Journal Articles - Computer and Information Technology - 2019

Permanent URI for this collection

http://repository.vlu.edu.vn:443/handle/123456789/100

Browse

Browsing Journal Articles - Computer and Information Technology - 2019 by Title

  • Publication
    OCR Error Correction for Unconstrained Vietnamese Handwritten Text
    ( 2019)
    Quoc-Dung Nguyen
    ;
    Duc-Anh Le
    ;
    Ivan Zelinka
    Post-processing is an essential step in detecting and correcting errors in OCR-generated texts. In this paper, we present an automatic OCR post-processing model which comprises both error detection and error correction phases for OCR output texts of unconstrained Vietnamese handwriting. We propose a hybrid approach of generating and scoring correction candidates for both non-syllable and real-syllable errors based on the linguistic features as well as the error characteristics of OCR outputs. We evaluate our proposed model on a Vietnamese benchmark database at the line level. The experimental results show that our model achieves 4.17% of character error rate (CER) and 9.82% of word error rate (WER), which helps improve both CER and WER of an attention-based encoder-decoder approach by 0.5% and 3.5% respectively on the VNOnDB-Line dataset of the Vietnamese online handwritten text recognition competition (VOHTR2018). These results outperform those obtained by various recognition systems in the VOHTR2018 competition.
  • Publication
    Resource Sharing and Segment Allocation Optimized Video Streaming over Multi-hop Multi-path in Dense D2D 5G Networks
    ( 2019)
    Quang-Nhat Tran
    ;
    Nguyen-Son Vo
    ;
    Minh-Phung Bui
    ;
    Van-Ca Phan
    ;
    Zeeshan Kaleem
    ;
    Trung Q. Duong
    The rapid increase in mobile users (MUs) and various video applications and services (VASs) poses a set of challenges to 5G networks. Although device-to-device (D2D) communications and resource reuse techniques can improve the streaming performance of VASs, they have not utilized the fact that there are many MUs with available sharing downlink resources, namely cellular users (CUs), and many videos cached in dense MUs to establish an efficient video streaming session. In this paper, we propose a downlink resource sharing and segment allocation (RSA) optimization problem and solve it for high video streaming performance over multi-hop multi-path (MHMP) in dense D2D 5G networks. Particularly, the RSA provides a maximum capacity for the CUs that share their downlink resources with the D2D hops in each path by finding the optimal downlink resource sharing-receiving pairs between the CUs and the D2D hops. Then, the segments of a video cached in different D2D helpers (DHs) are sent to the D2D requester (DR) over the MHMP. By finding which segments are allocated to which paths for sending, the reconstructed distortion of the received segments at the DR is minimized for high playback quality. Simulation results are shown to investigate the benefits of the proposed solution compared to other schemes without RSA.
  • Publication
    Social Aware Caching and Resource Sharing Optimization for Video Delivering in 5G Networks
    ( 2019)
    Bùi Minh Phụng
    ;
    Vo Nguyen Son
    ;
    Nguyen, T T
    ;
    Tran, Q N
    ;
    Tran, A T
    The proliferation demand of mobile users (MUs) for video contents, which will occupy up to 78% of data traffic by 2021, poses a serious challenge of system delivery capacity to the macro base stations (MBSs) and the small cell base stations, e.g., femtocell base stations (FBSs), in 5G networks. In this paper, we propose a social-aware caching and resource sharing (SCS) scheme that can help the MBSs and the FBSs relax the backhaul links and provide the MUs with high system delivery capacity. Particularly, we formulate an SCS optimization problem under the constraints on the number of replicas of each video cached in the FBSs and the target signal to interference plus noise ratio (SINR) of the cellular users (CUs) that share the downlink resources. This problem is then solved for maximum system delivery capacity by finding the best placements to cache the videos in the FBSs and the best device-to-device (D2D) pairs shared the same downlink resources with the CUs to offload the videos over D2D communications. Importantly, the behavior of MUs to access the videos and the social relationship of each D2D pair are considered in the SCS optimization problem to efficiently improve the system performance. Simulation results are shown to demonstrate the benefits of the proposed SCS scheme compared to other conventional schemes.
  • Publication
    Social-Aware Caching and Resource Sharing Maximized Video Delivery Capacity in 5G Ultra-Dense Networks
    ( 2019)
    Minh-Phung Bui
    ;
    Nguyen-Son Vo
    ;
    Sang Quang Nguyen
    ;
    Quang-Nhat Tran
    In 5G networks, a massive number of connections of high data rate services, e.g., video streaming services, certainly make the networks deteriorated because of extreme traffic congestion at the backhaul links of macro base stations (MBSs). Although ultra-dense networks (UDNs) have been considered as a promising architecture to stimulate the 5G networks, the congestion problem hampers the UDNs to provide mobile users (MUs), i.e., represented by device-to-device (D2D) pairs and shared downlink resource users (SUs), with high video delivery capacity. In this paper, we propose a social-aware caching and resource sharing (SCS) strategy for video streaming services in 5G UDNs. Particularly, we formulate the SCS problem by taking into account the social relationship of each D2D pair, the available storage of femtocell base stations (FBSs) and transmitters (TXs) of D2D pairs, the target signal to interference plus noise ratio (SINR) of SUs, and the popularity of videos. The SCS problem is then solved for 1) optimal number of caching copies of each video and optimal caching placements in the FBSs and 2) optimal resource sharing allocation between the SUs and the D2D pairs for D2D communications. This way, the workload at the backhaul links of the MBSs can be reduced. Simultaneously, an arbitrary MU can retrieve the videos alternately from the MBSs, FBSs, and TXs at high cache-hit ratio and maximum delivery capacity. Simulation results are analyzed to show the benefits of the proposed SCS strategy compared to other conventional schemes.