Title :
Silicon spike-based synaptic array and address-event transceiver
Author :
Vogelstein, R. Jacob ; Mallik, Udayan ; Cauwenberghs, Gert
Author_Institution :
Dept. of Biomed. Eng., John Hopkins Univ., Baltimore, MD, USA
Abstract :
An integrated array of 2,400 spiking silicon neurons, with reconfigurable synaptic connectivity and adjustable neural spike-based dynamics, is presented. At the system level, the chip serves as an address-event transceiver, with incoming and outgoing spikes communicated over an asynchronous event-driven bus. Internally, every cell implements a spiking neuron that models general principles of synaptic operation as observed in biological membranes. Synaptic conductance and synaptic reversal potential can be dynamically modulated for each event. The implementation employs mixed-signal charge-based circuits to facilitate digital control of system parameters and minimize variability due to transistor mismatch. In addition to describing the structure of the silicon neurons, we present experimental data characterizing the operation of the 3mm × 3 mm chip fabricated in 0.5 μm CMOS technology.
Keywords :
CMOS integrated circuits; microprocessor chips; neural chips; transceivers; 0.5 micron; CMOS technology; address-event transceiver; asynchronous event-driven bus; biological membranes; digital control; mixed-signal; reconfigurable synaptic connectivity; silicon spike; spiking silicon neurons; synaptic array; synaptic conductance; synaptic operation; synaptic reversal potential; system parameters; transistor mismatch; Biological system modeling; Biomembranes; CMOS technology; Cells (biology); Circuits; Neuromorphics; Neurons; Silicon; Transceivers; Virtual manufacturing;
Conference_Titel :
Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium on
Print_ISBN :
0-7803-8251-X
DOI :
10.1109/ISCAS.2004.1329585