Digital Library[ Search Result ]
Deep Ensemble Network with Explicit Complementary Model for Accuracy-balanced Classification
http://doi.org/10.5626/JOK.2019.46.9.941
One of the major evaluation metrics for classification systems is average accuracy, while accuracy deviation is another important performance metric used to evaluate various deep neural networks. In this paper, we present a new ensemble-like fast deep neural network, Harmony, that can reduce the accuracy deviation among categories without degrading the overall average accuracy. Harmony consists of three sub-models: the Target model, Complementary model, and Conductor model. In Harmony, an object is classified by using either the Target model or the Complementary model. The Target model is a conventional classification network for general categories, while the Complementary model is a classification network specifically for weak categories that are inaccurately classified by the Target model. The Conductor model is used to select one of the two models. The experimental results indicate that Harmony accurately classifies categories and also, reduces the accuracy deviation among the categories.
Traffic Steering System with Dual Connectivity for Video Streaming Services
Gi Seok Park, Hyunmin Noh, Jae Jun Ha, Hyung Jun Kim, Sang Heon Shin, Dong Hyun Kim, Jong Hwan Ko, Jeung Won Choi, Hwangjun Song
http://doi.org/10.5626/JOK.2019.46.6.577
In this paper, we propose a traffic steering system with dual connectivity to provide stable video streaming services for users by steering portion of the macrocell traffic into small cells. The proposed system achieves a good balance between fairness and social welfare in terms of video quality by allocating the radio resource of the macro base station. The user data flow is divided into two channels toward the macro base station and the small cell AP, and the users receive their data from both. In the proposed system, the fountain code is adopted to overcome practical issues in the dual connectivity. Moreover, the SDN is employed not only to rapidly react to time-varying network condition, but also to control network resources efficiently. The proposed system is implemented using NS-3. The simulation results show that the proposed system can achieve much better performance compared with existing traffic steering algorithms.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr