Optimum Source Reflection Coefficient Research Articles

Modeling of high‐frequency noise of silicon CMOS transistors for RFIC design

SUMMARYThis work presents high‐frequency noise measurements and compact modeling for a 90‐nm silicon complementary metal–oxide–semiconductor (CMOS) technology in terms of radio frequency (RF) figures of merit (FoMs). Minimum noise figure (NFmin), equivalent noise resistance (Rn), optimum source reflection coefficient (Γopt), thermal noise excess factor and a recently introduced FoM for common‐source low‐noise amplifier (LNA) design are presented from a circuit design perspective. For this purpose, the behavior of the above mentioned RF FoMs is investigated over the level of channel inversion, providing insight from a circuit design point of view. The EKV3 MOSFET advanced compact model has been used in Spectre simulator, and results are validated over a large range of frequencies, channel lengths and bias points, for both N‐type and P‐type MOS devices. Optimum performance is shown to be shifted from higher to lower levels of inversion in moderate inversion, when scaling from 240 nm down to 100 nm. This is of great significance considering that the demand for low‐power RF circuits becomes more and more imperative, as planar silicon CMOS technology is scaled to the deca‐nanometer regime. Copyright © 2014 John Wiley & Sons, Ltd.

Analytical Modeling of High-Frequency Noise Including Temperature Effects in GaN HEMTs on High-Resistivity Si Substrates

In this paper, analytical modeling of high-frequency noise was carried out including temperature effects in AlGaN/GaN high electron mobility transistors (HEMTs) on high-resistivity Si substrates over a wide temperature range from -50°C to 200°C. The device's microwave S-parameters and overall noise parameters, including minimum noise figure (NFmin), equivalent noise resistance (Rn) , and optimum source reflection coefficient (|⌈opt| and ∠⌈opt) were measured over the whole temperature range. Its small signal equivalent circuit parameters (ECPs) and internal noise source coefficients (P, R, and C) were extracted at each measurement temperature and their temperature variances were fitted using a quadratic relationship. An analytical model of the overall noise parameters is proposed based on Pucel's PRC theory. It is compared with two other commonly used analytical models and verified with the measured data, including temperature dependence. The feedback capacitance Cgd was found to be important to accurately simulate all the measured noise parameters over temperature.

A matrix formulation for noise transduction as a general case of noise measure

Conventional noise characteristics of an active device or circuit are given by the minimum noise figure, the optimum source reflection coefficient, and a noise resistance. The noise measure proposed extends the noise figure to include the available gain of the network, for the case of a conjugate output match, and gives values that minimize the noise power available from the network consistent with maximum available gain. Noise transduction follows as a general case of the noise measure by using the transducer gain with any output load. A new noise equation is derived that is solved for the minimum transduced noise, the optimum source impedance, and the optimum load impedance. These equations minimize the noise power delivered to the load impedance consistent with a maximum in the transducer gain.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>