Hierarchical Network-on-Chip and Traffic Compression for Spiking Neural Network Implementations
- 1 May 2012
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2012 IEEE/ACM Sixth International Symposium on Networks-on-Chip
Abstract
The complexity of inter-neuron connectivity is prohibiting scalable hardware implementations of spiking neural networks (SNNs). Traditional neuron interconnect using a shared bus topology is not scalable due to non-linear growth of neuron connections with the neural network size. This paper presents a novel hierarchical NoC (H-NoC) architecture for SNN hardware which addresses the scalability issue by creating a 3-dimensional array of clusters of neurons with a hierarchical structure of low and high-level routers. The H-NoC architecture also incorporates a spike traffic compression technique to exploit SNN traffic patterns, thus reducing traffic overhead and improving throughput on the network. In addition, adaptive routing capabilities between clusters balance local and global traffic loads to sustain throughput under bursting activity. Simulation results show a high throughput per cluster (3.33×10 9 spikes/second), and synthesis results using 65-nm CMOS technology demonstrate low cost area (0.587mm 2 ) and power consumption (13.16mW @100MHz) for a single cluster of 400 neurons, which outperforms existing SNN hardware strategies.Keywords
This publication has 20 references indexed in Scilit:
- Scalable network-on-chip architecture for configurable neural networksMicroprocessors and Microsystems, 2011
- Adaptive Routing Strategies for Large Scale Spiking Neural Network Hardware ImplementationsLecture Notes in Computer Science, 2011
- Neuromorphic silicon neuron circuitsFrontiers in Neuroscience, 2010
- A 118.4 GB/s Multi-Casting Network-on-Chip With Hierarchical Star-Ring Combined Topology for Real-Time Object RecognitionIEEE Journal of Solid-State Circuits, 2010
- A Multicast Routing Scheme for a Universal Spiking Neural Network ArchitectureThe Computer Journal, 2009
- Networks on chips: a new SoC paradigmComputer, 2002
- Exploring complex networksNature, 2001
- Route packets, net wiresPublished by Association for Computing Machinery (ACM) ,2001
- Estimation of the Number of Synapses in the Cerebral Cortex: Methodological ConsiderationsCerebral Cortex, 1999
- Collective dynamics of ‘small-world’ networksNature, 1998