Sparse Plane Wave Approximation of Acoustic Modes to Address Basis Mismatch

Low-frequency sound field reconstruction in an enclosed space has many applications where the plane wave approximation of acoustic modes plays a crucial role. However, the basis mismatch of the plane wave directions degrades the approximation accuracy. In this study, a two-stage method combining ℓ1-norm relaxation and parametric sparse Bayesian learning is proposed to address this problem. This method involves selecting sparse dominant plane wave directions from pre-discretized directions and constructing a parameterized dictionary of low dimensionality. This dictionary is used to re-estimate the plane wave complex amplitudes and directions based on the sparse Bayesian framework using the variational Bayesian expectation and maximization method. Numerical simulations show that the proposed method can efficiently optimize the plane wave directions to reduce the basis mismatch and improve acoustic mode approximation accuracy. The proposed method involves slightly increased computational cost but obtains a higher reconstruction accuracy at extrapolated field points and is more robust under low signal-to-noise ratios compared with conventional methods.

Noninteger Derivative Order Analysis on Plane Wave Reflection from Electro-Magneto-Thermo-Microstretch Medium with a Gravity Field within the Three-Phase Lag Model

The present research paper illustrates how noninteger derivative order analysis affects the reflection of partial thermal expansion waves under the generalized theory of plane harmonic wave reflection from a semivacuum elastic solid material with both gravity and magnetic field in the three-phase lag model (3PHL). The main goal for this study is investigating the fractional order impact and the applications related to the orders, especially in biology, medicine, and bioinformatics, besides the integer order considering an external effect, such as electromagnetic, gravity, and phase lags in a microstretch medium. The problem fractional form was formulated, and the boundary conditions were applied. The results were displayed graphically, considering the 3PHL model with magnetic field, gravity, and relaxation time. These findings were an explicit comparison of the effect of the plane wave reflection amplitude with integer derivative order analysis and noninteger derivative order analysis. The fractional order was compared to the correspondence integer order that indicated to the difference between them and agreement with the applications in biology, medicine, and other related topics. This phenomenon has more applications in relation to the biology and biomathematics problems.

Numerical analysis of wave propagating in a periodic layered structure

Controlling sound fields is a key technology for noise removal, acoustic lenses, energy harvesting, etc. This study investigated the control of sound field by a periodic layered structure. At first, we formulated the wave propagation in a periodic layered structure and proved that the wave fields constructed by the periodic boundary conditions are limited to plane wave modes with discretely different propagation directions. Numerical calculations clarified that the desired plane wave mode can be obtained in the transmitted wave through an intermediate thin-plate stacked region in a periodic layered structure, in which Lamb waves travel in each plate at different phase velocities and create phase difference at the exit of the intermediate thin-plate region. Further numerical investigations revealed that tuning frequency and length of the thin-plate region provides wave field more dominantly with a single wanted plane wave mode.

Wave-equation diffraction imaging using pseudo dip-angle gather

Diffraction images can directly indicate local heterogeneities such as faults, fracture zones, and erosional surfaces that are of high interest in seismic interpretation and unconventional reservoir development. We propose a new tool called pseudo dip-angle gather (PDAG) for imaging diffractors using the wave equation. PDAG has significantly lower computational cost compared with the classical dip-angle gather (DAG) due to using plane-wave gathers, a fast local Radon transform algorithm, and one-side decomposition assumption. Pseudo dip angle is measured from the vertical axis to the bisector of the plane-wave surface incident angle and scattered wave-propagation angle. PDAG is generated by choosing the zero lag of the correlation of the plane-wave source wavefields and the decomposed receiver wavefields. It reveals similar diffraction and reflection patterns to DAG, i.e. diffractions spreading as a flat event and reflections focused at a spectacular angle, while they may have dissimilar coverage for diffraction and different focused locations for reflection compared with that of DAG. A windowed median filter is then applied to each PDAG for extracting the diffraction energy and suppressing the focused reflection energy. Besides, the stacked PDAG can be used to evaluate the migration accuracy by measuring the flatness of the image gathers. Numerical tests on both synthetic and field data sets demonstrate that our method can efficiently produce accurate results for diffraction images.

Accurate Direction-of-Arrival Estimation Method based on Space-Time Modulated Metasurface

In this paper, we present a metasurface-based Direction of Arrival (DoA) estimation method that exploits the properties of space-time modulated reflecting metasurfaces to estimate in real-time the impinging angle of an illuminating monochromatic plane wave. The approach makes use of the amplitude unbalance of the received fields at broadside at the frequencies of the two first-order harmonics generated by the interaction between the incident plane wave and the modulated metasurface. Here, we first describe analytically how to generate the desired higher-order harmonics in the reflected spectrum and how to realize the breaking of the spatial symmetry of each order harmonic scattering pattern. Then, the one dimensional (1D) omnidirectional incident angle can be analytically computed using +1st and -1st order harmonics. The approach is also extended to 2D DoA estimation by using two orthogonally arranged 1D DoA modulation arrays. The accuracy of 1D DoA estimation is verified through full-wave numerical simulations. Compared to conventional DoA estimation methods, the proposed approach simplifies the computation and hardware complexity, ensuring at the same time estimation accuracy. The proposed method may have potential applications in wireless communications, target recognition, and identification.

Deep learning for enhancing multi-source reverse time migration

<p>Reverse time migration (RTM) is a technique used to obtain high-resolution images of underground reflectors; however, this method is computationally intensive when dealing with large amounts of seismic data. Multi-source RTM can significantly reduce the computational cost by processing multiple shots simultaneously. However, multi-source-based methods frequently result in crosstalk artifacts in the migrated images, causing serious interference in the imaging signals. Plane-wave migration, as a mainstream multi-source method, can yield migrated images with plane waves in different angles by implementing phase encoding of the source and receiver wavefields; however, this method frequently requires a trade-off between computational efficiency and imaging quality. We propose a method based on deep learning for removing crosstalk artifacts and enhancing the image quality of plane-wave migration images. We designed a convolutional neural network that accepts an input of seven plane-wave images at different angles and outputs a clear and enhanced image. We built 505 1024×256 velocity models, and employed each of them using plane-wave migration to produce raw images at 0°, ±20°, ±40°, and ±60° as input of the network. Labels are high-resolution images computed from the corresponding reflectivity models by convolving with a Ricker wavelet. Random sub-images with a size of 512×128 were used for training the network. Numerical examples demonstrated the effectiveness of the trained network in crosstalk removal and imaging enhancement. The proposed method is superior to both the conventional RTM and plane-wave RTM (PWRTM) in imaging resolution. Moreover, the proposed method requires only seven migrations, significantly improving the computational efficiency. In the numerical examples, the processing time required by our method was approximately 1.6% and 10% of that required by RTM and PWRTM, respectively.</p>

Deep learning for enhancing multi-source reverse time migration

<p>Reverse time migration (RTM) is a technique used to obtain high-resolution images of underground reflectors; however, this method is computationally intensive when dealing with large amounts of seismic data. Multi-source RTM can significantly reduce the computational cost by processing multiple shots simultaneously. However, multi-source-based methods frequently result in crosstalk artifacts in the migrated images, causing serious interference in the imaging signals. Plane-wave migration, as a mainstream multi-source method, can yield migrated images with plane waves in different angles by implementing phase encoding of the source and receiver wavefields; however, this method frequently requires a trade-off between computational efficiency and imaging quality. We propose a method based on deep learning for removing crosstalk artifacts and enhancing the image quality of plane-wave migration images. We designed a convolutional neural network that accepts an input of seven plane-wave images at different angles and outputs a clear and enhanced image. We built 505 1024×256 velocity models, and employed each of them using plane-wave migration to produce raw images at 0°, ±20°, ±40°, and ±60° as input of the network. Labels are high-resolution images computed from the corresponding reflectivity models by convolving with a Ricker wavelet. Random sub-images with a size of 512×128 were used for training the network. Numerical examples demonstrated the effectiveness of the trained network in crosstalk removal and imaging enhancement. The proposed method is superior to both the conventional RTM and plane-wave RTM (PWRTM) in imaging resolution. Moreover, the proposed method requires only seven migrations, significantly improving the computational efficiency. In the numerical examples, the processing time required by our method was approximately 1.6% and 10% of that required by RTM and PWRTM, respectively.</p>