Search for: All records

Shear wave splitting of teleseismic core phases such as SKS is commonly used to constrain mantle seismic anisotropy, a proxy for convective deformation. In plate boundaries, sharp lateral variations of splitting measurements near transform faults are often linked to deformation within a lithospheric shear zone below, but potential seismic waveform effects from heterogeneous structure on small scales may influence the interpretation. Here, we explore possible finite frequency effects on shear wave splitting near fault zones in a fully three‐dimensional anisotropic setting. We find that shear zones wider than 80 km, a scale set by the Fresnel zone, can be clearly detected, but narrower zones are less distinguishable. Near the edge of the shear zone, the combined effect of anisotropy and scattering generates false splitting measurements with large delay times and fast axis orientation approaching the back‐azimuth, a bias which can only be identified when records from different back‐azimuths are analyzed together. This substantiates that back‐azimuthal variations of splitting can arise not just from vertical layering but also lateral changes of anisotropic media. We also test the effects of shear zone edge geometry, epicentral distance, filtering frequency, crustal thickness, and sediment cover. Our study delineates the ability of shear wave splitting to resolve and investigate fault zones, and emphasizes the importance of good azimuthal coverage to correctly interpret observed anisotropy. Based on revisiting previous shear wave splitting and lithospheric deformation studies, we infer that many crustal fault zones are underlain by lithospheric shear zones at least 20 km wide. 

We present Q-functionals, an alternative architecture for continuous control deep reinforcement learning. Instead of returning a single value for a state-action pair, our network transforms a state into a function that can be rapidly evaluated in parallel for many actions, allowing us to efficiently choose high-value actions through sampling. This contrasts with the typical architecture of off-policy continuous control, where a policy network is trained for the sole purpose of selecting actions from the Q-function. We represent our action-dependent Q-function as a weighted sum of basis functions (Fourier, Polynomial, etc) over the action space, where the weights are state-dependent and output by the Q-functional network. Fast sampling makes practical a variety of techniques that require Monte-Carlo integration over Q-functions, and enables action-selection strategies besides simple value-maximization. We characterize our framework, describe various implementations of Q-functionals, and demonstrate strong performance on a suite of continuous control tasks. 

The highly sparse nature of propagation channels and the restricted use of radio frequency (RF) chains at transceivers limit the performance of millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems. Introducing reconfigurable antennas to mmWave can offer an additional degree of freedom on designing mmWave MIMO systems. This paper provides a theoretical framework for studying the mmWave MIMO with reconfigurable antennas. We present an architecture of reconfigurable mmWave MIMO with beamspace hybrid analog-digital beamformers and reconfigurable antennas at both the transmitter and the receiver. We show that employing reconfigurable antennas can provide throughput gain for the mmWave MIMO. We derive the expression for the average throughput gain of using reconfigurable antennas, and further simplify the expression by considering the case of large number of reconfiguration states. In addition, we propose a low-complexity algorithm for the reconfiguration state and beam selection, which achieves nearly the same throughput performance as the optimal selection of reconfiguration state and beams by exhaustive search.