• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.10
• /
• pp.4787-4807
• /
• 2016

Although the dense interconnection datacenter networks (DCNs) (e.g. FatTree) provide multiple paths and high bisection bandwidth for each server pair, the single-path TCP (SPT) and ECMP which are widely used currently neither achieve high bandwidth utilization nor have good load balancing. Due to only one available transmission path, SPT cannot make full use of all available bandwidth, while ECMP's random hashing results in many collisions. In this paper, we present OFPT, an OpenFlow based Parallel Transport framework, which integrates precise routing and scheduling for better load balancing and higher network throughput. By adopting OpenFlow based centralized control mechanism, OFPT computes the optimal path and bandwidth provision for each flow according to the global network view. To guarantee high throughput, OFPT dynamically schedules flows with Seamless Flow Migration Mechanism (SFMM), which can avoid packet loss in flow rerouting. Finally, we test OFPT on Mininet and implement it in a real testbed. The experimental results show that the average network throughput in OFPT is up to 97.5% of bisection bandwidth, which is higher than ECMP by 36%. Besides, OFPT decreases the average flow completion time (AFCT) and achieves better scalability.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1000-1010
• /
• 2019

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

• International Journal of High-Rise Buildings
• /
• v.8 no.2
• /
• pp.137-154
• /
• 2019

Singapore has seen a phenomenal and an unprecedented transformation from a swampland to a high density urban environment since its independence in 1965, made possible largely and single-handedly by the sustained efforts of its government. Indeed, urban space is a key vehicle for achieving urban social, environmental, economic, and cultural sustainability. The dense urban context in Singapore has seen an emergence and increase in elevated spaces in the form of sky-gardens, sky-bridges and sky-courts in a range of building types, seemingly seeking to tie together the different horizontal and vertical components of the city. This paper, therefore, examines the effectiveness of elevated urban spaces and pedestrian networks in Singapore and their ability to contribute to the horizontal to vertical transitions, and consequently to the urban vitality and accessibility. It does this through the analysis of two key developments: Marina Bay Sands and the Jurong Gateway. In particular, it considers the implications of certain constraints placed on urban spaces by their inherent location at height, in addition to the familiar privatization of public spaces, over-management of spaces, and their somewhat utilitarian characteristics. The paper argues that some of these issues may pose detrimental effects on the publicness of these spaces that in turn may lead to such spaces being underused and therefore adding redundancies and further stress to Singapore's urban land. Finally, the paper outlines key strategies that may help overcome the aforementioned issues, including the disjuncture associated with elevated spaces such that they may become a seamless extension of the urban spaces on ground.

• Electronics and Telecommunications Trends
• /
• v.37 no.2
• /
• pp.83-95
• /
• 2022

5G goes beyond people to serve indoor and outdoor companies and industries, as well as campuses such as halls, industrial complexes, educational institutions, stadiums, dense urban areas, rural areas, and government institutions. Therefore, a new approach to small cells is needed. Accordingly, 3GPP and Small Cell Forum are researching 5G small cell architecture; 3GPP, Small Cell Forum, and 5G Alliance for Connected Industries and Automation are also researching private networks tailored to meet the specific requirements of various companies and local governments. In particular, in the UK, a small cell-based technology is required for realizing the Joint Operator Technical Specifications-Neutral Host In-Building specification to cost-effectively secure indoor coverage. Further, the research on the SON(Self-Organizing Network) technology for small cells in 5G, where commercialization has begun, is required. The 5G-based small cell structure, private network, and Neutral Host In-Building and SON reviewed in this study are at the initial research stages; therefore, additional research is needed to secure the competitiveness of the small cell technology in 5G and Beyond 5G.

• Journal of Multimedia Information System
• /
• v.9 no.3
• /
• pp.177-182
• /
• 2022

A deep neural network (DNN) includes variables whose values keep on changing with the training process until it reaches the final point of convergence. These variables are the co-efficient of a polynomial expression to relate to the feature extraction process. In general, DNNs work in multiple 'dimensions' depending upon the number of channels and batches accounted for training. However, after the execution of feature extraction and before entering the SoftMax or other classifier, there is a conversion of features from multiple N-dimensions to a single vector form, where 'N' represents the number of activation channels. This usually happens in a Fully connected layer (FCL) or a dense layer. This reduced 2D feature is the subject of study for our analysis. For this, we have used the FCL, so the trained weights of this FCL will be used for the weight-class correlation analysis. The popular DNN models selected for our study are ResNet-101, VGG-19, and GoogleNet. These models' weights are directly used for fine-tuning (with all trained weights initially transferred) and scratch trained (with no weights transferred). Then the comparison is done by plotting the graph of feature distribution and the final FCL weights.

• Journal of Internet Computing and Services
• /
• v.23 no.5
• /
• pp.9-15
• /
• 2022

This paper proposes a coordinated multi-point communication (CoMP) method with channel selection to improve performance of a macro user equipment (MUE) in a dense small-cell network environment in a building located within coverage of a macro base station (MBS). In the proposed CoMP method, in order to improve the performance of the MUE located in the building, A small-cell base station (SBS) selects a channel with lower interference to the neighboring MUE and transmits appropriate signals to the MUE requiring CoMP. Simulation results show that the proposed CoMP method improves the performance of the MUE by up to 164% and 51%, respectivley, compared to a random channel allocation based traditional SBS network and CoMP method.

• International Journal of Computer Science & Network Security
• /
• v.22 no.9
• /
• pp.23-30
• /
• 2022

Glaucoma is a chronic neuropathy that affects the optic nerve which can lead to blindness. The detection and prediction of glaucoma become possible using deep neural networks. However, the detection performance relies on the availability of a large number of data. Therefore, we propose different frameworks, including a hybrid of a generative adversarial network and a convolutional neural network to automate and increase the performance of glaucoma detection. The proposed frameworks are evaluated using five public glaucoma datasets. The framework which uses a Deconvolutional Generative Adversarial Network (DCGAN) and a DenseNet pre-trained model achieves 99.6%, 99.08%, 99.4%, 98.69%, and 92.95% of classification accuracy on RIMONE, Drishti-GS, ACRIMA, ORIGA-light, and HRF datasets respectively. Based on the experimental results and evaluation, the proposed framework closely competes with the state-of-the-art methods using the five public glaucoma datasets without requiring any manually preprocessing step.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.644-647
• /
• 2004

Optical add/drop multiplexers (OADMs), one of the new network elements, will play a key role enabling greater connectivity and flexibility in the dense wavelength-division multiplexing (DWDM) networks. The importance of OADMs is that they allow the optical network to be local transmitting/extraction on a wavelength-by-wavelength basis to optimize traffic, efficient network utilization, network growth, and to enhance network flexibility. Also, the automatic assembly system of micro optical filters and fibers is a key technology in the development of optical modules with high functionality. Recently, one of remarkable tends in the development of optical communication industry is the miniaturization and integration of products. In this research, we have developed a system capable of automatic alignment of a film filter and a lensed fiber in order to improve the speed and losses in the optical fiber to filter alignment of optical modules. Using the developed automatic alignment system and silicon optical benches, we have fabricated the micro OADM and measured the insertion loss and aging effect.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.47-54
• /
• 2023

For Bengali music emotion classification, deep learning models, particularly CNN and RNN are frequently used. But previous researches had the flaws of low accuracy and overfitting problem. In this research, attention-based Conv1D and BiGRU model is designed for music emotion classification and comparative experimentation shows that the proposed model is classifying emotions more accurate. We have proposed a Conv1D and Bi-GRU with the attention-based model for emotion classification of our Bengali music dataset. The model integrates attention-based. Wav preprocessing makes use of MFCCs. To reduce the dimensionality of the feature space, contextual features were extracted from two Conv1D layers. In order to solve the overfitting problems, dropouts are utilized. Two bidirectional GRUs networks are used to update previous and future emotion representation of the output from the Conv1D layers. Two BiGRU layers are conntected to an attention mechanism to give various MFCC feature vectors more attention. Moreover, the attention mechanism has increased the accuracy of the proposed classification model. The vector is finally classified into four emotion classes: Angry, Happy, Relax, Sad; using a dense, fully connected layer with softmax activation. The proposed Conv1D+BiGRU+Attention model is efficient at classifying emotions in the Bengali music dataset than baseline methods. For our Bengali music dataset, the performance of our proposed model is 95%.

• Journal of the Korea Society for Simulation
• /
• v.18 no.3
• /
• pp.47-59
• /
• 2009

Due to the application-specific nature of wireless sensor networks, the sensitivity to such a requirement as data reporting latency may vary depending on the type of applications, thus requiring application-specific algorithm and protocol design paradigms which help us to maximize energy conservation and thus the network lifetime. In this paper, we propose a novel delay-adaptive sensor scheduling scheme for energy-saving data gathering which is based on a two phase clustering (TPC). The ultimate goal is to extend the network lifetime by providing sensors with high adaptability to the application-dependent and time-varying delay requirements. The TPC requests sensors to construct two types of links: direct and relay links. The direct links are used for control and forwarding time critical sensed data. On the other hand, the relay links are used only for data forwarding based on the user delay constraints, thus allowing the sensors to opportunistically use the most energy-saving links and forming a multi-hop path. Simulation results demonstrate that cluster-based delay-adaptive data gathering strategy (CD-DGS) saves a significant amount of energy for dense sensor networks by adapting to the user delay constraints.