Table of Content

31 July 2005, Volume 23 Issue 4

Previous Issue    Next Issue

Articles

Super Resolution Reconstruction of the Video Image Sequence Based on PDE Diffusion

HAN Yu-bing, WU Le-nan

2005, 23(4):  331-335. 

Asbtract ( 320 )   PDF (255KB) ( 69 )  

Related Articles | Metrics

An adaptive filter for video image sequence super resolution reconstruction is proposed on the basis of the weighted least square and PDE regularization in this paper.The motion compensation matrix and weighted matrix are thoroughly discussed and the recursive filter formula is deduced.Experimental results demonstrate the effectiveness of the proposed method which is both adaptive and robust.

Dynamic Routing Control Based on Energy Balancing in Underwater Acoustic Networks

ZHANG Hong-tao, JIANG Wei-dong, TONG Feng, LU Ji-ren

2005, 23(4):  336-340. 

Asbtract ( 277 )   PDF (226KB) ( 81 )  

Related Articles | Metrics

In order to solve the problem that power supply of some relay nodes in underwater acoustic networks is exhausted earlier due to overuse, a new dynamic routing control method is presented.Lifetime of nodes is estimated from prediction of the generated message rate and the remaining power.The least lifetime among all nodes is defined as an objective function.The optimal routing with a maximum objective function value is found by using genetic algorithm, which attempts to achieve uniform consumption of the power.Simulation results show that the proposed method can prevent nodes with less lifetime from being heavily used, therefore effectively increasing their lifetime.

Frequency Domain Audio Watermark Embedding Capable of Resisting MP3 Encoding

LIU Wei, WANG Shuo-zhong, ZHANG Xin-peng

2005, 23(4):  341-344. 

Asbtract ( 272 )   PDF (274KB) ( 75 )  

Related Articles | Metrics

Based on the psychoacoustic model Ⅱ and perceptual entropy (PE) used in MP3 encoding, watermark embedding is carried out by modifying the frames with PE above 1 800.By adding a random complex sequence to the spectral components in the medium and low bands in the frequency domain, a watermark is embedded.Extraction of the embedded data is achieved by using a frequency domain correlation approach without the original host signal.Experimental results show that the proposed scheme can achieve high imperceptibility and satisfactory robustness against MP3 encoding.

A Classifier Based on Singular Value Decomposition and Its Application to Face Recognition

HE Yun-hui, ZHAO Li, ZOU Cai-rong

2005, 23(4):  345-349. 

Asbtract ( 313 )   PDF (237KB) ( 67 )  

Related Articles | Metrics

A new kind of classifier based on singular value decomposition of face images which solves the small sample size problem in face recognition tasks is presented in this paper.The method of renewing classifier using the incorrectly classified face image to improve the performance of classifiers is also given.Experimental results show that high recognition rate can be obtained by using only one sample per person and that the renewal of classifiers using only one incorrectly classified sample can further improve the performance of classifiers.

An Iterative Detection Scheme for MIMO-OFDM Systems

HOU Yong-min, RONG Meng-tian

2005, 23(4):  350-354. 

Asbtract ( 390 )   PDF (220KB) ( 123 )  

Related Articles | Metrics

MIMO-OFDM system is a promising scheme for achieving high data rates over the future broadband wireless networks.For traditional MIMO-OFDM, the frequency diversity is not fully utilized.In this paper a novel iterative detection scheme for the MIMO-OFDM system which exploits both frequency and spatial diversity is proposed.Through iterative processing, the extrinsic information can be used to exploit the spatial-frequency diversity of the rich scattering environment, thereby achieving the capacity limit of the system.And for further improvement of the system's performance, a soft interference cancellation based receiver is presented, too. The analysis and simulation results show that the proposed detection scheme achieves an enhanced performance.

Minimum-Rate Capacity and Optimal Power Allocation for MIMO Fading Multiaccess Channels

PAN Cheng-kang, CAI Yue-ming, XU You-yun

2005, 23(4):  355-358. 

Asbtract ( 311 )   PDF (259KB) ( 58 )  

Related Articles | Metrics

The capacity region with average power constraints of MIMO fading multi-access channels based on minimum rate requirements is analyzed and the optimal power allocation strategy is suggested.The strategy is a two-step process, the allocation of the minimum power required to achieve the minimum rates and the optimal allocation of the excess power to maximize the ergodic rates.It is shown the minimum-rate capacity can be achieved by successive decoding.The numerical results show that, the strategy provides notable capacity gain and good tradeoff in performance.

Channel Estimation for MIMO-OFDM Systems Based on Parametric Channel Modeling

JING Ya, WANG Dong-ming, CHEN Ming, CHENG Shi-xin

2005, 23(4):  359-363. 

Asbtract ( 315 )   PDF (258KB) ( 124 )  

Related Articles | Metrics

A low complexity channel estimation algorithm based on the tracking of multipath time delays is presented for broadband multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) systems.The estimator's performance and computation complexity is analyzed.Simulation results show that the tracking of multipath time delays can improve the channel estimation performance and reduce the computation complexity by decreasing the dimension of estimated parameters.

Performance Oriented Customization of On-Chip Memory Capacity

PU Han-lai, LING Ming

2005, 23(4):  364-369. 

Asbtract ( 280 )   PDF (279KB) ( 53 )  

Related Articles | Metrics

A scheme to customize capacity of on-chip memory is presented, which establishes relationship between the on-chip memory capacity and the application performance by analyzing the most frequently accessed instructions or data blocks in a specific application.The best capacity is also calculated, and performance changes predicted.Experimental results show that, with the proposed scheme, the best capacity is generally smaller than 4% of the application size, and the performance is significantly improved, doubled on average.

Packet Tetrad Model: A New Method for Measuring Bottleneck Bandwidth

ZHU Lin, ZHAO Jin

2005, 23(4):  370-374. 

Asbtract ( 364 )   PDF (215KB) ( 84 )  

Related Articles | Metrics

Existing algorithms for measuring the network bottleneck bandwidth are mostly based on the packet pair model.Intrinsic error exists in packet pair model due to the effect of data link layer and physical layer. Incorrect results result due to the filter algorithm with a heavy network load.To resolve these problems, a new model called packet tetrad model is proposed.Because of the difference relation between the two packet pairs in a packet tetrad, disturbance in the data link layer and the physical layer may be eliminated.The packet tetrad model has a high accuracy in measurement.Results of simulated measurements under the ns-2 environment and the actual measurements of networks show that the packet tetrad mode is more accurate and robust than the packet pair model.

Coverage Directed Vector Generation for Functional Verification Using Genetic Algorithm

LUO Chun, YANG Jun, LING Ming

2005, 23(4):  375-379. 

Asbtract ( 359 )   PDF (237KB) ( 123 )  

Related Articles | Metrics

With a new algorithm for coverage directed vector generation, a major challenge in the verification on the register transfer level by simulation is addressed.The approach is based on genetic algorithm, which provides an efficient way in forming feedback from the coverage domain back to the generator.This technique has been applied to the verification of an industrial SoC module, showing encouraging results and indicating that the quality of the verification is improved.

Investigation of Quasi-saturation Effect in Small Size Power VDMOS

ZHANG Jun, CHENG Dong-fang, XU Zhi-ping

2005, 23(4):  380-383. 

Asbtract ( 352 )   PDF (232KB) ( 89 )  

Related Articles | Metrics

In small size power VDMOS transistors, the cause of quasi-saturation effect has been investigated and simulated by software.Due to the resistance distribution of R_j and R_ch in the device, carriers' drift velocity in MOS channel changes from saturation to unsaturation while the carriers' drift velocity in the autoecious JFET channel changes from unsaturation to saturation.When the device is in high V_GS, I_DS is independent of V_GS, resulting in the quasi-saturation effect.

Holographic Reconstruction and Prediction of Acoustic Scattering Field Using a Distributed Source Boundary Point Method

LI Wei-bing, CHEN Jian, BI Chuan-xing, CHEN Xin-zhao

2005, 23(4):  384-388. 

Asbtract ( 301 )   PDF (248KB) ( 79 )  

Related Articles | Metrics

A theoretical model for reconstruction and prediction of acoustic scattering fields is built based on the distributed source boundary point method (DSBPM).In this method, the transform matrices of acoustic quantities are formed with particular matrices at boundary nodes, hologram measurement points, reconstruction points and prediction points generated by the imaginary point sources, which is localized on the normal direction of the boundary nodes and a distance away from the analysis domain.Compared with NAH based on the boundary element method (BEM), variable interpolation, numerical calculation and disposal of singular integral are avoided in this method.It has advantages of fast calculation, high precision, good stability, simplicity and clarity in computation.

Estimating Instantaneous Frequency of Vortex Signal in Pulsing Flow

HUANG Yun-zhi, XU Ke-jun

2005, 23(4):  389-392. 

Asbtract ( 278 )   PDF (170KB) ( 85 )  

Related Articles | Metrics

A model of vortex signal in the state of pulsing flow is studied, and a new frequency measuring technique is proposed.Wavelet transform based on ⅡR filter banks is used to decompose vortex signals.A recursive wavelet algorithm using ⅡR filter banks is adopted to meet the requirement of real-time processing. The decomposed vortex signal is processed with Hilbert transform, and the instantaneous frequency (IF) is then estimated.Simulation results show that the proposed method provides satisfactory accuracy and real-time characteristic, therefore can be used to effectively process vortex signals in pulsing flow.

An Efficient Hierarchical Cluster Storage Structure of Multi-dimensional Data

YANG Ke-hua, HU Kong-fa, DONG Yi-sheng

2005, 23(4):  393-398. 

Asbtract ( 341 )   PDF (286KB) ( 69 )  

Related Articles | Metrics

In this article, the concept of hierarchical B⁺ tree and hierarchical cluster CUBE is put forward according to the hierarchy of dimension in the multi-dimensional schema which has been applied in the preserving and querying of multi-dimensional data.According to this concept, the querying and updating with the hierarchical information have been optimized.The roll-up, roll-down and rotating operation are also supported.Furthermore, the hierarchical structure enables the updating operation of the multi-dimension data's schema.

PBC: A Path-Based Method to Clustering XML Documents

LIANG Zuo-peng, YE Ning, DONG Yi-sheng

2005, 23(4):  399-403. 

Asbtract ( 254 )   PDF (234KB) ( 60 )  

Related Articles | Metrics

Most existing clustering techniques for XML are based on the concept of edit-distance and have two main disadvantages:(1) very high time complexity and (2) difficulty in understanding the description of the resulting clusters.In this paper, a novel approach called path-based clustering (PBC) is presented.Instead of comparing XML documents structure and clustering them directly, the paths contained in these documents are clustered.For each path, a cluster containing documents that have that path is formed.After that, clusters that contain similar sets of documents are combined.The resulting clusters contain documents that share a similar set of paths.Experimental results show the effectiveness and efficiency of this approach.

An Obstacle Avoidance Planning Scheme for Robot Based on Digital Potential Field in C-T Space

WANG Cong-qing, ZHAO Chang-jun

2005, 23(4):  404-407. 

Asbtract ( 265 )   PDF (198KB) ( 80 )  

Related Articles | Metrics

In this paper, an obstacle avoidance scheme for robot using digital potential field in C-T configuration-time (C-T) space is presented.In terms of a grid digital potential filed with repulsion potential energy and attractive potential energy in the C-T space, collision avoidance for robot to the dynamic obstacle is implemented using interpolation of discrete path in the C-T space.Simulation shows that, compared to conventional methods, the scheme has higher computation efficiency, and reduces much explicit computation in C space.The scheme is applicable to avoidance path planning for robot under a dynamic environment.

Grey Incidence Decision-Making with Incomplete Information

LUO Dang, LIU Si-feng

2005, 23(4):  408-412. 

Asbtract ( 242 )   PDF (224KB) ( 60 )  

Related Articles | Metrics

Based on the grey system theory and maximum entropy principle, the grey incidence decisionmaking with incomplete information is discussed by using analytical technique.A incidence degree coefficient formula for interval grey numbers is constructed, which extends classical grey incidence decision-making to the case of interval grey numbers.The case of project objective value with nullity is then discussed and the theory and method are given to fulfill the null values.Based on the satisfactory degree, an algorithm is established for grey interval incidence decision-making.This algorithm makes the decision-makig method more flexible and the result more practical.An example is used to demonstrate the rationality, practicability and validity of the algorithm.

Existence of Positive Solutions to a Class of Positone Boundary Value Problems

SU Xiao-hong, ZHENG Lian-cun, ZHANG Xin-xin

2005, 23(4):  413-416. 

Asbtract ( 244 )   PDF (176KB) ( 48 )  

Related Articles | Metrics

This paper deals with a class of singular Positone boundary value problems.Three sufficient conditions for the existence of positive solutions are established by using the nonlinear alternative fixed-point theorem.

Modeling Research of Production Logistics System Based on Fuzzy Graph Theory

WU Yi-sheng, LONG Wen, LI Xiang-quan, WANG Ning-sheng

2005, 23(4):  417-419. 

Asbtract ( 248 )   PDF (150KB) ( 66 )  

Related Articles | Metrics

A new method for modeling the production logistics system is put forward by applying the graph theory.The multi-directioned fuzzy graph and the generalized fuzzy matrix of production logistics system are established integrating the case.It indicates that the model can clearly and quantitatively describe the relation between the various steps of production logistics system.

Multidisciplinary Design Optimization: Its Methodology and Application to Vehicle Design

WU Bao-gui, HUANG Hong-zhong, YUAN Wei

2005, 23(4):  420-423. 

Asbtract ( 310 )   PDF (188KB) ( 92 )  

Related Articles | Metrics

In view of the special features in the design of a complicated vehicle system which involves many disciplines, a new methodology for the design of vehicles by using multidisciplinary design optimization (MDO) is presented, and a relevant analysis of MDO's application to vehicle design is also given in detail.

Thermal Deformation of Hollow Parts in an Uneven Steady-State Temperature Field

LUO Zai, FEI Ye-tai, MIAO En-ming

2005, 23(4):  424-427. 

Asbtract ( 270 )   PDF (182KB) ( 75 )  

Related Articles | Metrics

Hollow parts always work in an uneven steady-state temperature field.Causes of hollow part's thermal deformation include thermal stress and thermal expansion.An equation based on thermoelasticity method with a hollow part in an uneven steady-state temperature field is derived.The equation is verified by experimental results.

Research Notes

The Application of ADI-FDTD for Three-Dimensional Scattering

WANG Yu, YUAN Nai-chang

2005, 23(4):  428-431. 

Asbtract ( 337 )   PDF (180KB) ( 73 )  

Related Articles | Metrics

An implicit finite difference time domain method using the principle of alternating direction implicit technique is unconditionally stable, and the maximum time step size is not limited by the CFL condition, but rather by numerical errors.Compared with the conventional FDTD method, the time step size of ADI-FDTD can be much larger, so it takes less time to simulate.In this paper ADI-FDTD is applied to simulate the scattering of three-dimensional targets in vacuum and the settings of excitation and total-field/scattered-field connecting boundary condition are derived.Compared with FDTD, the results of ADI-FDTD have the same lev el of accuracy as conventional FDTD and the CPU cost is obviously reduced.

New Criteria for Stability of Linear Discrete Time Dynamical Systems

ZHU Wei, YANG Xiao-song

2005, 23(4):  432-434. 

Asbtract ( 447 )   PDF (135KB) ( 106 )  

Related Articles | Metrics

In this paper a result on distribution of the eigenvalues of real entry irreducible matrix whose module are equal to the spectral radius is presented, which is applied to give very simple proof for a theorem in ref.[2].Then a sufficient and necessary condition on the stability and asymptotical stability of linear discrete dynamical systems is given.

Video Transform of Naked-Eye Stereoscopic Display

LIANG Fa-yun, DENG Shan-xi, YANG Yong-yue

2005, 23(4):  435-437. 

Asbtract ( 280 )   PDF (151KB) ( 56 )  

Related Articles | Metrics

To match a specific stereoscopic device, image and video data for stereoscopic display must be organized in a specified format.A new data format and its transform are proposed for a new naked-eye stereoscopic display device.These are compatible with the data format of a conventional 3D display and able of meeting the requirement of naked-eye stereoscopic display.

Analysis of Thermal Optic Errors in Hot Rolling Train Wheel Measurements

YANG Yong-yue, DENG Shan-xi, HONG Zhan-yong

2005, 23(4):  438-440. 

Asbtract ( 284 )   PDF (158KB) ( 72 )  

Related Articles | Metrics

In measuring a hot rolling train wheel, optic rays are curved due to the non-uniform distribution of surrounding air temperature, leading to significant errors.Air temperature distribution is analyzed with a finite element method.Linear distribution of refractive index is calculated for certain measurement positions. Reasonable positions are chosen according to the amount and direction of curly light rays so that influence of the ray curvature is avoid.