The development of a shear horizontal surface acoustic wave (SH-SAW) liquid sensor with identification (ID) for wireless applications is described. Moreover, a novel feature of "write capability" for this sensor is proposed. The SH-SAW passive sensor has a delay line configuration with a sensing area. The sensor is fabricated on a lithium tantalate (LiTaO3) substrate, which is Y-cut with rotations specified between about 36° around the X-axis. The principle of detection is based on the attenuation and time delay between the signals. The sensor is encoded according to partial reflections of an interrogation signal by metal reflectors placed on both side of the interdigital transducers (IDT). The structure of the sensor, the principle of the wireless communication, the detection process and testing results are presented.

A semiconductor device noise model in the framework of semiclassical transport and Pauli's exclusion principle is presented. Terminal current noise is modeled as a direct consequence of electron scattering taking place inside the device at the microscopic level. The approach directly connects electron scattering rates of semiclassical transport theory with the current spectral density at the device terminals. It is shown that the spectral density of steady-state current fluctuations can be obtained from the transient solution of the Boltzmann transport equation with special initial conditions. This formulation is inherently suitable for deterministic solution techniques, for instance, the computationally efficient spherical harmonics method. Approximating the instantaneous value of the occupation number by the occupation probability, this model is able to account for Pauli's principle and at the same time describe the behavior of the electron ensemble in terms of independent entities. As a practical demonstration, the model is employed to compute the current noise spectral density due to generation recombination and acoustic and optical phonon scattering for bulk n-type silicon material. Additionally, in order to add more physical insight and to verify results, the model is also employed to compute the low-frequency current spectral density as a function of the electric field and temperature, respectively. The results show good agreement with low-frequency noise measurements reported in literature.

This paper presents a new parallel-in-time algorithm for the two dimensional transient simulation of SOI devices. With this approach, simulation in both space and time domains is performed in parallel As a result, the CPU time is reduced significantly from the conventional serial-in-time method. This new approach fully exploits the inherent parallelism of the finite difference formulation of the basic semiconductor device equations and the massively parallel architecture of SIMD computers. The space domain computations are inherently parallel due to the nature of our technique of solving the finite-difference equations. Time domain parallelism is achieved by shifting the potentials from previous time points to subsequent points one-step forward along the time axis with each Gummel iteration. This algorithm employs a fixed-point iteration technique, therefore a direct solution of matrix equations is avoided. The algorithm is especially suitable for the transient simulation of SOI devices that exhibit transient drain current overshoot. Numerical experiments show that the new parallel-in-time method is up to eight times faster than the conventional serial-in-time method in SOI transient simulations. The program is coded in CM Fortran for the Connection Machine.>