• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.9
• /
• pp.1688-1695
• /
• 2016

Underwater communication has multipath error because of reflection by sea-level and sea-bottom. The multipath of underwater channel causes signal distortion and error floor. The excessive multipath encountered in underwater communication channel creates inter symbol interference, which is a limiting factor to achieve a high data rate and bit error rate performance. Therefore, to increase throughput efficiency and improve performance, this paper consider FTN (Faster-than-Nyquist) signalling based on turbo equalization. FTN signalling is a technique of transmitting information at a rate higher than the allowed Nyquist limit. This paper presented efficient decoder structure of FTN transmission in the environment of multipath underwater channel and we compare the performance between FTN method and conventional punctured method in lake experimentation. As a results of lake experiment, we confirmed FTN method based on turbo equalization is applicable and efficiency in underwater communication.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.11a
• /
• pp.252-253
• /
• 2014

최근 채널 효율을 향상시키기 위하여 다중안테나(MIMO, multi-input multi-output) 기술이 연구되고 있다. 다중안테나 기술은 송 수신 안테나 개수를 증가시켜 수신 성능을 향상시키고 대역 효율을 향상시키지만 송신 안테나 개수를 늘려야만 대역 효율을 증가시킬 수 있으므로 기존 방송 시스템에 적용하기에는 비용이 많이 드는 단점이 있다. 이에 본 논문은 방송 시스템에 적합한 SIMO(single-input multi-output) 시스템에 FTN(faster than Nyquist)을 적용하여 대역 효율을 증가시키고 수신 성능을 향상시키고자 한다. 또한 MRC(maximum ratio combining) 및 경판정(hard decision)을 적용하여 수신 안테나 개수에 따른 수신 성능을 컴퓨터 실험을 통하여 알아본다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.5
• /
• pp.967-974
• /
• 2017

In this paper, we propose an inter-symbol interference (ISI) estimation scheme based on the maximum a posteriori (MAP) algorithm for faster-than-Nyquist (FTN) systems. Unfortunately, the ISI estimator based on the MAP algorithm requires relatively high computational complexity. To reduce the complexity of the MAP based ISI estimator, we propose a hybrid ISI estimation scheme based on the MAP and successive interference cancellation (SIC) algorithms. The proposed scheme not only offers good ISI estimation performances but also requires reasonably low complexity.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05c
• /
• pp.390-395
• /
• 2002

MPLS는 고속 전달, 트랙픽 엔지니어링 그리고 가상 사설망 서비스를 가능하게 한다. MPLS 레이블 스위치와 제어기 사이에서 동작하는 GSMP 프로토콜은 연결, 구성, 장애, 성능관리 및 동기화 기능을 제공하는 개방형 인터페이스 프로토콜이다. GSMP가 적용된 개형형 인터페이스의 망 관리 기능은 제어기 또는 레이블 스위치에 위치할 수 있다. 본 논문에서는 MPLS 레이블 스위치의 단순화 및 자원 이용의 효율성을 높이기 위하여 SNMP 에이전트가 제어기에 탑재되는 망 관리 모델을 제시하였다. MPLS-LSR 및 FTN MIB와 GSMP 메시지의 매핑 시나리오를 제시하였으며, 제어기의 구현을 통하여 본 논문에서 제시된 망 관리 모델의 실현성을 확인하였다.

• Proceedings of the IEEK Conference
• /
• 2002.06c
• /
• pp.169-172
• /
• 2002

In this paper, we consider the problem of embedding Fibonacci linear array, Fibonacci mesh, Fibonacci tree into Fibonacci circulants and between Fibonacci cubes and Fibonacci circulants. We show that the Fibonacci linear array of order n , Ln is a subgraph of the Fibonacci circulants of order n , En with En◎ Ln,n≥0 , the Fibonacci mesh of order (nt,n2), M(n,.nT)with S2n.1 f( M(n.れ)닌 M(n.1.n.1)), 52れ 늰( M(n.n.1)띤 M(H.n-1)) and the Fibonacci tree-lof order n, FT/sub n/ with ∑/sub n+3/⊇ FTn , n≥0, the Fibonacci tree-ll of order n , Tれ with ∑/sub n/⊇ Tn Fu퍼hermore, 낀e show that the Fibonacci cubes of order n , rn is subgraph of the Fibonacci circulants of order n , En and inversely rn can be embedded into En with expansion 1, dilation n -2 and congestion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1795-1811
• /
• 2019

In order to achieve rapid and accurate detection of vehicle objects in complex traffic conditions, we propose a novel vehicle detection method. Firstly, more contextual and small-object vehicle information can be obtained by our Joint Feature Network (JFN). Secondly, our Evolved Region Proposal Network (EPRN) generates initial anchor boxes by adding an improved version of the region proposal network in this network, and at the same time filters out a large number of false vehicle boxes by soft-Non Maximum Suppression (NMS). Then, our Mask Network (MaskN) generates an example that includes the vehicle occlusion, the generator and discriminator can learn from each other in order to further improve the vehicle object detection capability. Finally, these candidate vehicle detection boxes are optimized to obtain the final vehicle detection boxes by the Fine-Tuning Network(FTN). Through the evaluation experiment on the DETRAC benchmark dataset, we find that in terms of mAP, our method exceeds Faster-RCNN by 11.15%, YOLO by 11.88%, and EB by 1.64%. Besides, our algorithm also has achieved top2 comaring with MS-CNN, YOLO-v3, RefineNet, RetinaNet, Faster-rcnn, DSSD and YOLO-v2 of vehicle category in KITTI dataset.