The generalizability of a model linking illness characteristics to psychosocial well-being was tested in a cross-sectional study of 237 adults with type 2 diabetes. It was hypothesized that diabetic complications increase illness intrusiveness, which in turn increases depressive symptomatology either directly or indirectly by reducing personal control over health outcomes. Illness intrusiveness was defined as the result of disruptions of valued activities and interests due to constraints imposed by the illness. An excellent fit of this model to the data was found using structural equation modeling. The model explained 65% of the variance in depressive symptomatology. Assessment of an alternative model excluding personal control suggested that the extent to which diabetes intrudes in life, rather than diabetic complications per se or personal control, is a key factor in relation to depressive symptomatology in individuals with diabetes.

Cell Broadband Engine is a microprocessor that has very high processing performance and very high speed I/O to communicate with other devices such as system LSI and memory chips. Design optimization with statistical analysis was carried out.DOE matrix was made with dimension of signal traces and dielectric constant of the insulator in the package substrate. After major factors were detected, random sampling was done to make sure that the distribution of characteristic impedance of signal traces is within the specification. Passive characterization was carried out with test vehicle. Insertion loss and characteristic impedance is dependent on both temperature and water absorption. These environmental factors need to be taken into account when the design rule is determined. After design rule is fixed, circuit simulation for whole signal channel was carried out with considering impedance tolerance in both the package and the printed circuit board.DOE matrix was made and analyzed to determine the major factors on mid-frequency and low-frequency noise in power distribution. Inductance of package substrate and decoupling capacitor are the major effect on the midfrequency noise, and capacitance of on-module and on-PCB affect low frequency noise. The effect of on-chip parameters was also evaluated. Several types of capacitors were characterized to measure their parasitic parameters. 1.. IntroductionCell Broadband Engine is a multi-core microprocessor based on the state-of-the-art architecture. Details of the chip architecture and internal performance are described in otherFrom a packaging standpoint, Cell Broadband Engine is very challenging for both signal and power integrity. Cell Broadband Engine needs to communicate with other system LSIs and memory chips for input and output of processed data. The bandwidth of chip-to-chip and memory interface is quite large compared to other devices. It needs to increase both the I/O frequency and the number of signal pins. [4] Signal density in the package substrate needs to be increased. It will cause more coupling noise, and degrade both timing and voltage margin. Since I/O frequency is so high, noise issue is very critical to assure the operation of high performance microprocessor.Noise issue is also critical for power distribution design. Because of the high performance of Cell Broadband Engine, operating frequency of processing units is also increased. It means that simultaneous switching noise due to current switching in the processor core will be increased. However, operating voltage tends to be lowered to decrease power consumption. Package designers have to manage these two opposed issues.As stated above, the allowable margin to assure the performance is being decreased. For instance, characteristic impedance of high speed signals needs to be tightly controlled to avoid the loss caused by the reflection at the boundary of package and printed circuit board (PCB). Tight control of dimensions of signal trace is required, but too tight control will affect manufacturing yield a...

This work concerns the electrical properties of waveguide applicators consisting of flanged rectangular waveguides filled with a dielectric, used in medical applications (microwave thermography and local hyperthermia). The reflection coefficient and the near field configuration in lossy materials were obtained in some cases analytically and in some other cases numerically. The validity of these methods was verified experimentally. The study shows that the matching of the applicator and the penetration depth in a lossy material, such as a living tissue, depend not only on the tissue characteristics but also on the characteristics of the applicator itself.

No abstract

An experimental-analytical technique was developed to evaluate the noise produced inside various types of sonar domes under the effect of a turbulent flow, in a frequency range where the structure has a high modal density. The transfer functions between a punctual external force applied to the dome and the acoustic pressure at a point inside the cavity was measured on a scale model. This measure was made using a reciprocity technique with emitting hydrophones within the cavity and with accelerometers all over the dome's external surface. The noise level at a point inside the cavity is calculated from these experimental transfer functions and a turbulent boundary layer wall pressure model. The wave vector frequency spectrum model used is Chase's model [D. M. Chase, J. Sound Vib. 70, 29–67 (1980)], adjusted according to the local parameters of the flow. The results obtained with this method agree with experimental results. In order to test various types of structure, this technique on a scale model has considerable advantages compared with experimentations on a real structure: limited expense, experimental ease, and reliability.