Compact models for transient response of distortionless distributed resistance, inductance, capacitance and conductance (RLCG) interconnects are rigorously derived with clear physical insight. Signal integrity (SI) of distortionless interconnects including steady-state voltage and settling time is analyzed in detail while SI enhancement schemes in practical circuit design are also proposed.

An analytic potential model for undoped cylindrical gate-all-around (GAA) MOSFET with asymmetric gate underlap is developed in this paper. This model is derived by solving the Poisson's equation with the parabolic potential approximation, channel length transformation and conformal mapping. The analytic body center potential solution is presented. Compared with TCAD simulations, the proposed model shows good agreements, with different dimensions of the structure and varied bias conditions. The model here is appropriate for predicting the effect of gate misalignment or asymmetric gate underlap in GAA MOSFETs' design.

In this paper, a new iterative grid selection strategy for the Time-Mapped Harmonic Balance (TMHB) method is presented. This strategy is referred to the adaptive grid methods, which tracks the rapid transitions and constructs a smooth time-map function. The theoretical analysis and simulation results of this grid selection strategy are presented. The results demonstrate that this new strategy improves the numerical stability of the TMHB method and achieves accurate results based on the improved TMHB method.

In this paper we provide an insight into the drain current model for long-channel MOSFET devices. A new method to perform the integral of the rigorous Pao-Sah dual integral current is derived. From it, we demonstrate the error of the traditional charge sheet models in predicting the drain current compared with Pao-Sah's dual integral model, also provide the reason that Brews' charge sheet model fails to pass the self consistency tests reported previously. Three charge-sheet approximation models are tested in order to find a simple yet accurate drain current model for surface potential-based compact models.

In this paper, the band-structure of ultra-thin body (UTB) double-gate (DG) MOSFETs is calculated by empirical tight-binding model, and the the nonparabolicity effects are included in the model by the modified Schrodinger equation with the fitting parameters which are extracted from the band-structure by expanding the dispersion relationship in a power series up to the third order. The characteristics of UTB-DG MOSFET current are simulated and the results are compared by the non-equilibrium Green's function (NEGF) transport theory based on effective mass approximation (EMA), parabolic quantum confined masses (PQM) and nonparabolicity effect model (NPE). It is shown that the saturation current is overestimated in the models based on EMA and PQM compared with the model including NPE, and the value rises to 4% when Tsi equals to 3nm. The difference between NPE model and PQM model increases with increment of Vg, but less sensitive to Vds.

In this paper, a compact model for transient response of the dispersionless interconnects is rigorously derived with resistive, inductive and capacitive load terminations. The proposed compact models are verified by the HSPICE simulation result to be highly accurate. And the analysis of signal integrity such as delay and noise for the dispersionless interconnect is also demonstrated by the mans of the developed compact model.

The random dopant fluctuation (RDF) of double gate (DG) MOSFET based 6-T SRAM is investigated with three-dimensional (3-D) statistical simulation. The doping profile is generated by matlab and the threshold voltage variation due to RDF is obtained by device simulation. Then the performance of DG MOSFET based 6-T SRAM is evaluated by feeding the results into a compact DG MOSFET model using HSPICE Monte Carlo simulation. The results show that pull down transistor dominates static noise margin (SNM) fluctuation and access transistor dominates write margin (WM) fluctuation.

>The forward gated-diode method is used to extract the dielectric oxide thickness and body doping concentration of MOSFETs,especially when both of the variables are unknown previously.First,the dielectric oxide thickness and the body doping concentration as a function of forward gated-diode peak recombination-generation (R-G) current are derived from the device physics.Then the peak R-G current characteristics of the MOSFETs with different dielectric oxide thicknesses and body doping concentrations are simulated with ISE-Dessis for parameter extraction.The results from the simulation data demonstrate excellent agreement with those extracted from the forward gated-diode method.