Search

The combined use of multitaper estimation and wavelet thresholding has proved to be a very powerful digital- signal-processing algorithm in achieving the true power spectrum of signals that are peculiar to digital wireless communication systems and in ultimately carrying out accurate and repeatable power measurements. Nevertheless, the induced computational burden is too heavy. This paper mainly shows that GRID, which is currently one of the key technologies in deploying global-scale computing infrastructures, can make the aforementioned algorithm enforceable in a real-time context. A GRID computing-based approach for power measurement in digital wireless communication systems is proposed, which allows high levels of accuracy and repeatability in a significantly reduced measurement time. After a concise outline of the measurement algorithm, all of the adopted hardware and software strategies are described in detail. Major results concerning performance assessment and actual measurements are also given and discussed.

A novel finite-element method (FEM) floating-conductor formulation for the numerical electromagnetic analysis of electrostatic microactuators is presented and discussed in this paper. It is based on the assumption that the moving part of the microdevice can be modeled as a perfect conductor and leads to a considerable reduction of the mesh size, as the region occupied by the moving part does not need to be discretized at all. It is also shown that the developed formulation can be easily coupled with the well-known method of equivalent charges for the computation of the driving torque acting upon the rotor of the microdevice. This coupling is performed directly in terms of variables involved in the solution process of the final algebraic system and it allows acquiring torque information without any postprocessing operation. The proposed approach has been applied to the analysis of an electrostatic micromotor with radial field. When compared to simulations obtained with a commercial FEM electromagnetic code using a classical high-permittivity formulation, this method proves to be accurate and exhibits an overall reduction of preprocessing, processing and postprocessing time.

In this paper, it is shown that the problem of the electromagnetic field coupling to a line of finite length can be put into "a standard form" of dual integral equations. A general method for solving such dual integral equations is presented and discussed. It basically consists of a transformation of the first kind Fredholm equation into a linear system of algebraic equations by means of a suitable Neumann expansion of the unknown function. This procedure, carried out in the frequency domain, has allowed evaluating the current induced on the line, which has also been computed with the classical method of moment in order to have a validation of the procedure itself.