A programmable hardware for exploring spatiotemporal waves in real-time


Yeniçeri R. , Yalcm M. E.

2008 11th International Workshop on Cellular Neural Networks and their Applications, CNNA 2008, Cellular Nano-scale Architectures, Santiago de Compostela, Spain, 14 - 16 July 2008, pp.7 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/cnna.2008.4588638
  • City: Santiago de Compostela
  • Country: Spain
  • Page Numbers: pp.7

Abstract

In this demo, we will introduce a hardware based on Xilinx XC2VP30 FPGA chip on the Xilinx University Program Virtex-II Pro Development System for studying spatiotemporal waves in real-time. The network which is implemented on the FPGA is a two dimensional reaction-diffusion Cellular Neural Network consists of relaxation oscillators. The network can be programmed to generate spatiotemporal waves. Because of the real-time computation capability of the implementation, spatiotemporal evolution of the waves on the network can be monitor displaying the consecutive simulation results on a VGA monitor. The parameters and state variables can be accessed dynamically while the hardware is working. Hence the hardware can be programmed by the user to simulate the network with different parameters and initial conditions. Furthermore, user can stop the dynamics of the selected cells using fixed state condition such as in ACE16k CNN chip.