Journal of Shanghai University（Natural Science Edition）

All-Optical OFDM-PON Architecture Based on Centralized Comb Source for 125 Gbit/s Access Network

WANG Min,ZHANG Qian-wu,ZHANG Lin,SONG Ying-xiong,LI Ying-chun,LIN Ru-jian

2012, 18(1): 1-6. doi:10.3969/j.issn.1007-2861.2012.01.001

Asbtract ( 2831 )

PDF (2416KB) ( 1425 )

Related Articles | Metrics

OFDM (orthogonal frequency-division multiplexing)-PON (passive optical network) is considered a promising solution for the next generation PON-based access networks. Meanwhile, the released 40/100 Gbit/s Ethernet standard allows 100 Gbit/s fiber transmission to be realized in the near future. We propose a novel OFDM-PON architecture based on all-optical waveform and flat frequency comb generation for 100 Gbit/s optical access. The flat frequency comb generation provides an effective and costless way to generate all-optical OFDM signals and implement coherent detection in both downstream and upstream. Simulation results show robust and outstanding performance of the proposed system. 

Experiment of 2×2 Fused Fiber Surface-Enhanced Raman Scattering

CHEN Zhen-yi,CHEN Na,HU Ling,WANG Ting-yun,PANG Fu-fei,LIU Lin

2012, 18(1): 7-11. doi:10.3969/j.issn.1007-2861.2012.01.002

Asbtract ( 2468 )

PDF (2026KB) ( 1282 )

Related Articles | Metrics

Based on the optical fiber evanescent-wave excitation and surface-enhanced Raman scattering (SERS), we have carried out experimental studies on SERS characteristics of the 2×2 fused fiber coupler coated with Ag colloid particles. By stretching a pair of fused single-mode fibers with coating Ag colloid on its surface, the formed component can work as the surface-active substrate to perform Raman enhancement. With increase of the taper length, fiber core diameters decrease and less energy are bounded in the fiber. This enables stronger evanescent-wave to leak outward and become a Raman exciting source. In the experiments, the Raman spectra of R6G solutions has been detected at the coupling zone. The minimum detected concentration is 10^-8 mol/L.

Estimation of Fast Fading Channel for OFDM with Imperfect Carrier Synchronization

FANG Yong,ZHAO Wei-jie,WANG Min

2012, 18(1): 12-19. doi:10.3969/j.issn.1007-2861.2012.01.003

Asbtract ( 2421 )

PDF (1265KB) ( 1311 )

Related Articles | Metrics

A fast-fading-channel estimation algorithm is proposed in this paper for orthogonal frequency-division multiplexing (OFDM) systems with imperfect carrier synchronization. Specifically, diagonalizing decomposition is proposed for autocorrelation of a compact matrix of generalized channel impulse response (G-CIR). In such decomposition, only the diagonal matrix is random. A compressed expression of the G-CIR matrix and its equivalent channel model are derived. Finally, a least square (LS) estimators is formulated to estimate the compressed channel, from which the CIR matrix is reconstructed. Simulation results demonstrated that the proposed scheme has better estimation accuracy and lower bit error rate (BER) over the existing techniques.

Power Allocation Method for Amplify-and-Forward Cooperative Communication Systems with Multiple Relays

WU Zhuo,YANG Hong-bing,ZHENG Guo-xin

2012, 18(1): 20-25. doi:10.3969/j.issn.1007-2861.2012.01.004

Asbtract ( 2404 )

PDF (1777KB) ( 1500 )

Related Articles | Metrics

Power allocation has long been one of the most crucial technologies in wireless communications. Research on effective utilization of power has great significance for wireless systems. This paper proposes a novel power allocation scheme for the amplify-and-forward (AF) cooperative communication systems to minimize the total transmit power. The simulated annealing (SA) algorithm is introduced here to solve the optimization problem. Simulation results indicate that the proposed scheme outperforms that of equal power allocation scheme in terms of bit error rate and system capacity. 

Positive Definite Solution for a Class of Generalized Stein Equation

DUAN Xue-feng1,2,WANG Qing-wen2,CHANG Hai-xia3

2012, 18(1): 26-29. doi:10.3969/j.issn.1007-2861.2012.01.005

Asbtract ( 2204 )

PDF (669KB) ( 967 )

Related Articles | Metrics

Based on the elegant properties of the Thompson metric, we give a sufficient condition for the existence of positive definite solution for a class of generalized Stein equation. An iterative method is proposed to compute the positive definite solution, and its error estimate formula is also derived. Numerical examples show that this iterative method is feasible.

Shock Front Solution to 2-Dimensional Axially-Symmetrical Piston Problem of Pressure-Gradient Equations

DOU Yan-ping,TANG Zhong-fei,XU Yu-lan

2012, 18(1): 30-34. doi:10.3969/j.issn.1007-2861.2012.01.006

Asbtract ( 1875 )

PDF (699KB) ( 907 )

Related Articles | Metrics

In this paper, we consider the existence of local solutions to a 2-dimensional axially-symmetrical piston problem of pressure-gradient equations. As the problem is axially-symmetric, proper variables first are selected to transform the original problem into a 1-dimensional problem. By Taylor expansion, we construct an N-ordered approximate solution used as the first term in the Newton iterative method. With the Newton iteration and the energy estimation, we show the local existence of the shock front solution to the problem.

Two Conjugacy Classes of All Non-abelian Subgroups of Finite Group

ZHOU Zhi-hao,GUO Xiu-yun

2012, 18(1): 35-39. doi:10.3969/j.issn.1007-2861.2012.01.007

Asbtract ( 2136 )

PDF (689KB) ( 1318 )

Related Articles | Metrics

In this paper, we discuss the solvability of the finite groups all of whose conjugacy classes of non-abelian subgroups are no more than 3. And thus we study the non-p-groups whose conjugacy classes of non-abelian subgroups are equal to 2. This kind of groups is completely classified.

Minimal-Energy Graph for a Forest with a Given Number of Components

WANG Wen-huan

2012, 18(1): 40-42. doi:10.3969/j.issn.1007-2861.2012.01.008

Asbtract ( 2121 )

PDF (660KB) ( 1056 )

Related Articles | Metrics

The energy E(G) of a graph G is defined as the sum of absolute values of eigenvalues of G. Let F_n,q be a set of forests with n vertices and q components. For given n and q, we derive a graph with minimal energy in F_n,q.

Optimal Conditions and Algorithms for Global Optimization

DING Yi,CHENG Jian-qiang,WU Dong-hua

2012, 18(1): 43-47. doi:10.3969/j.issn.1007-2861.2012.01.009

Asbtract ( 2078 )

PDF (682KB) ( 932 )

Related Articles | Metrics

This paper presents another integral-level set method for finding the global maxima when the function is always greater than zero, and modifies the integral function according to the concept of weight. The algorithm and the convergence criteria for the method are also given. 

Numerical Algorithm for Fractional Calculus Based on Chebyshev Polynomial Approximation

CHEN An,LI Chang-pin

2012, 18(1): 48-53. doi:10.3969/j.issn.1007-2861.2012.01.010

Asbtract ( 2630 )

PDF (701KB) ( 1535 )

Related Articles | Metrics

Numerical algorithms based on Chebyshev polynomial approximation are proposed, including a numerical algorithm for fractional integral and a generalized numerical algorithm for fractional derivative. Error estimates are given. Numerical examples show effectiveness of the methods.

Implementation of Global Optimization and Comparison with Several Deviation Integral Algorithms

WANG Xiao-li,LIANG Ze-liang,YAO Yi-rong,ZHENG Quan

2012, 18(1): 54-58. doi:10.3969/j.issn.1007-2861.2012.01.011

Asbtract ( 2176 )

PDF (678KB) ( 1037 )

Related Articles | Metrics

In this paper, several kinds of deviation integrals for global optimization are constructed. Their global minimization is solved and studied. To implement 100 variables of the algorithm with the Monte-Carlo technique, different deviation integrals for global optimization are used. Numerical tests show that the different deviation integrals have advantages for different objective functions.

Cutting Plane Method for Solving Concave Minimization Programming Based on a New γ-Extension

LIU Lin-na,YANG Yong-jian,YU Feng

2012, 18(1): 59-63. doi:10.3969/j.issn.1007-2861.2012.01.012

Asbtract ( 1896 )

PDF (691KB) ( 1115 )

Related Articles | Metrics

We first briefly give the concept of concave minimization programming and the cutting plane method, and then propose the uncertain upper limited integral function method which will be used to solve γ constant in the γ-extension. We show that the proposed method only requires a finite number of iterations to reach a near-optimal solution that is just the global optimization. Implementation of the method is reported with satisfactory numerical results.

Group Classification of Finite Difference Scheme

PAN Shu-qin,RAN Zheng

2012, 18(1): 64-67. doi:10.3969/j.issn.1007-2861.2012.01.013

Asbtract ( 2135 )

PDF (997KB) ( 1252 )

Related Articles | Metrics

This paper presents an analysis on the group structure of finite difference scheme in computational fluid dynamics using the Lie group analysis method, together with the invariant solutions of the traditional schemes. The study will help understand the differential effects of discreteness on numerical dissipation and dispersion.

Equilibria Bifurcation in Coupled Planar of High-Speed Axially Moving Beams

HUANG Hui-chun1,DING Hu2,CHEN Li-qun2,3

2012, 18(1): 68-71. doi:10.3969/j.issn.1007-2861.2012.01.014

Asbtract ( 2295 )

PDF (1089KB) ( 989 )

Related Articles | Metrics

Equilibria of axially moving beams transversely and longitudinally coupled with the fixed boundary conditions are numerically studied in the supercritical transport speed ranges. In the supercritical regime, the pattern of equilibria consists of the straight configuration and of non-trivial solutions that bifurcate with transport speed. The numerical schemes are presented for the governing equation of coupled planar and the corresponding static equilibrium equation for non-trivial equilibrium solutions via the finite difference method under the fixed boundary conditions. A copper beam is treated as example to demonstrate the non-trivial equilibrium solutions. Numerical results indicate that the equilibrium of the coupled planar with the changing parameters.

Analytical Solution of Stress and Strain Distribution in a Trapezoidal Cross-Section Quantum-Wire Structure

HUANG Ling-feng1,XU Kai-yu1,2

2012, 18(1): 72-75. doi:10.3969/j.issn.1007-2861.2012.01.015

Asbtract ( 2558 )

PDF (1192KB) ( 1001 )

Related Articles | Metrics

Based on the isotropic elastic theory and using the integral operation for Green’s function, analytical solutions of the stress and strain fields inside and outside a trapezoidal cross-section quantum-wire structure in an infinite matrix are obtained. The influences of variable height for the quantum wire cross-section and initial mismatch strain on the strain distribution are discussed.

Proper Orthogonal Decomposition Method for Flow Analysis of Multi-element Airfoil

FAN Chen-lin,LI Xiao-wei

2012, 18(1): 76-82. doi:10.3969/j.issn.1007-2861.2012.01.016

Asbtract ( 2413 )

PDF (3829KB) ( 1960 )

Related Articles | Metrics

In-depth characterization and understanding of multi-element airfoil flow structure is important to the design of multi-element airfoil shapes. Based on numerical simulation, the proper orthogonal decomposition (POD) method was used for reconstruction and analysis of numerical results of the multielement airfoil. With changes in the attack angle, the main modes of the flow are extracted, and their weight function with attack angle variation obtained. According to the simulation of flow field with an overlapped grid, a method that avoids ineffective impact on the analysis results is used by sifting the points in snapshots, and returning other points to the original position. It is found that the pulse stream flow sites account for a relatively small proportion of the total energy. The energy spectrum of the overall convergence of POD shows a slow-down pattern after a large-scale flow structure. Flow field in most of the energy distribution is directly related to the large scale flow sites included in the low-order modes, while the higher modes represent a complex pulse structure.

Effects of Eccentricity of Casing on Collapse Resistance in Non-uniform In-situ Stresses

CHEN Zhan-feng1,2,ZHU Wei-ping1,2,DI Qin-feng1,2,TANG Ji-ping3,LIANG Hong-jun3

2012, 18(1): 83-86. doi:10.3969/j.issn.1007-2861.2012.01.017

Asbtract ( 2074 )

PDF (2294KB) ( 1105 )

Related Articles | Metrics

In conventional stratum, the casing strength is sufficient, and therefore the influence of casing eccentricity is generally less considered. However, in a deep well or in carbonate formations, the influence cannot be ignored. Considering eccentricity of casing in a circular hole, a finite element model is established and the effect of casing eccentricity on collapse resistance of casing analyzed. A contour map showing relationship between collapse resistance of casing and eccentricity is drawn, and a regression formula for eccentricity and the maximum casing stress is obtained. Simulation results show that, under non-uniform in-situ stress, the larger the casing eccentricity in hole, the greater the ability of casing resisting non-uniform loading in the maximum in-situ stress direction; and the larger the casing eccentricity in hole, the easier the casing damage becomes.

Dynamic Response Analysis of Interaction between Large Cylinder Structure, Soil and Wave

WANG De-jiang,YE Zhi-ming

2012, 18(1): 87-91. doi:10.3969/j.issn.1007-2861.2012.01.018

Asbtract ( 2104 )

PDF (1865KB) ( 1247 )

Related Articles | Metrics

Based on the complex dynamic characteristics of new shell structure, a 3-D nonlinear finite element model of interaction between large cylinder structure, soil and wave is established. The non-linear contact characteristics between shell structure and soil are simulated by setting contact element of zero thickness. The non-linear constitutive model of soil is used to simulate dynamic response analysis of the structure under the dynamic loads on waves. The distribution of stress in x- and y-directions, the changing law of stress with wave and buried depth are obtained. Finally the stress distribution function and reasonable size of cylinder structure are given.

Shear Strength Model of Reinforced Concrete Beam-Column Joints

CHEN Ling-li,WANG Jian

2012, 18(1): 92-97. doi:10.3969/j.issn.1007-2861.2012.01.019

Asbtract ( 3067 )

PDF (1507KB) ( 1646 )

Related Articles | Metrics

In this paper, three existing models for calculating the shear strength of the core of beam-column joint are introduced and analyzed. Calculated results show that the theoretical results do not match experimental results, with considerable dispersion. Based on the strut and truss model, a modified model is proposed, in which a softening coefficient of cracked concrete and a correction coefficient for high concrete joints are considered. Results of the modified model are close to experimental results, and are safe. The modified model can be applied to normal and high strength reinforced concrete joints.

Soil-Water and Deformation Characteristics of Unsaturated Shanghai Soft Clays

CHEN Zhen-xin,SUN De-an

2012, 18(1): 98-103. doi:10.3969/j.issn.1007-2861.2012.01.020

Asbtract ( 2112 )

PDF (1961KB) ( 1136 )

Related Articles | Metrics

Using the methods of pressure plate and filter paper to measure soil suction, the soil-water characteristic curves and shrinkage deformation curves of undisturbed Shanghai soft clays (layers ②, ③ and ④) and reconstituted Shanghai soft clays(layers ② and ③) are obtained from drying tests. It is found from the soil-water characteristic tests that the air-entering values of undisturbed Shanghai soft clays (layers ②, ③ and ④) and reconstituted Shanghai soft clays (layers ② and ③) are 150~180, 220~250, 650~800, 260~310, and 550~600 kPa, respectively. The process of soil shrinkage can be divided into three stages: elastic stage, elastoplastic stage and stage of shrinkage limit. When the imposed matric suction is small, the soil is at the elastic stage. When the large suction is applied, the shrinkage deformation becomes noticeable, and the soil is at the elastoplastic stage. At the stage of shrinkage limit, almost no deformation occurs during the increase of matric suction.

Element Partition Method with Mohr-Coulomb Criterion for Simulating Joint Propagation Subjected to Confining Stress

YANG Fan1,ZHANG Zhen-nan2

2012, 18(1): 104-110. doi:10.3969/j.issn.1007-2861.2012.01.021

Asbtract ( 2553 )

PDF (3094KB) ( 1171 )

Related Articles | Metrics

To explore and develop a simple and efficient method for simulating joint propagation, we take the element partition method (EPM) as a basic method and use the Mohr-Coulomb criterion in element partition. Since EPM takes advantage of geometric features of triangular elements to construct a special threenode contact element, it can simulate fracture propagation in the original mesh scheme. The Mohr-Coulomb criterion is a widely used failure criterion in geomechanics, which can accurately predict the failure behavior of geomaterials. Combination of EPM and Mohr-Coulomb criterion has advantages in simulating joint propagation. To validate the proposed method, failure processes of crack material with double crack subject to different confining stresses are simulated. The results show general agreement with experiments. Comparison between the simulated and experimental results suggests that the method can represent typical features of joint propagation and coalescence subject to confining stress.

Table of Content