Search

In underwater acoustic target recognition, the target signal is usually complex and the samples which are difficult to obtain also have some uncertain information. In order to effectively solve these problems, the evidence clustering recognition algorithm (TECRA) is presented. In this new method, the k-nearest neighbor are first determined by using the feature distance between the object and its neighbors in each class of the training set, and a reasonable initial basic belief assignments (bba's) for each target data are constructed by the improved k-nearest neighbor classification algorithm. Then the final global bba's of the target is obtained by optimizing the objective function of the algorithm. Finally the object can be recognized by the fusion result and the classification rule presented in the paper. Several experiments based on real underwater acoustic data sets are made to test the effectiveness of TECRA in comparison with some other methods. The results indicate that TECRA can effectively improve the recognition accuracy.

Due to the multipath effect, the direction of arrival (DOA) estimation performance for low-angle targets decreases greatly. To overcome this problem, this paper proposes a spatial differencing reconstruction based DOA estimation algorithm by using the received signal model of multiple input multiple output (MIMO) radar. Combining with the spatial diversity of MIMO radar, the proposed method can first use the multipath echo power to select the best signal. Then, a spatial differencing based iterative scheme is developed to reduce the effect of additive noise, resulting in a better estimation performance for low-angle targets. Simulation results show that the proposed method has better advantages in suppressing noise and improving estimation accuracy.

To deal with the problem of the difficult optimization search and expensive computational cost caused by large-scale design variables, the hierarchical optimization design system based on the global sensitivity analysis method is established in this paper. The M-OAT method is used to analyze the global sensitivity of the design variables, according to the sensitivity information to layer design variables, then optimize the design variables in each hierarchy. Through the study of the hierarchical optimization design of airfoils and wings, compared with the normal parameter optimization design system, the hierarchical optimization design system based on the global sensitivity analysis method can reduce effectively the number of design variables in a single optimization, reduce the difficulty of the optimization search, improve the convergence speed of the optimization, gain better optimization results at the same time. For optimization design with large-scale design variables, the hierarchical optimization design system based on the global sensitivity analysis method is a sort of effective ways of design.