The Fork-Join program consisting of K parallel tasks is a useful model for a large number of computing applications. When the parallel processor has multi-channels, later tasks may finish execution earlier than their earlier tasks and may join with tasks from other programs. This phenomenon is called exchangeable join (EJ), which introduces correlation to the task’s service time. In this work, we investigate the response time of multiprocessor systems with EJ with a new approach. We analyze two aspects of this kind of systems: exchangeable join (EJ) and the capacity constraint (CC). We prove that the system response time can be effectively reduced by EJ, while the reduced amount is constrained by the capacity of the multiprocessor. An upper bound model is constructed based on this analysis and a quick estimation algorithm is proposed. The approximation formula is verified by extensive simulation results, which show that the relative error of approximation is less than 5%.

In this paper we propose a two-layer emergent model for scalable swarm system. The first layer describes the individual flocking behavior to the local goal position (the center of minimal circumcircle decided by the neighbors in the positive visual set of individuals) resulting from the individual motion to one or two farthest neighbors in its positive visual set; the second layer describes the emergent aggregating swarm behavior resulting from the individual motion to its local goal position. The scale of the swarm will not be limited because only local individual information is used for modelling in the two-layer topology. We study the stability properties of the swarm emergent behavior based on Lyapunov stability theory. Simulations showed that the swarm system can converge to goal regions while maintaining cohesiveness.

A statistical algorithm for the reconstruction from time sequence echocardiographic images is proposed in this paper. The ability to jointly restore the images and reconstruct the 3D images without blurring the boundary is the main innovation of this algorithm. First, a Bayesian model based on MAP-MRF is used to reconstruct 3D volume, and extended to deal with the images acquired by rotation scanning method. Then, the spatiotemporal nature of ultrasound images is taken into account for the parameter of energy function, which makes this statistical model anisotropic. Hence not only can this method reconstruct 3D ultrasound images, but also remove the speckle noise anisotropically. Finally, we illustrate the experiments of our method on the synthetic and medical images and compare it with the isotropic reconstruction method.

Mean shift is a widely used clustering algorithm in image segmentation. However, the segmenting results are not so good as expected when dealing with the texture surface due to the influence of the textures. Therefore, an approach based on wavelet transform (WT), co-occurrence matrix (COM) and mean shift is proposed in this paper. First, WT and COM are employed to extract the optimal resolution approximation of the original image as feature image. Then, mean shift is successfully used to obtain better detection results. Finally, experiments are done to show this approach is effective.

Fingerprint segmentation is an important step in fingerprint recognition and is usually aimed to identify non-ridge regions and unrecoverable low quality ridge regions and exclude them as background so as to reduce the time expenditure of image processing and avoid detecting false features. In high and in low quality ridge regions, often are some remaining ridges which are the afterimages of the previously scanned finger and are expected to be excluded from the foreground. However, existing segmentation methods generally do not take the case into consideration, and often, the remaining ridge regions are falsely classified as foreground by segmentation algorithm with spurious features produced erroneously including unrecoverable regions as foreground. This paper proposes two steps for fingerprint segmentation aimed at removing the remaining ridge region from the foreground. The non-ridge regions and unrecoverable low quality ridge regions are removed as background in the first step, and then the foreground produced by the first step is further analyzed for possible remove of the remaining ridge region. The proposed method proved effective in avoiding detecting false ridges and in improving minutiae detection.

Location privacy means a user keeps his/her geographical location secret. If location information falls into the wrong hands, an adversary can physically locate a person. To address this privacy issue, Qi et al.(2004a; 2004b) proposed a special and feasible architecture, using blind signature to generate an authorized anonymous ID replacing the real ID of a legitimate mobile user. The original purpose of his architecture was to eliminate the relationship of authorized anonymous ID and real ID. We present an algorithm to break out Qi’s registration and re-confusion protocol, and then propose a new mechanism based on bilinear pairings to protect location privacy. Moreover we show that the administrator or third parity cannot obtain information on the legitimate user’s authorized anonymous ID and real ID in our proposed protocols.

The dilemma of the quantization parameter (QP) being involved in both rate control and rate-distortion optimization (RDO) prevents using the traditional rate control scheme. Although some rate control schemes are proposed to circumvent the dilemma, the inaccurate prediction model and improper bit allocation deter H.264 application on low bandwidth channel. To resolve this issue, this paper proposes a novel rate control scheme by considering the macroblock (MB) encoding complexity variation and buffer variation and by exploiting the spatio-temporal correlation sufficiently well. Simulations showed that this scheme improves the perceptual quality of the pictures with similar or smaller PSNR deviations when compared to that of rate control in JVT-O016.

This paper introduces a method to realize digital television transmission based on asynchronous transfer mode (ATM), a technology based on fast packet switching. Choosing the integral multiple length of an ATM cell payload to equal to the length of an MPEG transport stream packet, an MPEG transport stream packet can be inextenso loaded by several ATM cells and transformed into ATM cells. Using ATM virtual connection technology and B-ISDN, the interoperability between ATM and DTV may be realized, and DTV signal transmission may also be realized finally.

With the increase of high-speed network backbones, the performance of server’s network interface gradually becomes a pivotal factor. This study provides a method called Ethernet Links Trunking (ELT) technology for achieving efficient connectivity between backbones and servers, which provides higher bandwidth and availability of server network interface. The overview of the ELT technology and the results of performance experiment are presented in this paper. Findings showed that the network bandwidth can be scaled by multiple ELT technologies so that more reliable network connectivity can be guaranteed. Some crucial techniques such as Adapter Load Balancing (ALB) and Adapter Fault Tolerance (AFT) are presented in this paper. Experimental results showed that parallel channels of Fast Ethernet are both necessary and sufficient for supporting the data rates of multiple concurrent file transfers on file server.

This paper proposes an approach for rendering breaking waves out of large-scale of particle-based simulation. Moving particle semi-implicit (MPS) is used to solve the governing equation, and 2D simulation is expanded to 3D representation by giving motion variation using fractional Brownian motion (fBm). The waterbody surface is reconstructed from the outlines of 2D simulation. The splashing effect is computed according to the properties of the particles. Realistic features of the wave are rendered on GPU, including the reflective and refractive effect and the effect of splash. Experiments showed that the proposed method can simulate large scale breaking waves efficiently.

This paper discusses current research and development of fringe projection-based techniques. A system based on Fourier transform profilometry (FTP) is proposed for three-dimensional (3D) shape recovery. The system improves the method of phase unwrapping to gain accurate 3D shapes of objects. The method uses a region-growing algorithm for the path prediction guided by the quality map to increase the recovering accuracy and provides a fast and simple tool for 3D shape recovery. The shape measurement and data recovery are integrated to offer a new method of 3D modelling. Examples are presented to verify the feasibility of the proposed method.

The separation between s- and p-polarization components invariably affects thin film edge filters used for tilted incidence and is a difficult problem for many applications, especially for optical communication. This paper presents a novel design method to obtain edge filters with non-polarization at incidence angle of 45°. The polarization separation at 50% transmittance for a long-wave-pass filter and a short-wave-pass filter is 0.3 nm and 0.1 nm respectively. The design method is based on a broadband Fabry-Perot thin-film interference filter in which the higher or lower interference band at both sides of the main transmittance peak can be used for initial design of long-wave-pass filter or short-wave-pass filter and then can be refined to reduce the transmittance ripples. The spacer 2H2L2H or 2L2H2L of the filter is usually taken. Moreover, the method for expanding the bandwidth of rejection and transmission is explained. The bandwidth of 200 nm for both rejection region and transmission band is obtained at wavelength 1550 nm. In this way, the long-wave-pass and short-wave-pass edge filters with zero separation between two polarization components can easily be fabricated.

Colorimetric characterization is to transform the device-dependent responses to device-independent colorimetric values, and is usually conducted in CIEXYZ space. However, the optimal solution in CIEXYZ space does not mean the minimization of perceptual error. A novel method for colorimetric characterization of imaging device based on the minimization of total color difference is proposed. The method builds the transform between RGB space and CIELAB space directly using the downhill simplex algorithm. Experimental results showed that the proposed method performs better than traditional least-square (LS) and total-least-square (TLS) methods, especially for colors with low luminance values.

Adaptive modulation (AM) is an effective technique to approach the theoretical bound of multi-input and multi-output (MIMO) channel. In most previous studies, the AM parameters were obtained by maximizing the transmission rate for a given total transmit power. In this paper, a novel AM-MIMO algorithm is presented, which is based on minimizing total transmit power when the link’s QoS requirements are given. By taking the QoS requirements into account directly, the proposed algorithm not only makes the system more flexible, but also makes the cross layer design of wireless network easier. At last, the numerical results of the proposed scheme are presented.

The new 6-switch single-phase 5-level current-source inverter proposed in this paper was developed by properly simplifying the traditional 8-switch single-phase 5-level current-source inverter, and its operational principle was analyzed. Just like the problem of voltage-unbalance between different levels existing in voltage-source multilevel inverters, a similar problem of current-unbalance between different levels whether for the 8-switch single-phase 5-level current-source inverter, or for the new 6-switch 5-level current-source inverter also exists. A simple current-balance control method via DC current feedback is presented here to implement the current-balance control between different levels. And to reduce the output current harmonics, PWM control technique was used. Simulation and experimental results showed that this new 6-switch topology operates correctly and that the balance-inductor can almost equally distribute the total DC current.

This paper presents a wideband frequency-shift keying (FSK) demodulator suitable for a digital data transmission chain of wireless neural stimulation microsystems such as cochlear implants and retinal prostheses. The demodulator circuit derives a constant frequency clock directly from an FSK carrier, and uses this clock to sample the data bits. The circuit occupies 0.03 mm² using a 0.6 μm, 2M/2P, standard CMOS process, and consumes 0.25 mW at 5 V. This circuit was experimentally tested at transmission speed of up to 2.5 Mbps while receiving a 5~10 MHz FSK carrier signal in a cochlear implant system.

In this paper, Radial point collocation method (RPCM), a kind of meshfree method, is applied to solve convection-diffusion problem. The main feature of this approach is to use the interpolation schemes in local supported domains based on radial basis functions. As a result, this method is local and hence the system matrix is banded which is very attractive for practical engineering problems. In the numerical examination, RPCM is applied to solve non-linear convection-diffusion 2D Burgers equations. The results obtained by RPCM demonstrate the accuracy and efficiency of the proposed method for solving transient fluid dynamic problems. A fictitious point scheme is adopted to improve the solution accuracy while Neumann boundary conditions exist. The meshfree feature of the present method is very attractive in solving computational fluid problems.

In this paper, we study a class of Finsler metric in the form F=αexp(β/α)+εβ, where α is a Riemannian metric and β is a 1-form, ε is a constant. We call F exponential Finsler metric. We proved that exponential Finsler metric F is locally projectively flat if and only if α is projectively flat and β is parallel with respect to α. Moreover, we proved that the Douglas tensor of exponential Finsler metric F vanishes if and only if β is parallel with respect to α. And from this fact, we get that if exponential Finsler metric F is the Douglas metric, then F is not only a Berwald metric, but also a Landsberg metric.

We first prove that for a finite dimensional non-semisimple Hopf algebra H, the trivial H-module is not projective and so the almost split sequence ended with k exists. By this exact sequence, for all indecomposable H-module X, we can construct a special kind of exact sequence ending with it. The main aim of this paper is to determine when this special exact sequence is an almost split one. For this aim, we restrict H to be unimodular and the square of its antipode to be an inner automorphism. As a special case, we give an application to the quantum double D(H)=(H^op)^*>< H) of any non-semisimple Hopf algebra.

Modifying the knots of a B-spline curve, the shape of the curve will be changed. In this paper, we present the effect of the symmetric alteration of four knots of the B-spline and the NURBS surfaces, i.e., symmetrical alteration of the knots of surface, the extended paths of points of the surface will converge to a point which should be expressed with several control points. This theory can be used in the constrained shape modification of B-spline and NURBS surfaces.

A novel mobile robot simultaneous localization and mapping (SLAM) method is implemented by using the Rao-Blackwellized particle filter (RBPF) for monocular vision-based autonomous robot in unknown indoor environment. The particle filter combined with unscented Kalman filter (UKF) for extending the path posterior by sampling new poses integrating the current observation. Landmark position estimation and update is implemented through UKF. Furthermore, the number of resampling steps is determined adaptively, which greatly reduces the particle depletion problem. Monocular CCD camera mounted on the robot tracks the 3D natural point landmarks structured with matching image feature pairs extracted through Scale Invariant Feature Transform (SIFT). The matching for multi-dimension SIFT features which are highly distinctive due to a special descriptor is implemented with a KD-Tree. Experiments on the robot Pioneer3 showed that our method is very precise and stable.

Efficient lookup is essential for peer-to-peer networks and Chord is a representative peer-to-peer lookup scheme based on distributed hash table (DHT). In peer-to-peer networks, each node maintains several unidirectional application layer links to other nodes and forwards lookup messages through such links. This paper proposes use of bidirectional links to improve the lookup performance in Chord. Every original unidirectional link is replaced by a bidirectional link, and accordingly every node becomes an anti-finger of all its finger nodes. Both theoretical analyses and experimental results indicate that these anti-fingers can help improve the lookup performance greatly with very low overhead.

Conditional access system (CAS) is a key technical component in digital TV broadcasting through which TV operators manage the appropriate rights of different subscribers in order to protect their commercial benefits. The normal digital TV receiver can only receive and decode the pay TV programs scrambled by one specific CAS. In this paper, the authors proposed a smartcard conditional access interface (SCAI) scheme in order to make the digital TV receiver be a common receiving platform independent of any specific CAS employed at the broadcasting head-end. As a result, it only needs to include a common conditional access software package (CCAP) without any requirement of hardware modification in the receiver. Comparison between the two mentioned DVB-CI-based schemes showed that the cost of such kind receiver is greatly reduced. The main design points of the proposed scheme and its reference implementation’s architecture are presented in this paper. This scheme is also one of the candidate national standards for Chinese digital TV broadcasting industry.

In this paper, the Chung’s strong law of large numbers is generalized to the random variables which do not need the condition of independence, while the sequence of Borel functions verifies some conditions weaker than that in Chung’s theorem. Some convergence theorems for martingale difference sequence such as L_p martingale difference sequence are the particular cases of results achieved in this paper. Finally, the convergence theorem for A-summability of sequence of random variables is proved, where A is a suitable real infinite matrix.

Outsourcing software development has many advantages as well as inevitable risks. Of these risks, outsourcee selection is one of the most important. A wrong outsourcee selection may have severe adverse influence on the expected outcome of the project. We analyzed the risks involved in outsourcee selection and also provided methods to identify these risks. Using the principles of Analytical Hierarchy Process (AHP) and Cluster Analysis based on Group Decision Making, we established an index evaluation system to evaluate and select outsourcees. Real world applications of this system demonstrated its effectiveness in evaluating and selecting qualified outsourcees.