Applied Mechanics and Materials Vols. 651-653

Abstract: Variance analysis is a method commonly used in biological statistics. How to use the statistical analysis software for the analysis of variance to achieve the results of the rapid and scientific treatment, to obtain the correct conclusion, is an important part of agricultural research. This paper introduces how to use SPSS (Statistical Package for the Social Science or Statistic Products andService Solution) data analysis tool for the variance analysis method; the data analysis and processing of fast, accurate and intuitive; compared with Excel, the statistical analysis function of SPSS is more powerful, can help to improve the efficiency of data processing, but also reduce the experimental cost.

Abstract: Probability limit theory is not only one of the main branches of probability theory, but also important base of others branches and mathematical statistics. In this paper, we discuss the laws of large number and complete convergence of mixing sequence. The paper can be divided into the following two parts. In the first part, we introduce Cesáro uniform integrability and equivalent conditions for sequences of random variable. Then, under these conditions, we establish the weak law of large number and convergence for weighted sums of sequences of random variable of mixing sequence. As an especial example, we obtain the weak law of large number or convergence for mixing sequence. From the course, we can find that Cesáro uniform integrability is also an effective tool for researching the weak law of large number of mixing sequence .In the second part, we investigate a complete convergence theorem of weighted sums of mixing sequence, and prove that the existent results are especial examples for it.

Abstract: Modeled on data quality evaluating methodology, a standardized real-time rule framework is proposed. The framework is accurate and reliable for estimating aerospace experiment tracking information. By converting all the rule evaluating results that consisted of time-delay, error, range and etc. to a uniform questionable data count format, the framework presents an algorithm that takes both tracking source and station into account based on questionable data statistics for quantified information quality evaluating. The quantitative quality evaluation of measurement information problem can be solved via the proposed framework.

Abstract: Watermark technique is one of the technologies of digital product property rights protection. This article from the basic principle, implementation, application, the experimental results were introduced from the aspects such as watermark technology, and analyzes the digital water marking technology based on singular value decomposition.

Abstract: The moving of a population of agents in a closed rectangular cell is emulated by adding the bouncing of the agents at boundaries and the remembering of end boundaries to a basic model, in which only the alignment rule is used. The agents herd in only a few clusters, which move asymmetrically along the long side of the cell. The effects of the end-boundary information holding time , noise strength and number of agents on the formation and turning of the cluster are investigated. The results show that, the formation time of the cluster decreases with , but increases with and ; the turning time of the cluster slightly increases with and , but nearly does not change with .

Abstract: Nonlinear processing techniques are very important tools in signal and image processing. One of the most significant classes is based on order statistics, which depends on data sorting algorithm. The sorting algorithm is very time consuming and intractable for vector data. Therefore, it is extremely urgent to put forward a fast algorithm to compute the r-th order statistics without sorting operation. The ITM and ITTM filter proposed an iteration algorithm to approach the median, from the mean. It is reasonab to generalize to the other order statistics by building a new data set Xp which is composed by the elements of the original set X. It makes the r-th order statistics of X to be the median of Xp. Then, estimating the median of the new set Xp by the ITTM algorithm, the output is estimation value of x_(r).

Abstract: To solve the problem of standard particle swarm optimization (PSO) easy turn to premature convergence and poor ability in local search, this paper present a hybrid particle swarm optimization algorithm merging simulated annealing (SA) and mountain-climb. During the running time, the algorithm use the pso to find the global optimal position quickly, take advantage of the Gaussian mutation and mountain-climb strategy to enhance local search ability, and combine with SA to strengthen the population diversity to enable particles to escape from local minima. Test results on several typical test functions show that this new algorithm has a significant improve in searching ability and effectively overcome the premature convergence problem.

Abstract: Many methods of constructing S-box often adopt the classical chaotic equations. Yet study found that some of the chaotic equations exists drawbacks. Based on that, this paper proposed a new method to generate S-Box by improving the Logistic map and Henon map, and combining the real and imaginary part of complex produced by the Mandelbrot set. By comparing with several other S-boxes proposed previously, the results show the S-box here has better cryptographic properties. So it has a good application prospect in block ciphers.

Abstract: Hashing which maps data into binary codes in Hamming space has attracted more and more attentions for approximate nearest neighbor search due to its high efficiency and reduced storage cost. K-means hashing (KH) is a novel hashing method which firstly quantizes the data by codewords and then uses the indices of codewords to encode the data. However, in KH, only the codewords are updated to minimize the quantization error and affinity error while the indices of codewords remain the same after they are initialized. In this paper, we propose an optimized k-means hashing (OKH) method to encode data by binary codes. In our method, we simultaneously optimize the codewords and the indices of them to minimize the quantization error and the affinity error. Our OKH method can find both the optimal codewords and the optiaml indices, and the resulting binary codes in Hamming space can better preserve the original neighborhood structure of the data. Besides, OKH can further be generalized to a product space. Extensive experiments have verified the superiority of OKH over KH and other state-of-the-art hashing methods.

Abstract: The optimum vibrational resonance in a time-delay bistable system driven by bihiarmonic signals is discussed in this paper. The theoretically expression for the response amplitude gain of low frequency signal in the time-delay bistable system is deduced, and the effects of time delay parameter on the optimum vibrational resonance peak and the required amplitude of high frequency signal are investigated. It is shown that the optimum vibrational resonance can be achieved by adjusting the high frequency signal amplitude and time delay parameter jointly. Meanwhile, the optimum vibrational resonance appeared periodically with time delay parameter and the period is equal to the period of low-frequency signal. The amplitude of high-frequency signal required for the optimum vibrational resonance can be fixed or varied with different time delay parameter depending on the ratio of the frequencies between biharmonic signals.