Enhanced emulated digital CNN-UM (CASTLE) arithmetic cores

被引:0
|
作者
Hidviégi, T
Keresztes, P
Szolgay, P
机构
[1] Hungarian Acad Sci, Inst Comp & Automat, Analog & Neural Comp Syst Lab, H-1502 Budapest, Hungary
[2] Istvan Szechenyi Univ, H-9026 Gyor, Hungary
[3] Univ Veszprem, Image Proc & Neurocomp Dept, Veszprem, Hungary
关键词
cellular nonlinear/ neural networks (CNN); emulated digital architecture; optimized arithmetic core; CNN Universal Machine;
D O I
10.1142/S0218126603001136
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
An emulated digital CNN-UM (CASTLE) architecture was published few years ago.(1) Different emulated digital CNN-UM architectures are analyzed in the paper. These new modified architectures are optimized according to the silicon area, operating speed or dissipated power. A reconfigurable arithmetic core will also be shown in the paper, by which solution of the neighborhood size can be changed. An advanced CASTLE with pipe-lining is presented. The operation frequency is increased by using this solution in approximately 10 times.
引用
收藏
页码:711 / 738
页数:28
相关论文
共 50 条
  • [31] Configurable multi-layer CNN-UM emulator on FPGA
    Nagy, Z
    Szolgay, P
    CELLULAR NEURAL NETWORKS AND THEIR APPLICATIONS, 2002, : 164 - 171
  • [32] A QCIF resolution binary I/O CNN-UM chip
    Paasio, A
    Kananen, A
    Halonen, K
    Porra, V
    JOURNAL OF VLSI SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 1999, 23 (2-3): : 281 - 290
  • [33] NEROvideo: a general-purpose CNN-UM video processing system
    Mueller, Jens
    Mueller, Jan
    Tetzlaff, Ronald
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2016, 12 (04) : 763 - 774
  • [34] NEROvideo: a general-purpose CNN-UM video processing system
    Jens Müller
    Jan Müller
    Ronald Tetzlaff
    Journal of Real-Time Image Processing, 2016, 12 : 763 - 774
  • [35] An embedded CNN-UM Global Analogic Programming Unit implementation on FPGA
    Voroshazi, Zsolt
    Nagy, Zoltan
    Kiss, Andras
    Szolgay, Peter
    PROCEEDINGS OF THE 2006 10TH IEEE INTERNATIONAL WORKSHOP ON CELLULAR NEURAL NETWORKS AND THEIR APPLICATIONS, 2006, : 318 - +
  • [36] Artificial immune systems based sound event detection with CNN-UM
    Stolte, J
    Cserey, G
    Proceedings of the 2005 European Conference on Circuit Theory and Design, Vol 3, 2005, : 11 - 14
  • [37] CNN-UM based transversely isotropic elastic wave propagation simulation
    Sonkoly, Peter
    Noe, Istvan
    Carcione, J. M.
    Nagy, Zoltan
    Szolgay, Peter
    2007 EUROPEAN CONFERENCE ON CIRCUIT THEORY AND DESIGN, VOLS 1-3, 2007, : 284 - +
  • [38] The computational infrastructure of analogic CNN computing -: Part I:: The CNN-UM chip prototyping system
    Roska, T
    Zarándy, A
    Zöld, S
    Földesy, P
    Szolgay, P
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS, 1999, 46 (02): : 261 - 268
  • [39] Random number generator and Monte Carlo type simulations on the CNN-UM
    Ercsey-Ravasz, Maria
    Roska, Tamas
    Neda, Zoltan
    PROCEEDINGS OF THE 2006 10TH IEEE INTERNATIONAL WORKSHOP ON CELLULAR NEURAL NETWORKS AND THEIR APPLICATIONS, 2006, : 47 - +
  • [40] Object oriented motion-segmentation for video-compression in the CNN-UM
    Szirányi, T
    László, K
    Czúni, L
    Ziliani, F
    JOURNAL OF VLSI SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 1999, 23 (2-3): : 479 - 496