Modeling And Simulation Of Neural Information Transmission Based On Synaptic Plasticity

Information technology combined with technology advances, the electromagneticenvironment has become increasingly complex. Therefore, the drawbacks of traditionalelectromagnetic immunity such as grounding, filter masks become increasingly prominentin the information war. Therefore, how to effectively improve the reliability andadaptability of electronic system in complex electromagnetic environment has become aserious problem nowadays, no matter in social life or military fields.Based on the concept of electromagnetic bionic protection and the adaptive robustadvantages of organisms in a complex electromagnetic environment, In this thesis, wefocus on "what is the phenomenon of robust","What is the mechanism of robust","Thedegree of robust ability". The mechanism of neural information transmission, therelationships between synaptic plasticity and biological adaptive characteristic, the buildingand the roubusr analysis of adaptive neural network model were deeply researched. Themain contribution and innovations are as follows:First of all, the key role of synaptic in neural information transmission and processingwas studied. The relationship between synaptic plasticity and biological adaptivecharacteristic was revealed;Based on the resarch of the mechanism of action potential produce and transmissionand the key role of synaptic in neural information transmission and processing, the impactof synaptic plasticity on neuronal activity was quantified from the perspective ofmathematics and engineering, and the conclusion that synaptic plasticity is the key factorwhich influenced the biological adaptive characteristic was further revealed.Secondly, the adaptive robust network model was constructed and the adaptive robustability of this model was analyzed;Based on the STDP mechanism, fully connected adaptive feedforward neural networkswith single-output and multi-output were constructed respectively, the robust experimentswere simulated on the network with STDP plasticity mechanism combined with the realbiological neural network. The results show that the fully connected feedforward neuralnetwork constructed has a certain degree of adaptive robustness.Finally, the adaptive robustness ability of adaptive neural network model was verified, which is closely related with the STDP plasticity mechanism.The robust experiments were simulated on the network with STDP mechanism orwithout STDP plasticity mechanism, simultaneously, the relationship between the stabilityof the network we constructed and the STDP mechanisms was analyzed. The results showthat the adaptive robustness ability of the fully connected feedforward neural network weconstructed is closely related with the dynamic regulation under STDP plasticitymechanism.The research work lay the theoretical research foundation that introducing robustmechanism of biological system into electronic circuit design and enhancing the protectionability of electronic system in complex electromagnetic environment.