Single Module Identifiability in Linear Dynamic Networks With Partial Excitation and Measurement

Identifiability of a single module in network transfer functions is determined by whether particular function the can be uniquely distinguished within model set, on basis data. Whereas previous research has focused situations that all signals are either excited or measured, we develop generalized analysis results for situation partial measurement and excitation. As identifiability conditions typically require sufficient number external excitation signals, this article introduces novel structure such from unmeasured noise included, which leads to less conservative than relying measured only. More importantly, graphical developed verify global generic based topology dynamic network. Depending input output present four cover possible identification. These further lead synthesis approaches allocating selecting warrant identifiability. In addition, if satisfied only, indirect identification methods provide consistent estimate module. All obtained also extended multiple modules