Table of Content

26 September 2013, Volume 31 Issue 5

Previous Issue    Next Issue

Communication Engineering

Effective Access Selection Algorithm for Load Balance in Heterogeneous Wireless Networks

TANG Liang-rui1, YANG Zheng1, ZHAO Lin2

2013, 31(5):  441-447.  doi:10.3969/j.issn.0255-8297.2013.05.001

Asbtract ( 1406 )   PDF (2098KB) ( 969 )  

Related Articles | Metrics

 A dynamic access selection algorithm based on load balance is proposed for heterogeneous wireless networks. To adjust the network load to reach an ideal state, the algorithm not only considers user mobility, user satisfaction and available bandwidth of each network, but also considers the influences of selecting different access networks on the overall call blocking probabilities. Simulation results show that the proposed algorithm can dynamically adapt to the network load and lessen the unbalance of call blocking probabilities among candidate networks, and thus effectively reduces the average call blocking probability of the whole heterogeneous system.

Cognitive Radio Spectrum Allocation Based on an Improved Population Adaptive Artificial Bee Colony Algorithm

LI Xin-bin1, LIU Lei1, SHI Ai-wu1, MA Kai1, GUAN Xin-ping1,2

2013, 31(5):  448-453.  doi:10.3969/j.issn.0255-8297.2013.05.002

Asbtract ( 1745 )   PDF (3023KB) ( 975 )  

Related Articles | Metrics

To improve the exploiting ability and speed the convergence of the artificial bee colony (ABC),this paper proposes a dynamic acceleration population adaptive artificial bee colony (DA-PA-ABC) in which a dynamic acceleration operator and a population adaptive proportion operator are introduced. DA-PA-ABC
is applied to solve the spectrum allocation problem of cognitive radio, in which feasible solutions correspond to the allocation matrix of the TV spectrum allocation model. Network benefit and fairness are taken as the objective function of the problem to achieve reasonable allocation of the idle TV spectrum. Simulation results show that, compared with other methods, spectrum allocation based on DA-PA-ABC can maximize network benefits and ensure fairness of users.

Design of Double Codebook for Multi-cell Systems with Limited Feedback

ZHAO Wei1,2, YUAN Chao-wei2, LI Mei-ling2, LI Ying-xue2

2013, 31(5):  454-458.  doi:10.3969/j.issn.0255-8297.2013.05.003

Asbtract ( 1841 )   PDF (791KB) ( 924 )  

Related Articles | Metrics

We consider a two-cell Wyner model with limited feedback. Due to the existence of channel quantizing error, the system cannot reach the theoretical full multiplexing gain. This paper proposes to use feedback information of the phase between channel direction and perfect channel (PI) to achieve the full
multiplexing gain. To design phase of the feedback codebook, we derive the probability density function of phase, and design a non-uniform quantification codebook based on the PDF. Simulation results show that the proposed scheme can improve the system capacity by 2.5 dB as compared to the traditional CDI scheme. Compared to uniform quantification, the system capacity has 1 dB improvement.

Signal and Information Processing

Blind Recognition of Linear Block Codes under High Error Rate Condition

CHEN Jin-jie1,3, JI Tong-zhong2, YANG Jun-an1

2013, 31(5):  459-467.  doi:10.3969/j.issn.0255-8297.2013.05.004

Asbtract ( 2000 )   PDF (2280KB) ( 861 )  

Related Articles | Metrics

 For blind recognition of binary linear block codes in information interception, with the linear structure and checkout properties of block codes and characteristic weight distribution, this paper proposes a method of blind recognition of linear block codes under a high bit error rate condition based on matrix rank and the information entropy function of the weight distribution. The method can recognize code length using the information entropy of rank and precisely compute codes start bit based on the information entropy function of the weight distribution. According to the characteristic of systematic or non-systematic linear block codes,the generator matrix is accurately obtained using different calculation methods, and blind recognition of linear block codes is achieved. Simulation results show that this blind recognition method has good performance in the case of high BER.

High Speed High Maneuvering Target Variable Structure Multiple Model Algorithm Based on Connected Graph

XIAO Song1, LI Zhi-huai1, TAN Xian-si2, WANG Hong2, LI Chang-xi1

2013, 31(5):  468-474.  doi:10.3969/j.issn.0255-8297.2013.05.005

Asbtract ( 1718 )   PDF (3714KB) ( 832 )  

Related Articles | Metrics

 The traditional algorithm of variable structure multiple model based on adaptive digraph (ADVSMM) cannot quickly jump between model sets, and requires large amount of computation. To deal with the problem, a variable structure multiple model algorithm based on connected graph (CG-VSMM) to the high
speed high maneuvering targets is proposed. By analyzing the model sets conversion method of the traditional AD-VSMM, acceleration and angular velocity are selected as the model parameters to establish connectivity relationships between model sets connected graph. Models involved in the calculation are chosen depending on the target tactical situation. Partial models are adjusted according to the posterior probability of sub-model in real time. Monte Carlo simulation results show that the proposed method can achieve quick jump between models of VSMM algorithm. Tracking accuracy, computation complexity, and stability are improved. It can be used to track high speed high maneuvering targets.

Variable Step-Size Convex Combination of LMS Adaptive Filtering: Algorithm and Analysis

MIAO Jun1,2, RUI Guo-sheng1, ZHANG Yang1

2013, 31(5):  475-480.  doi:10.3969/j.issn.0255-8297.2013.05.006

Asbtract ( 1700 )   PDF (1131KB) ( 1041 )  

Related Articles | Metrics

The existing variable step-size convex combination of LMS (VSCLMS) algorithm needs to predetermine the behavioral parameters. To avoid this, the paper proposes a new variable step-size adaptive filter with analytical minimization of the ensemble-averaged mean-square weight error. Instead of variable step-size parameters in the original VSCLMS, the proposed algorithm uses a constant step-size parameter derived based on steady-state MSWE minimization. Theoretical analysis and simulations show that the proposed algorithm has a better tracking performance in the presence of noise and in a time-varying and even non-stable environment. Besides, it converges fast and is stable in the convergence process, and is better in these aspects as compared with the original VSCLMS and CLMS algorithms.

Blind Identification of Underdetermined Mixing Matrix Using an Outlier Detection Method

DONG Tian-bao

2013, 31(5):  481-487.  doi:10.3969/j.issn.0255-8297.2013.05.007

Asbtract ( 1480 )   PDF (4471KB) ( 1093 )  

Related Articles | Metrics

This paper focuses on the mixing matrix estimation in underdetermined blind source separation.A method for detecting the single source points in the time-frequency (TF) domain is given to reduce the effect of multiple source points in the TF domain. A method of outlier detection is used to remove outliers
in the single source points to improve accuracy of mixing matrix estimation. A subtractive clustering method is used to cluster the single source points so that the number of sources and the mixing matrix are obtained simultaneously. Experiments on speech signals show that the proposed algorithm can estimate the mixing matrix with high accuracy and robustness as compared to the other two mixing matrix estimation algorithms based on single source point detection in the TF domain.

Single Sample Face Recognition with Virtual Samples and Linear Subspace Feature Extraction

MENG Yi-fei1, YUAN Xue1, WEI Xue-ye1, QIN Fei-zhou2, QIN Qing-nu1

2013, 31(5):  488-494.  doi:10.3969/j.issn.0255-8297.2013.05.008

Asbtract ( 1844 )   PDF (4952KB) ( 1817 )  

Related Articles | Metrics

 A method of reference model illumination mapping is proposed to deal with the problem of one sample per person. To enhance the classification information of single training sample, we extend virtual images generated from the given single training image. Using discrete wavelet transform (DWT), the low-frequency band is processed to map the illumination information from the reference model to create a virtual sample. Principle-component-analysis plus linear-discriminant-analysis (PCA+LDA) is performed on the virtual training set to extract features. Experiments are performed on the Yale B and extended Yale B facial image database. The results show that, compared with the PCA feature extraction with single sample, recognition rate is significantly higher using the proposed method.

Construction and Application of Blur Invariants Based on Orthogonal Fourier-Mellin Moment

DAI Xiu-bin1, LIU Tian-liang2, SHU Hua-zhong3, LUO Li-min3

2013, 31(5):  495-502.  doi:10.3969/j.issn.0255-8297.2013.05.009

Asbtract ( 1290 )   PDF (3265KB) ( 1129 )  

Related Articles | Metrics

We propose a new set of blur invariants based on orthogonal Fourier-Mellin moments to effectively recognize blurred images. We establish the relationship between orthogonal Fourier-Mellin moments of the original and distorted images. We then use this relationship to construct blur invariants of orthogonal
Fourier-Mellin moment. The new blur invariants are used as feature descriptors to recognize blurred images. Experimental results show that，with different kind of blur degradation and different densities of noises introduced into the images，better invariance and higher recognition rates can be obtained by using the proposed invariants as compared to other blur invariants.

Extracting Missile Targets in Ultraviolet Image with Strong Noise and Burst Interference

GAO Qi-na1,2, ZHU Ying1,2, XIAO Yang3, KOU Jin-feng3

2013, 31(5):  503-511.  doi:10.3969/j.issn.0255-8297.2013.05.010

Asbtract ( 1617 )   PDF (13309KB) ( 1161 )  

Related Articles | Metrics

A knotty problem in ultraviolet (UV) images processing is how to cancel the strong noise and burst interferences coming from UV CCD sensors and high gain amplifiers. The strong noise and burst interferences will influence detecting the missile targets in the UV images effectively. To solve the problem, without changing the existing UV CCD sensors, this paper develops a 3-D recursive filter for UV image sequences based on a 3-D IIR filter, which is stable and it extracts the faint targets fast and clearly. The proposed 3-D IIR filtering algorithm composes of a 2-D IIR spatial filter and 1-D IIR time filter. The proposed 3-D IIR filtering algorithm is of less computation amount for UV image sequences and it can realize fast or real-time filtering for UV images. The stability test algorithms for the 2-D spatial IIR and 1-D time IIR are provided. Simulation experiments of UV image processing verify that the proposed 3-D IIR filter and related algorithms are correct，effective and pragmatical.

Fast and Robust Measurement of Pavement Ruts

CHEN Xiao-yu1,2, LEI Bo3

2013, 31(5):  512-518.  doi:10.3969/j.issn.0255-8297.2013.05.011

Asbtract ( 1407 )   PDF (5639KB) ( 1330 )  

Related Articles | Metrics

This paper proposes a technique for fast detection of pavement ruts based on laser triangulation. For the real-time measurement of ruts, transversal projection is applied to confine the valid region of laser curve, followed by segmental line fitting to eliminate gross errors. Bezier curve fitting is applied to form
a continuous smooth curve representing the road profile. To improve the measuring precision, instead of obtaining parameters of laser triangulation equations and calculating object coordinates of the laser curve, a calibration grid construction method based on triangulated irregular networks (TIN) and the back-calculation algorithm from image points to their object coordinates are proposed. Experimental results show that the algorithm is capable of sampling the road surface 80 times per second. The results of automatic measurements are in general agreement with that of the manual measurements. Therefore, the proposed method is effective and reliable for pavement rut measurement.

Normalization of Face Illumination Based on Edge-Preserving Filter and Visually Optimal Criterion

ZHANG Er-hu, MOU Yong-qiang, CHEN Wan-jun

2013, 31(5):  519-525.  doi:10.3969/j.issn.0255-8297.2013.05.012

Asbtract ( 1778 )   PDF (2627KB) ( 1413 )  

Related Articles | Metrics

Inspired by the edge-preserving filter used in computational photography, we propose a method of illumination normalization for face recognition under varying lighting conditions. The brightness layer of a face image is accurately decomposed into a reflectance layer and a shading layer with a weighted least square filter. The histogram of the reflectance layer is remapped to that of sample images selected based on a visually optimal criterion to obtain an image with the best visual quality. Experimental results on the YALE-B and CMU-PIE face databases demonstrate effectiveness of the proposed method under varying illumination conditions.

Local Mean Pattern Texture Descriptor for Gesture Recognition

DING You-dong1, PANG Hai-bo2,3, WU Xue-chun2, WEI Xiao-cheng2

2013, 31(5):  526-532.  doi:10.3969/j.issn.0255-8297.2013.05.013

Asbtract ( 1901 )   PDF (6937KB) ( 1116 )  

Related Articles | Metrics

 This paper presents an improved local binary pattern (LBP) descriptor, known as the local mean pattern (LMP), to classify static gestures. We select original gesture images, nonlinear illumination images,Gaussian blurred images, and images contaminated by salt and pepper noise, calculate their LMP, LBP and
local angular phase (LAP) descriptor. The gentle_Adaboost classification algorithm is used for training and verifying these gesture features. The descriptor makes full use of correlation and difference of pixel gray values in certain regions. It is a good description for the characteristics of different gesture images. Experiment results show that LMP descriptor outperforms LBP and LAP. Classification accuracy of LMP descriptor reaches 95%.  The descriptor is robust to nonlinear illumination and Gaussian blur.

Encoding of Distributed Compressive Video Sampling

WU Ming-hu1,2, ZHU Xiu-chang1, GAN Zong-liang1

2013, 31(5):  533-536.  doi:10.3969/j.issn.0255-8297.2013.05.014

Asbtract ( 1655 )   PDF (864KB) ( 1766 )  

Related Articles | Metrics

This article introduces some typical encoding schemes for distributed compressive video sampling (DCVS) and compares with the channel coding based distributed video coding (DVC). The theoretical basis of DCVS encoding is described, and the existing frameworks of DCVS analyzed. Different coding tools are compared. The key techniques including sparse description of video sequences, reconstruction algorithm, and rate control schemes are analyzed.

Electronic Engineering

Gate Replacement with Input Vector Constraint to Mitigate Circuit Aging

LI Yang1,2, LIANG Hua-guo3, TAO Zhi-yong1

2013, 31(5):  537-543.  doi:10.3969/j.issn.0255-8297.2013.05.015

Asbtract ( 1311 )   PDF (1412KB) ( 1167 )  

Related Articles | Metrics

To mitigate NBTI-induced circuit aging and enhance circuit reliability, gate replacement with input vector constraint during standby is proposed. The potential critical paths are found by using dynamic and static NBTI-aware static timing analysis. The critical gates are then found by considering path correlation. An
input vector is generated to make the critical gates in a recovery phase to the largest extent. Gate replacement is applied to the critical gates beyond control. Experiment results with ISCAS benchmark circuit demonstrate that the average gate replacement rate is reduced to 9.68%, and the average delay improvement is increased to 39.65% with the circuit timing margin 5%.

Analysis for Probe-Fed Microstrip Antennas in Multilayered Spherical Media Based on MoM

YU Tao, YIN Cheng-you, LIU Hai-yi

2013, 31(5):  544-550.  doi:10.3969/j.issn.0255-8297.2013.05.016

Asbtract ( 1451 )   PDF (5195KB) ( 792 )  

Related Articles | Metrics

Formulas for calculating the impedance matrix of microstrip antenna in a multilayered spherical media using Rao-Wilton-Glission (RWG) basis functions based on MoM are derived. Division of pairs of triangles is adopted to calculate the elements of impedance matrix at the junction of the probe and patch.
The complex line integral singularity can be avoided by substitution. Asymptotic items of the dyadic Green’s functions of the source and field points located on the same and different layers are extracted to accelerate convergence of the infinite series. The characteristic of a microstrip antenna in a multilayered spherical media is calculated. The results of input impedance are in good agreement with the published data, showing validity of the analysis method.