Identification of biological neurons using adaptive observers

摘要

This paper is to investigate the use of adaptiveobservers for the modeling of biological neurons and net-works. Assuming that a neuron can be modeled as acontinuous-time nonlinear system, it is possible to deter-mine its unknown parameters using adaptive observer,based on the concept of adaptive synchronization. Thesame technique can be extended for the identification of anentire biological neural network. Some conventionalobserver designs are studied in this paper and satisfactoryresults are obtained, yet with some restrictions. To furtherextend the applicability of adaptive observers for themodeling process, a new design is suggested. It is based ona combination of linear feedback control approach and thedynamical minimization algorithm. The effectiveness ofthe designed adaptive observer is confirmed withsimulations.