Abstract:The uncertainty of parameters may greatly affect structural analysis of solid propellant grain. Nonpropabilistic convex set theoretic models, combining viscoelastic finite element method, are developed to study the uncertainty of structures. Incremental method is applied to solving the hereditary integrals. Perturbation method is used to predict the variability of structural response. The developed method is suitable for uncertain structural analysis of solid propellant grain. By comparing with stochastic structural analysis, relations between the two methods have been found.

Abstract:As to the formation flying with the reference satellite running on the circle orbit, the J₂ effect is analyzed. On this basis, a method is presented, which can greatly mitigate the breakage of the J₂ term by adjusting the semi-major axis of the flying-around satellite. Furthermore,the cost required for formation-keeping is evaluated under impulse assumption. Numerical results show that satellite formations with semi-major axis adjustments will last much longer, and save much more fuel.

Abstract:By simplifying the sensible volume of semiconductor devices, the energy deposited by the cosmic high-energy protons in the sensible volume of semiconductor devices is calculated. Then by using the Weibull function derived by the ground heavy ion SEU experiments, the SEU rates of several devices by the shielding of 33 mm Aluminum in the CRRRES satellite orbit are calculated. The results are compared and analyzed.

Abstract:The stress-ongitudinal strain curves are tested using electrical measuring technique. The relations of transverse strain and longitudinal strain are also examined. Some properties such as σ_0.2，σ_b,E and μ are derived. Testing results indicate that transverse strain has distinct jumpiness when stress exceeds proportional elongation point. The properties of the material are different from others.The fracture toughness of surface crack is also tested.

Abstract:The synthesis of 4-hydroxy-3-methoxyphenyl-oxime has been made from vanillin, hydroxylamine hydrochloric acid and sodium hydroxide The factors of reaction process include theses,such as temprature, reagents, strong bases and mole rate of three crude materials.The results show that the best mole rate of three reagents is for 1:1.2:1.1,and the product ratio is for 97.4%The reactive process and mechanism of synthesis has been discussed.The synthesis product has been corroborated after the measurement of melting point and analysis of its IR,¹HNMR,TIC-MS.

Abstract:A sol-gel process catalyzed by oxalic acid is adopted to prepare SiC precursors from the raw materials of tetrathoxysilane(TEOS) and phenolic resin, and this precursors is homogeneous without the impurity of sulfur and chlorine. Sintered with a certain heating program, the obtained precursor is converted into SiC. The effect of the temperature on the hydrolyze and the influences of the amount of water, catalyzer, Nickel nitrate on the gel time are studied.The result shows that the sol-gel process is best at 40℃ and exactly after 24 hr. While the ratio of H₂O:TEOS(mol) is 0.28 and oxalic acid:TEOS(mol) is 0.01, the gel time is shortest. The further study indicates that β-SiC is converted when the catalyzer of the nickel nitrate is mixed in.

Abstract:Bootstrap filtering algorithm is a recursive Bayesian estimation algorithmSince in this algorithm the probability density function of the state to be estimated is approximated by a series of samples, it can be applied to the circumstance of nonlinear system model and observation model ,even non-Gaussion noise .The bootstrap filter is compared with the Extended Kalman Filter(EKF),the simulation results have shown that the performance of the bootstrap filter is better than that of EKF

Abstract:This paper introduces a BPSK carrier synchronization method and its performance analysis for software implementation in all digital receiverThis method is adapt to systems in which both time and memory are limited.The software methods can overcome many weaknesses of the traditional methods and is fit for implementation on Digital Signal Processors(DSPs)

Abstract:Based on the active service low-resolution radar,the approach to target classification and recognition is studiedThe methods of target classification and recognition are proposed according to the information of target waveforms and time-spectrum.The feature extracted by target waveform is analyzed and the useful conclusion is given.The experiment result demonstrates that this recognition method is fine to classify big and small aircraft and recognize single plane and multiple planes

Abstract:In the high-level synthesis of application-specific DSP, pipeline scheduling is a difficult and urgent problem Concerning the global optimization of functional units, registers and interconnections, we present a novel algorithm called directed searching pipeline scheduling, which is an iterative algorithm utilizing heuristic information. The heuristic information speeds up the searching process, and the weakness of deterministic algorithms, which are vulnerable to trip in local optimal, poor at hill-climbing, is overcome in this algorithm. Typical design examples show that the performance of this transformation-based algorithm has reached or surpassed the best performance of pipeline scheduling algorithms reported so far.

Abstract:In the paper,we analyze the processing efficiency of single processor in the cluster bus multiprocessors system,then we bring forward some suggestions about memory structure in the systemFinally,we compare the processing performance of cluster bus system, bus shared system and distributed bus system.

Abstract:Parametric methods are the important techniques for the implementation of Radio Frequency Interference (RFI) suppression in Ultra-WideBand Synthetic Aperture Radar (UWB-SAR). In order to evaluate the practical performance of RFI suppression algorithms, several primary algorithms are compared by the interference suppression capability and pulse compression performance. The results with real data from our experimental UWB-SAR show that parametric algorithms can greatly improve the signal-to-interference ratio. Furthermore, the algorithms based on sinusoidal model are a little better than those based on AR model.

Abstract:A layered-stucture simulation training model is put forward on the basis of the characters of communication equipment and network,and the demand of communication service training.The modeling methods and program technologies of its layers are developed and analysized.And then,some simulation training systems of communication equipment and network are represented,which show that this model is easy to program,comprehend maintain and extend the system functions for a simulation training system of communication equipment and network.

Abstract:Researchers have developed two general categories of intrusion detection, i.e. misuse detection and anomaly detection, which differ at model construction. Signature-based misuse detection, which can detect the well-known attacks, will do nothing when new attack comes. Even traditional anomaly detection can catch some new attacks, the learning process overly relying on the training data sets which contain either purely clean normal data or correctly labeled data makes it useless in most cases. To solve such a problem, a novel clustering based method, capable of proces sing training data sets without type label and/or containing unknown intrusion data, is presented in this paper. After the normalization of network connection data, cluster centroids which is null at first can be obtained gradually and automatically through comparing the distance between data instances and the predefined cluster width , and each data instance can be then classified into the cluster which has the minimum distance with it. To ensure that the clusters can best represent the data distribution, cluster centroids also can be dynamically adjusted according to data instances contained in this cluster. With the classified data instances, the anomaly data clusters can be easily identified using normal cluster ratio , therefore performing the real-time detecting of each real network connection datum. Experiment result shows that this method can not only detect some new attacks, from network connection data sets, with low false positive rate, but also tolerate more general data sets.

Abstract:Hardware-software partitioning is a key problem in Hardware-software co-design of embedded system. This paper describes an approach based on genetic algorithm. Two methods, simulated annealing and selecting an individual in light of the probability, are used to enhance the algorithm's self-adaptability, efficiency and global optimization. It is validated via the partitioning results that the approach is stable, effective and efficient.

Abstract:In this paper, we design and implement fast multiplierless approximations of the discrete cosine transform(DCT) with the lifting scheme, named the BinDCT, which avoids floating-point operations. This algorithm is derived from the plane rotation-based algorithm, which reduces 3 lifting steps to 2 lifting steps via the plane rotation. So the arithmetic operation cost is reduced greatly.

Abstract:A modified type of preconditioner is constructed with the help of local block factorization of block tridiagonal matrices. Then the existence and the properties are analyzed. For the standard 5-point matrices, which are derived from the 2-D Laplace operator, the actual condition numbers of the preconditioned matrices are computed. The result shows that the condition number is proportioned to the square root of the order of the matrix. What's more, the longer the step of the local factorization, the smaller the coefficient is. Then efficient implementations of the preconditioners are focused on and three of them provided. Finally lots of experiments are performed for the constructed preconditioners and the well-known effective ones on the personal computer with main frequency of 550MHz and memory of 256M. The matrices in these experiments include the standard five point ones, and the ones derived from a 2-D elliptic operator with discontinuous coefficients. The results also show that the preconditioners are more efficient than the other tested ones.

Abstract:This paper makes an investigation on the algorithm of reconstructing a 3-dimentional CAD model from 2-dimentional photos. By a set of points picked on an object's 2-dimentional photos taken in different direction and position, the movement parameters of the camera and the 3-dimentional geometric information and CAD model of the object can be calculated with the algorithm. The algorithm makes sufficient considerations on the constraints of manufacturablity in CAD modelling, is accurate and robust, and will be useful in the fields of reverse engineering, computer vision and robots.

Abstract:The 2-step analysis method is the most widely used one in the statistical analysis for constant-stress accelerated life testing under Weibull distribution. Aiming at the deficiency of the 2-step analysis method, this paper develops a new analysis method based on constructed data. The method takes the relativity between Weibull distribution parameters into account, and introduces the 2nd. estimation for characteristic life of Weibull distribution. The method exceeds the 2-step analysis on model fitting optimization, and it improves the precision of estimation for accelerated life testing in turn. Moreover, the method is more convenient for software realization and engineering application.

Abstract:Decoupled mechanisms are easy to control and apt to reach high accuracy. With the development of mechanism, especially the emergence of parallel mechanism, many ones don’t hold this merit. Especially, this merit is not an intrinsic property of parallel mechanisms. It's necessary to study their simple motions and name these motions decoupling motions. This kind of movement links very closely to motion control and pathway planning. By the use of screw theory, the definition, classification of decoupling motion, together with the necessary condition and several sufficient conditions for mechanism to realize this kind of motion are given. Finally a mechanism design example of a 2-DOF rotating mechanism is given as an example. It incarnates our research work on mechanism design and mechanism analysis.

Abstract:This paper analyses the various singular configurations of parallel manipulators using differential geometry. In order to perform an effect control of parallel manipulators at singularity, an over-control method is discussed. Finally this paper tests the over-actuated PD strategy on the two degree-of-freedom planar parallel robot and simulations verify the method proposed.

Abstract:In this paper,the sequential estimation of the parameter of normal distribution is studied.At first,the Bayes double sampling sequential estimation method is constructed.And then,as the requirment for testing analysis,the recursive algorithm of the estimator of distribution parameter is given.Thus,it can easily be realized in engineering applications.

Abstract:The reflux is one of the main factors to cause the dielectric surface flashover. The possibility of producing the reflux is analyzed by the use of Maxwell equations and electron motion equations. Then the PIC method is used to simulate the production process of the e beam electrons reflux in foil-less diode and some simulation results are presented. Finally some suggestions are given for the optimum design of the diode by choosing proper guiding magnetic field and diode structure.

Abstract:In order to find a calculation method for the design of Vlasov-type mode convertor, the electromagnetic radiation field of Vlasov radiator is calculated with vector diffraction theory by constructing a proper model, and the field scattered by reflector is also calculated. Numerical far-field radiation patterns accord basically with the measured results, which shows that it is useful for the design and analysis of Vlasov-type mode convertor.