基于SVM法线算法的特征基因提取

鉴于特征提取在基因分类中扮演的重要作用,提出了一种基于支持向量机法线算法的特征提取方法。首先,使用线性支持向量机对训练样本进行训练,求出支持向量;然后,根据SVM法线算法计算各特征的权重,来获得法线向量,剔除数据集中权重较小的冗余特征基因;最后,使用线性支持向量机分类器实现对有效特征基因子集的分类。用急性白血病数据集和结肠癌数据集进行了实验,结果表明:该方法识别效果良好,适应性强。

一种改进的线性支持向量机的特征筛选算法

针对SVM法线特征筛选算法仅考虑法线对特征筛选的贡献,而忽略了特征分布对特征筛选的贡献的不足,在对SVM法线算法进行分析的基础上,基于特征在正、负例中出现概率的不同提出了加权SVM法线算法,该算法考虑到了法线和特征的分布。通过试验可以看出,在使用较小的特征空间时,与SVM法线算法和信息增益算法相比,加权SVM法线算法具有更好的特征筛选性能。

基于工业机器人的复杂曲面路径规划及组合抛光工艺研究

复杂曲面工件抛光加工中的路径规划及工艺的选择直接影响表面加工质量。针对目前机器人传输速率慢、容量受限以及对复杂曲面加工路径规划能力较差等问题,建立了一套机器人离线编程和远程控制系统,使用pose类型数据为载体传输位置坐标、欧拉角、速度等加工信息,传输速度相较一般信息传输模式提高了85.5%;同时,实现了对复杂曲面CAD模型的3D点云数据的提取及平滑插值以实现更高精度的表面重构。路径规划中提出12点采样耦合求解方法,解决了3D点云中法线误差影响机器人运行稳定的问题。结合以上控制及求解手段,通过采用发泡聚氨酯与不同粒径颗粒的组合加工工艺对6620铝合金叶片研磨抛光,实现了粗糙度从138.11 nm到2.05 nm的高效超光滑抛光加工。

结合有限元和非刚体运动估计的左心室应变分析

提出一种直观地分析左心室在心脏收缩期的应变情况的方法.首先利用心脏核磁共振图像数据对左心室进行三维实体建模,再进行有限元单元划分;然后,使用非刚体运动估计算法——单位法线及高斯曲率混合算法,在左心室收缩期相邻时刻的有限元模型表面上寻找最相似单元结点作为对应单元结点,计算各对应结点间的位移,并用α剪枝平均对位移过大的可疑结点进行处理;最后,利用弹塑性力学中应变与位移间的几何关系求解左心室表面各单元结点的应变,获得位移和应变的彩色云图.实验结果表明,该方法能直观、有效地反映左心室表面在整个心脏收缩期的形变情况和应变分布.

SPEEDING-UP RE-SAMPLED ALGORITHM IN RAY CASTING VOLUME RENDERING OF MEDICAL IMAGES

Ray casting algorithm can obtain a better quality image in volume rendering, however, it exists some problems, such as powerful computing capacity and slow rendering speed. How to improve the re-sampled speed is a key to speed up the ray casting algorithm. An algorithm is introduced to reduce matrix computation by matrix transformation characteristics of re-sampling points in a two coordinate system. The projection of 3-D datasets on image plane is adopted to reduce the number of rays. Utilizing boundary box technique avoids the sampling in empty voxel. By extending the Bresenham algorithm to three dimensions, each re-sampling point is calculated. Experimental results show that a two to three-fold improvement in rendering speed using the optimized algorithm, and the similar image quality to traditional algorithm can be achieved. The optimized algorithm can produce the required quality images, thus reducing the total operations and speeding up the volume rendering.

MARS:A General Multilayer Area Router

Based on a ripped-up and rerouted methodology,a multilayer area detailed router is presented by using simulated evolution technique.A modified maze algorithm is also performed for the single net.

Supersymmetric Sawada-Kotera-Ramani Equation: Bilinear Approach

In this paper, using the Hirota＇s bilineax method, we consider the N = 1 supersymmetric Sawada-Kotera- Ramani equation and obtain the Bazcklund transformation of it. Its one- and two-supersoliton solutions axe obtained and N-supersoliton solutions for N ≥ 3 are given under the condition kiξj = kjξi.

Optimal subcarrier allocation for cognitive radio with multi-user OFDM and distributed antenna

The subcarrier allocation problem in cognitive radio（CR）networks with multi-user orthogonal frequency-division multiplexing（OFDM）and distributed antenna is analyzed and modeled for the flat fading channel and the frequency selective channel,where the constraint on the secondary user（SU）to protect the primary user（PU）is that the total throughput of each PU must be above the given threshold instead of the ＂interference temperature＂.According to the features of different types of channels,the optimal subcarrier allocation schemes are proposed to pursue efficiency（or maximal throughput）,using the branch and bound algorithm and the 0-1 implicit enumeration algorithm.Furthermore,considering the tradeoff between efficiency and fairness,the optimal subcarrier allocation schemes with fairness are proposed in different fading channels,using the pegging algorithm.Extensive simulation results illustrate the significant performance improvement of the proposed subcarrier allocation schemes compared with the existing ones in different scenarios.

AN IMPLEMENTATION OF FAST ALGORITHM FOR ELLIPTIC CURVE CRYPTOSYSTEM OVER GF(p)

The design and implementation of fast algorithms related to Elliptic Curve Cryptography (ECC) over the field GF(p), such as modular addition, modular subtraction, point addition, point production, choice of embedding plaintext to a point, etc. are given. A practical software library has been produced which supports variable length implementation of the ECCbased ElGamal cryptosystem. More importantly, this scalable architecture of the design enables the ECC being used in restricted platforms as well as high-end servers based on Intel Pentium CPU. Applications such as electronic commerce security, data encryption communication, etc.are thus made possible for real time and effective ECC.

Multiple sub-array beamspace CAATI algorithm for multi-beam bathymetry system

This paper extends CAATI （Computed Angle-of-Arrival Transient Imaging） technique of Multi-angle Swath Bathymetry Sidesean Sonar （MSBSS） into Multi-Beam Bathymetry Sonar （MBBS） and presents a new Multiple Sub-array Beamspaee - CAATI （MSB-CAATI） algorithm. The method not only can achieve high resolution seafloor mapping in the whole wide swath, but also can work well in complex acoustic environments or geometries. Simulation results and processing results of sea-experiment data prove the validity and superiority of the algorithm.

A SPARSE PROJECTION CLUSTERING ALGORITHM

A clustering algorithm based on Sparse Projection (SP), called Sparse Projection Clus- tering (SPC), is proposed in this letter. The basic idea is applying SP to project the observed data onto a high-dimensional sparse space, which is a nonlinear mapping with an explicit form and the K-means clustering algorithm can be therefore used to explore the inherent data patterns in the new space. The proposed algorithm is applied to cluster a complete artificial dataset and an incomplete real dataset. In comparison with the kernel K-means clustering algorithm, the proposed algorithm is more efficient.

Upper bound solution for active thrust of nonhomogeneous cohesive soil considering effects of intermediate principal stress

Considering the variation of cohesion along the depth,the upper bound solution of active earth pressure for a rough inclined wall with sloped backfill is formulated based on a log-spiral failure mechanism.For a more accurate prediction,the influence of intermediate principal stress is taken into consideration using the unified strength theory.Converting the search for the active pressure to an optimization problem,the most critical failure surface can be located by a natural selection-based gravitational search algorithm(GSA).The proposed method is validated compared with existing methods for noncohesive and cohesive cases and proved to be more accordance with the limit equilibrium solution.The influences of the variation of soil cohesion and intermediate principal stress on active earth pressure coefficient are then fully studied.It can be concluded that both the variations of soil cohesion and intermediate principal stress have a significant influence on the active earth pressure coefficient.

Novel Interaction Solutions to Kadomtsev-Petviashvili Equation

In this paper,based on the forms and structures of Wronskian solutions to soliton equations,a Wronskianform expansion method is presented to find a new class of interaction solutions to the Kadomtsev-Petviashvili equation.One characteristic of the method is that Wronskian entries do not satisfy linear partial differential equation.

A New Power Allocation Algorithm in Cognitive Radio Networks

Most resource allocation algorithms are based on interference power constraint in cognitive radio networks.Instead of using conventional primary user interference constraint,we give a new criterion called allowable signal to interference plus noise ratio(SINR) loss constraint in cognitive transmission to protect primary users.Considering power allocation problem for cognitive users over flat fading channels,in order to maximize throughput of cognitive users subject to the allowable SINR loss constraint and maximum transmit power for each cognitive user,we propose a new power allocation algorithm.The comparison of computer simulation between our proposed algorithm and the algorithm based on interference power constraint is provided to show that it gets more throughput and provides stability to cognitive radio networks.

BFGS Method for Linear Programming

In this paper we propose an algorithm based on the BFGS Quasi-Newton method to solve a linear program. The choice of this method is justified by its theoretical efficiency, the ease to determine a descent direction and its fast convergence towards an optimal solution. Our proposed method is compared with Newton＇s method for linear program named lpnew, widely used as an optimization algorithm for classification problems.

Fault detection of large-scale process control system with higher-order statistical and interpretative structural model

Nonlinear characteristic fault detection and diagnosis method based on higher-order statistical(HOS) is an effective data-driven method, but the calculation costs much for a large-scale process control system. An HOS-ISM fault diagnosis framework combining interpretative structural model(ISM) and HOS is proposed:(1) the adjacency matrix is determined by partial correlation coefficient;(2) the modified adjacency matrix is defined by directed graph with prior knowledge of process piping and instrument diagram;(3) interpretative structural for large-scale process control system is built by this ISM method; and(4) non-Gaussianity index, nonlinearity index, and total nonlinearity index are calculated dynamically based on interpretative structural to effectively eliminate uncertainty of the nonlinear characteristic diagnostic method with reasonable sampling period and data window. The proposed HOS-ISM fault diagnosis framework is verified by the Tennessee Eastman process and presents improvement for highly non-linear characteristic for selected fault cases.

A NOVEL SUPER-RESOLUTION BEAMFORMING ALGORITHM

A novel simply-structured hybrid smart antenna system suitable to be used in ad-hoc network terminals is proposed in this letter. The super-resolution beamforming algorithm is also pre-sented based on the system using DOA estimation results. The algorithm can switch the beamforming to the direction of the expected signal and get the best transmitting performance after the pre-beamforming of the Butler matrix. The shifting value formulas are presented to obtain the best SNR when there is no interfering signal and to acquire the highest Signal to Interference Ratio (SIR) as there is one interfering signal. When there are more than one interfering signals,the pre-beamforming feature of the Butler matrix can also suppress the interfering signals. Simulation results verified the algorithm.

Nonlinear inversion for electrical resistivity tomography based on chaotic DE-BP algorithm

Nonlinear resistivity inversion requires efficient artificial neural network(ANN)model for better inversion results.An evolutionary BP neural network(BPNN)approach based on differential evolution(DE)algorithm was presented,which was able to improve global search ability for resistivity tomography 2-D nonlinear inversion.In the proposed method,Tent equation was applied to obtain automatic parameter settings in DE and the restricted parameter Fcrit was used to enhance the ability of converging to global optimum.An implementation of proposed DE-BPNN was given,the network had one hidden layer with 52 nodes and it was trained on 36 datasets and tested on another 4 synthetic datasets.Two abnormity models were used to verify the feasibility and effectiveness of the proposed method,the results show that the proposed DE-BP algorithm has better performance than BP,conventional DE-BP and other chaotic DE-BP methods in stability and accuracy,and higher imaging quality than least square inversion.

Superlinear Convergence of a Smooth Approximation Method for Mathematical Programs with Nonlinear Complementarity Constraints

Mathematical programs with complementarity constraints(MPCC) is an important subclass of MPEC.It is a natural way to solve MPCC by constructing a suitable approximation of the primal problem.In this paper,we propose a new smoothing method for MPCC by using the aggregation technique.A new SQP algorithm for solving the MPCC problem is presented.At each iteration,the master direction is computed by solving a quadratic program,and the revised direction for avoiding the Maratos effect is generated by an explicit formula.As the non-degeneracy condition holds and the smoothing parameter tends to zero,the proposed SQP algorithm converges globally to an S-stationary point of the MPEC problem,its convergence rate is superlinear.Some preliminary numerical results are reported.

Regularization Semismooth Newton Method for P_0-NCPs with Non-monotone Line Search

Based on the generalized Fischer-Burmeister function, Chen et al in 2008 put forward a regularization semismooth Newton method for solving the nonlinear complementarity problem with a P0-function. In this paper, we investigate the above algorithm with the monotone line search replaced by a non-monotone line search. It is shown that the non-monotone algorithm is well-defined, and is globally and locally superlinearly convergent under standard assumptions.