A novel method based on the relevance vector machine（RVM） for the inverse scattering problem is presented in this paper.The nonlinearity and the ill-posedness inherent in this problem are simultaneously considered.The nonlinearity is embodied in the relation between the scattered field and the target property,which can be obtained through the RVM training process.Besides,rather than utilizing regularization,the ill-posed nature of the inversion is naturally accounted for because the RVM can produce a probabilistic output.Simulation results reveal that the proposed RVM-based approach can provide comparative performances in terms of accuracy,convergence,robustness,generalization,and improved performance in terms of sparse property in comparison with the support vector machine（SVM） based approach.
A parabolic equation method （PEM）-based discrete algorithm is proposed and is used to obtain the field distribution in the evaporation duct space. This method not only improves the computing speed, but also provides the flexibility to adjust the simulation accuracy. Numerical simulation of the wave propagation in the oceanic waveguide structure is done. In addition, the initial field distribution and progressive steps are determined. The loss model in the waveguide is solved through the numerical solution. By comparing the characteristics of the radio wave propagation in the duct and in the normal atmospheric structure, we analyses the radio transmission over the horizon detection in the oceanic waveguide.
