Authors
Zining Xiong, Jingru Sun, Jiqi He, Fei Yu, Yichuang Sun
Published in
Neural networks : the official journal of the International Neural Network Society. Volume 205. Issue Pt B. Pages 109690.
Abstract
Many existing memristive synaptic models merge presynaptic and postsynaptic activities into a single aggregated stimulus, making their distinct functional roles difficult to interpret. To address this issue, this paper proposes a memristive synapse with a pre-driven and post-gated state-update mechanism, in which presynaptic activity drives synaptic-state evolution while postsynaptic activity nonlinearly modulates the direction and strength of the update. The proposed synapse is embedded into a compact two-neuron memristive Hopfield neural network to investigate how this functional decomposition influences network dynamics. Equilibrium and stability analysis, bifurcation diagrams, Lyapunov exponents, and multistability analysis reveal variable-wing chaotic attractors, parameter-dependent transitions, and coexisting dynamical states. FPGA implementation verifies the hardware realizability of the proposed system, while a color-image encryption scheme demonstrates the application potential of the generated chaotic sequences. These results show that the proposed framework provides a functionally interpretable modeling route for incorporating differentiated pre/post synaptic regulation into memristive Hopfield dynamics.
PMID:
42870404
Bibliographic data and abstract were imported from PubMed on 14 Oct 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 15
- Comments 0