Table of Content

31 May 2006, Volume 24 Issue 3

Previous Issue    Next Issue

Articles

An Iterative Joint Channel Estimation and Symbol Detection Algorithm Based on RLS Adaptive Filter

ZHANG Peng, BI Guang-guo, CAO Xiu-ying, XIAO Hai-yong, JIN Xiu-feng

2006, 24(3):  221-226. 

Asbtract ( 306 )   PDF (306KB) ( 105 )  

Related Articles | Metrics

Taking into account channel characteristics of high-speed mobile communication and using an RLS adaptive filter capable of tracking time variations of input statistics under non-stationary conditions, the paper presents an iterative joint channel-estimation and symbol-detection algorithm based on RLS adaptive predictor.Calculating initial channel values of each iterative estimation recursively, successive iteration times are reduced and convergence values become more accurate.Simulation results show that it improves convergence and system performance dramatically compared to conventional algorithms.

Kernel Nearest Features Classifier for Face Recognition

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

2006, 24(3):  227-231. 

Asbtract ( 301 )   PDF (236KB) ( 92 )  

Related Articles | Metrics

Kernel-based nearest feature classifiers are proposed in this paper, which can directly classify original face images and need not extract features beforehand.Besides, two KNFS methods are proposed, one is generalization of KNFP and the other employs KPCA to construct nonlinear feature subspace.Experimental results on ORL and YALE database demonstrate the feasibility of the proposed methods.

Speech Enhancement Using Sub-band Spectral Analysis

XU Jing-bo, YU Hong-tao, RAN Chong-sen

2006, 24(3):  232-235. 

Asbtract ( 260 )   PDF (191KB) ( 75 )  

Related Articles | Metrics

This paper describes an efficient algorithm for enhancement of speech signals, which is based on estimation of spectral amplitude using an overlap-add FFT filter bank.It can be adjusted to different kinds of noise by means of two parameters.The speech enhancement system requires neither speech pause detection for noise estimation nor VAD (voice activity detection), and overcomes disadvantages existing in single channel noise suppression algorithms.

Improved Coarse Symbol Synchronization Algorithms for OFDM Systems

LEI Jun, WU Le-nan

2006, 24(3):  236-239. 

Asbtract ( 303 )   PDF (185KB) ( 90 )  

Related Articles | Metrics

Symbol beginnings estimated by conventional methods utilizing the cyclic prefix (CP) continually vary with the path energies in OFDM systems, therefore ISI and ICI are inevitable.An improved coarse symbol synchronization algorithm impervious to delay spread is proposed in this paper.Only employing the rear of CP, timing positions of the proposed method concentrate on the first path in multi path environments.Simulation results in MW/SW channels show excellent performance of this approach.

Analysis and Application of CDR Circuit Using Oversampling

YIN Yong-sheng, HU Yong-hua, GAO Ming-lun

2006, 24(3):  240-244. 

Asbtract ( 423 )   PDF (242KB) ( 153 )  

Related Articles | Metrics

In serial data communication domain, clock and data recovery (CDR) using oversampling is a low-cost "fast retiming" technique that can be implemented as monolithic circuits using the standard digital CMOS process.This paper presents a CDR circuit based on oversampling.The design uses 4x oversampling rate, that is, an FSM is constructed to get four equidistant samples per bit cell.Data are then extracted from input data dreams using a majority decision algorithm.Simulation results show that the specified bit error rate oriented to USB2.0 is achieved in the presence of at least 1/4 bit cell jitter.It is believed that the analysis and design technique are useful for other serial data communication of computer peripherals.

A Novel Automatic Amplitude Control Circuit for VCO

SUN Wen, LUO Lan, WU Xuan, YUN Ting-hua

2006, 24(3):  245-249. 

Asbtract ( 292 )   PDF (235KB) ( 84 )  

Related Articles | Metrics

A new automatic amplitude control circuit for VCO, charge-pump AAC, is presented in this paper.Phase noise performance better than traditional methods is achieved due to its inherent characteristic.Stability analysis based on a small signal model in the S domain is presented.The simulation results indicate that, compared to analog AAC, 14.4 dB phase noise improvement is obtained at 10 kHz offset using the charge-pump AAC.

A Scalable Load-Balancing Algorithm for Distributed Data Stream Management Systems

WANG Jin-dong, ZHOU Liang, ZHANG Lei, DING Qiu-lin

2006, 24(3):  250-255. 

Asbtract ( 270 )   PDF (278KB) ( 70 )  

Related Articles | Metrics

In distributed data stream processing systems, data streams are processed in a number of computers in a collaborative manner.Overload on a part of nodes caused by input data stream fluctuation may weaken data processing ability of the entire system.Aimed at this practical problem and based on the characteristics of distributed data stream processing system and shortcomings of existing load balancing methods, a scalable load-balancing algorithm is proposed. This algorithm can deal with the load-balancing requirement of data stream processing systems composed of computers with different processing ability.Numerical experiments show the effectiveness of the proposed algorithm.

Detection and Repairing Method for Outliers over Data Streams

WANG Yong-li, XU Hong-bing, DONG Yi-sheng, QIAN Jiang-bo, LIU Xue-jun

2006, 24(3):  256-261. 

Asbtract ( 273 )   PDF (287KB) ( 103 )  

Related Articles | Metrics

Outlier detection and repairing in a data stream environment are discussed.An online detection method for outliers over data streams, called amnesia Kalman filtering (AKF), is proposed.In order to identify outliers, an improved Kalman filter is applied with the amnesia factor to forecast data-value in the future times tamp. A novel online adaptive repairing method for outliers over data streams, called adaptive interpolating wavelet (AdaptiveIW), is then proposed.The AdaptiveIW applies a method of variable-resolution interpolation, named the interpolation wavelet with adaptive resolution, to repair outliers, which determines interpolation resolution based on the number of continuous outliers.It adapts to different requested precision for out lier repairing over evolving data streams very well.Experimental results on actual power load data show that this method can provide precise instantaneous detection and accurate repairing for outliers over data streams.

Decomposition Coordination in Weapon Target Assignment Problems

LIU Yi-an, CHEN Song-can, WANG Shi-tong

2006, 24(3):  262-264. 

Asbtract ( 223 )   PDF (142KB) ( 55 )  

Related Articles | Metrics

Weapon-target assignment (WTA) problems may be separately decomposed by K sub-optimal problems based on the number K of resources being protected with decomposition coordination method in the large-scale system theory.The coordinator is then used to coordinate solutions of all sub-problems in order to get the best results of the overall weapon target assignment problem.Simulation results indicate that the method is effective and can be performed in real time.

Development of WebGIS Based on WMS Specifications and Integration with Water Environment Information Service Platform

XIE Jian-cang, WANG Xiao-hui, LI Jian-xun, LUO Jun-gang

2006, 24(3):  265-269. 

Asbtract ( 220 )   PDF (232KB) ( 72 )  

Related Articles | Metrics

Introducing WMS specifications of openGIS consortium (OGC), the key technology and developing plan of WebGIS based on OGC WMS discuss specifications.The specific development process and method are provided, and the WebGIS system is established.Meanwhile, seamless integration of the information service platform for monitoring Yellow River water environment using eXo portal technology developing is realized.It well solves the problem of combining two different kinds of information approaches, namely, information service platform and WebGIS, so that functions of the information service platform are enhanced.

A New Knowledge Model Supporting Network Teaching of Meta-cognitive Capacity

SHEN Jun

2006, 24(3):  270-277. 

Asbtract ( 245 )   PDF (391KB) ( 65 )  

Related Articles | Metrics

The goal of a knowledge model is to define the type and structure of knowledge used in task execution.In network teaching, a knowledge model directly deducts think and theory of teaching and drives design of instructions, organization of teaching activity and test of teaching.Building of knowledge model must reflects the related teaching think and strategies.Referring to the modern research findings in the cognitive science, this paper proposes a concept of knowledge application pattern, explicitly differenced basic knowledge (explicit knowledge) and knowledge application pattern (implicit knowledge), and gives a new knowledge model based on pattern and its construction.Using this model, one can implement the instructions of basic knowledge as well as meta-cognitive capacity.Finally, the goal of individual network teaching is reached.The theoretical model of network teaching is first discussed, and the principle and formal representation of the new knowledge model, and the construction policy and algorithm of individual cognitive discussed. Example and application analyses are presented.

A Method for Rational Approximation of Offset Curves

GUO Qing-wei

2006, 24(3):  278-282. 

Asbtract ( 302 )   PDF (228KB) ( 100 )  

Related Articles | Metrics

The rational approximation of offset curve of Bézier curve is obtained when a polynomial is used to approximate the norm of parametric speed of Bézier curve.The obtained rational curve and the offset curve of Bézier curve have the same higher derivates at the endpoints.If the method is combined with the subdivision method of curve, the rational spline curve with higher-order continuity is presented which is the approximation of offset curve.Numerical examples are given to show effectiveness of the method.The results are compared with other methods, and conclusions are made based on the numerical examples.

A Complete Damped Algorithm with Fuzzy Controller Based on Strapdown AHRS

DU Ya-ling, LIU Jian-ye, JIANG Yong

2006, 24(3):  283-287. 

Asbtract ( 293 )   PDF (281KB) ( 86 )  

Related Articles | Metrics

A novel complete damped algorithm with a fuzzy controller is presented in this paper.According to the control theory, the system adds different damp feedback in three loops, which uses magnetic heading degree and velocities to damp errors.Because there are limits in using level damp networks, a fuzzy controller is used to decide whether the network is used.Simulation and experimental results show that the complete damped algorithm with a fuzzy controller can damp most Schuler and Foucault oscillations, and efficiently improves precision of strap down AHRS.

Multi-parameter Performance Ranking of Aeroengines Based on Fuzzy Information Entropy Method

ZHANG Hai-jun, ZUO Hong-fu, LIANG Jian

2006, 24(3):  288-292. 

Asbtract ( 238 )   PDF (212KB) ( 68 )  

Related Articles | Metrics

Performance ranking of civil aviation engine based on surveillance information can provide the degree of engine deterioration and stagger difference, make optimal maintenance plan, and reduce maintenance cost.Therefore it has become a critical issue in improving fleet management.However rational methods for evaluating engine's overall performance in civil aviation management are unavailable.This paper establishes a theoretical model and proposes an algorithm to distribute all parameter weights based on a fuzzy equivalent aggregation technique and attribute information entropy of rough set theory when multi-attribute weight information are unknown.As an example, the algorithm has been verified in performance ranking of ten CF6 engines from a civil aviation company.The result shows that the method is objective and effective.

GPS/SST/SINS Integrated Navigation System

KANG Guo-hua, LIU Jian-ye, ZHU Yan-hua, XIONG Zhi

2006, 24(3):  293-297. 

Asbtract ( 357 )   PDF (236KB) ( 136 )  

Related Articles | Metrics

An integrated navigation system of GPS/SST/SINS is presented in this paper.This system uses an improved centralized Kalman filter to perform data fusion between the multi-sensor outputs and estimate the optimal navigation result.In this system the attitude of the strap down star tracker (SST), the position and velocity of GPS, and the correspondence information of SINS are filtered to improve navigation precision of attitude, velocity and position.The system is then used in a ballistic missile model to implement a simulated platform of GPS/SST/SINS integrated navigation. The result shows that the system can effectively enhance navigation precision and the filter is stable and reliable.

Intelligent Altitude Sensor Based on Artificial Neural Networks Using Multilayer Perceptron

QU Guo-fu, LIU Hong-zhao

2006, 24(3):  298-301. 

Asbtract ( 282 )   PDF (187KB) ( 65 )  

Related Articles | Metrics

Based on the atmospheric pressure and temperature measured by a piezoresistive sensor and a temperature sensor, and their relations with altitude, a novel algorithm for sensor design is proposed.The pressure and temperature sensors are treated as two inputs of a multi-layered perceptron in an artificial neural network, and the altitude as the output.To get weights and offsets for each layer, the ANN is trained with standard data using the MOBP algorithm, which can be implemented on DSP.It can produce accurate altitude at any given pressure and temperature.Experiments indicate that the new sensor provides a higher accuracy than using the curve fitting technique, and have advantages in suppression of sensors' temperature shift and time shift.

Fuzzy Reliability Analysis Based on AFOSM Method

DONG Yu-ge, WANG Chun-xian, CHO Hyun-deog, KWON Jong-wan

2006, 24(3):  302-306. 

Asbtract ( 294 )   PDF (219KB) ( 70 )  

Related Articles | Metrics

The general formulae of equivalent probability density function, equivalent mean value and equivalent standard deviation are given by means of the transformation from fuzzy variables to equivalent random variables.Specific formulae of equivalent mean value and equivalent standard deviation are obtained when the membership functions of the fuzzy variables are commonly used ones.Based on the formula of limit state, fuzzy reliability in complex cases is computed by the advanced first order second moment method in the convectional reliability theory, and the reliability index to measure fuzzy reliability is obtained.An example is given to show the effectiveness and applicability of the proposed method.

Quantitative Analysis for Crack Signals Based on Search Approximation in the Conjugate Direction

LIU Mei-quan, XU Zhang-sui, WANG Jian-bin

2006, 24(3):  307-310. 

Asbtract ( 219 )   PDF (194KB) ( 67 )  

Related Articles | Metrics

Magnetic charge model for V-shaped crack and measurement equation are established.The optimization method of search approximation in the conjugate direction on quantitative analysis for crack is adopted.The problem in quantitative analysis for leakage magnetic field is solved.Testing result indicates that this algorithm converges rapidly to real solutions.The algorithm is an effective non-distructive testing method to realize quantitative testing for crack.

Application of Rigid Plastic FEM in the Analysis of Orthogonal Cutting

LI Yan, ZHU Jiang-xin, XIAO Ji-ming, LI Shu-juan, LI Peng-yang

2006, 24(3):  311-315. 

Asbtract ( 237 )   PDF (262KB) ( 53 )  

Related Articles | Metrics

Lagrange method for the static large deformation analysis is used to establish a rigid-plastic FEM (RPFEM) analysis model based on the orthogonal cutting process of 2D plain strain hypothesis.Based on this model, an RPFEM analysis program is developed to simulate deformation of the cutting process.The whole process of orthogonal cutting is analyzed.The rule of separation between cut chips and the work piece, the criterion of separation between chip and reamer, etc., are discussed.Using the arithmetic of combining the total average weighted value of areas and that of volume, transfer of parameters between the new and old grids are realized when the grids are further divided.Simulation experiments of aluminum work pieces under normal temperature are performed, showing validity of the numeric method.

Solution to Rectangular Slab Forging with Bulge between Two Parallel Platens Using Strain Rate Vector Analysis

ZHAO De-wen, WANG Lei, FANG Qi, LIU Xiang-hua

2006, 24(3):  316-319. 

Asbtract ( 270 )   PDF (186KB) ( 66 )  

Related Articles | Metrics

A linear integration of strain rate vector is developed.First, effective strain rate for slab forging with bulge is expressed in terms of two-dimensional strain rate vector.Its inner-product is integrated term by term.Second, sum of the integrated results is obtained, and an analytical solution to the stress factor obtained from a bulge formula.It turns out that the expression of power by the above inner-product integration is the same as that of the traditional immediate integration.Also, the calculated result of total forging pressure by formula in the present paper is in general agreement with that by Avitzur's formula, both 4.2% higher than the measured value in a press test.

Service-Oriented Architecture of Information Integration System for Assessment of Enterprise Logistics Performance

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

2006, 24(3):  320-324. 

Asbtract ( 239 )   PDF (240KB) ( 63 )  

Related Articles | Metrics

Enterprise logistics activities exist among the functional departments, suppliers, sellers and customers. Enterprise logistics service is a result of the integration of logistics activities.Therefore it is difficult to construct a log istics performance assessment system.The novel service-oriented idea and the system integration theory are used in this paper to construct architecture of information integration system for enterprise logistics performance assessment based on the relation between logistics performance assessment system and other application systems in the enterprise.The architecture gives a new way for establishing comprehensive and objective information integration system of logistics performance assessment.

A Novel Approach to Initial Vector Construction in Iterative Target Transformation Factor Analysis

LIU Guang-jun, GAO Hong-tao

2006, 24(3):  325-330. 

Asbtract ( 249 )   PDF (274KB) ( 46 )  

Related Articles | Metrics

Iterative target transformation factor analysis (ITTFA), a type of self-modeling curve resolution (SMCR), is commonly applied in spectra decomposition.Construction of favorable initial iterative vectors is the key to ITTFA.A novel approach is proposed to construct better initial vectors based on the chromatographic information provided by evolving factor analysis (EFA).A region containing the peak position at maximum height is determined.Several elements in the region in each initial vector, instead of one element, are initialized as 1.It is unnecessary to determine the exact peak position at maximum height for the resolution of partly overlapping chromatographic profiles, which may avoid the divergence brought about by determination of the peak position at maximum height.In addition, it can give acceptable resolution for the embedded peaks.It was applied to resolve 2-dimensional simulated data and GC/MS data for liquor analysis, the resolutions were reasonable and improved.