F. Franze

Using a coordination language to specify and analyze systems containing mobile components

New computing paradigms for network-aware applications need specification languages able to deal with the features of mobile code-based systems. A coordination language provides a formal framework in which the interaction of active entities can be expressed. A coordination language deals with the creation and destruction of code or complex agents, their communication activites, as well as their distribution and mobility in space. We show how the coordination language PoliS offers a flexible basis for the description and the automatic analysis of architectures of systems including mobile entities. Polis is based on multiple tuple spaces and offers a basis for defining, studying, and controlling mobility as it allows decoupling mobile entities from their environments both in space and in time. The pattern-matching mechanism adopted for communication helps in abstracting from addressing issues. We have developed a model-checking technique for the automatic analysis of PoliS specifications. In the article we show how this technique can be applied to mobile code-based systems

A coordination model to specify systems including mobile agents

A coordination model provides a formal framework in which the interaction of active entities that we call agents can be expressed. A coordination model deals with the creation and destruction of agents, their communication activities, their distribution and mobility in space, as well as the synchronization and distribution of their actions over time. We show how a coordination model called PoliS offers a flexible basis for the description and the analysis of architectures of systems including mobile agents. We have developed a model checking technique for the automatic analysis of PoliS specifications.

Thermofluid Analysis of Ultra Low Power Hotplates for a MOX Gas Sensing Device

This work presents a full three-dimensional finite-element multiphysics simulation of the conjugate heat transfer for a gas sensing device composed by a two-element array of ultra low power (ULP) metal oxide semiconductor (MOX) sensors operated in a miniaturized sampling chamber. The heat equation in a solid, the Poisson equation for the electric potential and the incompressible Navier-Stokes and energy equations for a fluid have been solved in a coupled manner. Validation of the simulation results has been performed comparing the simulated power dissipated by the array with a set of experimental data under different operating conditions. A maximum relative error of less than 7% between the simulations and the experiments has been obtained without application of any fitting strategy on the physical properties. A negligible effect on the power dissipated by the sensor, in presence of volumetric fluxes in the sampling chamber, has been observed both numerically and experimentally. Finally, a real operational condition has been simulated and examined.

Construction of VHDL-AMS simulator in Matlab

We describe the digital kernel implementation of the simulator S.A.M.S.A., a tool for the simulation of VHDL-AMS systems in Matlab/sup /spl trade//. The digital kernel was validated by simulating different systems. In particular we will show the simulation of a low-power multistage decimator for a wideband /spl Delta/-/spl Sigma/ analog to digital converter (ADC). This example was correctly simulated given the same results as other VHDL commercial tools.

3D Mesh Generation with Wavelet-Driven Adaptivity

In this paper we show the effectiveness on 3-dimensional domains of a wavelet-based adaptive method (WAM), which is able to drive a progressive adaptation of computational meshes suited for semiconductor device simulation in order to capture sensible regions. An improved two-step wavelet analysis is performed on some relevant physical variables, allowing for a highly selective refinement of domain regions with stringent resolution requirements for simulation accuracy; moreover, a mesh quality control algorithm ensures a smooth grading of element sizes and eliminates bad configurations affecting convergence. Complex real structures can thus be handled with negligible computational overhead and with no skilled control from the user. Simulation results related to a 3D p-n diode and to different nMOSFET driver geometries demonstrate the capability of the proposed automatic tool to ensure good convergence and accuracy properties with considerable advantages over reference manually-constructed meshes of much larger size

Automatic Wavelet Localization and Adaptive Meshing of Physical Relevances in Device Simulation

In this paper we report how a wavelet-based adaptive method (WAM) able to automatically detect sensible regions is integrated within CAD softwares for device simulation. The wavelets localization property drives a procedure which constructs an anisotropic grid, whose density remarks the internal structure of physical relevances. The mesh generation is performed considering the necessity of a Delaunay triangulation which avoids the presence of obtuse angles: for this reason Steiner points are inserted. Ad hoc procedures render the adaptation gradual: there is no redistribution of grid points and yet the passage between refined and coarse regions is obtained with a smooth grading. Such a grid construction produces smooth characteristics, even for a fully dynamic adaptation in critical quasi-stationary simulations. 2D practical examples are discussed to validate the proposed approach

Wavelet-based adaptive mesh generation for device simulation

In this paper we propose a new method based on wavelet analysis, to define the discretization grid used to solve semiconductor devices PDE systems. We will show that the achieved adaptively refined grid, can be fruitfully used to get a solution with the same level of accuracy of a reference case, which has a considerable larger number of points. The proposed algorithm is given by an automatic procedure that requires neither the user-feedback control nor an in-depth physical knowledge of the problem to be solved, and it is enough accurate to describe all important physical effects encountered in a number of practical situations. 2D practical examples will be discussed (a power p-n junction and a planar MOSFET) to demonstrate (i) how the wavelet-refined grids preserve geometrical and physical consistency, and (ii) the efficiency and reliability of the proposed approach. It seems also useful to notice that, due of its simplicity, the proposed approach can be readily and efficiently extended to perform multidimensional analysis.

3D simulation of conjugate heat transfer of ULP hotplates for a MOX gas sensing device

This work presents a full three dimensional finite element multi-physics simulation of the conjugate heat transfer for a gas sensing device composed by a two-element array of Ultra Low Power (ULP) Metal Oxide Semiconductor (MOX) sensors operated in a miniaturized sampling chamber. The heat equation in a solid, the Poisson equation for the electric potential and the incompressible Navier-Stokes and energy equations for a fluid have been solved in a coupled manner. Validation of the simulation results has been performed comparing the power dissipated by the array with a set of experimental data under different operating conditions. Good agreement between the simulations and the experiments has been obtained. A small effect on the power dissipated by the sensor, in presence of volumetric fluxes in the sampling chamber, has been observed both numerically and experimentally.

Wavelet Adaptivity for 3-D Device Simulation

A new technique has been implemented with the aim of providing an adaptive tool for the generation of computational meshes for 3-D semiconductor device simulation. The core of the proposed wavelet-based adaptive method (WAM) is a refinement algorithm based on the wavelet transform, which gives an estimation of solution regularity and progressively adapts the grid by increasing the resolution only in regions where the important physical phenomena take place. WAM is inserted into a validation tool that provides the interfacing filters with both the solver and a meshing engine. Additional features assure the quality of the generated meshes by increasing the selectivity properties of the refinement tool, preventing numerical instabilities or artifacts, such as abrupt variations of I-V curves between successive bias points during quasi-stationary simulations. Simulation results are provided, which show the effectiveness of the proposed approach as a means to guide the automatic refinement of the discretization grid, preserving accuracy with negligible computational overhead and no skilled control from the user.

A tool for the integration of new VHDL-AMS models in SPICE

In this work we describe a tool for VHDL-AMS modeling within SPICE based simulators. The user can write new SPICE models using VHDL-AMS standard, the proposed tool will automatically build the code to be embedded inside the SPICE simulator. A Smart Power band-gap reference voltage circuit and a recipe to build an advanced MOS model are presented as examples to demonstrate the effectiveness of our approach.