论文信息 - PCNN neurocomputers - Event driven and parallel architectures

PCNN neurocomputers - Event driven and parallel architectures

The simulation of large spiking neural networks (PCNN) especially for vision purposes is limited by the computing power of general purpose computer systems [5,9,10]. Therefore, the simulation of real world scenarios requires dedicated simulator systems. This article presents architectures of software and hardware implementations for PCNN simulator systems. The implementations are based on a common event driven approach using spike events for communication and processing flow. Furthermore, parallel approaches utilizing the spike event computing are introduced for simulation acceleration. Implementations of software simulators on workstation clusters and parallel computers and hardware accelerators based on FPGAs, ASICs and DSPs are described. The presented results demonstrate the capability to simulate large vision networks close to real world/real time requirements.

Carsten Wolff | Cyprian Grassmann | Tim Schönauer

[1] T. Schoenauer,et al. MASPINN: novel concepts for a neuroaccelerator for spiking neural networks , 1999, Other Conferences.

[2] U. Ruckert,et al. ParSPIKE-a parallel DSP-accelerator for dynamic simulation of large spiking neural networks , 1999, Proceedings of the Seventh International Conference on Microelectronics for Neural, Fuzzy and Bio-Inspired Systems.

[3] Heinrich Klar,et al. A SIMD/dataflow architecture for a neurocomputer for spike-processing neural networks (NESPINN) , 1996, Proceedings of Fifth International Conference on Microelectronics for Neural Networks.

[4] Joachim K. Anlauf,et al. Distributed, Event Driven Simulation of Spiking Neural Networks , 1998, NC.

[5] Georg Hartmann,et al. Biology-Inspired Early Vision System for a Spike Processing Neurocomputer , 2000, Biologically Motivated Computer Vision.

[6] Christopher J. Bishop,et al. Pulsed Neural Networks , 1998 .

[7] Joachim K. Anlauf,et al. Fast Digital Simulation of Spiking Neural Networks and Neuromorphic Integration with Spikelab , 1999, Int. J. Neural Syst..

[8] C. Spengler,et al. Contour Segmentation with Recurrent Neural Networks of Pulse-Coding Neurons , 1997, CAIP.

[9] Tim Schönauer,et al. Simulation of Spiking Neural Networks on Different Hardware Platforms , 1997, ICANN.

[10] Martin Schäfer,et al. Simulation of spiking neural networks -- architectures and implementations , 2002, Neurocomputing.