Intrinsic noise characteristics of gallium nitride high electron mobility transistors

Wide bandgap gallium nitride high electron mobility transistors (GaN HEMTs) have recently been developed for microwave sources and amplifiers. An experimental and numerical modeling effort is presented to study noise mechanisms within GaN HEMTs. This allows an equivalent-circuit model to be established, facilitating the extraction of intrinsic noise and also microwave circuit design. Intrinsic noise sources in GaN HEMTs are extracted and studied. Using measured microwave noise and scattering parameter data, the gate and drain noise sources and their correlation are determined using a small-signal equivalent-circuit representation. This model correctly predicts the measured frequency-dependent noise. Three noise mechanisms are identified in these devices, namely, those due to velocity fluctuation, gate leakage, and traps. Transistor nonlinearities are examined to assess their influence on intrinsic noise properties using an equivalent-circuit model that captures all physical noise sources. The noise sources in the nonlinear equivalent-circuit model are determined using low-frequency spectrum analyzer and microwave noise figure-meter data, while the other circuit elements are extracted from bias-dependent scattering parameter measurements and analytic nonlinear parameter expressions. This model correctly predicts the near-carrier sideband amplitude and phase noise simultaneously when the device is operated in the nonlinear regime. A numerical approach to simulate the intrinsic noise sources within transistors is described, and the impact of spatial correlation between local fluctuations is investigated. Using a two-dimensional numerical device solver, spectral densities for the gate and drain noise current sources and their correlation are evaluated using a Green\u27s function approach, an equivalent of Shockley\u27s impedance field method. Case studies with an AlGaN/GaN high electron mobility transistor are supported by measurement data. While local velocity fluctuations are correlated, the assumption of using uncorrelated diffusion noise source is found to be valid for the FET geometry studied