Department of Electrical and Electronics Engineering2024-11-0920130090-677810.1109/TCOMM.2013.042313.1207992-s2.0-84959473493http://dx.doi.org/10.1109/TCOMM.2013.042313.120799https://hdl.handle.net/20.500.14288/7135Communication between neurons occurs via transmission of neural spike trains through junctional structures, either electrical or chemical synapses, providing connections among nerve terminals. Since neural communication is achieved at synapses, the process of neurotransmission is called synaptic communication. Learning and memory processes are based on the changes in strength and connectivity of neural networks which usually contain multiple synaptic connections. In this paper, we investigate multiple-access neuro-spike communication channel, in which the neural signal, i.e., the action potential, is transmitted through multiple synaptic paths directed to a common postsynaptic neuron terminal. Synaptic transmission is initiated with random vesicle release process from presynaptic neurons to synaptic paths. Each synaptic channel is characterized by its impulse response and the number of available postsynaptic receptors. Here, we model the multiple-access synaptic communication channel, and investigate the information rate per spike at the postsynaptic neuron, and how postsynaptic rate is enhanced compared to single terminal synaptic communication channel. Furthermore, we analyze the synaptic transmission performance by incorporating the role of correlation among presynaptic terminals, and point out the postsynaptic rate improvement.EngineeringElectrical and electronic engineeringTelecommunicationsJournal Article321219100032Q15302