Unsupervised feature selection for sensor time-series in pervasive computing applications

The paper introduces an efficient feature selection approach for multivariate time-series of heterogeneous sensor data within a pervasive computing scenario. An iterative filtering procedure is devised to reduce information redundancy measured in terms of time-series cross-correlation. The algorithm is capable of identifying nonredundant sensor sources in an unsupervised fashion even in presence of a large proportion of noisy features. In particular, the proposed feature selection process does not require expert intervention to determine the number of selected features, which is a key advancement with respect to time-series filters in the literature. The characteristic of the prosed algorithm allows enriching learning systems, in pervasive computing applications, with a fully automatized feature selection mechanism which can be triggered and performed at run time during system operation. A comparative experimental analysis on real-world data from three pervasive computing applications is provided, showing that the algorithm addresses major limitations of unsupervised filters in the literature when dealing with sensor time-series. Specifically, it is presented an assessment both in terms of reduction of time-series redundancy and in terms of preservation of informative features with respect to associated supervised learning tasks.     

Scanning Near-Field Ellipsometry Microscopy: imaging nanomaterials with resolution below the diffraction limit

We introduce a simple Scanning Near-Field Ellipsometer Microscopy (SNEM) setup to address the rapidly increasing need for simple, routine optical imaging techniques with resolution well below the diffraction limit. Our setup is based on the combination of commercially available atomic force microscope (AFM) and ellipsometry equipment with gold-coated AFM tips to obtain near-field optical images with a demonstrated resolution below λ/10. AFM topographical data, obtained in contact mode, and near-field optical data were acquired simultaneously using a combined AFM-ellipsometer. The highly enhanced field due to lightning-rod effects and localized surface plasmons excited at the end of the gold-coated tip allowed us to resolve and identify metallic nanoparticles embedded in poly(methyl methacrylate) as well as microphases in microphase-separated block copolymer films.     

Effects and Mechanism of Pre-ozonation on Sand Filtration Performance

Diatoms were shown to be responsible for clogging the sand filtration tanks in the waterworks of Beijing treating reservoir source water. The effect of pre-ozonation on the performance of the sand filter was investigated using a continuous-flow (200 L h^?1) pilot experimental system. At the optimum ozone dose of 1.0 mg L^?1, the filter run time was prolonged 2.7–2.8-fold, the number of 2–5 μm diameter particles decreased from 353–829 ml^?1 to 53–269 ml^?1 and the turbidity in the filtrate decreased from >0.2 NTU to < 0.1 NTU. Laboratory tests showed that ozone was able to break the cells of the diatoms, including Fragilaria sp., Diatoma sp. and Navicula sp., into smaller particles, and the optimum ozone dose was also 1.0 mg L^?1. Decomposition of algae is likely to be the main reason for the significantly improved performance of filtration.     

Tunable PLGA‐Based Nanoparticles Synthesized Through Free‐Radical Polymerization

A new class of PLGA‐based materials suitable for the production of biodegradable NPs via free‐radical polymerization is proposed. ROP of lactide and glycolide using HEMA is carried out; macromonomers (HEMA‐L_xG_yA) with short PLGA chain ends are successfully produced. Afterwards, monomer‐starved semi‐batch emulsion polymerization of these HEMA‐L_xG_yA macromonomers is performed to obtain NPs with a low polydispersity index and a controlled size. The stability and degradability of NPs is confirmed from degradation studies that verify the possibility of tuning the degradation time by changing macromonomer characteristics such as chain length and composition.

     

Towards integrated photonic interposers for processing octave-spanning microresonator frequency combs

Microcombs—optical frequency combs generated in microresonators—have advanced tremendously in the past decade, and are advantageous for applications in frequenc...     

Comparative study on the influence of surface characteristics on de-icing evaluation

Journal of Materials Science - A comparative study of de-icing evaluation methods was conducted in this work, and their variations in response to surface characteristics were investigated. The...