Isometry-Based Shape-Adaptive Fractal Coding for Images

Fractal image coding is an effective method to eliminate the image redundancy through piecewise self-transformability. The fractal code consists of a set of contractive affine transforms. To improve the performance when a range block experiences large error, we usually partition the range block into square or nonsquare subrange blocks for two- or multilevel fractal coding. In this paper, we find an inherent property of fractal coding that can be used to decide the edge orientation of a range block. Then this property is used for shape-adaptive fractal coding (SAFC). In SAFC, the top-level range block is partitioned into square or nonsquare (rectangle or triangle) subrange blocks for multilevel fractal encoding. Here, the maximum size of the range block can be the same as that of the whole image size while the minimum size is 4×4. In SAFC, no additional computations are required to obtain the edge orientation of a range block. Instead, we propose an edge-orientation detector, where the edge orientation of a range block is obtained during the fractal encoding process. According to our simulation results, SAFC can reduce the bit rate requirement of the conventional fractal coding scheme.     

Distortion Minimization with Fast Local Search for Fractal Image Compression

Optimal fractal image coding is an NP-hard combinatorial optimization problem, which consists of finding in a finite set of contractive affine mappings one whose unique fixed point is closest to the original image. Current fractal image schemes are based on a greedy suboptimal algorithm known as collage coding. In a previous paper, Hamzaoui, Hartenstein, and Saupe proposed a local search algorithm that iteratively improves an initial solution found by collage coding. For a standard fractal scheme based on quadtree image partitions, peak-signal-to-noise ratio (PSNR) gains are up to 0.8 dB. However, the algorithm is time-consuming because it involves many iteration steps, each of which requires the computation of the fixed point of an affine mapping. In this paper, we provide techniques that drastically reduce the complexity of the algorithm. Moreover, we show that the algorithm is also successful with a state-of-the-art fractal scheme based on highly adaptive image partitions.     

Image Compression Based on Multistage Vector Quantization

This paper presents a new three-stage vector quantization system for the compression of images. It uses some simple schemes including error block classifier, search order coding (SOC), and index vector coding. The error block classifier preserves the edge blocks and discards the psychovisually redundant texture blocks in the last stage. The index vector coding encodes the combination of quantization indexes of the last two stages, and the SOC is used for encoding the quantization index of the first stage. The proposed system can achieve better compression performance than the conventional multistage vector quantization systems.     

一种基于自适应Canny算子的舰船红外图像边缘检测方法

提出了一种用于舰船红外图像边缘检测的自适应Canny算子边缘检测方法。先利用图像的局部最大标准差值为参考值设定高斯滤波尺度参数,然后采用基于梯度幅度直方图的类间方差最大法计算Canny算子的高低闽值,从而实现边缘的自动检测。实验结果表明,采用该方法检测出的舰船边缘清晰连续,噪声较少,优于传统的Canny算子边缘检测方法,并且具有很好的工程适用性。     

Low Complexity Minimum Mean Square Error Channel Estimation for Adaptive Coding and Modulation Systems

Performance of the Adaptive Cod- ing and Modulation （ACM） strongly depends on the retrieved Channel State Information （CSI）, which can be obtained using the chan- nel estimation techniques relying on pilot sy- mbol transmission. Earlier analysis of methods of pilot-aided channel estimation for ACM systems were relatively little. In this paper, we investigate the performance of CSI prediction using the Minimum Mean Square Error （MMSE） channel estimator for an ACM system. To solve the two problems of MMSE： high compu- tational operations and oversimplified assum- ption, we then propose the Low-Complexity schemes （LC-MMSE and Recursion LC-MMSE （R-LC-MMSE））. Computational complexity and Mean Square Error （MSE） are presented to evaluate the efficiency of the proposed algo- rithm. Both analysis and numerical results sh- ow that LC-MMSE performs close to the well- known MMSE estimator with much lower com- plexity and R-LC-MMSE improves the appli- cation of MMSE estimation to specific cir- cumstances.     

一种面向智慧协同网络的自适配路由策略研究

现有互联网网络体系和机制相对"静态"和"僵化",缺乏支持智慧网络的有效机制.要从根本上解决现有互联网存在的严重弊端,必须创建新的网络理论体系.本文在智慧协同网络"三层"、"两域"体系结构下,针对"网络组件层"的路由自适配问题展开研究,提出基于生物启发的转发网络族群自适配路由策略,实现族群内路由组件之间的智慧协调、动态重构和优化决策,有效解决现有路由策略的静态、僵化等问题.通过数学分析证明:如果设定模型参数μ∈(0,1),提出的自适配路由策略可以始终保持稳定性.最后,通过原型系统验证了提出自适配路由策略是切实可行的,能够提高网络的承载业务数量和提升用户体验.     

DVB-S2 ACM技术中的信噪比估计

为了在DVB-S2系统自适应编码调制(ACM)技术中实现对信道信噪比的精确、高效估计,提出了一种基于子空间分解的数据辅助类信噪比估计算法。计算机仿真结果表明,在较宽的信噪比范围内,该算法的性能优于数据辅助类的最大似然估计算法和经典的基于子空间分解的ED算法,估计精度高,计算复杂度低,非常适合在DVB-S2系统ACM技术中应用。     

Adaptive Coding and Modulation for Mobile Wireless Access Via High Altitude Platforms

In this paper we are concerned with broadband wireless access via high altitude platform system, providing the Internet access and broadband multimedia services to passengers equipped with WLAN terminals connecting through a collective terminal mounted on the train. The main challenge in such scenario is the development of efficient and reliable radio interface for the broadband communication link in the mobile wireless access segment. We are focusing on performance analysis of the adaptive coding and modulation scheme in the communication link between a high altitude platform and a collective terminal on-board moving train. In order to increase the reliability of the communication system in a fading environment we also exploit space and platform diversity. The proposed approach significantly increases the throughput of the wireless access system, while bit error rate remains below the target value regardless of the considered propagation environment.Tomaz Javornik received his B.Sc., M.Sc. and Ph.D. degrees in electrical engineering from the University of Ljubljana, Slovenia, in 1987, 1990 and 1993, respectively. He joined the Jozef Stefan Institute in 1987, where he currently works as a researcher in the Department of Digital Communications and Networks. He is involved in the study of digital radio-relay systems, modulation techniques, coding, adaptive signal processing and digital mobile communication systems.Mihael Mohorcic received B.Sc., M.Sc. and Ph.D. degrees in Electrical Engineering from the University of Ljubljana, Slovenia, in 1994, 1998 and 2002, respectively, and M.Phil. degree in Electrical Engineering from University of Bradford, UK, in 1998. He is a research fellow in the Department of Digital Communications and Networks at the Jozef Stefan Institute. In 1996/1997 he spent 12 months as a Visiting Scientist at University of Bradford, Bradford, UK. His research interests include development and performance evaluation of network protocols and architectures for mobile and wireless communication systems, and resource management in satellite and high altitude platforms networks with the emphasis on routing algorithms and traffic engineering. He is a member of IEEE.Ales Svigelj received his B.Sc., M.Sc. and Ph.D. degrees in Electrical Engineering from the University of Ljubljana, Ljubljana, Slovenia, in 1997, 2000 and 2003 respectively. He is a research associate in the Department of Digital Communications and Networks at the Jozef Stefan Institute. In 2000/2001 he spent one year as a visiting researcher at Leeds Metropolitan University in Leeds, UK. He has participated in several national and international projects. His research interests concern the development of telecommunications systems, network protocols and architectures for satellite, high altitude platforms and terrestrial mobile communication systems. In 2004 he was awarded with The Jozef Stefan Golden Emblem Prize for outstanding contributions made to science in Doctoral theses in the field of natural sciences in Slovenia.Igor Ozimek received his B.Sc., M.Sc. and Ph.D. degrees in electrical engineering from the University of Ljubljana, Slovenia in 1980, 1988 and 1993, respectively. Since 1980 he has been with the Jozef Stefan Institute, Ljubljana, where he works as a researcher. His current interests include digital communications,DSP processing and computer networks.Gorazd Kandus received B.Sc., M.Sc. and Ph.D. degrees in Electrical Engineering from the University of Ljubljana, Ljubljana, Slovenia, in 1971, 1974 and 1991, respectively. He is currently the head of the Department of Digital Communications and Networks at the Jozef Stefan Institute and a Professor at the Faculty of Electrical Engineering, Computer Science and Information Technology, University of Maribor. He spent one year at Worchester Polytechnic Institute, Worchester, MA, as a Fulbright Fellow and 5 months as a Visiting Scientist at the University of Karlsruhe, Germany. His main research interests include design and simulation of mobile and wireless communication systems and development of new telecommunication services. He is a member of IEEE and Upsilon Pi Epsilon.     

自适应编码调制的一种高效半盲检测算法

给出一种基于协议信息的自适应编码调制高效半盲检测算法,该算法同时采用信号多序列抽取和信号多频点滑动匹配两种方法,能够有效克服中频调制信号最佳采样点偏差和频率偏差的影响,可以准确高效地识别出信号类型、信道编码方式、调制类型。实验表明,该算法在低信噪比下就能达到很高的识别率,性能明显优于传统模式识别的检测算法,具有较好的实用前景。     

VLC网络中兼具功率与时隙分配的自适应干扰管理机制

为了更好解决目前可见光通信(visible light communication,VLC)中干扰管理方案存在的动态优化问题,提出了一种兼具优化功率分配与时隙分配的自适应干扰管理机制。首先,在每个时隙依据用户的位置建立每个用户的接入点(access point,AP)协作集,寻找所有由最多数量互不干扰用户组成的极大独立集,以此自适应地避免同频干扰;在每个时隙为每个极大独立集采用改进的线性注水功率分配算法为信道自适应地分配发送功率,以此优化每个极大独立集的用户和速率;基于用户和速率、速率公平性及时延公平性的归一化优先因子,选出具有最大优先因子的候选极大独立集,其中包含的用户在该时隙被调度。通过仿真且与代表性文献中的算法比较可得,本文提出的自适应干扰管理与优化资源分配方案在网络频谱利用率、能效、用户速率公平性与时延公平性方面具有明显优势。     

基于自适应窗口固定及传播的 多尺度纹理图像分割

 本文提出一种基于自适应窗口固定及传播的多尺度纹理图像分割方法,在小波域隐马尔科夫树模型(WHMT)的初始分割基础上,根据分割粗尺度上的区域一致性好,细尺度上的边缘准确的特点,利用上下尺度像素之间以及本层邻域像素的马尔科夫性,标记出图像的一致性区域和边缘区域,将一致性区域固定,类标直接下传到下一尺度,边缘区域则利用邻域信息确定出上文权值背景传播到下一尺度,与下尺度一起共同指导图像分割,从而很好的保持了区域均匀性和边缘准确性.同时根据纹理图像区域聚集性的特性,利用基于多项式展开和置信区间交叉(LPA_-ICI)方法找出各类区域聚集的物理位置中心,融入上下文权值背景中,使得指导分割策略能够更好的进行.实验表明,对于合成纹理图像来说,本文提出的多尺度融合算法在均匀区域内部及区域边界都大为改善,而且无须进行参数的训练,使算法快速的完成.