Fiber optic light collection system for scanning-tunneling-microscope-induced light emission

We report a compact light collection scheme suitable for retrofitting a scanning tunneling microscope (STM) for STM-induced light emission experiments. The approach uses a pair of optical fibers with large core diameters and high numerical apertures to maximize light collection efficiency and to moderate the mechanical precision required for alignment. Bench tests indicate that efficiency reduction is almost entirely due to reflective losses at the fiber ends, while losses due to fiber misalignment have virtually been eliminated. Photon-map imaging with nanometer features is demonstrated on a stepped Au(111) surface with signal rates exceeding 10(4) counts/s.     

Context-Sensitive Visibility

An improvement to the so-called visual verification approach is presented. Visual verification is a method for checking the correctness of the behaviour of a reactive or concurrent system. It shares a great deal of common ground with ordinary formal state space verification, but is more user-friendly. This is because the user does not need to specify in detail the properties that the system must satisfy to be correct. Instead, the user only lists the atomic actions that are relevant for the property. Computer tools are used to obtain a graphical representation which is a summary of all possible alternative behaviours of the system, and the user then analyses the result. The improvement presented in this article allows the user to pick a region of the graphical representation and investigate it in more detail, without being overwhelmed by the details outside the region. The improvement is illustrated by analysing the livelocks in a model of the alternating bit protocol.     

Using linear algebra in decomposition of Farkas interpolants

The use of propositional logic and systems of linear inequalities over reals is a common means to model software for formal verification. Craig interpolants constitute a central building block in this setting for over-approximating reachable states, e.g. as candidates for inductive loop invariants. Interpolants for a linear system can be efficiently computed from a Simplex refutation by applying the Farkas’ lemma. However, these interpolants do not always suit the verification task—in the worst case, they can even prevent the verification algorithm from converging. This work introduces the decomposed interpolants, a fundamental extension of the Farkas interpolants, obtained by identifying and separating independent components from the interpolant structure, using methods from linear algebra. We also present an efficient polynomial algorithm to compute decomposed interpolants and analyse its properties. We experimentally show that the use of decomposed interpolants in model checking results in immediate convergence on instances where state-of-the-art approaches diverge. Moreover, since being based on the efficient Simplex method, the approach is very competitive in general.

     

An improved extreme learning machine model for the prediction of human scenarios in smart homes

One of the main objectives of smart homes is healthcare monitoring and assistance, especially for elderly and disabled people. Therefore, an accurate prediction of the inhabitant behavior is very helpful to provide the required assistance. This work aims to propose a prediction model that satisfies the accuracy as well as the rapidity of the learning phase. To do so, we propose to improve the existing extreme learning machine (ELM) model by defining a recurrent form. This form ensures a temporal relationship of inputs between observations at different time steps. The new model uses feedback connections to the input layer from the output layer which allows the output to be included in the long-term prediction. A recurrent dynamic network, with feedback connections of the output of the network, is proposed to predict the future series representing future activities of the inhabitant. The resulting model, called Recurrent Extreme Learning Machine (RELM), provides the ability to learn the human behavior and ensures a good balance between the learning time and the prediction accuracy. The input data is based on the real data representing the activities of persons belonging to the profile of first level (i.e. P ₁) as measured by the dependency model called Functional Autonomy Measurement System (SMAF) used in the geriatric domain. The experimental results reveal that the proposed RELM model requires a minimum time during the learning phase with a better performance compared to existing models.     

Breakdown of Zero-Energy Quantum Hall State in Graphene in the Light of Current Fluctuations and Shot Noise

We have investigated the cross-over from Zener tunneling of single charge carriers to avalanche type of bunched electron transport in a suspended graphene Corbino disk in the zeroth Landau level. At low bias, we find a tunneling current that follows the gyrotropic Zener tunneling behavior. At larger bias, we find an avalanche type of transport that sets in at a smaller current the larger the magnetic field is. The low-frequency noise indicates strong bunching of the electrons in the avalanches. On the basis of the measured low-frequency switching noise power, we deduce the characteristic switching rates of the avalanche sequence. The simultaneous microwave shot noise measurement also reveals intrinsic correlations within the avalanche pulses and indicate a decrease in correlations with increasing bias.     

Power saving through state retention in IGZO‐TFT AMOLED displays for wearable applications

We present a qHD (960 × 540 with three sub‐pixels) top‐emitting active‐matrix organic light‐emitting diode display with a 340‐ppi resolution using a self‐aligned IGZO thin‐film transistor backplane on polyimide foil with a humidity barrier. The back plane process flow is based on a seven‐layer photolithography process with a CD = 4 μm. We implement a 2T1C pixel engine and use a commercial source driver IC made for low‐temperature polycrystalline silicon. By using an IGZO thin‐film transistor and leveraging the extremely low off current, we can switch off the power to the source and gate driver while maintaining the image unchanged for several minutes. We demonstrate that, depending on the image content, low‐refresh operation yields reduction in power consumption of up to 50% compared with normal (continuous) operation. We show that with the further increase in resolution, the power saving through state retention will be even more significant.     

C-terminal truncations of a thermostable Bacillus stearothermophilus {alpha}-amylase

A series of truncated proteins from a thermostable Bacillusstearothermophilus -amylase was prepared to study the importanceof the extension in the C-terminus compared with other liquefyingBacillus -amylases. The mutations introducing new translationtermination sites shortened the 515 amino acid residue-longwild type enzyme by 17, 32, 47, 73 or 93 residues. The longerthe truncation, the lower the specific activity of the enzyme.Only the two longest mutant proteins were active: the specificactivity of the 498 residue variant was 97% and protein 483was 36% that of the parental enzyme. The K_m values of starchhydrolysis changed from 1.09 for wild type enzyme to 0.35 and0.21 for mutants 498 and 483, respectively, indicating alteredsubstrate binding. The mutant enzymes had almost identical pHand temperature optima with the wild type amylase, but enhancedthermal stability and altered end product profile. The consequencesof the truncation to the structure and function of the enzymeswere explored with molecular modeling. The liquefying amylasesseem to require {small tilde}480 residues to be active, whereasthe C-terminal end of B.stearothermophilus amylase is requiredfor increased activity.     

Mineralogical and elemental composition of fly ash from pilot scale fluidised bed combustion of lignite, bituminous coal, wood chips and their blends

The chemical and mineralogical composition of fly ash samples collected from different parts of a laboratory and a pilot scale CFB facility has been investigated. The fabric filter and the second cyclone of the two facilities were chosen as sampling points. The fuels used were Greek lignite (from the Florina basin), Polish coal and wood chips. Characterization of the fly ash samples was conducted by means of X-ray fluorescence (XRF), inductive coupled plasma-optical emission spectrometry (ICP-OES), thermogravimetric analysis (TGA), particle size distribution (PSD) and X-ray diffraction (XRD). According to the chemical analyses the produced fly ashes are rich in CaO. Moreover, SiO₂ is the dominant oxide in fly ash with Al₂O₃ and Fe₂O₃ found in considerable quantities. Results obtained by XRD showed that the major mineral phase of fly ash is quartz, while other mineral phases that are occurred are maghemite, hematite, periclase, rutile, gehlenite and anhydrite. The ICP-OES analysis showed rather low levels of trace elements, especially for As and Cr, in many of the ashes included in this study compared to coal ash from fluidised bed combustion in general.     

Modelling non-stationary noise with spectral factorisation in automatic speech recognition

Speech recognition systems intended for everyday use must be able to cope with a large variety of noise types and levels, including highly non-stationary multi-source mixtures. This study applies spectral factorisation algorithms and long temporal context for separating speech and noise from mixed signals. To adapt the system to varying environments, noise models are acquired from the context, or learnt from the mixture itself without prior information. We also propose methods for reducing the size of the bases used for speech and noise modelling by 20–40 times for better practical applicability. We evaluate the performance of the methods both as a standalone classifier and as a signal-enhancing front-end for external recognisers. For the CHiME noisy speech corpus containing non-stationary multi-source household noises at signal-to-noise ratios ranging from +9 to ?6 dB, we report average keyword recognition rates up to 87.8% using a single-stream sparse classification algorithm.