Communication theoretic analysis of the synaptic channel for cortical neurons

Abstract

In this paper, we develop a realistic model of the synaptic multiple-input single-output (MISO) communication channel for cortical neurons. The synaptic channel weights change adaptively according to the rules of spike timing-dependent plasticity (STDP) to enable learning and memory within neuronal connections. We calculate the ergodic capacity of the synaptic multiple-input multiple-output (MIMO) communication channel, and investigate its performance using the statistical properties of neuro-spike communication. Moreover, we analyze the communication performance of synaptic channels in terms of decoding error probability, and define a lower bound on the synaptic multiple-input single-output (MISO) communication channel.