“…For the DD model this slow convergence has been frequently reported e.g. [9] and it persists for the HD model, where the coupling between the equations is even stronger as will be discussed later. Hence some authors e.g.…”
Section: Nonlinear Relaxation Schemessupporting
confidence: 60%
“…As discussed already in [9] this is sufficient to resolve the coupling within the inversion channel.…”
Section: A Solution With An Accuracy Of About 4 (Or About 100 MV Fomentioning
Until now all nonlinear relaxation schemes that have been used (e.g.[1],[6],[7],[8]) for hydrodynamic MOS simulations showed only low convergence speed. In this paper a new fast converging non1inea.r relaxation scheme for this class of simulation problems is presented.
“…For the DD model this slow convergence has been frequently reported e.g. [9] and it persists for the HD model, where the coupling between the equations is even stronger as will be discussed later. Hence some authors e.g.…”
Section: Nonlinear Relaxation Schemessupporting
confidence: 60%
“…As discussed already in [9] this is sufficient to resolve the coupling within the inversion channel.…”
Section: A Solution With An Accuracy Of About 4 (Or About 100 MV Fomentioning
Until now all nonlinear relaxation schemes that have been used (e.g.[1],[6],[7],[8]) for hydrodynamic MOS simulations showed only low convergence speed. In this paper a new fast converging non1inea.r relaxation scheme for this class of simulation problems is presented.
“…This approach converges asymptotically linear with the simulation time similar to the convergence behavior of the Gummel loops in the drift-diffusion based TCAD device simulators [1,15]. Thus the proposed approach is much faster than a Monte Carlo algorithm which gains numerical accuracy proportional to the square root of the CPU time [16].…”
mentioning
confidence: 59%
“…This approach converges linearly with simulation time similar to the Gummel loop in the classical drift-diffusion based TCAD device simulators [1,15], and thus much faster than an MC algorithm with its square root dependence [16]. In addition, it yields a truly stationary solution.…”
Section: Convergence Enhancement Methods For the Iteration Loopmentioning
confidence: 87%
“…[1,15]), the carrier density is defined on the direct grid and the current density on the adjoint grid. Similarly the unknowns g m of the projected BTE (55) are defined either on the direct or adjoint grid depending on the parity of m. The "densities" (m even) are defined on the direct grid and the "fluxes" (m odd) on the adjoint.…”
Section: Discretization Of the Projected Btementioning
In this paper numerical aspects of deterministic multisubband device simulations are presented for strained double gate PMOSFETs including magnetotransport. The simulations are based on a self-consistent solution of the multisubband Boltzmann transport equation (BTE), 6 × 6 k · p Schrödinger equation (SE) and Poisson equation (PE). For accurate and efficient calculation of the subband structure, an efficient discretization of the 2D k-space combined with a monotonic cubic spline interpolation is employed. The multisubband BTE is solved with a deterministic method based on a Fourier expansion of the distribution function. The Fourier series is found to converge rapidly for nanoscale double gate PMOSFETs. A convergence enhancement method for the Gummel type SE-PE-BTE loop by solving the BTE-PE simultaneously is proposed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.