• Acoustic transparency and opacity using Fano Interferences in Metamaterials

      Khelif, A.; Amin, A. Elayouch M.; Farhat, M.; Addouche, M.; Bagci, Hakan (2015-08-04) [Presentation]
      We investigate both experimentally and theoretically how to generate the acoustical analogue of the Electromagnetically Induced Transparency. This phenomenon arises from Fano resonances originating from constructive and destructive interferences of a narrow discrete resonance with a broad spectral line or continuum. Measurements were realized on a double-cavity structure by using a Kundt’s Tube. Transmission properties reveal an asymmetric lineshape of the transmission that leads to acoustic transparency.
    • Adaptive multi-channel downlink assignment for overloaded spectrum-shared multi-antenna overlaid cellular networks

      Radaydeh, Redha Mahmoud; Alouini, Mohamed-Slim; Qaraqe, Khalid (Qatar Foundation Annual Research Forum Proceedings, Hamad bin Khalifa University Press (HBKU Press), 2012-10-19) [Presentation]
      Overlaid cellular technology has been considered as a promising candidate to enhance the capacity and extend the coverage of cellular networks, particularly indoors. The deployment of small cells (e.g. femtocells and/or picocells) in an overlaid setup is expected to reduce the operational power and to function satisfactorily with the existing cellular architecture. Among the possible deployments of small-cell access points is to manage many of them to serve specific spatial locations, while reusing the available spectrum universally. This contribution considers the aforementioned scenario with the objective to serve as many active users as possible when the available downlink spectrum is overloaded. The case study is motivated by the importance of realizing universal resource sharing in overlaid networks, while reducing the load of distributing available resources, satisfying downlink multi-channel assignment, controlling the aggregate level of interference, and maintaining desired design/operation requirements. These objectives need to be achieved in distributed manner in each spatial space with as low processing load as possible when the feedback links are capacity-limited, multiple small-cell access points can be shared, and data exchange between access points can not be coordinated. This contribution is summarized as follows. An adaptive downlink multi-channel assignment scheme when multiple co-channel and shared small-cell access points are allocated to serve active users is proposed. It is assumed that the deployed access points employ isotropic antenna arrays of arbitrary sizes, operate using the open-access strategy, and transmit on shared physical channels simultaneously. Moreover, each active user can be served by a single transmit channel per each access point at a time, and can sense the concurrent interference level associated with each transmit antenna channel non-coherently. The proposed scheme aims to identify a suitable subset of transmit channels in operating access points such that certain limits on the aggregate interference or number of serving access points are satisfied, while reducing the load of processing. The applicability of the results for some scenarios, including the identification of interference-free channels in operating access points is explained. Numerical and simulations results are shown to clarify achieved gains with the use of the proposed scheme under various operating conditions.
    • Adaptive Strategies in Date Palm Revealed by Confocal Imaging Technologies

      Xiao, Ting Ting; Blilou, Ikram (2018-06-22) [Presentation]
      Date palm are confronted by harsh environmental conditions and have therefore adapted various strat-egies to survive the hostile environment. To unravel the underlying mechanisms of adaptation to desert conditions we conducted a detailed analysis of date palm tissue anatomy at different developmental stag-es. Using confocal imaging we reveal new anatomical features and complex structures in roots, shoot and leaves explaining strategies of adaptation of date palm to desert conditions.
    • Advances in Spectral Methods for UQ in Incompressible Navier-Stokes Equations

      Le Maitre, Olivier (2014-01-06) [Presentation]
      In this talk, I will present two recent contributions to the development of efficient methodologies for uncertainty propagation in the incompressible Navier-Stokes equations. The first one concerns the reduced basis approximation of stochastic steady solutions, using Proper Generalized Decompositions (PGD). An Arnoldi problem is projected to obtain a low dimensional Galerkin problem. The construction then amounts to the resolution of a sequence of uncoupled deterministic Navier-Stokes like problem and simple quadratic stochastic problems, followed by the resolution of a low-dimensional coupled quadratic stochastic problem, with a resulting complexity which has to be contrasted with the dimension of the whole Galerkin problem for classical spectral approaches. An efficient algorithm for the approximation of the stochastic pressure field is also proposed. Computations are presented for uncertain viscosity and forcing term to demonstrate the effectiveness of the reduced method. The second contribution concerns the computation of stochastic periodic solutions to the Navier-Stokes equations. The objective is to circumvent the well-known limitation of spectral methods for long-time integration. We propose to directly determine the stochastic limit-cycles through the definition of its stochastic period and an initial condition over the cycle. A modified Newton method is constructed to compute iteratively both the period and initial conditions. Owing to the periodic character of the solution, and by introducing an appropriate time-scaling, the solution can be approximated using low-degree polynomial expansions with large computational saving as a result. The methodology is illustrated for the von-Karman flow around a cylinder with stochastic inflow conditions.
    • Application of hierarchical matrices for partial inverse

      Litvinenko, Alexander (2013-11-26) [Presentation]
      In this work we combine hierarchical matrix techniques (Hackbusch, 1999) and domain decomposition methods to obtain fast and efficient algorithms for the solution of multiscale problems. This combination results in the hierarchical domain decomposition (HDD) method, which can be applied for solution multi-scale problems. Multiscale problems are problems that require the use of different length scales. Using only the finest scale is very expensive, if not impossible, in computational time and memory. Domain decomposition methods decompose the complete problem into smaller systems of equations corresponding to boundary value problems in subdomains. Then fast solvers can be applied to each subdomain. Subproblems in subdomains are independent, much smaller and require less computational resources as the initial problem.
    • Application of Parallel Hierarchical Matrices and Low-Rank Tensors in Spatial Statistics and Parameter Identification

      Litvinenko, Alexander (2018-03-12) [Presentation]
      Part 1: Parallel H-matrices in spatial statistics 1. Motivation: improve statistical model 2. Tools: Hierarchical matrices 3. Matern covariance function and joint Gaussian likelihood 4. Identification of unknown parameters via maximizing Gaussian log-likelihood 5. Implementation with HLIBPro. Part 2: Low-rank Tucker tensor methods in spatial statistics
    • Application of Parallel Hierarchical Matrices in Spatial Statistics and Parameter Identification

      Litvinenko, Alexander (2018-04-20) [Presentation]
      Parallel H-matrices in spatial statistics 1. Motivation: improve statistical model 2. Tools: Hierarchical matrices [Hackbusch 1999] 3. Matern covariance function and joint Gaussian likelihood 4. Identification of unknown parameters via maximizing Gaussian log-likelihood 5. Implementation with HLIBPro
    • Bayesian techniques for fatigue life prediction and for inference in linear time dependent PDEs

      Scavino, Marco (2016-01-08) [Presentation]
      In this talk we introduce first the main characteristics of a systematic statistical approach to model calibration, model selection and model ranking when stress-life data are drawn from a collection of records of fatigue experiments. Focusing on Bayesian prediction assessment, we consider fatigue-limit models and random fatigue-limit models under different a priori assumptions. In the second part of the talk, we present a hierarchical Bayesian technique for the inference of the coefficients of time dependent linear PDEs, under the assumption that noisy measurements are available in both the interior of a domain of interest and from boundary conditions. We present a computational technique based on the marginalization of the contribution of the boundary parameters and apply it to inverse heat conduction problems.
    • Bio-optical characterization in an ultra-oligotrophic region: the North central Red Sea

      Kheireddine, Malika; Jones, Burton (2015-04) [Presentation]
      Until recently, satellite-derived ocean color observations have been the only means of evaluating optical variability of the Red Sea. During a cruise in autumn 2014, we investigated the variability of Inherent Optical Properties (IOPs) in the North Central Red Sea (NCRS) with a particular focus on the particulate backscattering coefficient, bbp, and colored dissolved organic matter, CDOM, absorption. To our knowledge, these are some of the measurements of these properties in the Red Sea. The IOPs are derived from the concentration and physical properties of suspended particles in the ocean. They provide a simple description of the influence of these particles on the light within the water column. Bio-optical relationships found for ultra-oligotrophic waters of the six stations sampled significantly depart from the mean standard relationships provided for the global ocean, showing the peculiar character of the Red Sea. These optical anomalies relate to the specific biological and environmental conditions occurring in the Red Sea ecosystem. Specifically, the surface specific phytoplankton absorption coefficients are lower than the values predicted from the global relationships due to a high proportion of relatively large sized phytoplankton. Conversely, bbp values are much higher than the mean standard values for a given chlorophyll-a concentration, Chl a. This presumably results from the influence of highly refractive submicrometer particles of Saharan origin in the surface layer of the water column.
    • Bromination of graphene: a new route to making high performance transparent conducting electrodes with low optical losses

      Mansour, Ahmed; Amassian, Aram; Tanielian, Minas H. (Carbon Nanotubes, Graphene, and Emerging 2D Materials for Electronic and Photonic Devices VIII, SPIE-Intl Soc Optical Eng, 2015-09-03) [Presentation]
      The high optical transmittance, electrical conductivity, flexibility and chemical stability of graphene have triggered great interest in its application as a transparent conducting electrode material and as a potential replacement for indium doped tin oxide. However, currently available large scale production methods such as chemical vapor deposition produce polycrystalline graphene, and require additional transfer process which further introduces defects and impurities resulting in a significant increase in its sheet resistance. Doping of graphene with foreign atoms has been a popular route for reducing its sheet resistance which typically comes at a significant loss in optical transmission. Herein, we report the successful bromine doping of graphene resulting in air-stable transparent conducting electrodes with up to 80% reduction of sheet resistance reaching ~180 Ω/ at the cost of 2-3% loss of optical transmission in case of few layer graphene and 0.8% in case of single layer graphene. The remarkably low tradeoff in optical transparency leads to the highest enhancements in figure of merit reported thus far. Furthermore, our results show a controlled increase in the workfunction up to 0.3 eV with the bromine content. These results should help pave the way for further development of graphene as potentially a highly transparent substitute to other transparent conducting electrodes in optoelectronic devices.
    • Changing Dimensions of Librarianship: users, collections, systems and services

      Vijayakumar, J.K. (2015-01) [Presentation]
      Through this presentation, I am trying to explain the developments of new generation libraries and connecting them to the contexts of an emerging spectrum of new librarianship and library transitions taking place; such as changes in user expectations, library spaces, systems, collections and marketing metrics.
    • Characterization of genomic variations in SNPs of PE_PGRS genes reveals deletions and insertions in extensively drug resistant (XDR) M. tuberculosis strains from Pakistan

      Kanji, Akbar; Hasan, Zahra; Ali, Asho M.; McNerney, Ruth; Mallard, Kim E.; Coll, Francesc; Hill-Cawthorne, Grant A.; Pain, Arnab; Nair, Mridul; Clark, Taane G.; Zaver, Ambreen; Jafri, Sana M Wasim; Hasan, Rumina (International Journal of Mycobacteriology, Elsevier BV, 2015-03) [Presentation]
      Background: Mycobacterium tuberculosis (MTB) PE_PGRS genes belong to the PE multi-gene family. Although the function of the members of the PE_PGRS multi-gene family is not yet known, it is hypothesized that the PE_PGRS genes may be associated with genetic variability. Material and methods: Whole genome sequencing analysis was performed on (n= 37) extensively drug resistant (XDR) MTB strains from Pakistan which included Central Asian (n= 23), East African Indian (n= 2), X3 (n= 1), T group (n= 3) and Orphan (n= 8) MTB strains. Results: By analyzing 42 PE_PGRS genes, 111 SNPs were identified, of which 13 were non-synonymous SNPs (nsSNPs). The nsSNPs identified in the PE_PGRS genes were as follows: 6, 9, 10 and 55 present in each of the CAS, EAI, Orphan, T1 and X3 XDR MTB strains studied. Deletions in PE_PGRS genes: 19, 21 and 23 were observed in 7 (35.0%) CAS1 and 3 (37.5%) in Orphan XDR MTB strains, while deletions in the PE_PGRS genes: 49 and 50 were observed in 36 (95.0%) CAS1 and all CAS, CAS2 and Orphan XDR MTB strains. An insertion in PE_PGRS6 gene was observed in all CAS, EAI3 and Orphan, while insertions in the PE_PGRS genes 19 and 33 were observed in 19 (95%) CAS1 and all CAS, CAS2, EAI3 and Orphan XDR MTB strains. Conclusion: Genetic diversity in PE_PGRS genes contributes to antigenic variability and may result in increased immunogenicity of strains. This is the first study identifying variations in nsSNPs, Insertions and Deletions in the PE_PGRS genes of XDR-TB strains from Pakistan. It highlights common genetic variations which may contribute to persistence.
    • Clustering based gene expression feature selection method: A computational approach to enrich the classifier efficiency of differentially expressed genes

      Abusamra, Heba; Bajic, Vladimir B. (BMC Genomics, Springer Nature, 2016-07-20) [Presentation]
      The native nature of high dimension low sample size of gene expression data make the classification task more challenging. Therefore, feature (gene) selection become an apparent need. Selecting a meaningful and relevant genes for classifier not only decrease the computational time and cost, but also improve the classification performance. Among different approaches of feature selection methods, however most of them suffer from several problems such as lack of robustness, validation issues etc. Here, we present a new feature selection technique that takes advantage of clustering both samples and genes. Materials and methods We used leukemia gene expression dataset [1]. The effectiveness of the selected features were evaluated by four different classification methods; support vector machines, k-nearest neighbor, random forest, and linear discriminate analysis. The method evaluate the importance and relevance of each gene cluster by summing the expression level for each gene belongs to this cluster. The gene cluster consider important, if it satisfies conditions depend on thresholds and percentage otherwise eliminated. Results Initial analysis identified 7120 differentially expressed genes of leukemia (Fig. 15a), after applying our feature selection methodology we end up with specific 1117 genes discriminating two classes of leukemia (Fig. 15b). Further applying the same method with more stringent higher positive and lower negative threshold condition, number reduced to 58 genes have be tested to evaluate the effectiveness of the method (Fig. 15c). The results of the four classification methods are summarized in Table 11. Conclusions The feature selection method gave good results with minimum classification error. Our heat-map result shows distinct pattern of refines genes discriminating between two classes of leukemia.
    • CO2 Sink/Source in the Indonesian Seas

      Kartadikaria, Aditya R.; Watanabe, Atsushi; Nadaoka, Kazuo; Prayitno, Hanif; Adi, Novi; Suharsono, Suharsono; Muchtar, Muswerry; Triyulianti, Iis; Setiawan, Agus; Suratno, Suratno; Khasanah, Elly (2015-04) [Presentation]
      Two distinct CO2 sink/source characteristics appeared from the compiled observed data 1984-2013 in the tropical Indonesian seas. The western part persistently emits CO2 to the atmosphere, while the eastern is rather dynamic which emits and absorbs smaller amount of CO2 to and from atmosphere, respectively. The segregation is proximal to the virtual Wallace line, where in the continental shelf is located. Lower salinity and higher silicate condition in the western part influenced the higher pCO2 condition in Java Sea. Temperature is found to have a limited influence to control different characteristic in the west and east, but SST change of 2.0 0C during La Ninã condition effectively reduced the source amount of CO2 by 50% compared to Normal year condition. Yet, during La Ninã, higher wind speed increases CO2 flux twice compared to Normal year. In the continental shelf area where CO2 sink area is found, 29 years data showed that pCO2 trend is increasing ±0.6-3.8 μatm/year. From this study, the overall areas have a significant source of CO2 of approximately 10 - 24 μatm.
    • Computable error estimates for Monte Carlo finite element approximation of elliptic PDE with lognormal diffusion coefficients

      Hall, Eric; Haakon, Hoel; Sandberg, Mattias; Szepessy, Anders; Tempone, Raul (2016-01-09) [Presentation]
      The Monte Carlo (and Multi-level Monte Carlo) finite element method can be used to approximate observables of solutions to diffusion equations with lognormal distributed diffusion coefficients, e.g. modeling ground water flow. Typical models use lognormal diffusion coefficients with H´ older regularity of order up to 1/2 a.s. This low regularity implies that the high frequency finite element approximation error (i.e. the error from frequencies larger than the mesh frequency) is not negligible and can be larger than the computable low frequency error. We address how the total error can be estimated by the computable error.
    • Computational Design of Urban Layouts

      Wonka, Peter (2015-10-07) [Presentation]
      A fundamental challenge in computational design is to compute layouts by arranging a set of shapes. In this talk I will present recent urban modeling projects with applications in computer graphics, urban planning, and architecture. The talk will look at different scales of urban modeling (streets, floorplans, parcels). A common challenge in all these modeling problems are functional and aesthetic constraints that should be respected. The talk also highlights interesting links to geometry processing problems, such as field design and quad meshing.
    • A CRIS in the Desert: The Implementation of Pure at KAUST: A Case Study in Information Exchange

      Grenz, Daryl M.; Lery, Thibaut L.; Ward, Manus; Mastoraki, Eirini; Baessa, Mohamed A. (Procedia Computer Science, Elsevier BV, 2016-06-10) [Presentation, Conference Paper]
      The integration of research information systems with existing university processes has tended towards information exchange models in which the CRIS ingests information from existing systems and takes on functions that were previously distributed across several independent solutions. This paper draws upon the experience of the implementation of a CRIS at the King Abdullah University of Science and Technology (KAUST) to posit a model in which functions remain distributed so as to take advantage of the strengths of each system. The functions discussed include institutional reporting, publications tracking, preservation of research outputs, provision of public access, researcher identity and profiling, and metrics analysis. The systems reviewed include a CRIS (Pure), a locally developed publications tracking system, a hosted DSpace repository, a locally developed ORCID integration, and a metrics dashboard (PlumX). The interactions between these systems forms a network of services to our research community, with each node connected to several others, and we discuss how we arrived at the current arrangement, as well as its drawbacks and advantages. The still limited use of standard data exchange formats like CERIF XML is discussed as a constraint that increases the costs of adding to and maintaining the network of services. At the same time we look at how increased standardization should make this distributed approach sustainable, allowing institutions like ours to mix and match complementary systems to achieve an optimal set of research information services for our needs.
    • DAIET: a system for data aggregation inside the network

      Sapio, Amedeo; Abdelaziz, Ibrahim; Canini, Marco; Kalnis, Panos (Proceedings of the 2017 Symposium on Cloud Computing - SoCC '17, ACM Press, 2017-09-27) [Presentation]
      Many data center applications nowadays rely on distributed computation models like MapReduce and Bulk Synchronous Parallel (BSP) for data-intensive computation at scale [4]. These models scale by leveraging the partition/aggregate pattern where data and computations are distributed across many worker servers, each performing part of the computation. A communication phase is needed each time workers need to synchronize the computation and, at last, to produce the final output. In these applications, the network communication costs can be one of the dominant scalability bottlenecks especially in case of multi-stage or iterative computations [1].