• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.375-390
• /
• 2024

Edge computing architecture has effectively alleviated the computing pressure on cloud platforms, reduced network bandwidth consumption, and improved the quality of service for user experience; however, it has also introduced new security issues. Existing anomaly detection methods in big data scenarios with cloud-edge computing collaboration face several challenges, such as sample imbalance, difficulty in dealing with complex network traffic attacks, and difficulty in effectively training large-scale data or overly complex deep-learning network models. A lightweight deep-learning model was proposed to address these challenges. First, normalization on the user side was used to preprocess the traffic data. On the edge side, a trained Wasserstein generative adversarial network (WGAN) was used to supplement the data samples, which effectively alleviates the imbalance issue of a few types of samples while occupying a small amount of edge-computing resources. Finally, a trained lightweight deep learning network model is deployed on the edge side, and the preprocessed and expanded local data are used to fine-tune the trained model. This ensures that the data of each edge node are more consistent with the local characteristics, effectively improving the system's detection ability. In the designed lightweight deep learning network model, two sets of convolutional pooling layers of convolutional neural networks (CNN) were used to extract spatial features. The bidirectional long short-term memory network (BiLSTM) was used to collect time sequence features, and the weight of traffic features was adjusted through the attention mechanism, improving the model's ability to identify abnormal traffic features. The proposed model was experimentally demonstrated using the NSL-KDD, UNSW-NB15, and CIC-ISD2018 datasets. The accuracies of the proposed model on the three datasets were as high as 0.974, 0.925, and 0.953, respectively, showing superior accuracy to other comparative models. The proposed lightweight deep learning network model has good application prospects for anomaly traffic detection in cloud-edge collaborative computing architectures.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.8
• /
• pp.1647-1653
• /
• 2004

In a contention-based wireless LAN protocol, maximization of channel utilization and fair bandwidth allocations are main topics to deal with. But it is very difficult to achieve these two goals simultaneously. Many studies have been done to achieve these goals. In this paper we propose a control mechanism to support fair transmissions among traffic classes in IEEE 802.11e Wireless LAN. The proposed algorithm uses short-term and long-term transmission times of each traffic classes to control their $CW_{min}$ for fairness. The proposed algorithm don't need to know the exact number of nodes in the networks to support fairness as other studies do. Furthermore any modifications in AP and mobile hosts are not required.

• Journal of Communications and Networks
• /
• v.8 no.2
• /
• pp.228-233
• /
• 2006

Survivability is a very important requirement for the deployment of broadband networks because out of service links can affect volumes of traffic even if it is a very short time. And the data paths of broadband networks, which are critical for traffic engineering, are always necessary to be well protected. The procedure of protection or restoration for a path is initiated when failure is detected within the working path. In order to minimize the influence on transmission quality caused by the failure of links and to provide a definite time for the recovery from the failure, the protection switching time (PST) should be carefully considered in the path arrangement. Several researches have been devoted to construct the protection and restoration schemes of data paths over dense wavelength division multiplexing (DWDM) networks, however, there was rare research on the design of data paths with guaranteed protection switching time. In this paper, the PST-guaranteed scheme, which is based on the concept of short leap shared protection (SLSP), for the arrangement of data paths in DWDM networks is proposed. The proposed scheme provides an efficient procedure to determine a just-enough PST-guaranteed backup paths for a working path. In addition to selecting the PST-guaranteed path, the network cost is also considered in a heuristic manner. The experimental results demonstrate that the paths arranged by the proposed scheme can fully meet the desired PST and the required cost of the selected path is competitive with which of the shared path scheme.

• Journal of Korean Society of Transportation
• /
• v.34 no.5
• /
• pp.421-436
• /
• 2016

Public-Private-Partnerships tend to be committed high project development cost and recover the cost through future revenue during the operation period. In general, long-term concession can bring on more revenue to private investors, but short-term concession less revenue due to the short recovering opportunities. The concession period is usually determined by government in advance or by the private sectors's proposal although it is a very crucial factor for the PPPs. Accurate traffic forecasting should be most important in planing and evaluating the operation period in that the forecasted traffic determines the project revenue with user fees in PPPs. In this regards, governments and the private investors are required to consider the traffic forecast risk when determining concession period. This study proposed a model for the optimal concession period in the PPPs transportation projects. Monte Carlo simulation was performed to find out the optimal concession period while traffic forecast uncertainty is considered as a project risk under the expected return of the private sector. The simulation results showed that the optimal concession periods are 17 years and 21 years at 5.5% and 7% discount level, respectively. This study result can be applied for the private investors and/or any other concerned decision makers for PPPs projects to set up a more resonable concession period.

• Journal of Haehwa Medicine
• /
• v.20 no.1
• /
• pp.127-135
• /
• 2011

Objectives : The purpose of this study was to investigate the effect of treatment with Sacro-Occipital Technique block on musculoskeletal pain caused by traffic accident by analysis of the Visual Analogue Scale(VAS), Neck Disability Index(NDI), Pain Disability Index(PDI), Oswestry Low back Pain Disability Index(ODI) and Short Form - McGill Pain Qusetionnaire (SF-MPQ). Methods : This study carried out on 18 patients who have received hospital treatment in Daejeon Univ. Dun-San Oriental Hospital. Control group got acupunture-therapy, herbal medication, physical therapy and Experimental group got all the therapies and treatment with Sacro-Occipital Technique block. We measured VAS, NDI, PDI, ODI and SF-MPQ on 1st day and 7 days later. Results : After being treated by our methods, Both group were improved in VAS, NDI, PDI, ODI, and SF-MPQ. Especially, Experimental group was significantly meaningful improved in VAS, PDI, and ODI. Control group was significantly meaningful improved in VAS and SF-MPQ. But, differences between control and experimental group were nonsignificant. Conclusions : The results suggest that treatment with Sacro-Occipital Technique block is not significantly meaningful but gives a positive impact on musculoskeletal pain caused by traffic accident. But further long term study in a large scale is needed.

• International Journal of Highway Engineering
• /
• v.16 no.4
• /
• pp.127-134
• /
• 2014

PURPOSES : To operate more efficient traffic management system, it is utmost important to detect the change in congestion level on a freeway segment rapidly and reliably. This study aims to develop classification method of congestion change type. METHODS: This research proposes two classification methods to capture the change of the congestion level on freeway segments using the dedicated short range communication (DSRC) data and the vehicle detection system (VDS) data. For developing the classification methods, the decision tree models were employed in which the independent variable is the change in congestion level and the covariates are the DSRC and VDS data collected from the freeway segments in Korea. RESULTS : The comparison results show that the decision tree model with DSRC data are better than the decision tree model with VDS data. Specifically, the decision tree model using DSRC data with better fits show approximately 95% accuracies. CONCLUSIONS : It is expected that the congestion change type classified using the decision tree models could play an important role in future freeway traffic management strategy.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.32 no.5
• /
• pp.35-42
• /
• 2004

This study examines the traffic noise reduction effects of the diverse efficiency features of a sunken square, and investigates the reduction effects and characteristics of noise alterations according to the construction conditions in order to verify its possibility as a new noise reduction facility. The measured data were analyzed. As a result, (1) the traffic noise from the roads around the sunken squares exceeded the regulatory standards; (2) the sunken squares were protected from noise; (3) the sunken spaces yielded a greater reduction by 6∼7dB than the flat ground thanks to the diffraction effect; and (4) there were fine changes to noise according to the conditions of space construction. In short, a sunken square has the potential to afford new applications as a noise reduction facility in a highly concentrated downtown with the existing defects solved. In addition, its use for effective noise control is evident based on the noise characteristics of a space. Future studies need to include efforts to provide clearer quantitative data, as well as research on shape, layout, and materials.

• Journal of Korean Society of Steel Construction
• /
• v.22 no.3
• /
• pp.209-218
• /
• 2010

The fatigue phenomenon, which is induced by stress accumulation due to the repeated loading of vehicles in the long term, is one of the main factors of the span of life of a steel bridge. In this paper, the effects of traffic properties on the fatigue life of ordinary short- and medium-span steel plate girder bridges that are exposed to relatively large dynamic effects are investigated. From the analysis, it was known that the fatigue life of the bridge becomes shorter with increasing traffic volume and number of large vehicles, and is affected by the weights of the vehicles. Based on the analysis results, a new parameter that can represent the traffic property that affects the fatigue life of the subject bridge is suggested, and the validity of the parameter is confirmed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.2
• /
• pp.647-669
• /
• 2016

Multiple attribute for decision making including user preference will increase the complexity of route selection process. Various approaches have been proposed to solve the optimal route selection problem. In this paper, multi attribute decision making (MADM) algorithms such as Simple Additive Weighting (SAW), Weighted Product Method (WPM), Analytic Hierarchy Process (AHP) method and Total Order Preference by Similarity to the Ideal Solution (TOPSIS) methods have been proposed for acoustic signature based optimal route selection to facilitate user with better quality of service. The traffic density state conditions (very low, low, below medium, medium, above medium, high and very high) on the road segment is the occurrence and mixture weightings of traffic noise signals (Tyre, Engine, Air Turbulence, Exhaust, and Honks etc) is considered as one of the attribute in decision making process. The short-term spectral envelope features of the cumulative acoustic signals are extracted using Mel-Frequency Cepstral Coefficients (MFCC) and Adaptive Neuro-Fuzzy Classifier (ANFC) is used to model seven traffic density states. Simple point method and AHP has been used for calculation of weights of decision parameters. Numerical results show that WPM, AHP and TOPSIS provide similar performance.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2002.05a
• /
• pp.418-425
• /
• 2002

Bluetooth is an emerging technology expected to provide users with short range, low cost, pico-cellular wireless connectivity. The access to the medium for Bluetooth is based on a Master driven Time Division Duplexing (TDD) scheme. A slave transmits packets in the reverse slot only after the master polls the slave (or transmits a packet to the slave) in a forward slot. The master transmits packets to a slave in even slots while the slave transmits packets to the master in an odd slot. The way in which the master schedules packets transmission to slaves or polls them determines system performance. In this paper. we propose a traffic adaptive MAC scheduling scheme for Bluetooth. The proposed scheme adopts the ISAR (Intelligent Segmentation and Reassembly) policy, which adjusts the packet size to the traffic patterns, to adapt the polling frequency to the traffic conditions. Also for achieving fairness among master-slave connections our scheme includes a priority policy assinging prioritised service tlimes to each connection. By considering a scenario where a Bluetooth master is used as wireless access point to the Internet, we show that our scheme improve the system throughput and average queue delay with regard to a naive Round Robin (RR) scheme.