A Study on Global and Local Optimization Techniques for TCAD Analysis Tasks

Abstract

Abstract-We evaluate optimization techniques to reduce the necessary user interaction for inverse modeling applications as they are used in the technology computer-aided design field. Four optimization strategies are compared. Two well-known global optimization methods, simulated annealing and genetic optimization, a local gradient-based optimization strategy, and a combination of a local and a global method. We rate the applicability of each method in terms of the minimal achievable target value for a given number of simulation runs and in terms of the fastest convergence. A brief overview over the three used optimization algorithms is given. The optimization framework that is used to distribute the workload over a cluster of workstations is described. The actual comparison is achieved by means of an inverse modeling application that is performed for various settings of the optimization algorithms. All presented optimization algorithms are capable of evaluating several targets in parallel. The best optimization strategy that is found is used in the calibration of a model for silicon self-interstitial cluster formation and dissolution.

Numerical Simulation of Static Noise Margin for a Six-Transistor Static Random Access Memory Cell with 32nm Fin-Typed Field Effect Transistors

“…The results are used in the calculation of SNM for the SRAM with different devices. To consider the quantum mechanical effects in the mixed-mode simulation, 3D density-gradient is solved together with the drift-diffusion transport model for electrical characteristics of planar MOSFETs and SOI FinFETs [13,14,15,17,20]. The computed device characteristics are connected to the SRAM's circuit simulation.…”

“…1a, is examined with a mixedmode simulation. The three-dimensional (3D) device simulation is performed to calculate device current-voltage (I − V ) characteristics by solving a set of density-gradient drift-diffusion equations [13,14] in the solution of coupled device and circuit equations [15,16,17]. For SRAM with the 32nm planar MOSFETs and SOI FinFETs, shown in the insets of Fig.1b, SNM is explored and compared.…”

Inverse modeling of FIB milling by dose profile optimization

“…Furthermore it is desirable not only to predict the outcome of a specific experiment, but also to generate a spatial dose distribution that results in an arbitrarily chosen target topography through the process of inverse modeling. Inverse modeling is not unique to FIB simulation [11,12]. The inverse modeling approaches that have been applied to FIB milling are either quasi-stationary, or non-stationary approaches, the difference being found in the variations of the pixel dwell times with time in the non-stationary approaches.…”

Classification of analog synthesis tools based on their architecture selection mechanisms

“…Whereas a designer can make proposals based on his or her experience or on analysis, most modern methods for designing analog and mixedsignal systems include an optimization algorithm [2,12,13].…”