• Numerical simulation of the magnetic cement induction response in the borehole environment

      Eltsov, Timofey; Patzek, Tadeusz W. (Society of Exploration Geophysicists, 2018-09-12)
      We present a technique for the detection of integrity of the magnetic cement behind resistive fiberglass casing. Numerical simulations show that an optimized induction logging tool allows one to detect small changes in the magnetic permeability of cement through a non-conductive casing in a vertical (or horizontal) well. Changes in magnetic permeability influence mostly the real part of the vertical component of magnetic field. The signal attenuation is sensitive to a change of magnetic properties of the cement. Our simulations show that optimum separation between the transmitter and receiver coils ranges from 0.25 to 0.6 meters, and the most suitable magnetic field frequencies vary from 0.1 to 10 kHz. Our goal is to build cheap, long-lasting, low-temperature (<150°C) geothermal wells with water recirculation.
    • Highly-Sensitive Magnetic Tunnel Junction Based Flow Cytometer

      Amara, Selma; Bu, Ride; Alawein, Meshal; Alsharif, Nouf; Khan, Mohammed Asadullah; Wen, Yan; Zhang, Xixiang; Kosel, Jurgen; Fariborzi, Hossein (IEEE, 2018-08-20)
      Flow cytometers are important instruments for biological and biomedical analyses. These instruments are large and expensive, and researchers are continuously striving to come up with smaller, cheaper, and more energy-efficient flow cytometers. In this work, we present a highly-sensitive magnetic tunnel junction (MTJ) based flow cytometer. An externally magnetized magnetic beads labeling cells were placed above an MTJ sensor that can measure the stray field surrounding the beads. It was found that each time labeled cells pass through the sensitive area of the sensor, a peak of signal was observed. The results demonstrate a novel MTJ based flow cytometer design approach for accurate detection of magnetically labeled cells.
    • Flexible Magnetoresistive Sensors for Guiding Cardiac Catheters

      Hawsawi, M.; Amara, S.; Mashraei, Y.; Almansouri, A.; Mohammad, H.; Sevilla, G. Torres; Jakob, G.; Jaiswal, S.; Klaui, M.; Haneef, A.; Saoudi, A.; Hussain, M.; Kosel, J. (IEEE, 2018-08-20)
      Cardiac catheterization is a procedure, in which a long thin tube that is called a “catheter” is inserted into the heart for diagnosis or treatment. Due to the excessive use of x-ray doses and contrast agents for orientation detection during the surgery, there is a need to find a better alternative. This paper presents magnetic tunnel junction sensors on flexible Si attached to the catheter tip for orientation detection during minimally invasive surgeries. Due to the small size of catheters, extreme minimization in terms of size, weight, thickness and power consumption is needed for any device implemented on it. The fabricated flexible magnetic tunnel junctions fulfill those requirements with size, thickness, weight and power consumption of 150 μm 2 , 12 μm, 8 μg and 0.15 μW, respectively, while still providing a high sensitivity of 4.93 %/Oe. The sensors can be bent with up to 500 μm in diameter, which is more than needed for even the smallest catheters of size 1 mm (3 Fr) in diameter. This result is a stepping-stone towards the development of a versatile and low-cost smart catheter system that can help surgeons navigate inside the heart while minimizing the side effects.
    • Going beyond 10-meter, Gbit/s underwater optical wireless communication links based on visible lasers

      Shen, Chao; Guo, Yujian; Sun, Xiaobin; Liu, Guangyu; Ho, Kang-Ting; Ng, Tien Khee; Alouini, Mohamed-Slim; Ooi, Boon S. (IEEE, 2017-12-13)
      Laser diode based visible light communication link with data rate of 2.2 Gbps over a 12-meter underwater channel is reported using non-return-to-zero on-off keying modulation scheme. The underwater optical wireless communication link beyond 10-meter can be achieved using visible lasers.
    • Reliable detection of abnormal ozone measurements using an air quality sensors network

      Harrou, Fouzi; Dairi, Abdelkader; Sun, Ying; Senouci, Mohamed (IEEE, 2018-06-14)
      Ozone pollution is one of the most important pollutants that have a negative effect on human health and the ecosystem. An effective statistical methodology to detect abnormal ozone measurements is proposed in this study. We used a Deep Belief Network model to account for nonlinear variation of ground-level ozone concentrations, in combination with a one-class support vector machine, for detecting abnormal ozone measurement. We assessed the efficiency of this methodology by using real data from a network of air quality monitoring systems in Isère, France. Results demonstrated the capability of the proposed strategy to identify abnormalities in ozone measurements.
    • Mitigating Network Side Channel Leakage for Stream Processing Systems in Trusted Execution Environments

      Bilal, Muhammad; Alsibyani, Hassan; Canini, Marco (ACM Press, 2018-06-20)
      A crucial concern regarding cloud computing is the confidentiality of sensitive data being processed in the cloud. Trusted Execution Environments (TEEs), such as Intel Software Guard extensions (SGX), allow applications to run securely on an untrusted platform. However, using TEEs alone for stream processing is not enough to ensure privacy as network communication patterns may leak information about the data. \n \nThis paper introduces two techniques -- anycast and multicast --for mitigating leakage at inter-stage communications in streaming applications according to a user-selected mitigation level. These techniques aim to achieve network data obliviousness, i.e., communication patterns should not depend on the data. We implement these techniques in an SGX-based stream processing system. We evaluate the latency and throughput overheads, and the data obliviousness using three benchmark applications. The results show that anycast scales better with input load and mitigation level, and provides better data obliviousness than multicast.
    • TopPPR

      Wei, Zhewei; He, Xiaodong; Xiao, Xiaokui; Wang, Sibo; Shang, Shuo; Wen, Ji-Rong (ACM Press, 2018-05-25)
      Personalized PageRank (PPR) is a classic metric that measures the relevance of graph nodes with respect to a source node. Given a graph G, a source node s, and a parameter k, a top-k PPR query returns a set of k nodes with the highest PPR values with respect to s. This type of queries serves as an important building block for numerous applications in web search and social networks, such as Twitter's Who-To-Follow recommendation service. Existing techniques for top-k PPR, however, suffer from two major deficiencies. First, they either incur prohibitive space and time overheads on large graphs, or fail to provide any guarantee on the precision of top-k results (i.e., the results returned might miss a number of actual top-k answers). Second, most of them require significant pre-computation on the input graph G, which renders them unsuitable for graphs with frequent updates (e.g., Twitter's social graph). To address the deficiencies of existing solutions, we propose TopPPR, an algorithm for top-k PPR queries that ensure at least ? precision (i.e., at least ? fraction of the actual top-k results are returned) with at least 1-1/n probability, where ? ? (0, 1] is a userspecified parameter and n is the number of nodes in G. In addition, TopPPR offers non-trivial guarantees on query time in terms of ?, and it can easily handle dynamic graphs as it does not require any preprocessing. We experimentally evaluate TopPPR using a variety of benchmark datasets, and demonstrate that TopPPR outperforms the state-of-the-art solutions in terms of both efficiency and precision, even when we set ? = 1 (i.e., when TopPPR returns the exacttop-k results). Notably, on a billion-edge Twitter graph, TopPPR only requires 15 seconds to answer a top-500 PPR query with ? = 1.
    • Robust Asymmetric Recommendation via Min-Max Optimization

      Yang, Peng; Zhao, Peilin; Zheng, Vincent W.; Ding, Lizhong; Gao, Xin (ACM Press, 2018-07-02)
      Recommender systems with implicit feedback (e.g. clicks and purchases) suffer from two critical limitations: 1) imbalanced labels may mislead the learning process of the conventional models that assign balanced weights to the classes; and 2) outliers with large reconstruction errors may dominate the objective function by the conventional $L_2$-norm loss. To address these issues, we propose a robust asymmetric recommendation model. It integrates cost-sensitive learning with capped unilateral loss into a joint objective function, which can be optimized by an iteratively weighted approach. To reduce the computational cost of low-rank approximation, we exploit the dual characterization of the nuclear norm to derive a min-max optimization problem and design a subgradient algorithm without performing full SVD. Finally, promising empirical results demonstrate the effectiveness of our algorithm on benchmark recommendation datasets.
    • Folding and stretching a thermoelectric generator

      Albettar, Mohammed A.; Prieto Rojas, Jhonathan; Conchouso, David; Arevalo, Arpys; Singh, Devendra; Foulds, Ian; Hussain, Muhammad M.; Rehman, Mutee (SPIE, 2018-05-08)
      As we are at the verge of entering the era of Internet-of-Things (IoT), there is a clear need to produce continuous power supply to the huge amount of electronic devices that must be wirelessly interconnected and operate uninterruptedly. At the same time, new mechanical constrains arise from the fact that these devices should be ubiquitous, which leads to the need of lightweight and mechanical compliance to any shape or surface. As an important renewable energy source, a mechanically adaptable thermoelectric generator (TEG) can make use of the usually wasted thermal differences between ambient and technology-users to power-up such systems. With this idea in mind, we have developed a simple approach to fabricate TEGs, based on commonly available substrates (paper or polymers) and assisted through simple folding and cutting techniques (born from origami and kirigami) to form strategic structures (serpentine, helical, spiral, etc.) with the mechanical advantage of foldability and over 100% demonstrated stretchability. The use of these methods and structures allows the mechanical reconfigurability of the devices to, for example, increase the temperature difference in a TEG, thus its power, or allow a more efficient use of area and therefore increase the power density. We will discuss the strategies to effectively integrate folding and cutting techniques with common materials and the basic TEG configuration, as well as demonstrate the devices' implementation and characterization. Finally, we believe our simple integration approach offers an interesting and versatile methodology, which can be easily extrapolated to new materials and technologies for a greater variety of applications.
    • Convergence Analysis of Gradient Descent for Eigenvector Computation

      Xu, Zhiqiang; Cao, Xin; Gao, Xin (International Joint Conferences on Artificial Intelligence Organization, 2018-07-05)
      We present a novel, simple and systematic convergence analysis of gradient descent for eigenvector computation. As a popular, practical, and provable approach to numerous machine learning problems, gradient descent has found successful applications to eigenvector computation as well. However, surprisingly, it lacks a thorough theoretical analysis for the underlying geodesically non-convex problem. In this work, the convergence of the gradient descent solver for the leading eigenvector computation is shown to be at a global rate O(min{ (lambda_1/Delta_p)^2 log(1/epsilon), 1/epsilon }), where Delta_p=lambda_p-lambda_p+1>0 represents the generalized positive eigengap and always exists without loss of generality with lambda_i being the i-th largest eigenvalue of the given real symmetric matrix and p being the multiplicity of lambda_1. The rate is linear at (lambda_1/Delta_p)^2 log(1/epsilon) if (lambda_1/Delta_p)^2=O(1), otherwise sub-linear at O(1/epsilon). We also show that the convergence only logarithmically instead of quadratically depends on the initial iterate. Particularly, this is the first time the linear convergence for the case that the conventionally considered eigengap Delta_1= lambda_1 - lambda_2=0 but the generalized eigengap Delta_p satisfies (lambda_1/Delta_p)^2=O(1), as well as the logarithmic dependence on the initial iterate are established for the gradient descent solver. We are also the first to leverage for analysis the log principal angle between the iterate and the space of globally optimal solutions. Theoretical properties are verified in experiments.
    • Task-Guided and Semantic-Aware Ranking for Academic Author-Paper Correlation Inference

      Zhang, Chuxu; Yu, Lu; Zhang, Xiangliang; Chawla, Nitesh V. (International Joint Conferences on Artificial Intelligence Organization, 2018-07-05)
      We study the problem of author-paper correlation inference in big scholarly data, which is to effectively infer potential correlated works for researchers using historical records. Unlike supervised learning algorithms that predict relevance score of author-paper pair via time and memory consuming feature engineering, network embedding methods automatically learn nodes' representations that can be further used to infer author-paper correlation. However, most current models suffer from two limitations: (1) they produce general purpose embeddings that are independent of the specific task; (2) they are usually based on network structure but out of content semantic awareness. To address these drawbacks, we propose a task-guided and semantic-aware ranking model. First, the historical interactions among all correlated author-paper pairs are formulated as a pairwise ranking loss. Next, the paper's semantic embedding encoded by gated recurrent neural network, together with the author's latent feature is used to score each author-paper pair in ranking loss. Finally, a heterogeneous relations integrative learning module is designed to further augment the model. The evaluation results of extensive experiments on the well known AMiner dataset demonstrate that the proposed model reaches significant better performance, comparing to a number of baselines.
    • Mining Streaming and Temporal Data: from Representation to Knowledge

      Zhang, Xiangliang (International Joint Conferences on Artificial Intelligence Organization, 2018-07-05)
      In this big-data era, vast amount of continuously arriving data can be found in various fields, such as sensor networks, network management, web and financial applications. To process such data, algorithms are usually challenged by its complex structure and high volume. Representation learning facilitates the data operation by providing a condensed description of patterns underlying the data. Knowledge discovery based on the new representations will then be computationally efficient, and to certain extent be more effective due to the removal of noise and irrelevant information in the step of representation learning. In this paper, we will briefly review state-of-the-art techniques for extracting representation and discovering knowledge from streaming and temporal data, and demonstrate their performance at addressing several real application problems.
    • Bandit Online Learning on Graphs via Adaptive Optimization

      Yang, Peng; Zhao, Peilin; Gao, Xin (International Joint Conferences on Artificial Intelligence Organization, 2018-07-05)
      Traditional online learning on graphs adapts graph Laplacian into ridge regression, which may not guarantee reasonable accuracy when the data are adversarially generated. To solve this issue, we exploit an adaptive optimization framework for online classification on graphs. The derived model can achieve a min-max regret under an adversarial mechanism of data generation. To take advantage of the informative labels, we propose an adaptive large-margin update rule, which enjoys a lower regret than the algorithms using error-driven update rules. However, this algorithm assumes that the full information label is provided for each node, which is violated in many practical applications where labeling is expensive and the oracle may only tell whether the prediction is correct or not. To address this issue, we propose a bandit online algorithm on graphs. It derives per-instance confidence region of the prediction, from which the model can be learned adaptively to minimize the online regret. Experiments on benchmark graph datasets show that the proposed bandit algorithm outperforms state-of-the-art competitors, even sometimes beats the algorithms using full information label feedback.
    • A Novel Energy Stable Numerical Scheme for Navier-Stokes-Cahn-Hilliard Two-Phase Flow Model with Variable Densities and Viscosities

      Feng, Xiaoyu; Kou, Jisheng; Sun, Shuyu (Springer International Publishing, 2018-06-12)
      A novel numerical scheme including time and spatial discretization is offered for coupled Cahn-Hilliard and Navier-Stokes governing equation system in this paper. Variable densities and viscosities are considered in the numerical scheme. By introducing an intermediate velocity in both Cahn-Hilliard equation and momentum equation, the scheme can keep discrete energy law. A decouple approach based on pressure stabilization is implemented to solve the Navier-Stokes part, while the stabilization or convex splitting method is adopted for the Cahn-Hilliard part. This novel scheme is totally decoupled, linear, unconditionally energy stable for incompressible two-phase flow diffuse interface model. Numerical results demonstrate the validation, accuracy, robustness and discrete energy law of the proposed scheme in this paper.
    • Adaptive Time-Splitting Scheme for Nanoparticles Transport with Two-Phase Flow in Heterogeneous Porous Media

      El-Amin, Mohamed F.; Kou, Jisheng; Sun, Shuyu (Springer International Publishing, 2018-06-12)
      In this work, we introduce an efficient scheme using an adaptive time-splitting method to simulate nanoparticles transport associated with a two-phase flow in heterogeneous porous media. The capillary pressure is linearized in terms of saturation to couple the pressure and saturation equations. The governing equations are solved using an IMplicit Pressure Explicit Saturation-IMplicit Concentration (IMPES-IMC) scheme. The spatial discretization has been done using the cell-centered finite difference (CCFD) method. The interval of time has been divided into three levels, the pressure level, the saturation level, and the concentrations level, which can reduce the computational cost. The time step-sizes at different levels are adaptive iteratively by satisfying the Courant-Friedrichs-Lewy (CFL<1) condition. The results illustrates the efficiency of the numerical scheme. A numerical example of a highly heterogeneous porous medium has been introduced. Moreover, the adaptive time step-sizes are shown in graphs.
    • A Compact and Efficient Lattice Boltzmann Scheme to Simulate Complex Thermal Fluid Flows

      Zhang, Tao; Sun, Shuyu (Springer International Publishing, 2018-06-12)
      A coupled LBGK scheme, constituting of two independent distribution functions describing velocity and temperature respectively, is established in this paper. Chapman-Enskog expansion, a procedure to prove the consistency of this mesoscopic method with macroscopic conservation laws, is also conducted for both lattice scheme of velocity and temperature, as well as a simple introduction on the common used DnQb model. An efficient coding manner for Matlab is proposed in this paper, which improves the coding and calculation efficiency at the same time. The compact and efficient scheme is then applied in the simulation of the famous and well-studied Rayleigh-Benard convection, which is common seen as a representative heat convection problem in modern industries. The results are interesting and reasonable, and meet the experimental data well. The stability of this scheme is also proved through different cases with a large range of Rayleigh number, until 2 million.
    • Study on Numerical Methods for Gas Flow Simulation Using Double-Porosity Double-Permeability Model

      Wang, Yi; Sun, Shuyu; Gong, Liang (Springer International Publishing, 2018-06-12)
      In this paper, we firstly study numerical methods for gas flow simulation in dual-continuum porous media. Typical methods for oil flow simulation in dual-continuum porous media cannot be used straightforward to this kind of simulation due to the artificial mass loss caused by the compressibility and the non-robustness caused by the non-linear source term. To avoid these two problems, corrected numerical methods are proposed using mass balance equations and local linearization of the non-linear source term. The improved numerical methods are successful for the computation of gas flow in the double-porosity double-permeability porous media. After this improvement, temporal advancement for each time step includes three fractional steps: (i) advance matrix pressure and fracture pressure using the typical computation; (ii) solve the mass balance equation system for mean pressures; (iii) correct pressures in (i) by mean pressures in (ii). Numerical results show that mass conservation of gas for the whole domain is guaranteed while the numerical computation is robust.
    • Study on an N-Parallel FENE-P Constitutive Model Based on Multiple Relaxation Times for Viscoelastic Fluid

      Li, Jingfa; Yu, Bo; Sun, Shuyu; Sun, Dongliang (Springer International Publishing, 2018-06-12)
      An N-parallel FENE-P constitutive model based on multiple relaxation times is proposed in this paper, which aims at accurately describing the apparent viscosity of viscoelastic fluid. The establishment of N-parallel FENE-P constitutive model and the numerical approach to calculate the apparent viscosity are presented in detail, respectively. To validate the performance of the proposed constitutive model, it is compared with the conventional FENE-P constitutive model (It only has single relaxation time) in estimating the apparent viscosity of two common viscoelastic fluids: polymer and surfactant solutions. The comparative results indicate the N-parallel FENE-P constitutive model can represent the apparent viscosity of polymer solutions more accurate than the traditional model in the whole range of shear rate (0.1 s–1000 s), and the advantage is more noteworthy especially when the shear rate is higher (10 s–1000 s). Despite both the proposed model and the traditional model can’t capture the interesting shear thickening behavior of surfactant solutions, the proposed constitutive model still possesses advantage over the traditional one in depicting the apparent viscosity and first normal stress difference. In addition, the N-parallel FENE-P constitutive model demonstrates a better applicability and favorable adjustability of the model parameters.
    • LES Study on High Reynolds Turbulent Drag-Reducing Flow of Viscoelastic Fluids Based on Multiple Relaxation Times Constitutive Model and Mixed Subgrid-Scale Model

      Li, Jingfa; Yu, Bo; Zhang, Xinyu; Sun, Shuyu; Sun, Dongliang; Zhang, Tao (Springer International Publishing, 2018-06-12)
      Due to complicated rheological behaviors and elastic effect of viscoelastic fluids, only a handful of literatures reporting the large-eddy simulation (LES) studies on turbulent drag-reduction (DR) mechanism of viscoelastic fluids. In addition, these few studies are limited within the low Reynolds number situations. In this paper, LES approach is applied to further study the flow characteristics and DR mechanism of high Reynolds viscoelastic turbulent drag-reducing flow. To improve the accuracy of LES, an N-parallel FENE-P constitutive model based on multiple relaxation times and an improved mixed subgrid-scale (SGS) model are both utilized. DR rate and velocity fluctuations under different calculation parameters are analyzed. Contributions of different shear stresses on frictional resistance coefficient, and turbulent coherent structures which are closely related to turbulent burst events are investigated in details to further reveal the DR mechanism of high Reynolds viscoelastic turbulent drag-reducing flow. Especially, the different phenomena and results between high Reynolds and low Reynolds turbulent flows are addressed. This study is expected to provide a beneficial guidance to the engineering application of turbulent DR technology.
    • Trajectory Optimization of Energy Efficient FSOC-UAV with Atmospheric and Geometric Loss

      Park, Ki-Hong; Lee, Ju-Hyung; Alouini, Mohamed-Slim; Ko, Young-Chai
      This paper address the issues of optimizing unmanned aerial vehicle (UAV) using free-space optical communication (FSOC). In particular, we examine how the beam divergence and the atmospheric attenuation of FSOC channel affect the flight of energy efficient UAVs. Especially based on geometric loss and also the effect of rain, snow, and fog on FSOC, we showed the energy efficient trajectory of UAV.