Electrical Engineering

Future wireless communications systems require better performance, lower cost, and compact RF front-end footprint. The RF front-end module development and its level of integration are, thus, continuous challenges. In most of the presently used microwave integrated circuit technologies, it is difficult to integrate the passives efficiently with required quality. Another critical obstacle in the design of passive components, which occupy the highest percentage of integrated circuit and circuit board real estate, includes the effort to reduce the module size. These issues can be addressed with multilayer substrate technology. A multilayer organic (MLO)-based process offers the potential as the next generation technology of choice for electronic packaging. It uses a cost effective process, while offering design flexibility and optimized integration due to its multilayer topology. We present the design, model, and measurement data of RF-microwave multilayer transitions and integrated passives implemented in a MLO system on package (SOP) technology. Compact, high inductors, and embedded filter designs for wireless module applications are demonstrated for the first time in this technology.

In this paper, various architectures of 3D compact microwave balanced to unbalanced (balun) transformers for Bluetooth/WiFi antenna applications are successfully designed and optimized using the Design of Experiments (DOE) approach. Two different multilayer topologies, one microstrip and one stripline, are investigated on Low Temperature Co-fired Ceramic (LTCC) substrate. The design goals for both baluns are perfectly balanced outputs from 2 to 3 GHz and a resonant frequency of exactly 2.4 GHz. It is demonstrated, using only eight simulations, that perfectly balanced outputs are not possible under the given conditions in the case of the microstrip balun.

Accent is the pattern of pronunciation which can identify a speaker's linguistic, social or cultural background. It is an important source of inter-speaker variability, and a particular problem for automated speech recognition. Current approaches to identification of speaker accent may require specialised linguistic knowledge or analysis of the particular speech contrasts, and often extensive pre-processing. An accent classification system using time-based segments of Mel Frequency Cepstral Coefficients and with Support Vector Machines is studied for a small corpus of two accents of English. Over one-to four-second samples from three topics, accuracy is up to 75% to 97.5% with very high recall and precision. Its use with mis-matched content is at best 85%, with a tendency towards majority-class classification if the accent groups are significantly imbalanced.

The refractive indices of rubidium hydrogen selenate are measured for several wavelengths at room temperature, and the transmission spectrum is measured from the UV (240 nm) to the near IR (2000 nm). The orientation of the optical indicatrix with respect to the crystal axes is also determined at several wavelengths in the visible range. Using a new method based on the deflection of light by the ferroelastic domain structure, we also determine refined values for birefringence at several wavelengths. Finally, the dispersion of the three birefringences in the range 450-900 nm is deduced from polarimetric measurements. This set of results yields complete knowledge of the linear optical characteristics required for interpretation of the electro-optical and nonlinear optical properties of this compound.