• Title/Summary/Keyword: Network Enhancement

Search Result 735, Processing Time 0.029 seconds

TCP Performance Enhancement over the Wireless Networks by Using CPC and ZWSC (CPC와 ZWSC를 이용한 무선 망에서의 TCP 성능 향상 방안)

  • Lee, Myung-Sub;Park, Young-Min;Chang, Joo-Seok;Park, Chang-Hyeon
    • IEMEK Journal of Embedded Systems and Applications
    • /
    • v.1 no.1
    • /
    • pp.24-30
    • /
    • 2006
  • With the original Transmission Control Protocol(TCP) design, which is particularly targeted at the wired networks, a packet loss is assumed to be caused by the network congestion. In the wireless environment where the chances to lose packets due to transmission bit errors are not negligible, though, this assumption may result in unnecessary TCP performance degradation. In these days, many papers describe about wireless-TCP which has suggested how to avoid congestion control when packet loss over the wireless network. In this paper, an enhancement scheme is proposed by modifying SNOOP scheme. To enhance the original SNOOP scheme, CPC(Consecutive Packet Control) and ZWSC(Zero Window Size Control) are added. The invocation of congestion control mechanism is now minimized by knowing the cause of packet loss. We use simulation to compare the overhead and the performance of the proposed schemes, and to show that the proposed schemes improve the TCP performance compares to SNOOP by knowing the cause of packet loss at the base station.

  • PDF

On the enhancement of the learning efficiency of the adaptive back propagation neural network using the generating and adding the hidden layer node (은닉층 노드의 생성추가를 이용한 적응 역전파 신경회로망의 학습능률 향상에 관한 연구)

  • Kim, Eun-Won;Hong, Bong-Wha
    • Journal of the Institute of Electronics Engineers of Korea TE
    • /
    • v.39 no.2
    • /
    • pp.66-75
    • /
    • 2002
  • This paper presents an adaptive back propagation algorithm that its able to enhancement for the learning efficiency with updating the learning parameter and varies the number of hidden layer node by the generated error, adaptively. This algorithm is expected to escaping from the local minimum and make the best environment for the convergence of the back propagation neural network. On the simulation tested this algorithm on three learning pattern. One was exclusive-OR learning and the another was 3-parity problem and 7${\times}$5 dot alphabetic font learning. In result that the probability of becoming trapped in local minimum was reduce. Furthermore, the neural network enhanced to learning efficient about 17.6%~64.7% for the existed back propagation. 

Sound Enhancement with Generative Adversarial Network under Noise Conditions (잡음 환경에서 Generative Adversarial Network를 이용한 소리 음질 향상)

  • Choi, Yongju;Lee, Jonguk;Wang, Huasang;Park, Daihee;Chung, Yongwha
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2018.10a
    • /
    • pp.673-676
    • /
    • 2018
  • 4차 산업혁명이 도래하면서 정보 통신 기술 및 융합 기술의 발전에 힘입어 소리 데이터를 이용한 연구가 활발하게 진행되고 있다. 소리 데이터를 이용한 학술적 프로토타입 연구들을 실제 환경에서 운용하기 위해서는 소리 취득 시 발생하는 다양한 잡음 환경에서도 원시 데이터(raw data)에 근접한 정보를 취득할 수 있는 시스템의 강인함이 보장되어야 한다. 본 논문에서는 SEGAN(Speech Enhancement Generative Adversarial Network) 모델을 활용하여, 전처리 및 후처리 과정이 필요 없이 원시 데이터를 대상으로 하는 end-to-end 방식의 소리 음질 향상 시스템을 제안한다. 제안하는 시스템은, 축산업 분야의 돼지 호흡기 질병 소리 데이터를 이용하여 실험하였으며, 여러 가지 잡음 상황(인위적인 잡음, 실제 환경 잡음)에서 소리 음질이 개선됨을 실험적으로 검증하였다.

Night-to-Day Road Image Translation with Generative Adversarial Network for Driver Safety Enhancement (운전자 안정성 향상을 위한 Generative Adversarial Network 기반의 야간 도로 영상 변환 시스템)

  • Ahn, Namhyun;Kang, Suk-Ju
    • Journal of Broadcast Engineering
    • /
    • v.23 no.6
    • /
    • pp.760-767
    • /
    • 2018
  • Advanced driver assistance system(ADAS) is a major technique in the intelligent vehicle field. The techniques for ADAS can be separated in two classes, i.e., methods that directly control the movement of vehicle and that indirectly provide convenience to driver. In this paper, we propose a novel system that gives a visual assistance to driver by translating a night road image to a day road image. We use the black box images capturing the front road view of vehicle as inputs. The black box images are cropped into three parts and simultaneously translated into day images by the proposed image translation module. Then, the translated images are recollected to original size. The experimental result shows that the proposed method generates realistic images and outperforms the conventional algorithms.

Generating a Retinex-based Reflectance Image from a Low-Light Image Using Deep Neural Network (심층 신경망을 이용한 저조도 영상에서 Retinex 기반 반사 영상 생성)

  • Kim, Wonhoi;Hwang, In-Chul;Kim, Manbae
    • Journal of Broadcast Engineering
    • /
    • v.24 no.1
    • /
    • pp.87-96
    • /
    • 2019
  • Improvement of low-light image mainly focuses on the contrast enhancement. Many researches have been carried out for brightness enhancement, contrast improvement and illumination reduction. Recently, the aforementioned approaches have been replaced by artificial neural networks. This paper proposes a methodology that can replace the Retinex-based reflectance image acquisition by deep neural network. Experiments carried out on 102 low-light images validated the feasibility of the replacement by producing PSNR=30.8682(db) and SSIM=0.4345.

Design of Speech Enhancement U-Net for Embedded Computing (임베디드 연산을 위한 잡음에서 음성추출 U-Net 설계)

  • Kim, Hyun-Don
    • IEMEK Journal of Embedded Systems and Applications
    • /
    • v.15 no.5
    • /
    • pp.227-234
    • /
    • 2020
  • In this paper, we propose wav-U-Net to improve speech enhancement in heavy noisy environments, and it has implemented three principal techniques. First, as input data, we use 128 modified Mel-scale filter banks which can reduce computational burden instead of 512 frequency bins. Mel-scale aims to mimic the non-linear human ear perception of sound by being more discriminative at lower frequencies and less discriminative at higher frequencies. Therefore, Mel-scale is the suitable feature considering both performance and computing power because our proposed network focuses on speech signals. Second, we add a simple ResNet as pre-processing that helps our proposed network make estimated speech signals clear and suppress high-frequency noises. Finally, the proposed U-Net model shows significant performance regardless of the kinds of noise. Especially, despite using a single channel, we confirmed that it can well deal with non-stationary noises whose frequency properties are dynamically changed, and it is possible to estimate speech signals from noisy speech signals even in extremely noisy environments where noises are much lauder than speech (less than SNR 0dB). The performance on our proposed wav-U-Net was improved by about 200% on SDR and 460% on NSDR compared to the conventional Jansson's wav-U-Net. Also, it was confirmed that the processing time of out wav-U-Net with 128 modified Mel-scale filter banks was about 2.7 times faster than the common wav-U-Net with 512 frequency bins as input values.

Single Low-Light Ghost-Free Image Enhancement via Deep Retinex Model

  • Liu, Yan;Lv, Bingxue;Wang, Jingwen;Huang, Wei;Qiu, Tiantian;Chen, Yunzhong
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.15 no.5
    • /
    • pp.1814-1828
    • /
    • 2021
  • Low-light image enhancement is a key technique to overcome the quality degradation of photos taken under scotopic vision illumination conditions. The degradation includes low brightness, low contrast, and outstanding noise, which would seriously affect the vision of the human eye recognition ability and subsequent image processing. In this paper, we propose an approach based on deep learning and Retinex theory to enhance the low-light image, which includes image decomposition, illumination prediction, image reconstruction, and image optimization. The first three parts can reconstruct the enhanced image that suffers from low-resolution. To reduce the noise of the enhanced image and improve the image quality, a super-resolution algorithm based on the Laplacian pyramid network is introduced to optimize the image. The Laplacian pyramid network can improve the resolution of the enhanced image through multiple feature extraction and deconvolution operations. Furthermore, a combination loss function is explored in the network training stage to improve the efficiency of the algorithm. Extensive experiments and comprehensive evaluations demonstrate the strength of the proposed method, the result is closer to the real-world scene in lightness, color, and details. Besides, experiments also demonstrate that the proposed method with the single low-light image can achieve the same effect as multi-exposure image fusion algorithm and no ghost is introduced.

CNN based dual-channel sound enhancement in the MAV environment (MAV 환경에서의 CNN 기반 듀얼 채널 음향 향상 기법)

  • Kim, Young-Jin;Kim, Eun-Gyung
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.23 no.12
    • /
    • pp.1506-1513
    • /
    • 2019
  • Recently, as the industrial scope of multi-rotor unmanned aerial vehicles(UAV) is greatly expanded, the demands for data collection, processing, and analysis using UAV are also increasing. However, the acoustic data collected by using the UAV is greatly corrupted by the UAV's motor noise and wind noise, which makes it difficult to process and analyze the acoustic data. Therefore, we have studied a method to enhance the target sound from the acoustic signal received through microphones connected to UAV. In this paper, we have extended the densely connected dilated convolutional network, one of the existing single channel acoustic enhancement technique, to consider the inter-channel characteristics of the acoustic signal. As a result, the extended model performed better than the existed model in all evaluation measures such as SDR, PESQ, and STOI.

Effects of Social Media Utilization on Labor Union Social Capital in South Korea

  • Lee, Ji-Heon;Jung, Hoe-Kyung
    • International journal of advanced smart convergence
    • /
    • v.6 no.2
    • /
    • pp.34-50
    • /
    • 2017
  • This study delved into the effects of labor union members' social media utilization for the formation of labor union social capital. Specifically, this study aimed to identify the effects of labor union-related social media use and participation on the labor union's social capital formation through quantitative and qualitative research. It set up trust, network, and participation as social capital components and as dependent variables. Network, in particular, was divided into bonding and bridging aspect. There is the correlation between labor union-related social media use and the formation of labor union social capital. As participation in the group type social media operated by a labor union becomes more active, evaluation on labor union social capital throughout trust, network, and participation is higher. Especially, the correlation between bonding network and bridging network was high. This proves that a labor union's bond enhancement does not result in the labor union's selfishness, but it can build a cooperative system with an external network.

Development of High Performance Backlight Unit Employing EEFL

  • Yoo, Hyeong-Suk;Kang, Moon-Shik;Lim, Jong-Sun;Lee, Keun-Woo;Oh, Weon-Sik;Park, Jong-Dae;Kang, Sung-Chul
    • 한국정보디스플레이학회:학술대회논문집
    • /
    • 2002.08a
    • /
    • pp.835-837
    • /
    • 2002
  • The 17" Backlight Unit (BLU) employing twelve EEFLs (External Electrode Fluorescent Lamp) has been developed for LCD-TV The characteristics of the EEFL BLU without dual brightness enhancement film (DBEF) were equivalent to those of CCFL (Cold cathode Fluorescent Lamp) BLU employing eight CCFLs with DBEF. Luminance, power consumption and uniformity were 12,000nits, 32watt and 80%, respectively. The inverter of EEFL Backlight Unit is composed of 2 transformers and driven by the sinusoidal waveform.

  • PDF