Control of Synchronization Patterns in Neural-like Boolean NetworksWhen: Thursday, March 28, 2013 at 3:00 pm
Where: DA 5th fl
Speaker: David Rosin
Organization: Research Scientist, Physics Department, Duke University
Sponsor: CCNR Seminar
Synchronization patterns have been observed in neurological time-delay networks of directed ring topologies and community structures. We study these networks experimentally using artificial neurons built with autonomous logic gates and heterogeneous link time delays on a field-programmable gate array (FPGA). We observe a transition in the network synchronization dynamics for different refractory periods of the artificialneurons (http://arxiv.org/abs/1211.0318). When the refractory period is comparable to the mean link time delays or the heterogeneity of the link time delays, cluster synchronization patterns are altered or suppressed, respectively. This mechanism allows us to control the network dynamics by adjusting the refractory period of only a small number of driver nodes identified by their in-degree. Our result may have implications for synchronization phenomena in biological neural networks in the brain because of the dependence of the refractory period of neurons on the concentration of drugs and hormones in the blood stream.