Applied Mechanics and Materials Vols. 427-429

Abstract: Stability, fast-settling and anti-noise abilities are the three necessary parameters to measure the performance of Phase locked lock (PLL). By using of a LPF with strong noise-suppression ability and replacing multiplier of PD with an adder, an improved PLL is proposed in this paper. Compared with the Costas Loop and Saber's PLL, the proposed PLL has better phase tracking ability, shorter settling time, and stronger anti-noise ability. It is expected that our work will be helpful for developing PLL with lower-price and higher-performance, and then contribute to more applications in communication.

Abstract: In this paper we propose a method to search curves in pre-processed images based on tensor voting. Given an image that has been binarized and thinned to single-pixel representations, the method performs a five-neighbor searching process which takes into account the turning angle of the pixels on a certain curve. At last the method is tested on several real images.

Abstract: Automatic focusing is one of the key technology of robot vision and digital video-systems, while play an important role in determining the quality of image. The performance of focusing depends on whether the evaluation function has unbiasedness, unimodality and noise resistance. This paper proposes a new evaluation function algorithm by improving image clarity-evaluation function of the traditional neighborhood difference operator. Compared with the existing algorithm, the results of experiments demonstrated the new algorithm has a good sensitivity, timeliness, good anti-noise ability and stability during the automatic focusing process.

Abstract: Functional magnetic resonance imaging (fMRI) has become one of the important tools of functional connectivity studies of the human brain. Fuzzy clustering method (FCM) is a common method for analysis of FMRI data. Traditional FCA methods measure the similarity between the BOLD time course of a centroid and the ones of all other voxels in the brain on the basis of Pearson correlation coefficient. This article puts forward a multi-voxel-based similarity measure, an improved RV (IRV) measure, which takes the hypothesis into account that the function homogeneous voxels of brain volume are spatially clustered within a local region. Experimental validation is presented through four visual fMRI data sets which shows that the IRVFCA method not only has improved the speed of FCA, but has comparatively raised the accuracy of the method.

Abstract: In order to overcome the non-uniform property of time domain signal which is obtained in the process of track irregularity detection, design a set of equidistant signal sampling system to collect signal. Use the EMD method based on the criterion of continuous mean square error de-nosing. By integrating the signal get the space trajectory of vector. Experiments show that the track irregularities detection accuracy has been greatly improved.

Abstract: In this paper we present an improved dynamic spectrum allocation algorithm based on the intelligence of Q-learning. The state space, action space and reward function of the algorithm are built, and, the agents are guided to perform actions through designing the reward function. Numerical simulation results show that the proposed algorithm can improve system throughput efficiently compared to other algorithms. Facing the status of spectrum resources is tension and spectrum utilization is low, it can also boost the spectrum using condition in the future.

Abstract: the theory for maneuvering target tracking is significant to national defense and civil application. The filtering algorithm is one of important components in maneuvering target tracking. After the model of the maneuvering target is built, state vectors in the model are forecast and estimated through relevant filtering algorithms. The Unscented Kalman filtering is a novel filtering algorithm specially used for the nonlinear system, which is characterized by easy implementation, good generality, stable performance and so forth. Compared with the traditional Extended Kalman Filtering algorithm, the filtering algorithm can achieve less tracking error and higher tracking precision.

Abstract: The study proposed an improved NSCT fusion method based on the infrared and visible light images characteristics and fusion requirement. This paper improved the high-frequency coefficient and low-frequency coefficient fusion rules. The low-frequency sub-band images adopted the pixel feature energy weighted fusion rule. The high-frequency sub-band images adopted the neighborhood variance feature information fusion rule. The fusion experiment results show that this algorithm has good robustness. It could effectively extract edges and texture information. The fused images have abundance scene information and clear target. So this algorithm is an effective infrared and visible image fusion method.

Abstract: Based on the wavelet transform, this study introduced the theory of the compressed sensing algorithm. Then proposed a wavelet transform based compressed sensing algorithm by the better sparse representation ability of the wavelet transform on the image. Finally, this algorithm was compared with the DCT and wavelet transform algorithm. The experiment results show that the reconstructed image quality has a significant improvement. Especially, this algorithm has better effect on the images with rich curve.

Abstract: . Topic Model is one of the important subfields in Data Mining, which has been developed very quickly and has been applicated in many fields in recent years. Many researchers have been engaged in this field. In this paper, we introduce the BNB process based on Beta and Negative Binomial distribution, using the hierarchical distribution instead of Dirichlet in LDA. And we give the expression of parameter estimation used by Gibbs sampling. Then, BNB process is applicated in the text topic classification. We design experiments to decide the numbers of topics and compare the BNB process with LDA. Experiment results show that the BNB process has better performance over LDA in English Dataset, but they have almost the same result in Chinese micro-blog topic classification. Finally we analyze the problem and give the idea in further research.