Search

We report ALD-Ru thin films using a new carbonyl-based Ru precursor and compare the results with various nonoxidizing reactants such as H 2 , NH 3 , CO molecule and NH 3 , N 2 , H 2 , N 2 +H 2 mixture plasma. In addition, we also investigatd ALD-Ru process with post annealing and in-situ periodic plasma treatment to improve ALD-Ru film quality. The Ru film was deposited at the deposition temperature ranging from 100 to 150 C and at the chamber pressure from 1 to 50 Torr. In the case of using N 2 and H 2 mixture plasma as a reactant, it was possible to deposit Ru film even at a substrate temperature of 100C and the film resistivity of ∼ 70 µΩ·cm was deposited after annealing and in-situ plasma treatment. It was shown that ALD-Ru film with the resistivity of ∼ 25 µΩ·cm could be prepapred using H 2 molecules at 150C with increasing chamber pressure up to 30 Torr. The step coverage of ALD-Ru films deposited using H 2 molecuels and at the chamber pressure of 30 Torr was excellent, around 100 % at dual trench structure with aspect ratio of ∼ 6.3. It should be noted that ALD-Ru films deposited using H 2 molecules contained no oxygen and carbon impurities while Ru films using deposited diluted O 2 has oxygen and carbon impurities by secondary ion mass spectrometry depth profile. And electro-plating of Cu is successfully possible on ALD-Ru deposited using H 2 molecules at 30 Torr.

A goal of this paper is to suggest a framework to infer the situation using IOT sensor data. To do so, the framework adopts contexts which were derived from the learning results of multiple deep neural networks for IOT sensor data and carries out hierarchical clustering of contexts in terms of the spatio-temporality. With the learned dendrogram, the most appropriate situation is inferred from case-based reasoning depending on the similar time and location. The result of reasoning is stored in a case memory and this can contribute to learning of a case base. The primary contribution of this paper is the situation reasoning under consideration for spatio-temporality that is a characteristic of IOT sensor data. Also, we performed experiments to show the superiority of our framework. The experimental results are not bad for a first attempt. In further research, if the algorithms are improved, we can expect better results.

In this study, we are proposing a medical group decision making using Health Service Knowledge Management (HSKM) with interactively extracted context from patients and doctors by sharing of health information. Especially, in the area of tele-healthcare, it is necessary to share distributed medical information to make medical decisions because it is not easy to make a decision without any face-to-face contact between patients and doctors. Occasionally, medical teams often need to share medical information with far-off clinicians to care patients much more. For such a reason, it is sufficient to develop HSKM for healthcare systems. The developed HSKM is applied to make a medical decision for patients, which is based on Health Service Ontology and Case-based reasoning (CBR). Health Service Ontology is based on medical information like symptoms, diseases, departments and doctors knowledge. Medical records are stored as cases in knowledge-base. The cases are applied to find appropriate medical cases in the medical group decision making using CBR. To extract a medical decision, patients and doctors' interactive contexts are considered during the making a medical decision. The contexts are operated as constraints to reach the medical goal using Constraints Satisfaction Problem (CSP). We used an example to discuss the necessity of the proposed HSKM and we expect the implementation and verification of the superiority of the system in further research.

This paper propose a new balance control method of spherical robot. When motors generate turning force, Spherical robot can maintain the balance for roll using the moment of inertia and angular acceleration. This method which is fixed to the center axis by bearing has zero angular acceleration, so spherical robot maintains the balance from disturbance and external impact. For the measurement of roll slope, this system used the gyro and acceleration sensor and used the kalman filter because of exact measurement of gyro and acceleration sensor which have a noise.

As cyber-crimes using personal information such as ID theft are increasing, there is a need for appropriate technology or law to protect privacy. To this end, the Korean Government established the Privacy Act on March 29th 2011. The Privacy Act prescribes a specification for dealing with privacy with the intention to protect personal information from being collected, leaked, misused, or abused so that it can improve rights and interests of the nation and eventually realize the dignity and value of man. The United States, Japan, Canada, and several countries of the EU have their own privacy law being established or revised. Although there must be differences depending on the circumstances of each country, the ultimate goal of the privacy law should be the same. Consequently, there might be the same or similar technical protection required by all these countries. Between the increasing interest in protecting personal information and the establishment of the Privacy Act, many industries are having relevant products released one after another. Customers without knowledge of the law and the product types cannot decide what they need. This paper intends to derive necessary security functions of a personal information security system based on the Common Criteria and analyze the limit of the products in order to make guidelines for privacy and information protection system.

Dynamic adaptation is to create proper information from the context of the end user exposed to a ubiquitous computing environment. Thereby, we propose a Context Ontology-based personalized recommendation in the ubiquitous computing Environment (CORALE) framework to process dynamic adaptation considering social context as a kind of tacit knowledge. The CORALE uses two types of reasoning for the dynamic adaptation, (i) we create a case including an end user's context and decision-making result and reuse the case to recommend a newly requested context, (ii) we adopt constraint satisfaction problem solving algorithm when additional adaptation is needed according to the change in the current context or end user' preference. To prove the necessity of the CORALE framework, we illustrate motivating scenarios.

This paper proposes the localization scheme using ultrasonic beacons in a multi-block workspace. Indoor localization schemes using ultrasonic sensors have been widely studied due to their cheap price and high accuracy. However, ultrasonic sensors are susceptible to environmental noises from their propagation characteristics. On account of their decay phenomena when they are transmitted over a long distance, ultrasonic sensors are not suitable for application in large indoor environments. To overcome these shortages of ultrasonic sensors while emphasizing their advantages, a multi-block approach has been proposed by dividing the indoor space into several blocks with multiple beacons in each block. This approach, however, is hard to divide into several blocks when beacons are not installed in a certain pattern, and in case of having newly installed beacons, all blocks placement is reconstructed. Therefore, this paper proposes a real time localization scheme to estimate the position of mobile robot without effecting beacons placement. Beacon scheduling algorithm has been developed to select the optimal beacons according to robot position and beacon arrangement for the mobile robot navigation. The performance of the proposed localization system is verified through simulations and real experiments.

High-k/metal gate (HKMG) compatible high performance Source/Drain (S/D) stress-memorization-technology (SMT) is presented. Channel stress generated by SMT can be simulated by using mask-edge dislocation model, which is consistent with the measured actual channel stress. Extremely deep pre-amorphization-implant (PAI) for SMT creates multiple mask-edge dislocations under S/D region, which enhances short-channel mobility by 40∼60%. Finally, more than 10% short channel drive current gain is achieved with additional S/D extension optimization.

To move a obstacle from a certain position to the predefined destination, robot needs various kinds of elements, like control algorithm, hardware, etc. General human can selects the distance between him and destination by numerical supervised learning, and establish the moving plan to the destination, generally called ‘Trajectory planning.’ He(or she) can make a trajectory planning against the obstacles which is placed on the path to the destination. The kinds of obstacle can be active obstacle, which is moving on the trajectory, or passive obstacle, which is stationary one. For the kinds of stationary obstacle is height difference between road and sidewalk, stairs, a ramp. And the active obstacles include a human, a vehicle that shows active motion. To make a trajectory planning for the stationary obstacle, autonomous mobile robot can be assume that it is placed at the center of the map, and from the distance information between autonomous mobile robot and obstacles using the advanced fuzzy controller, mobile robot can make a trajectory planning. But in case of active moving obstacle, there are many components and information is needed to process because its moving trace should be considered in real time. As mentioned above, for the proposed algorithm, various kinds of things should be considered to move to the destination. Obviously, autonomous driving of the mobile robot can provide conveniences to the human-being with the various kinds of aspects. In this paper, obstacle avoidance algorithm for the trajectory planning using advanced fuzzy controller is described in detail, and demonstrated proposed algorithm through the real experiment.

In the work, microfluidic device consisting of an interdigitated microarray (IDA) electrode was developed for a rapid, and separation-free immuno-sensors based on a manipulation technique of microparticles by dielectrophoresis (DEP). A poly-dimethylsiloxane (PDMS) substrate with microfluidic channel was placed on the IDA plate to allow to fabricating the device. On applying AC voltage to the IDA in a negative DEP (n-DEP) frequency region, goat anti-mouse immunoglobulin (anti-mouse IgG)-immobilized microparticles moved to the surface of PDMS substrate placed above the IDA by n-DEP force to accumulate at the designated areas of the PDMS surface, where anti-mouse IgG was precoated. When the fluorescence microparticles bearing anti-mouse IgG were suspended in an analyte (mouse IgG) solution, the microparticles trapped the analyte to form microparticle-conjugates. The conjugates were accumulated and captured at the designated areas of the PDMS surface via antibody-antigen-antibody (sandwich) reaction. The captured microparticles were detected selectively by fluorescence measurements at the focused, designated areas regardless of the presence of uncaptured microparticles in the suspended solution. Thus, the separation and washing-out steps, usually required for conventional immunoassay, are eliminated in the presented procedure. Since the formation of the sandwich structures was accelerated significantly by n-DEP, as short as 30 sec was enough to detect the immunoreaction at the surface. The fluorescence intensity of the captured microparticles at the designated area increased with the analyte in the range, 0.01 ~ 10 ng/mL. The present procedure realizes a rapid, sensitive and separation-free immunoassay in a simple device.

In electrical impedance tomography (EIT), the internal resistivity distribution of the unknown object is estimated using the boundary voltage data induced by different current patterns using various reconstruction algorithms. This paper presents a reconstruction algorithm based modified extended Kalman filter technique that is able to track fast change in the impedance distribution. The computer simulation for the 32 channels synthetic data shows that the reconstruction performance of the proposed scheme is improved compared to that of the conventional Kalman filter algorithm at the expense of slightly increased computational burden.

In this paper, observability properties of a multiantenna GPS measurement system for the estimation of errors in INS are presented. It is shown that time-invariant INS error models are observable with measurements from at least three GPS antennas on the vehicle. There is at least one unobservable mode with two antennas. There are three unobservable modes with one antenna. It is also shown that time-varying INS error models are instantaneously observable with measurements from three GPS antennas. A numerical simulation results are given to verify the effectiveness of the multiantenna measurement system on the INS error estimation. In the simulation, a GPS measurement system is considered in which a trade-off between computational load and accuracy of estimation is achieved.