A reconfigurable spiking neural network is implemented in a 0.5µm CMOS digital tiny-chip. The connection weights are uploaded to registers on the ASIC. These weights are learned off-line, using combined simulated annealing and genetic algorithm. Large computational power and many simulations create small powerful networks that are adapted to interact with the environment. These configurations are swapped in and out of the ASIC to cope with varying situations and increase robustness. The network has been successfully tested with a simulated robot in a maze and can be extended for target recognition.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A reconfigurable spiking neural network digital ASIC simulation and implementation


    Contributors:


    Publication date :

    2009-07-01


    Size :

    1540198 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Spiking Neural Network-based Flight Controller

    Arana, Diego Chavez / Garcia A., Omar A. / Garcia Carrillo, Luis Rodolfo et al. | IEEE | 2024


    Spiking Neural Network E-Nose classifier chip

    Abdel-Aty-Zohdy, H S / Allen, J N / Ewing, R L | IEEE | 2010


    Unterhaltungselektronik: ASIC fur Digital Radio Mondial

    British Library Online Contents | 2007


    An ASIC Digital Video Processing Library

    Tomasino, A. / Kruse, S. / Kelly, F. | British Library Online Contents | 1991


    Cognitive processing using spiking neural networks

    Allen, Jacob N. / Abdel-Aty-Zohdy, Hoda S. / Ewing, Robert L. | IEEE | 2009