• Title, Summary, Keyword: Convergence

Search Result 28,513, Processing Time 0.107 seconds

Acoustic Feedback and Noise Cancellation of Hearing Aids by Deep Learning Algorithm (심층학습 알고리즘을 이용한 보청기의 음향궤환 및 잡음 제거)

  • Lee, Haeng-Woo
    • The Journal of the Korea institute of electronic communication sciences
    • /
    • v.14 no.6
    • /
    • pp.1249-1256
    • /
    • 2019
  • In this paper, we propose a new algorithm to remove acoustic feedback and noise in hearing aids. Instead of using the conventional FIR structure, this algorithm is a deep learning algorithm using neural network adaptive prediction filter to improve the feedback and noise reduction performance. The feedback canceller first removes the feedback signal from the microphone signal and then removes the noise using the Wiener filter technique. Noise elimination is to estimate the speech from the speech signal containing noise using the linear prediction model according to the periodicity of the speech signal. In order to ensure stable convergence of two adaptive systems in a loop, coefficient updates of the feedback canceller and noise canceller are separated and converged using the residual error signal generated after the cancellation. In order to verify the performance of the feedback and noise canceller proposed in this study, a simulation program was written and simulated. Experimental results show that the proposed deep learning algorithm improves the signal to feedback ratio(: SFR) of about 10 dB in the feedback canceller and the signal to noise ratio enhancement(: SNRE) of about 3 dB in the noise canceller than the conventional FIR structure.

The Effects of Bojungikgi-tang on Streptozotocin-induced Diabetic Gastroparesis Rat Model (보중익기탕(補中益氣湯)이 streptozotocin 유발 당뇨병성 위부전마비 백서에 미치는 영향)

  • Kang, Yun-Mi;Kim, Hyo-Jung;Park, Yun-Beom;Jeong, Chan-Mun;Ham, Seong-Ho;Yang, Woong-Mo;An, Hyo-Jin
    • The Korea Journal of Herbology
    • /
    • v.34 no.6
    • /
    • pp.45-55
    • /
    • 2019
  • Objective : Diabetic gastroparesis is a complication that is defined as delayed gastric emptying and upper gastrointestinal symptoms and often occurs in long-standing diabetic patients. Bojungikgi-tang (BJT) is a traditional oriental herbal formula that has long been used for the treatment of digestive disorders. The purpose of this study was to investigate the effects of BJT on streptozotocin (STZ)-induced diabetic gastroparesis rat model. Methods : Sprague-Dawley (SD) male rats (250-270g) were divided into 13 groups including normal group, STZ-induced diabetic control group, BJT diet (7 various concentrations), and insulin-, glibenclamide-, metformin-treated group were used for the experiments for the comparison. Diabetic gastroparesis was induced by intraperitoneal injection of STZ. The water intake, food intake, body weights and fasting blood glucose levels were measured. After 4 weeks the animals were sacrificed and gastrin, leptin, insulin, hemoglobin A1C (HbA1c), lactate, lactate dehydrogenase (LDH), bilirubin, creatinine, albumin and lipid levels were evaluated. Results : Intraperitoneal injection of BJT for 4 weeks resulted in increased levels of gastrin in blood and decreased leptin and lactate concentration in STZ-induced diabetic gastroparesis rat model. BJT did not affect insulin, fasting glucose, HbA1c, and lipid levels in STZ-induced diabetic gastroparesis rat model. Conclusion : These results indicated that BJT would have protect effect on diabetic gastroparesis through the improvement effect of gastric motility and fatigue syndrome in STZ-induced diabetic rats. This study shows that BJT might be effective for treatment of diabetes and its complications such as gastroparesis.

Study on Structural Integrity and Dynamic Characteristics of Knuckle Parts of KTX Anti-Roll Bar (KTX 고속열차 안티롤바 너클부의 동특성 및 구조 안전성 평가)

  • Jeon, Kwang Woo;Shin, Kwang Bok;Kim, Jin Woo;Jeong, Yeon Il
    • Transactions of the Korean Society of Mechanical Engineers A
    • /
    • v.37 no.8
    • /
    • pp.1035-1041
    • /
    • 2013
  • To evaluate the structural integrity and dynamic characteristic of the knuckle part of a KTX anti-roll bar, an experimental and a numerical approach were used in this study. In the experimental approach, the acceleration and strain data for the knuckle parts of the KTX and KTX-SANCHUN anti-roll bar were respectively measured to evaluate and compare its structural dynamic characteristics under the operating environments of the Honam line. In the numerical approach, the evaluation of its structural integrity was conducted using LS-DYNA 3D, and then, the reliability of the finite element model used was ensured by a comparative evaluation with the experiment. The numerical results showed that the stress and velocity field of the knuckle part composed of a layered structure of a thin steel plate and rubber were more moderate than those of the knuckle part made of only a thick steel block owing to the reduction of relative contact between the knuckle and the connecting rod. It was found that the knuckle part made of a thin steel plate and rubber was recommended as the best solution to improve its structural integrity resulting from the elastic behavior of the KTX anti-roll bar being enabled under a repeating external force.

Study on the growth of 4H-SiC single crystal with high purity SiC fine powder (고순도 SiC 미분말을 적용한 4H-SiC 단결정 성장에 관한 연구)

  • Shin, Dong-Geun;Kim, Byung-Sook;Son, Hae-Rok;Kim, Moo-Seong
    • Journal of the Korean Crystal Growth and Crystal Technology
    • /
    • v.29 no.6
    • /
    • pp.383-388
    • /
    • 2019
  • High purity SiC fine powder with metal impurity contents of less than 1 ppm was synthesized by improved carbothermal reduction process, and the synthesized powder was used for SiC single crystal growth in RF heating PVT device at temperature above 2,100℃. In-situ x-ray image analyzer was used to observe the sublimation of the powder and single crystal growth behavior during the growth process. SiC powder was used as a source of single crystal growth, exhausted from the outside of the graphite crucible at the growth temperature and left graphite residues. During the growth, the flow of raw materials was concentrated in the middle and influenced the growth behavior of SiC single crystals. This is due to the difference in temperature distribution inside the crucible due to the fine powder. After the single crystal growth was completed, the single crystal ingot was cut into a 1 mm thick single crystal substrate and finely polished using a diamond abrasive slurry. A dark yellow 4H-SiC was observed overall of single crystal substrate, and the polycrystals generated in the outer part may be caused by the incorporation of impurities such as the bubble layer mixed in the process of attaching the seed crystal to the seed holder.

Optimal Location Modeling for Elementary Student's Care facility using Public Data (공공데이터를 활용한 초등학생 돌봄시설의 최적입지 선정)

  • Lee, Ji-Won;Kim, Ji-Young;Yu, Ki-Yun;Yang, Sung-Chul
    • Journal of Cadastre & Land InformatiX
    • /
    • v.49 no.2
    • /
    • pp.109-122
    • /
    • 2019
  • The expansion of double-income households is increasing the social interest in child care. In particular, children's entrance into elementary school is considered to be the main cause of women's career break as well as childbirth. This study proposes an optimal location selection method for caring facilities for elementary school students. As a candidate for care facilities, we selected existing child care facilities. We proposed a dual structure evaluation method that considers locational characteristics as well as mathematical optimization when selecting the optimal location. The experiment was conducted in Songpa-gu, Seoul. A total of 36 optimal locations were selected from a total of 258 candidate facilities. First, the evaluation criteria were established using public data, and the primary candidate facilities were selected by ranking the location scores. At this time mesh resampling method was used to integrate various public data into one. Next, the final care facilities were selected using the p-median method. The results chosen are not only the optimal location considering total distance but also satisfy various location criteria considering the characteristics of the care facility. We expect that the proposed method will contribute to public data convergence or utilization and it will be helpful for policy decision when selecting the optimal location for public facilities.

A Study on Improving Availability of Open Data by Location Intelligence (위치지능화를 통한 공공데이터의 활용성 향상에 관한 연구)

  • Yang, Sungchul
    • Journal of Cadastre & Land InformatiX
    • /
    • v.49 no.2
    • /
    • pp.93-107
    • /
    • 2019
  • The open data portal collects data created by public institutions and opens and shares them according to related laws. With the activation of the Fourth Industrial Revolution, all sectors of our society are demanding high quality data, but the data required by the industry has not been greatly utilized due to the lack of quantity and quality. Numerous data collected in the real world can be implemented in cyber physical systems to simulate real-world problems, and alternatives to various social issues can be found. There is a limit to being provided. Location intelligence is a technology that enables existing data to be represented in space, enabling new value creation through convergence. In this study, to present location intelligence of open data, we surveyed the status of location information by data in open data portal. As a result, about 60% of the surveyed data had location information and the representative type was address. Appeared. Therefore, by suggesting location intelligence of open data based on address and how to use it, this study aimed to suggest a way that open data can play a role in creating future social data-based industry and policy establishment.

Automatic Classification by Land Use Category of National Level LULUCF Sector using Deep Learning Model (딥러닝모델을 이용한 국가수준 LULUCF 분야 토지이용 범주별 자동화 분류)

  • Park, Jeong Mook;Sim, Woo Dam;Lee, Jung Soo
    • Korean Journal of Remote Sensing
    • /
    • v.35 no.6_2
    • /
    • pp.1053-1065
    • /
    • 2019
  • Land use statistics calculation is very informative data as the activity data for calculating exact carbon absorption and emission in post-2020. To effective interpretation by land use category, This study classify automatically image interpretation by land use category applying forest aerial photography (FAP) to deep learning model and calculate national unit statistics. Dataset (DS) applied deep learning is divided into training dataset (training DS) and test dataset (test DS) by extracting image of FAP based national forest resource inventory permanent sample plot location. Training DS give label to image by definition of land use category and learn and verify deep learning model. When verified deep learning model, training accuracy of model is highest at epoch 1,500 with about 89%. As a result of applying the trained deep learning model to test DS, interpretation classification accuracy of image label was about 90%. When the estimating area of classification by category using sampling method and compare to national statistics, consistency also very high, so it judged that it is enough to be used for activity data of national GHG (Greenhouse Gas) inventory report of LULUCF sector in the future.

Duty Cycle Scheduling considering Delay Time Constraints in Wireless Sensor Networks (무선네트워크에서의 지연시간제약을 고려한 듀티사이클 스케쥴링)

  • Vu, Duy Son;Yoon, Seokhoon
    • The Journal of The Institute of Internet, Broadcasting and Communication
    • /
    • v.18 no.2
    • /
    • pp.169-176
    • /
    • 2018
  • In this paper, we consider duty-cycled wireless sensor networks (WSNs) in which sensor nodes are periodically dormant in order to reduce energy consumption. In such networks, as the duty cycle interval increases, the energy consumption decreases. However, a higher duty cycle interval leads to the increase in the end-to-end (E2E) delay. Many applications of WSNs are delay-sensitive and require packets to be delivered from the sensr nodes to the sink with delay requirements. Most of existing studies focus on only reducing the E2E delay, rather than considering the delay bound requirement, which makes hard to achieve the balanced performance between E2E delay and energy consumption. A few study that considered delay bound requirement require time synchronization between neighboring nodes or a specific distribution of deployed nodes. In order to address limitations of existing works, we propose a duty-cycle scheduling algorithm that aims to achieve low energy consumption, while satisfying the delay requirements. To that end, we first estimate the probability distribution for the E2E delay. Then, by using the obtained distribution we determine the maximal duty cycle interval that still satisfies the delay constraint. Simulation results show that the proposed design can satisfy the given delay bound requirements while achieving low energy consumption.

Development and Suggestion of Integrated Collaboration System to Revitalize Community-Based National Science Museums (지역사회 기반 전국과학관 활성화를 위한 통합이용제도 개발 및 제언)

  • Park, Young Shin;Mun, Kingju;Hwang, Yohan
    • Journal of the Korean Society of Earth Science Education
    • /
    • v.12 no.3
    • /
    • pp.274-290
    • /
    • 2019
  • In order for the science museum to play a role as a foundation for strengthening the national creative capability, which is the core of the advanced national development paradigm, it is important to gather the capacities of the nation's science museums and establish a platform to cooperate in a shared value system. Science museum is an independent operating system, and there is a lack of strong connections among national, public and private science museums. The existing integrated collaboration system of science museums-centered can be studied first and promoted to develop programs for the free school year according to a specific topic. The same system of science museum-inclusion which link local cultural institutions or cultural places as science culture program were also studied to do the same purpose. On the basis of problems drawn from studies of integrated collaboration systems of each participating science museum, we proposed a convergence integrated collaboration systems of science museum-centered and science museum-inclusion. To this end, data were collected from practitioners of 7 representing science museums including 5 national ones. In order to suggest improvements, we also contacted five international science museums to collect the exemplary cases. Considering the regional characteristics, science museum-inclusion integrated collaboration systems considering the cultural characteristics and the science museum-centered integrated collaboration systems for free school semesters, were developed and tried by practitioners who participated in this research. It was found that integrated collaboration system can be more activated for the community. This suggests that support from the national level or at the level of regional autonomy is essential and the connection with the curriculum is necessary for the integrated collaboration system program. Finally, professional experts such as program development or commentator can be a decisive role.

The Impact of Perceived Economic Value and Personal Characteristics on Electric Vehicle Purchase Intention - For residents of Jeju as a special district for electric vehicles - (전기차에 대한 지각된 경제적 가치 및 개인적 특성이 구매의도에 미치는 영향에 관한 연구 -전기차 특구지역인 제주지역 주민을 대상으로-)

  • Shim, Soo-Min;Kim, Hyang Mi;Son, Sang-Hoon
    • Journal of Digital Convergence
    • /
    • v.18 no.2
    • /
    • pp.163-174
    • /
    • 2020
  • The market for electric vehicles is growing due to the public's interest in the environment and the expansion of electric vehicle support projects in terms of government policy. This study surveyed 2,332 people in Jeju, one of the nation's representative areas of electric vehicles, and the higher the perceived value in terms of the total cost of automobile ownership for electric vehicles, the higher the intention to purchase electric vehicles. The higher the level of knowledge and attachment, the higher the intention to purchase electric vehicles. While many previous studies considered economic value mainly as price, the study was conducted to approach economic value in terms of total cost of ownership. Marketing practitioners also look for practical contributions in that they can propose price framing so that customers can judge the economic value of the electric vehicle as a strategic way to increase the intention to purchase the electric vehicle, rather than just the purchase price. can see. In addition, the same research should be conducted in various regions besides Jeju, so that the research results can be generalized.