Narrow Results

This paper presents a survey on the topical aspects of the noise resistance in field-effect transistors (FET) at microwave frequencies. Such noise parameter represents the sensitivity of the device noise figure to the departure from the minimum noise condition and is therefore important in all low-noise applications. The performance of the noise resistance in FETs has been reviewed since the first noise modeling analysis of short-gate devices were presented in the early '70s. The authors also comment and compare their own results on this subject as obtained by extensive experimental activity in the field of noisy device characterization vs. frequency, bias and temperature conditions.

A system identification approach for the analysis of electrochemical noise data is proposed. The most common techniques, used by many investigators, are based on: (i) the ratio of sample standard deviations, which gives no information about the frequency dependence of the electrode impedance, or (ii) Power Spectral Density estimates, which deliver the modulus of the spectrum with large variations at the lowest frequencies. Phase is, almost invariably, not included. In this work, the electrochemical cell is modeled by an input-output model. With the application of system identification techniques, it is possible to identify values of the parameters of the system model. It is shown that this approach delivers a description of the system under study with: smooth electrode impedance curves, and magnitude and phase information. Some results obtained with the most common electrochemical noise analysis techniques are presented for comparison with the proposed approach. A theoretical limitation of the proposed approach appears if a perfect symmetry between both electrodes is considered.