994 .

[9]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[10]  John J. Hopfield,et al.  Simple 'neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit , 1986 .

[11]  Gerhard Galán,et al.  A New Input-Output Function for Binary Hopfield Neural Networks , 1999, IWANN.

[12]  R.J.T. Morris,et al.  Neural network control of communications systems , 1994, IEEE Trans. Neural Networks.

[13]  H. T. Nguyen,et al.  A neural network implementation of an input access scheme in a high-speed packet switch , 1989, IEEE Global Telecommunications Conference, 1989, and Exhibition. 'Communications Technology for the 1990s and Beyond.

[14]  Zheng Tang,et al.  Hopfield Neural Network with Hysteresis for Maximum Cut Problem , 2004 .

[15]  Stephen M. Walters,et al.  Neural network architecture for crossbar switch control , 1991 .

[16]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[17]  Jerry M. Mendel,et al.  The hysteretic Hopfield neural network , 2000, IEEE Trans. Neural Networks Learn. Syst..