This paper investigates the presence and the effects of multiple minim
a in the loss function to be minimized in time-delay estimation method
s based on least squares. It is shown how in most cases a low-pass fil
tering action is required on data used in the identification phase in
order to estimate the proper delay value. The results presented are de
rived by means of a frequency-domain interpretation of the least-squar
es technique. The paper does not focus on a specific estimation algori
thm, rather it presents a scenario to accommodate many recently publis
hed identification methods. Some numerical examples illustrate the mai
n results.