Table of Content

20 January 2010, Volume 28 Issue 1

Previous Issue    Next Issue

Communication Engineering

Design of Quasi-cyclic LDPC Code for Fast Encoding

FAN Jun1, XIAO Yang2

2010, 28(1):  1-8. 

Asbtract ( 1817 )   PDF (693KB) ( 2015 )  

Related Articles | Metrics

Since the generator matrices of the LDPC code are always non-sparse, it is hard to store a large generator matrix when the code-length is long. Moreover, time complexity of encoding with a generator matrix is high. In order to resolve this problem, we present a new design of quasi-cyclic (QC) LDPC code for fast encoding. Unlike other QC codes, the parity check matrices of the proposed QC codes not only have the quasi-cycle characteristic, but also contain sub-matrices with the quasi-diagonal structures. So the parity check matrices have approximate lower triangular structures. According to Richardson’s fast encoding algorithm, the parity check matrices with these structures can realize fast encoding by the parity check matrices. Therefore we can avoid the difficulties of storage difficulty and time complexity. Simulation in AWGN channels shows that the BER performance of the proposed codes is excellent in different code-lengths and code-rates.

Improved Extended Min-sum Decoding Algorithm of LDPC Codes over GF(q)

XU Ba1, ZHANG Zhong-ming1, HE Ying-liang2, ZHANG Er-yang1

2010, 28(1):  9-13. 

Asbtract ( 1913 )   PDF (590KB) ( 2075 )  

Related Articles | Metrics

To overcome the drawback of the extended min-sum algorithm (EMS), an improved EMS decoding algorithm is proposed, which is designed for LDPC over GF(q). The algorithm adaptively chooses the rank of FHT in each iteration step according to the average variance of bit nodes’ probability pairs. It differs from the original EMS algorithm in which the rank of FHT for each iteration step is a constant. Moreover, for oscillating variable nodes, the message of the previous iteration is added to the current message in the iterative procedure to reduce loss of performance and effects of oscillation. Experimental results show that the proposed algorithm can achieve better performance and converge faster than EMS with the same decoding complexity.

Blind Separation and Detection of Chaotic DS Signals Based on ICA

XU Wei1;3, HU Tie-hua2, ZHANG Rui-fang1, GUO Jing-bo3

2010, 28(1):  14-18. 

Asbtract ( 1745 )   PDF (644KB) ( 1578 )  

Related Articles | Metrics

Chaotic DS signal is a kind of spread spectrum signal generated by directly modulating information signal with a chaotic signal. Communications using chaotic DS signals instead of conventional direct sequence spread spectrum signals have many advantages such as larger capacity and low probability of intercept. However, detection of chaotic DS signals is difficult. We propose a method to detect chaotic DS signals by calculating the largest Lyapunov exponent of the time series after being separated from the received signals using an ICA blind signals separation algorithm. Simulation results show feasibility of this method under Gaussian noise and multi-path channel interferences.

Signal and Information Processing

Bi-mode Based Error Concealment for Stereoscopic Images

SONG Li-juan, ZHU Xiu-chang

2010, 28(1):  19-23. 

Asbtract ( 1540 )  

Related Articles | Metrics

A new error concealment scheme based on bi-mode for stereoscopic images is proposed, which
makes use of correlation between the two images and error concealments in monoscopic images based on
different local characteristics. Experimental results show that the proposed scheme has good performance
of error concealment for stereoscopic images. The average PSNR is increased by 1 dB as compared to the
monoscopic mode.

Image Saliency Detection Based on Contrast Features and Local Sharpness

YU Zhi-ming, WANG Shuo-zhong, ZHANG Xin-peng, LIU Ting-ting

2010, 28(1):  24-31. 

Asbtract ( 1690 )   PDF (6333KB) ( 2142 )  

Related Articles | Metrics

To extract important areas in an image that attract visual attention, a saliency detection scheme using both contrast features and local sharpness in the image is proposed. The local sharpness is related to the focusing status, which reflects the human judgment of the scene being imaged. Therefore, the presented method integrates two different visual characteristics, which we call low- and high-level human visual attention (HVA). The globe and local color contrast in the image arouse the low-level HVA, whereas the local sharpness affects the high-level HVA. Two saliency maps are thus generated with color classification and detailed area detection, respectively, and combined. Experiments show that integration of the saliency maps can provide good correspondence to HVA areas for different images.

Improved EMD Algorithm and Its Application to Striation Distance Measurement

LI Ling1, LI Ming1;2, LU Yu-ming2

2010, 28(1):  32-37. 

Asbtract ( 1534 )   PDF (2078KB) ( 2088 )  

Related Articles | Metrics

Empirical mode decomposition (EMD) is a useful method in nonlinear and non-stationary signal analysis. We propose an improvement to the EMD method to reduce the end effects based on linear prediction. Data are extended by linear prediction until a local extreme appears at each end. The same procedure is applied to each intrinsic mode function. Experiments show that the proposed method works well in restraining the end effects of EMD compared to other methods. Because striation distance in fatigue fracture image is quasi-periodical, striation distance in the horizontal and vertical directions are calculated based on improved EMD algorithm, and the real striation distances obtained through triangle conversion. The striation distance measurement of actual images shows that application of EMD to striation distances measurement is feasible.

Speaker Verification Based on GMM-UBM with Embedded Auto-associate Neural Network

CHEN Cun-bao, ZHAO Li

2010, 28(1):  38-43. 

Asbtract ( 1824 )   PDF (551KB) ( 1897 )  

Related Articles | Metrics

This paper proposes to embed an auto-associate neural network (AANN) in the Gaussian mixed
model-universal background model (GMM-UBM) for speaker verification. The scheme integrates the merits of
both GMM and AANN. GMM and AANN are trained as a whole in terms of maximum likelihood (ML), and
spatial information of the feature vectors is disclosed. AANN reshapes distribution of the data and improves
similarity of intra-class data. In the training, parameters of GMM and AANN are updated alternately. Experimental
results show that the proposed method together with TNorm can provide improvement of verification
rate by 26% over the baseline GMM-UBM.

Hash Function Construction Based on Chaotic System of Coupled Map Lattice

CHENG Yan-yun, SONG Yu-rong

2010, 28(1):  44-48. 

Asbtract ( 1832 )   PDF (589KB) ( 1836 )  

Related Articles | Metrics

A novel scheme for constructing Hash functions is proposed based on the chaotic coupled map lattice (CCML). The initial value, the parameters and the coupled factor of the chaotic dynamical system are modulated by the message plain text blocks, which are mapped directly into the parameter space of CCML. Theoretic analysis and numerical simulation results show that the chaotic initial values, parameter and coupled factor of the CCML-Hash algorithm have high sensitivity. It has good performance such as excellent one-way property, confusion and diffusion characteristics, and strong collision resistance.

Identification of Array Signal Parameters Based on Matrix Decomposition

LIU Xu, XU Zong-ze, LEI Lei

2010, 28(1):  49-55. 

Asbtract ( 1740 )   PDF (545KB) ( 1764 )  

Related Articles | Metrics

We study parameter identification of uniform linear array signals in this work. The received signal matrix is uniquely decomposed under the Vandermonde structure constrains. Parameters of the signal received from uniform linear antennae (ULA) are then identified by exploiting the permutation lemma. The result is extended to the case of dependent sources and multipath transmission by exploiting the property of Khatri-Rao
product structure. The matrix-decomposition-based identification results of array signal parameters provide a necessary condition for the effectiveness of parameter estimation algorithms. They are independent of specific parameter estimation algorithm, but only depend on the structure property of array manifold and signals.

New Adaptive Threshold Filtering for Fiber Optic Gyroscopes

LIU Ying1;2, LI Yan1, XU Jin-tao2

2010, 28(1):  56-59. 

Asbtract ( 1848 )   PDF (643KB) ( 1593 )  

Related Articles | Metrics

Through Lipschitz index analyzsis of changes in wavelet factor of the signal and noise in different discrete scales, the signal model of fiber optic gyroscope (FOG) is established. Based on the discrete wavelet transform, a new adaptive filtering algorithm is studied. This algorithm can automatically adjust the threshold values of the wavelet coefficients at different scales in accordance with the energy level of FOG’s output signal.
Thus, a new threshold function is obtained, which weakens wavelet coefficients lower than the threshold and maximally reserves the true signal using a multinomial. Compared with the soft and hard threshold function, the new threshold function shows perfect performance with different SNR. Simulation shows that, compared with traditional fixed threshold wave filtering, the new method can effectively eliminate noise, and improve
FOG’s bias stability, random walking and other technique parameters.

Electronic Engineering

Multi-resolution Approach to Periodicity Detection Based on Wavelet Transform

XU Zhan-yang1;2, Charles ZHAN3, ZHANG Shun-yi1

2010, 28(1):  60-64. 

Asbtract ( 1752 )   PDF (639KB) ( 1452 )  

Related Articles | Metrics

In this paper, we present a novel approach of periodicity detection using wavelet technology.Adaptive wavelet denoising is applied to the PV data. The redundant dyadic discrete wavelet transform is used to decompose the PV oscillation and detect the extreme wavelet coefficients which used to reconstruct the PV signal to avoid signal distortion at different resolution scales. Calculating the indexes based on the post-processing of extreme coefficients, periodicity at the scales selected is worked out, and the PV oscillation periodicity can be obtained.

Computer Science and Applications

Small-Area Implementation of RSA Encryption Engine

LIU Zheng-lin, GUO Wen-ping, HUO Wen-jie, ZOU Xue-cheng

2010, 28(1):  65-71. 

Asbtract ( 1636 )   PDF (613KB) ( 2529 )  

Related Articles | Metrics

Due to the complexity of the popular asymmetric-key encryption algorithm RSA, the hardware implementation
has a too large overhead to be used in resource-constrained systems. In order to solve this problem, an RSA
encryption engine based on 256 bit data width processor is designed, which greatly reduces the area required by RSA.
Synthesis results show that, in addition to the basic function implementation, the improved RSA design reduces the area
by 55.63% with respect to SLE66CX160S of Siemens. It has 24 k gates count with a maximum clock frequency of 100
MHz. The implemented RSA engine meets the design requirements.

Control and System

Self-Organizing Fuzzy Neural Network-Based Actuator Fault Estimation for Satellite Attitude Systems

CHENG Yue-hua1;2, JIANG Bin1, YANG Ming-kai1, GAO Zhi-feng1

2010, 28(1):  72-76. 

Asbtract ( 1891 )   PDF (608KB) ( 1536 )  

Related Articles | Metrics

Weights and nodes of a self-organizing fuzzy neural network (SOFNN) can be updated online for network structure optimization. This paper studies a robust fault diagnostic approach based on two SOFNNs for a class of satellite attitude dynamics. The designed SOFNN1 is used to estimate uncertainties and external perturbations of fault-free satellite attitude dynamics, whose output is chosen as a referenced threshold of fault
detection. Based on SOFNN1, SOFNN2 is constructed to estimate actuator faults occurring in the satellite attitude dynamics. Simulation results demonstrate that SOFNN has good dynamics performance in estimating actuator faults for the considered dynamics with external noise and system parameter uncertainties. Compared with fixed-structured FNN, the proposed SOFNN has advantages in estimation speed.

Lateral Reference Trajectory Planning of Terminal Area Energy Management for Reusable Launch Vehicles

ZHANG Jun, HUANG Yi-min, YANG Yi-dong

2010, 28(1):  77-82. 

Asbtract ( 1645 )   PDF (1133KB) ( 1534 )  

Related Articles | Metrics

Abstract: According to the initial position of reusable launch vehicles (RLV) in terminal area energy management (TAEM), the lateral reference trajectory is planned to erase the initial position errors using the SNAKE algorithm. The course of lateral reference trajectory planning can be divided into two phases: erasing the lateral position error and erasing the longitudinal error. In the former phase, the longitudinal error is decreased. For erasing the longitudinal error, we propose to erase the error by proper trajectory mode according to its size. The lateral trajectory planning algorithm provides a kernel algorithm for 3-D TAEM trajectory generation which is a basis of onboard trajectory generation of RLVs’ TAEM guidance.

Improved Multi-objective Genetic Algorithm with Application to PID Optimization Design

LIU Nan-nan1, SHI Yu1, CHENG Wei-ping2, QIN Fu-gao3;4

2010, 28(1):  83-89. 

Asbtract ( 2090 )   PDF (692KB) ( 1797 )  

Related Articles | Metrics

We propose a multi-objective optimization genetic algorithm, which uses a new method to calculate crowding distance and improves the comparative method of non-domination. Double elitism-mechanism is introduced to improve efficiency of evolution and solution quality, and more effectively increase diversity of the solution. The algorithm is applied to optimal design of PID. In this way, the system is capable of considering
requirements for quickness, reliability and robustness. A satisfactory solution is selected in Pareto optimum set according to the requirements of the present system. Simulation results indicate effectiveness of the proposed algorithm.

Large Travel Range Nanopositioning Actuation without Overshooting for Tubular Permanent Magnet Synchronous Linear Motor

XU Cong-yu, YU Xiao-fen, CHENG Ling-li

2010, 28(1):  90-94. 

Asbtract ( 1711 )   PDF (1042KB) ( 1657 )  

Related Articles | Metrics

Abstract: To reduce electromagnetic disturbance of nanopositioning caused by direct actuation of cylindrical
permanent magnet synchronous linear motor (TPMSLM), the actuation mode of rotor and stator’s eccentric
installation of TPMSLM is studied. The actuation mode changes the interior space magnetic field distribution
of TPMSLM, therefore the axial electromagnetic force containing electromagnetic disturbance is decreased and
the radial electromagnetic force containing electromagnetic disturbance is produced. When guide damping is
added, a large amount of electromagnetic disturbance can be absorbed. Further more, to prevent large positioning
overshooting, a PI control algorithm with variable coefficients is studied, which changes dynamical
and static actuation characteristics of system in real-time to supply optimal control parameters for large travel
range actuation and nanopositioning actuation. Experimental results indicate that, using the eccentric actuation
mode and variable coefficient PI control algorithm, large travel range positioning actuation without
overshooting and accurate positioning with errors less than 10 nm are achieved.

Suppression of Wing Rock with Adaptive Neural Control

YUAN Suo-zhong

2010, 28(1):  95-98. 

Asbtract ( 2049 )   PDF (576KB) ( 1718 )  

Related Articles | Metrics

We propose to control the nonlinear wing rock motion of fighter aircraft using an adaptive nonlinear control strategy based on neural networks. An approximate dynamic inversion method is used to linearize the dynamics of wing rock. A linear controller is designed to stabilize the error dynamics. An adaptive neural network is used to cancel the inversion error. Simulation results of a slender delta wing combat aircraft model
at different angle of attack show that the control design can effectively suppress wing rock.

Two-Step Fusion Algorithm for ADS/INS Angles of Attack on Transonic Flight

LI Rui-jia, LI Rong-bing, LIU Jian-ye, MENG Bo

2010, 28(1):  99-105. 

Asbtract ( 2093 )   PDF (914KB) ( 1685 )  

Related Articles | Metrics

The angle of attack (AOA) is an important state parameter of flight. It is also a necessary parameter in flight control and navigation systems. The modern fighters have increasing demands on the AOA accuracy. The performance of air data systems (ADS) seriously declines under transonic conditions. The accuracy of ADS-AOA is seriously affected. In view of air data sensor configuration features of testing plane, a transonic ADS-INS AOA two-step fusion algorithm based on variable parameters complementary filter and neural network is designed in this paper to achieve fusion of INS and ADS. The divergent transonic INS vertical velocity is resolved. Variable parameters complementary filter and neural network are used to fuse and correct ADS-INS AOA. The result approaches the true AOA with stability and reliability. Actual flight data are used to validate the two-step fusion algorithm. The results show that the AOA after fusion and correction can eliminate violent fluctuations of ADS-AOA under transonic condition. Its tendency coincides with the true AOA with small error.

MRF Locking Damper for Micrometer/Nanometer Measurement

MA Wen-ping, YU Xiao-fen, CHENG Ling-li, YU Qing

2010, 28(1):  106-110. 

Asbtract ( 1823 )   PDF (785KB) ( 1767 )  

Related Articles | Metrics

A locking damper based on the new functional materials, magnetorheological fluids, is studied. It can reduce vibration of driving motor and make a micro-impact locking for the working table. The magnetic circuit is simulated with finite element analysis and the property is calibrated using a reciprocating experimental system. It is used in Nano-CMM for nano-positioning experiments, showing that positioning steady-state fluctuation of the working table is reduced from 20 nm to less than 5 nm when the MRF locking damper is used. It ensures the nano-degree positioning steady-state fluctuation in Nano-CMM.