A parallel DSP-based neural network emulator with CMOS VLSI packet switching hardware
- 1 January 1994
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636862,p. 381-391
- https://doi.org/10.1109/asap.1994.331787
Abstract
This work describes a parallel neural network emulator which uses standard DSPs and application-specific VLSI communication processors with an integrated hardware routing algorithm. The use of DSPs as programmable processing elements enables the emulation of different types of neurons including biologically inspired models with learnable synaptic weights and delays, variable neuron gain, and static and dynamic thresholding. Locally interconnected communication processors attached to each DSP can span up a 2D- or 3D-computing grid and thus form a highly parallel network topology capable of global packet switched routing.Keywords
This publication has 5 references indexed in Scilit:
- Globally stable neural robot control capable of payload adaptationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Adaptive filter for two-dimensional Gabor transformation and its implementationIEE Proceedings I Communications, Speech and Vision, 1993
- A Recurrent Trajectory Storage Network with Parceling of the WorkspacePublished by Springer Nature ,1993
- A CMOS-array-computer with on-chip communication hardware developed for massively parallel applicationsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1991
- Prevention of Deadlocks in Packet-Switched Data Transport SystemsIEEE Transactions on Communications, 1981