No abstract

Combining the widely convergent homotopy method applied to the inversion process of operator identification with the Tikhonov regularization method for ill-posed problem, a new Widely Convergent Generalized Pulse-Spectrum Technique (WCGPST) for 2-D wave equation inversion is constructed. And constraint inversion is preformed with data of well logs. Many numerical simulations and tests of anti-noise indicate that this method is effective.

No abstract

We investigate the problem of estimating the velocity in a two-dimensional acoustic wave equation, which plays an important role in geological survey. The forward problem is discretized using finite-difference methods and the estimation is formulated as a least-square minimization problem with a regularization term. To reduce the computational burden, a nonlinear multigrid method is applied to solve this inverse problem. In the multigrid inversion process, in order to make the objective functionals at different scales compatible, they are dynamically adjusted. In this way, the necessary condition of "the optimal solution should be the fixed point of multigrid inversion" can be met. The stable and fast regularized Gauss-Newton method is applied to each grid. The results of numerical simulations indicate that the proposed method can effectively reduce the required computation, improve the inversion results, and have the anti-noise ability.

With the extensive application of artificial intelligence technology in 5G and Beyond Fifth Generation (B5G) networks, it has become a common trend for artificial intelligence to integrate into modern communication networks. Deep learning is a subset of machine learning and has recently led to significant improvements in many fields. In particular, many 5G-based services use deep learning technology to provide better services. Although deep learning is powerful, it is still vulnerable when faced with 5G-based deep learning services. Because of the nonlinearity of deep learning algorithms, slight perturbation input by the attacker will result in big changes in the output. Although many researchers have proposed methods against adversarial attacks, these methods are not always effective against powerful attacks such as CW. In this paper, we propose a new two-stream network which includes RGB stream and spatial rich model (SRM) noise stream to discover the difference between adversarial examples and clean examples. The RGB stream uses raw data to capture subtle differences in adversarial samples. The SRM noise stream uses the SRM filters to get noise features. We regard the noise features as additional evidence for adversarial detection. Then, we adopt bilinear pooling to fuse the RGB features and the SRM features. Finally, the final features are input into the decision network to decide whether the image is adversarial or not. Experimental results show that our proposed method can accurately detect adversarial examples. Even with powerful attacks, we can still achieve a detection rate of 91.3%. Moreover, our method has good transferability to generalize to other adversaries.