Search

A chip-level single-event latchup (SEL) estimation methodology is introduced, and its accuracy is demonstrated on the test vehicles manufactured in a state-of-the-art technology for various process options. Novel test structures have been designed and built to make them accessible to accelerated beam testing for the purpose of calibrating the SEL rate as a function of layout design style and use condition. The calibrated model and the SEL assessment strategy are capable of accurately accounting for variations in circuitry and layouts across complex products. The 64-MeV monoenergetic proton and broad-spectrum neutron beam testing results establish that the novel compact model accurately projects full-chip SEL rates, and hence, it can be used to guide layout optimizations and mitigate (or eliminate) component-level SEL rates. [ABSTRACT FROM AUTHOR]

We report on measured neutron radiation induced soft error rates (SER) for an academic, full-custom, low-power 75 kb SRAM macro in a commercial 28 nm CMOS process protected with an adjacent-bit error correction circuit (ECC). In low voltage (0.5 V) power saving mode, measurements show an 189x improvement in SER over an unprotected baseline array and a 5x improvement over a traditional single-error-correcting-double-error-detecting (SEC-DED) code using the same number of parity check-bits. Further, simulation results show an over 2x reduction in SER compared to other multi-bit ECCs while using three fewer check-bits. Chip measurement results show that the low voltage retention mode provides a 89.7% reduction in cell leakage current/bit compared to the nominal supply voltage. All radiation measurement data has been collected at the TRIUMF Neutron Irradiation Facility in Vancouver, Canada. [ABSTRACT FROM PUBLISHER]

The objective of this study is to apply machine learning classification to predict building characteristics from electricity smart meter data for the purpose of building stock characterization. Given that there are no publicly available large-scale residential electric smart meter data sets with detailed building characteristics, an open-source virtual smart meter (VSM) data set is used. The VSM data consists of electricity consumption profiles for 200,000 homes with 21 known characteristics, which are used to train predictive models with linear discriminant analysis (LDA). The classification accuracy (CA) is determined for a variety of scenarios where the meter data aggregation and period are varied. The CA depends on the parameter to be classified (the class), the number of data points per building (the features) and the number of buildings used for classification. Reliable classification results are obtained when the number of buildings exceeds the number of features by a significant margin. An application of the developed predictive models to a small data set of 30 real houses illustrates the usefulness of the method but also the challenges in achieving a generalized model with virtual data. [ABSTRACT FROM AUTHOR]

Web-based instruction shows great promise toward enriching the student learning experience. One particular area of interest is providing tutorial material and practice problems online so that classroom lecture time can be better utilized. However, the time and cost to develop full tutoring systems can be prohibitive. The project presented in this paper shows that low-cost online modules can be developed to complement existing course delivery methods. The key to the design is limiting the type of tutoring and focusing on instructional challenges involving the repetition of concepts that are introduced in the course lectures. For introductory programming courses, these challenges primarily involve the difficulties inherent in learning the syntax of a particular programming language and gaining sufficient familiarity with programming concepts, such as loops, conditional statements, and simple algorithms. The set of online modules was developed to reduce the need for repetition of these concepts during lectures. Thus, students benefit as they can gain knowledge and comprehension of these concepts at their own pace as they actively engage the tutorials and self-check exercises. The modules were used as an enhancement for an introductory programming course taught in C++ to first year university students, some of whom had little or no programming experience. Feedback from students and instructors shows that the modules were useful and aided student learning. [ABSTRACT FROM AUTHOR]