Search

Link prediction is an important task in many domains, including health sciences, biology, recommender systems, social networks, and many more. It is one of the problems residing within the intersection of knowledge graphs and machine learning. Link prediction aims to discover unknown links between entities in a graph using various techniques. However, due to the size of knowledge graphs today and their complexity, it is a challenging and time-consuming task. In this work, we present RegPattern2Vec, a method to effectively sample a large knowledge graph to learn node embeddings, while capturing the semantic relationships between graph nodes with minimum prior knowledge and human involvement. Our results show that the link prediction using RegPattern2Vec outperforms related graph embedding approaches on large-scale and complex knowledge graphs.

Maintaining transient stability during grid disturbances in an islanded Hybrid AC/DC Microgrid (HMG) with high penetration of Renewable Energy Sources (RES) exhibiting diverse dynamic characteristics, is a challenging task. This paper investigates an efficient control strategy based on the Battery Energy Storage System (BESS) for the resilience enhancement of the HMG under various operating conditions and grid disturbances. A coordinated control strategy for the wind turbine reactive current requirement has been adopted. In order to evaluate the robustness of the proposed control mechanism, several disruption cases have been considered as small and large signal disturbance case studies. Simulations have been carried out in the MATLAB\Simulink environment. The time-domain simulation results show that the proposed strategy enhances the transient stability of the system during HMG disturbances.

A Knowledge Graph (KG) is a machine-readable, labeled graph-like representation of human knowledge. As the main goal of KG is to represent data by enriching it with computer-processable semantics, the knowledge graph creation usually involves acquiring data from external resources and datasets. In many domains, especially in biomedicine, the data sources continuously evolve, and KG engineers and domain experts must not only track the changes in KG entities and their interconnections but introduce changes to the KG schema and the graph population software. We present a framework to track the KG evolution both in terms of the schema and individuals. KGdiff is a software tool that incrementally collects the relevant meta-data information from a KG and compares it to a prior version the KG. The KG is represented in OWL/RDF/RDFS and the meta-data is collected using domain-independent queries. We evaluate our method on different RDF/OWL data sets (ontologies).

Spatial computing experiences are physically constrained by the geometry and semantics of the local user environment. This limitation is elevated in remote multi-user interaction scenarios, where finding a common virtual ground physically accessible for all participants becomes challenging. Locating a common accessible virtual ground is difficult for the users themselves, particularly if they are not aware of the spatial properties of other participants. In this paper, we introduce a framework to generate an optimal mutual virtual space for a multi-user interaction setting where remote users' room spaces can have different layout and sizes. The framework further recommends movement of surrounding furniture objects that expand the size of the mutual space with minimal physical effort. Finally, we demonstrate the performance of our solution on real-world datasets and also a real HoloLens application. Results show the proposed algorithm can effectively discover optimal shareable space for multi-user virtual interaction and hence facilitate remote spatial computing communication in various collaborative workflows., Comment: Accepted at 2020 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)

This paper presents a novel coordinated control strategy based on Modulated Series Dynamic Braking Resistor (MSDBR) and Interlinking Converter (IC) for the Fault Ride Through (FRT) capability enhancement of an islanded Hybrid AC/DC Microgrid (HMG). A simple fault detection algorithm is proposed based on the positive sequence voltage and line fault current magnitudes of the AC grid. The MSDBR is mainly employed to suppress the fault current and evacuate excessive active power during the fault. In order to control the AC voltage sags and swells on the MSDBR's post-operation, a supplementary reactive current control is applied to the IC. Simulations have been performed by the MATLAB\Simulink software. Both symmetrical and asymmetrical faults are considered on the line connecting the load in the AC sub-grid. The time-domain simulation results show that the proposed strategy efficiently suppresses the fault current while providing positive sequence voltage recovery. Thus, the proposed strategy is effective for the HMG system.

New devices and applications such as wearables, low power sensor networks, and IoT require new concepts for energy supply. In this paper energy harvesting concepts for low power applications and a new type of a chemical, chip-based energy storage system for large power applications are presented. Combining such concepts, scalable autonomous power supply can be obtained which can be monolithically integrated in Si-based microelectronic devices or microsystems.

A tunable light source for photo acoustic (P A) sensing applications is presented. Tunability of the light source is based on the tilt angle depending reflectance spectrum of a photonic crystal. The 1D photonic crystals with pronounced peak at a given central wavelength were electrochemically etched in form of rugate filters and served as main optical component for the optical system. Effective medium approximation was used to simulate and adjust the rugate filter properties, both in the visible (VIS) and mid-infrared (MIR) spectral range. Fabrication of rugate filters is described, optical properties and tilt angle dependence of the filters are presented. A tunable light source system was developed using the manufactured rugate filter and a controllable tilt mechanism. The system properties are analyzed and optimized for photoacoustic applications, especially for detection of different gases. The system was successfully used for PA detection of methane.

Some sources transmit periodic and quasi periodic sparse pulse trains in the environment and a number of sensors might receive them through a single channel simultaneously. It is usually our interest to know which pulse belongs to which source. This identification process has wide applications in communications, radar system, medical applications, and neural systems. Blind source separation (BSS) is one solution for this problem. This paper proposed a geometrical approach to solve BSS problem when observations are whitened data and are obtained from the linear mixtures of 2-valued sparse signals (such as sparse pulse trains). In other words, the proposed approach aims to estimate a rotation matrix, and then to recover the independent sparse components from the whitened data. There are two assumptions in this work: First, all components of the source vector are non-negative 2-valued sparse signals. Second, all entries of the mixing matrix are positive. We have also provided some numerical simulations to illustrate the proposed method's performance. The results show that the performance is affected by the SNR value in sensors.

Recent advances in imaging, 3D processing, Graphic hardware and network have led to rapid increase in 3D geometric models. This rapid increase in 3D data necessitate retrieval effective techniques for management and reuse. The main challenge in retrieval algorithms of 3D objects is to create an algorithm which is robust to transformations and variations. Accordingly we need an algorithm that can provide high distinction features from 3D objects. Therefore, the aim of the present research is the retrieval of non-rigid 3D shapes using a hybrid algorithm based on local descriptors. According to the experiments the proposed algorithm have been compared with known methods of 3D shapes retrieval. The results of the experiments prove the remarkable performance of proposed algorithm.

This paper reports on a novel method for controlled release of huge amount of energy stored in a functionalized porous silicon (PS) on a silicon chip. This is achieved by adjusting the size and the separation distance between locally defined areas of PS. Such an approach allows multiple sequential or single explosions with defined energy outputs. The confinement and sequential explosions were studied for different shapes and sizes of PS areas with different separation distances between them. A critical separation distance between PS sites is found for which the sequential explosions are achieved.

Extending leg length in a biped entity may improve some gait factors such as speed, worsening some others. This study is undertaken to better understand the kinetic and kinematic aspects of human gait following changes in the geometry of lower extremities. An artificial structure is built to be attached and fixed to one's leg. Using a two-sensor motion capture system practical walking data with the extended leg is collected for three different walking patterns, for which gait parameters are calculated. Particularly, joints internal forces and torques as well as the ground reaction force are targeted, compared and discussed.

RoboCup champions aimed at improving artificial intelligence methods. Competition formats include elimination, double elimination and round-robin. In soccer simulation league an agents' energy and knowledge in each game is independent of previous or simultaneous games. So when the format is round-robin the result is fair and the chronological order of holding games doesn't affect the result. In this paper, by using three heuristic functions, a new intelligent method for providing a dynamic table for specifying chronological order of holding games is presented. By using this method the important and determinative games were held at the end of the competitions. It can make championships more exciting and avoid extra games after round-robin; Therefore total evaluation specify the winner, the ranks are fairer and the RoboCup champions get closer to its' scientific goals. The evaluation showed that this dynamic method improved excitement in comparison to static methods and was 66% closer to the most exciting state evaluation.

Some periodic and quasi-periodic pulse trains are emitted by different sources in the environment and a number of sensors receive them through a single channel simultaneously. We are often interested in separating these pulse trains for source identification at sensors. This identification process is termed as deinterleaving pulse trains. Deinterleaving pulse trains has wide applications in communications, radar systems, neural systems, biomedical engineering, and so on. This paper studies the deinterleaving problem with the assumption that both sources and sensors are fixed. In this study, the problem of deinterleaving pulse trains is modeled as a blind source separation (BSS) problem. To solve the BSS problem, we propose a novel geometry-based producer that has not been discussed in the literature yet. The proposed method has superiority over the previous ones in a number of aspects. First, it is a computationally simple method. Second, the proposed algorithm is capable of deinterleaving similar pulse trains. Third, it is able to separate pulse trains with complex pulse repetition interval (PRI) modulations. Finally, the algorithm's performance is not influenced by missing and spurious pulses. Some numerical simulations are provided in order to illustrate the effectiveness of the proposed method.

In this paper, a hybrid DC circuit breaker based on natural Zero Current Switching for DC microgrid is proposed. In the DC networks, arc quenching is a dramatic challenge due to lack of zero crossing of current, which results in corrosion and reduction of circuit breaker lifetime. The hybrid circuit breaker conducts the current using a mechanical switch in normal condition. In fault condition, counter current of a discharging capacitor makes mechanical switch to experience zero crossing condition. Therefore, the arc can be quenched safely. Also, the fault current with relatively higher impedance can be interrupted using zero voltage switching. The proposed breaker structure is described and its performance is evaluated using different simulations.

One of the challenging tasks in face recognition is illumination variation. In this paper, we propose an encoding domain correlation technique for face recognition to handle the illumination problem. This technique is based on sparse representation of optical encryption of training images to form unconstrained minimum average correlation energy (UMACE) filter. Averages authentication rates for YaleB and PIE database are increased rather than other methods because illumination variations in the plain face images, do not affect encrypted images with uniform histograms. Another advantage for the proposed method is its high security. Because encrypted version of facial images are used rather than plain images at the receiver.

This paper proposes a new approach to detect generalized-roughness bearing faults via motor voltage monitoring which has not been studied in the literature yet. The proposed method detects the bearing faults using analysis of instantaneous frequency of voltage signals. Our proposed method is superior to previous ones in a number of aspects. First, obtaining voltage signals is easier than other signals such as vibration, sound pressure, and temperature. Second, it is computationally simple method. Third, our proposed method is non-invasive. The Experimental results show that these impacts are so strong that even simple time-domain indices are able to detect bearing defects. The experimental results confirm the effectiveness of the proposed method.

Moore's Law has governed advances of silicon technology for more than four decades, providing a tremendous reduction in cost per transistor. Device geometry, packing density, speed performance, and manufacturing cost of a single transistor have scaled together-according to Dennard scaling rule. Approaching the sub-10nm era and beyond, Moore's Law faces serious challenges in the near future (5–6 years). Device geometry/density/performance/cost will not scale simultaneously anymore. What are the new scaling rules for logic and memory? Researchers are racing to address 3 scenarios: 1) extending silicon, 2) beyond silicon, and 3) beyond CMOS.

This work presents a bonding mechanism between needle-like surfaces for room temperature Si-Si direct bonding similar to the Velcro-principle, a fully CMOS compatible approach suitable for system integration using Si-motherboard concept. The proposed bonding model is superior to other presented models since it considers humidity effect and the deformation mechanism of the needles during the bonding.

In a “call for leadership” panel at ISSCC, we will be seeking leaders' perspectives on the future of discontinuous innovation in the semiconductor industry. An ensemble of visionaries, experts and CEOs will discuss the opportunities and challenges for innovation in our industry. Of primary concern is the reduction in funding available for new semiconductor ventures. Are the escalating NRE costs of ASICs providing a barrier to new entrants? What advice would our distinguished panellists give to entrepreneurs thinking of starting a new semiconductor company?

Velcro-like (needle like) surfaces based on porous silicon can result in strong permanent bonding at room temperature with capability of multiple bonding and un-bonding similar to Velcro® principle. However, understanding of the interaction mechanisms between such surfaces is crucial for employing this type of bonding for wafer or chip bonding and IC packing applications. In this paper, a simple Si technology to create Velcro-like surfaces is described and characterized, and the interaction mechanisms between such surfaces are modelled based on Van der Waals force approach.

In today's information technology (IT) era, a major part of the costs is being spent on computational needs. Enterprises are in efforts to increase their Return on Investment (ROI) and individuals are trying to reduce their costs. In this regard, cloud computing which emerges as a fifth utility can reduce costs and enhance performance of IT solutions. A large number of companies and institutions are dealing with cloud related issues as a provider or user. Due to the fact that cloud computing services have been proposed in recent years, organizations and individuals face with various challenges and problems such as how to migrate applications and software platforms into cloud and how to ensure security of migrated applications and etc. Given that many different definitions of cloud computing is presented in many publications and projects, a concrete and clear definition for cloud computing considering its characteristics, models and services is provided in this paper. In addition, current challenges and open issues in cloud computing is discussed in details and further recommendations and roadmaps for scientific activities and researches as well as potentials for improvements in this area from scientific and commercial points of view are given in this paper.

A distinguished panel of experts from global industry and academia will debate the nature of systems that will drive the killer applications of 2020. Can we really forecast these future killer applications? We have had the computing revolution, then communications, and now, the sensor revolution. Are sensors driving the future killer applications? Are there any other revolutions in sight? What circuits and system innovations can ISSCC bring forward to enable the next killer applications? What unique technology elements, device structures and memory architectures will be required? Should we continue with silicon technology and seek breakthroughs in system architecture, algorithms, SoC integration and packaging? Or, are we preparing for “beyond silicon” technologies? And if so, which beyond-silicon technologies are realisitic for the future, versus forever being technologies of the future?