• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.5C
• /
• pp.536-541
• /
• 2007

In video coding standard H.264/AVC, motion estimation using multiple reference pictures improves compression efficiency but the efficiency depends upon image content not the number of reference pictures. So, the motion estimation includes a large amount of computation of no worth according to image. This paper proposes fast motion estimation algorithm that removes worthless computation in the motion estimation using multiple reference pictures. The proposed algorithm classifies a block into valid and invalid blocks for the multiple reference pictures and removes the workless computation by applying a single reference picture to the invalid block. To estimate the proposed algorithm's performance, image quality, bit rate, and motion estimation time are compared with ones of the conventional algorithm in the reference software JM 9.5. The simulation results show that the proposed algorithm can considerably save about 38.67% the averaged motion estimation time while keeping the image quality and the bit rate, whose are average values are -0.02dB and -0.77% respectively, as good as the conventional algorithm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.11
• /
• pp.2093-2099
• /
• 2016

This paper describes a design of crypto-processor that supports multiple block cipher algorithms of PRESENT, ARIA, and AES. The crypto-processor integrates three cores that are PRmo (PRESENT with mode of operation), AR_AS (ARIA_AES), and AES-16b. The PRmo core implementing 64-bit block cipher PRESENT supports key length 80-bit and 128-bit, and four modes of operation including ECB, CBC, OFB, and CTR. The AR_AS core supporting key length 128-bit and 256-bit integrates two 128-bit block ciphers ARIA and AES into a single data-path by utilizing resource sharing technique. The AES-16b core supporting key length 128-bit implements AES with a reduced data-path of 16-bit for minimizing hardware. Each crypto-core contains its own on-the-fly key scheduler, and consecutive blocks of plaintext/ciphertext can be processed without reloading key. The crypto-processor was verified by FPGA implementation. The crypto-processor implemented with a $0.18{\mu}m$ CMOS cell library occupies 54,500 gate equivalents (GEs), and it can operate with 55 MHz clock frequency.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.4B
• /
• pp.725-733
• /
• 2000

In this paper, we propose a modified boundary block merging technique for the texture coding of MPEG-4. We propose an ORP(Optimized Region Partitioning) method that partition the VOP-based reference position to minimize the number of coding blocks. The merging possibility is improved by adding +90。and -90。 Rotation merging. We propose a MRM(Multiple Rotation Merging) method which applies the rotation merging in the order of 180。, +90。and -90。. If a pair of boundary blocks has low correlation, existing BBM's padding technique is not efficient. Our padding after merging method gives better result even if it has low correlation. The proposed method showed 5 ~8(%) coding bit reduction at the same PSNR values compared to BBM method.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2671-2673
• /
• 2002

본 논문에서는 주성분분석(principal component analysis; 이하 PCA)기법을 이용, 이차원 영상을 분류하여 다중해상도에서 기존의 그레이 블록 거리(grey block distance; GBD, 이하 GBD)알고리즘과 비교하여 이차원 영상간의 상대적 식별을 더 용이하게 하기 위한 새로운 GBD 알고리즘 방법을 제안한다. 이 제시된 방법은 다중해상도에서 기존의 GBD 알고리즘과 비교해서 영상이 급격히 변화하는 부분의 정보를 잃지 않게 개선할 수 있었다. 모의 실험 결과로부터 기존의 GBD 알고리즘에 비하여 상대적 식별이 더 용이함을 확인하였다.

• Proceedings of the Optical Society of Korea Conference
• /
• 2001.02a
• /
• pp.216-217
• /
• 2001

광섬유와 평면 도파로를 결합한 도파로 구조에서 광 결합 현상은 다양한 응용 가능성 때문에 이론적으로 또는 실험적으로 많은 연구가 되어오고 있다. 광섬유는 단일모드를 이용하고 적당한 곡률이 주어진 블록에 고정한 후 연마하여 클래딩의 일부를 제거하고, 그 상부에 다중모드 평면 도파로(Multi-Mode Overlay Waveguide; MMOW)를 구성한다. 제작된 결합기 구조에서 평면 도파로는 광섬유의 감쇠장 결합(Evanescent field coupling)을 유도하게 되므로, 소자의 특성은 평면 도파로의 구조와 조건에 의해 결정된다. (중략)

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.06a
• /
• pp.357-358
• /
• 2013

H.264 등 동영상 압축 표준에서는 비디오 신호의 시간적 중복 데이터를 제거하기 위해 움직임 추정/보상을 수행한다. 또한 움직임 추정/보상의 정확성을 향상하기 위해 다중 참조프레임을 지원한다. 여러 장의 참조 프레임 중 현재 블록과 가장 유사한 참조 프레임 영역으로부터 움직임 추정/보상을 수행하여 보다 정확한 예측에 의해 잔차신호의 크기가 감소하게 되고, 그 결과 부호화 효율이 더욱 개선되었다. 본 논문에서는 다중 참조 프레임을 사용한 움직임 추정/보상의 효율을 유지하면서도 참조프레임을 나타내는 참조프레임 인덱스 비트를 줄여주어 부호화 효율을 더욱 개선하는 방법을 제안한다. 본 논문에서 제안하는 방법은 움직임 추정/보상 시, 각각의 참조 프레임에서 움직임 추정/보상에 사용되는 예측화소들을 비교하여 다중 참조 프레임이 효과가 있다고 판단 되는 경우에만 다중 참조 프레임 움직임 추정/보상을 수행하고, 다중 참조 프레임이 효과가 없다고 판단 되는 경우에는 단일 참조 프레임 움직임 추정/보상을 적응적으로 수행하였다. 실험결과 제안하는 방법은 다중 참조 프레임 인덱스 부호화에 소요되는 비트를 절감하면서도 부호화 효율을 유지함을 확인 할 수 있었다. 제안하는 방법은 동영상 압축 코덱에 적용되어 압축 성능을 더욱 향상 할 수 있다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.4
• /
• pp.1072-1086
• /
• 1998

Transform, subband, and wavelet transform decompositions are powerful linear transformation tools in image coding because of their decorrelating effects on image pixels, the concentration of energy in a few coefficients, their multirat/multiresolution framework, and their frequency splitting, which allows for efficient coding matched to the statistics of each frequency band and to the characteristics of the human visulal system. Thus, a growingbody of research has bee performed to extend these tools in various kinds of modified formations. Hence, in this paper, an overall survey to achieve a general view on these transformation tools have been attempted. Starting from basic tools such as orthogonal transforms, lapped transforms, QMF(quadrature mirror filter) subband filter banks, and wavelet transforms, their hierarchical extensions, vector extensions, and linear time-varying extensions are investugated in detail.

• Journal of Satellite, Information and Communications
• /
• v.6 no.1
• /
• pp.19-27
• /
• 2011

In this paper, we propose the data stream allocation algorithms for maximizing sum capacity of downlink multiuser MIMO (Multiple-input Multiple-output) systems with BD (Block Diagonalization). The conventional BD precoding algorithms maximize the capacity by controlling power against channel gain of each user. In multiuser MIMO systems, however, the number of data streams for each user can be used to as another control parameter, which determines the capacity. This paper proposes the data stream allocation algorithm of BD for increasing capacity in multiuser MIMO systems. The proposed algorithm allocates unequal bit stream to each user based on channel matrix of each user for maximizing sum capacity. It is proved that proposed algorithm can achieve the significantly improved sum capacity by computer simulation.

• Journal of Internet Computing and Services
• /
• v.14 no.6
• /
• pp.41-47
• /
• 2013

GP-GPUs are general purposed GPUs for numerical computation based on multiple threads which are originally for graphic processing. GP-GPUs provide cache memory in a form of shared memory which user programs can access directly, unlikely typical cache memory. In this research, we implemented the parallel block LU decomposition program to utilize cache memory in GP-GPUs. The parallel blocked LU decomposition program designed with Nvidia CUDA C run 7~8 times faster than nun-blocked LU decomposition program in the same GP-GPU computation environment.

• The KIPS Transactions:PartA
• /
• v.17A no.6
• /
• pp.275-280
• /
• 2010

In this paper, we discuss and propose an efficiently parallelized block cipher algorithm on the CELL BE processor. With considering the heterogeneous feature of the CELL BE architecture, we apply different encoding/decoding methods to PPE and SPE and improve the throughput. Our implementation was fully tested, with execution results showing achievement of high throughput, capable of supporting as high network speed as 2.59 Gbps. Compared to various parallel implementations on multi-core systems, our approach provides speedup of 1.34 in terms of encoding/decoding speed.