Searcharxiv⌕ Search

arXiv subjects

Weiqiang Zhu

Publications and source records attributed to Weiqiang Zhu.

At least 19 recordsLinked to original sources
arXiv2026

Communication-efficient Coordinated RSS-based Distributed Passive Localization via Drone Cluster

Recently, passive unmanned aerial vehicle (UAV) localization has become popular due to mobility and convenience. In this paper, we consider a scenario of using distributed drone cluster to estimate the position of a passive emitter via received signal strength (RSS). First, a distributed majorizeminimization (DMM) RSS-based localization method is proposed. To accelerate its convergence, a tight upper bound of the objective function from the primary one is derived. Furthermore, to reduce communication overhead, a distributed estimation scheme using the Fisher information matrix (DEF) is presented, with only requiring one-round communication between edge UAVs and center UAV. Additionally, a local search solution is used as the initial value of DEF. Simulation results show that the proposed DMM performs better than the existing distributed Gauss-Newton method (DGN) in terms of root of mean square error (RMSE) under a limited low communication overhead constraint. Moreover, the proposed DEF performs much better than MM in terms of RMSE, but has a higher computational complexity than the latter.

eess.SP↗
arXiv2026

An ADMM-Based Geometric Configuration Optimization in RSSD-Based Source Localization By UAVs with Spread Angle Constraint

Deploying multiple unmanned aerial vehicles (UAVs) to locate a signal-emitting source covers a wide range of military and civilian applications like rescue and target tracking. It is well known that the UAVs-source (sensors-target) geometry, namely geometric configuration, significantly affects the final localization accuracy. This paper focuses on the geometric configuration optimization for received signal strength difference (RSSD)-based passive source localization by drone swarm. Different from prior works, this paper considers a general measuring condition where the spread angle of drone swarm centered on the source is constrained. Subject to this constraint, a geometric configuration optimization problem with the aim of maximizing the determinant of Fisher information matrix (FIM) is formulated. After transforming this problem using matrix theory, an alternating direction method of multipliers (ADMM)-based optimization framework is proposed. To solve the subproblems in this framework, two global optimal solutions based on the Von Neumann matrix trace inequality theorem and majorize-minimize (MM) algorithm are proposed respectively. Finally, the effectiveness as well as the practicality of the proposed ADMM-based optimization algorithm are demonstrated by extensive simulations.

eess.SP↗
arXiv2026

Earthquake Aftershock Forecasting using Conditional Generative Models

Forecasting how aftershocks evolve in space and time after a large earthquake is a central problem in statistical seismology and underpins operational earthquake forecasting. Existing forecasting methods rest on statistical point-process models such as the epidemic-type aftershock sequence (ETAS) and Reasenberg-Jones models, which prescribe a fixed decay in time and an isotropic kernel in space. They match the average Omori-Utsu and Gutenberg-Richter statistics well but do not capture the fault-controlled spatial patterns of real sequences or the productivity that varies among sequences. Neural point-process models relax these fixed forms but keep the event-by-event view and have not consistently surpassed ETAS on common benchmarks. Rather than modeling individual events as a point process, we recast aftershock forecasting as conditional generation of a spatiotemporal field. We develop QuakeGen, a diffusion model that generates the evolving fields of aftershock rate and maximum magnitude conditioned on recent seismicity and the forecasting horizon. The same conditional generative framework can be trained on rich seismic catalogs to forecast global aftershock sequences and regional daily seismicity, and could further condition on physical fields such as fault geometry or geodetic deformation. On global sequences, the data-driven approach outperforms the operational USGS Reasenberg-Jones forecast, recovering the fault-controlled, anisotropic spatial structure that fixed kernels cannot express. On daily forecasting, QuakeGen also matches the well-tuned ETAS baselines, which neural point-process models have yet to surpass on the regional benchmark. Conditional generative modeling, which has transformed prediction in fields as diverse as weather forecasting and protein structure prediction, holds the potential to forecast more accurately how earthquake sequences unfold in space and time.

physics.geo-ph↗
arXiv2026

VORA: Rapid Association of Earthquake Phases from Local to Global

Earthquake phase association, which groups seismic phase arrivals into common origins, is a key step towards more complete and reliable seismicity catalogs. It has become a challenging task because of the massive phase datasets produced from dense seismic networks and advanced phase picking methods. Here we present VORA (Voronoi tessellation- and Origin-time-based Rapid Associator), an efficient and scalable earthquake phase associator that treats association as an unsupervised spatio-temporal clustering problem. Specifically, VORA depends on two primary constraints: the estimated earthquake origin time (temporal) and seismic station adjacency (spatial). Recent advances in deep learning models have enabled detection of S- and P-phases with similar effectiveness, making it straightforward to estimate corresponding earthquake origin times from candidate phase pairs given a prescribed range of velocity ratios or directly from raw waveforms. We then leverage the Voronoi diagram to define each station's neighbors, cluster the origin times across neighboring stations, and apply an optional sub-clustering step to separate overlapping events. Benchmarks on synthetic and real datasets show that VORA achieves the fastest runtime and maintains robust performance (high recall and precision) even under intense seismicity. Applying it to a two-week global-scale pick dataset covering the 2019 Ridgecrest earthquake sequence further demonstrates that the same framework scales from local networks to a global station set. VORA requires no training and generalizes across local-to-global regions, varying velocity models, and evolving network geometries, helping to meet the growing demands of expanding seismic networks and the increasing volume of automated phase picks.

physics.geo-ph↗
arXiv2026

Long-term marine acoustic and seismic monitoring using distributed acoustic sensing and deep learning

The ocean remains one of the least instrumented parts of Earth, and many geophysical, biological, and anthropogenic signals go undetected for lack of instrumentation. Distributed acoustic sensing (DAS) can transform submarine fiber-optic cables into dense seafloor sensor arrays, but extracting diverse signals from massive DAS recordings remains challenging. Here we present DASNet, a deep learning framework that detects, classifies, and picks arrival times of diverse marine signals in continuous DAS data. Applied to nearly four years of Seafloor Fiber-Optic Array in Monterey Bay recordings, DASNet identifies more than 620,000 events. These detections reveal local earthquakes; distant earthquake- and volcanic-eruption-generated T-waves from the southwestern Pacific and mid-ocean ridge systems; more than 510,000 blue and fin whale calls with seasonal and interannual variability consistent with hydrophone records; and vessel traffic near the cable. Together, these results show that submarine fiber-optic cables combined with deep learning enable scalable, high-resolution ocean monitoring.

physics.geo-ph↗
arXiv2025

Joint inversion for Vp, Vp/Vs of the San Fransico Bay Area using ADTomo

This article presents a new seismological tomography method based on the fast sweeping method and advanced seismic phase picking techniques to study the complex geological structures of the San Francisco Bay Area. By calculating the eikonal equation using the fast-sweeping method, this study obtains travel time information and gradient data under a given velocity structure. With an automatic differentiation algorithm to calculate gradients of the loss function and the L-BFGS algorithm to achieve optimization, the velocity model is iteratively adjusted to minimize the loss function. The P wave to S wave velocity ratio obtained through joint inversion is more reliable than the velocity ratio obtained directly by dividing the P wave and S wave velocity models. Compared to traditional inversion, the velocity ratio here does not require the same P wave and S wave travel path, thus improving the accuracy of the velocity ratio. The method was applied to the San Francisco Bay Area, a region with complex geological structures and significant seismic activities. This region, a focal point of research interest, also provides abundant seismic data for this study. We use deep learning methods to automatically pick seismic phases, acquiring abundant P wave and S wave travel time information. This project obtains the S wave velocity structure and P wave to S wave velocity ratio results for the first time. Compared to previous inversion results, the P wave velocity model here demonstrates higher resolution and better geological correspondence. High and low velocity anomalies also align well with geological maps, providing reasonable explanations in terms of lithology.

physics.geo-ph↗
arXiv2025

Towards End-to-End Earthquake Monitoring Using a Multitask Deep Learning Model

Seismic waveforms contain rich information about earthquake processes, making effective data analysis crucial for earthquake monitoring, source characterization, and seismic hazard assessment. With rapid developments in deep learning, the state-of-the-art approach in artificial intelligence, many neural network models have been developed to enhance earthquake monitoring tasks, such as earthquake detection, phase picking, and phase association. However, most current efforts focus on developing separate models for each specific task, leaving the potential of an end-to-end framework relatively unexplored. To address this gap, we extend an existing phase picking model, PhaseNet, to create a multitask framework. This extended model, PhaseNet+, simultaneously performs phase arrival-time picking, first-motion polarity determination, and phase association. The outputs from these perception-based models can then be processed by specialized physics-based algorithms to accurately determine earthquake location and focal mechanism. The multitask approach is not limited to the PhaseNet model and can be applied to other state-of-the-art phase picking models, ultimately improving seismic monitoring through a more unified and efficient approach.

physics.geo-ph↗
arXiv2025

California Earthquake Dataset for Machine Learning and Cloud Computing

The San Andreas Fault system, known for its frequent seismic activity, provides an extensive dataset for earthquake studies. The region's well-instrumented seismic networks have been crucial in advancing research on earthquake statistics, physics, and subsurface Earth structures. In recent years, earthquake data from California has become increasingly valuable for deep learning applications, such as Generalized Phase Detection (GPD) for phase detection and polarity determination, and PhaseNet for phase arrival-time picking. The continuous accumulation of data, particularly those manually labeled by human analysts, serves as an essential resource for advancing both regional and global deep learning models. To support the continued development of machine learning and data mining studies, we have compiled a unified California Earthquake Event Dataset (CEED) that integrates seismic records from the Northern California Earthquake Data Center (NCEDC) and the Southern California Earthquake Data Center (SCEDC). The dataset includes both automatically and manually determined parameters such as earthquake origin time, source location, P/S phase arrivals, first-motion polarities, and ground motion intensity measurements. The dataset is organized in an event-based format organized by year spanning from 2000 to 2024, facilitating cross-referencing with event catalogs and enabling continuous updates in future years. This comprehensive open-access dataset is designed to support diverse applications including developing deep learning models, creating enhanced catalog products, and research into earthquake processes, fault zone structures, and seismic risks.

physics.geo-ph↗
arXiv2025

Robust Earthquake Location using Random Sample Consensus (RANSAC)

Accurate earthquake location, which determines the origin time and location of seismic events using phase arrival times or waveforms, is fundamental to earthquake monitoring. While recent deep learning advances have significantly improved earthquake detection and phase picking, particularly for smaller-magnitude events, the increased detection rate introduces new challenges for robust location determination. These smaller events often contain fewer P- and S-phase picks, making location accuracy more vulnerable to false or inaccurate picks. To enhance location robustness against outlier picks, we propose a machine learning method that incorporates the Random Sample Consensus (RANSAC) algorithm. RANSAC employs iterative sampling to achieve robust parameter optimization in the presence of substantial outliers. By integrating RANSAC's iterative sampling into traditional earthquake location workflows, we effectively mitigate biases from false picks and improve the robustness of the location process. We evaluated our approach using both synthetic data and real data from the Ridgecrest earthquake sequence. The results demonstrate comparable accuracy to traditional location algorithms while showing enhanced robustness to outlier picks.

physics.geo-ph↗
arXiv2024

QuakeFormer: A Uniform Approach to Earthquake Ground Motion Prediction Using Masked Transformers

Ground motion prediction (GMP) models are critical for hazard reduction before, during and after destructive earthquakes. In these three stages, intensity forecasting, early warning and interpolation models are corresponding employed to assess the risk. Considering the high cost in numerical methods and the oversimplification in statistical methods, deep-learning-based approaches aim to provide accurate and near-real-time ground motion prediction. Current approaches are limited by specialized architectures, overlooking the interconnection among these three tasks. What's more, the inadequate modeling of absolute and relative spatial dependencies mischaracterizes epistemic uncertainty into aleatory variability. Here we introduce QuakeFormer, a unified deep learning architecture that combines these three tasks in one framework. We design a multi-station-based Transformer architecture and a flexible masking strategy for training QuakeFormer. This data-driven approach enables the model to learn spatial ground motion dependencies directly from real seismic recordings, incorporating location embeddings that include both absolute and relative spatial coordinates. The results indicate that our model outperforms state-of-the-art ground motion prediction models across all three tasks in our research areas. We also find that pretraining a uniform forecasting and interpolation model enhances the performance on early warning task. QuakeFormer offers a flexible approach to directly learning and modeling ground motion, providing valuable insights and applications for both earthquake science and engineering.

physics.geo-ph↗
arXiv2023

Phase Neural Operator for Multi-Station Picking of Seismic Arrivals

Seismic wave arrival time measurements form the basis for numerous downstream applications. State-of-the-art approaches for phase picking use deep neural networks to annotate seismograms at each station independently, yet human experts annotate seismic data by examining the whole network jointly. Here, we introduce a general-purpose network-wide phase picking algorithm based on a recently developed machine learning paradigm called Neural Operator. Our model, called PhaseNO, leverages the spatio-temporal contextual information to pick phases simultaneously for any seismic network geometry. This results in superior performance over leading baseline algorithms by detecting many more earthquakes, picking more phase arrivals, while also greatly improving measurement accuracy. Following similar trends being seen across the domains of artificial intelligence, our approach provides but a glimpse of the potential gains from fully-utilizing the massive seismic datasets being collected worldwide.

physics.geo-ph↗
arXiv2023

Seismic Arrival-time Picking on Distributed Acoustic Sensing Data using Semi-supervised Learning

Distributed Acoustic Sensing (DAS) is an emerging technology for earthquake monitoring and subsurface imaging. The recorded seismic signals by DAS have several distinct characteristics, such as unknown coupling effects, strong anthropogenic noise, and ultra-dense spatial sampling. These aspects differ from conventional seismic data recorded by seismic networks, making it challenging to utilize DAS at present for seismic monitoring. New data analysis algorithms are needed to extract useful information from DAS data. Previous studies on conventional seismic data demonstrated that deep learning models could achieve performance close to human analysts in picking seismic phases. However, phase picking on DAS data is still a difficult problem due to the lack of manual labels. Further, the differences in mathematical structure between these two data formats, i.e., ultra-dense DAS arrays and sparse seismic networks, make model fine-tuning or transfer learning difficult to implement on DAS data. In this work, we design a new approach using semi-supervised learning to solve the phase-picking task on DAS arrays. We use a pre-trained PhaseNet model as a teacher network to generate noisy labels of P and S arrivals on DAS data and apply the Gaussian mixture model phase association (GaMMA) method to refine these noisy labels to build training datasets. We develop a new deep learning model, PhaseNet-DAS, to process the 2D spatial-temporal data of DAS arrays and train the model on DAS data. The new deep learning model achieves high picking accuracy and good earthquake detection performance. We then apply the model to process continuous data and build earthquake catalogs directly from DAS recording. Our approach using semi-supervised learning provides a way to build effective deep learning models for DAS, which have the potential to improve earthquake monitoring using large-scale fiber networks.

physics.geo-ph↗
arXiv2023

Machine Learning Methods for Inferring the Number of UAV Emitters via Massive MIMO Receive Array

To provide important prior knowledge for the DOA estimation of UAV emitters in future wireless networks, we present a complete DOA preprocessing system for inferring the number of emitters via massive MIMO receive array. Firstly, in order to eliminate the noise signals, two high-precision signal detectors, square root of maximum eigenvalue times minimum eigenvalue (SR-MME) and geometric mean (GM), are proposed. Compared to other detectors, SR-MME and GM can achieve a high detection probability while maintaining extremely low false alarm probability. Secondly, if the existence of emitters is determined by detectors, we need to further confirm their number. Therefore, we perform feature extraction on the the eigenvalue sequence of sample covariance matrix to construct feature vector and innovatively propose a multi-layer neural network (ML-NN). Additionally, the support vector machine (SVM), and naive Bayesian classifier (NBC) are also designed. The simulation results show that the machine learning-based methods can achieve good results in signal classification, especially neural networks, which can always maintain the classification accuracy above 70\% with massive MIMO receive array. Finally, we analyze the classical signal classification methods, Akaike (AIC) and Minimum description length (MDL). It is concluded that the two methods are not suitable for scenarios with massive MIMO arrays, and they also have much worse performance than machine learning-based classifiers.

eess.SP↗
arXiv2023

Neural mixture model association of seismic phases

Seismic phase association is the task of grouping phase arrival picks across a seismic network into subsets with common origins. Building on recent successes in this area with machine learning tools, we introduce a neural mixture model association algorithm (Neuma), which incorporates physics-informed neural networks and mixture models to address this challenging problem. Our formulation assumes explicitly that a dataset contains real phase picks from earthquakes and noise picks resulting from phase picking mistakes and fake picks. The problem statement is then to assign each observation to either an earthquake or noise. We iteratively update a set of hypocenters and magnitudes while determining the optimal class assignment for each pick. We show that by using a physics-informed Eikonal solver as the forward model, we can impose stringent quality control on surviving picks while maintaining high recall. We evaluate the performance of Neuma against several baseline algorithms on a series of challenging synthetic datasets and the 2019 Ridgecrest, California sequence. Neuma outperforms the baselines in precision and recall for each of the synthetic datasets. Furthermore, it detects an additional 3285 more earthquakes than the best baseline on the Ridgecrest dataset (13.5%), while substantially improving the hypocenters.

physics.geo-ph↗
arXiv2022

QuakeFlow: A Scalable Machine-learning-based Earthquake Monitoring Workflow with Cloud Computing

Earthquake monitoring workflows are designed to detect earthquake signals and to determine source characteristics from continuous waveform data. Recent developments in deep learning seismology have been used to improve tasks within earthquake monitoring workflows that allow the fast and accurate detection of up to orders of magnitude more small events than are present in conventional catalogs. To facilitate the application of machine-learning algorithms to large-volume seismic records, we developed a cloud-based earthquake monitoring workflow, QuakeFlow, that applies multiple processing steps to generate earthquake catalogs from raw seismic data. QuakeFlow uses a deep learning model, PhaseNet, for picking P/S phases and a machine learning model, GaMMA, for phase association with approximate earthquake location and magnitude. Each component in QuakeFlow is containerized, allowing straightforward updates to the pipeline with new deep learning/machine learning models, as well as the ability to add new components, such as earthquake relocation algorithms. We built QuakeFlow in Kubernetes to make it auto-scale for large datasets and to make it easy to deploy on cloud platforms, which enables large-scale parallel processing. We used QuakeFlow to process three years of continuous archived data from Puerto Rico, and found more than a factor of ten more events that occurred on much the same structures as previously known seismicity. We applied Quakeflow to monitoring frequent earthquakes in Hawaii and found over an order of magnitude more events than are in the standard catalog, including many events that illuminate the deep structure of the magmatic system. We also added Kafka and Spark streaming to deliver real-time earthquake monitoring results. QuakeFlow is an effective and efficient approach both for improving realtime earthquake monitoring and for mining archived seismic data sets.

physics.geo-ph↗
arXiv2021

Integrating Deep Neural Networks with Full-waveform Inversion: Reparametrization, Regularization, and Uncertainty Quantification

Full-waveform inversion (FWI) is an accurate imaging approach for modeling velocity structure by minimizing the misfit between recorded and predicted seismic waveforms. However, the strong non-linearity of FWI resulting from fitting oscillatory waveforms can trap the optimization in local minima. We propose a neural-network-based full waveform inversion method (NNFWI) that integrates deep neural networks with FWI by representing the velocity model with a generative neural network. Neural networks can naturally introduce spatial correlations as regularization to the generated velocity model, which suppresses noise in the gradients and mitigates local minima. The velocity model generated by neural networks is input to the same partial differential equation (PDE) solvers used in conventional FWI. The gradients of both the neural networks and PDEs are calculated using automatic differentiation, which back-propagates gradients through the acoustic PDEs and neural network layers to update the weights of the generative neural network. Experiments on 1D velocity models, the Marmousi model, and the 2004 BP model demonstrate that NNFWI can mitigate local minima, especially for imaging high-contrast features like salt bodies, and significantly improves the inversion in the presence of noise. Adding dropout layers to the neural network model also allows analyzing the uncertainty of the inversion results through Monte Carlo dropout. NNFWI opens a new pathway to combine deep learning and FWI for exploiting both the characteristics of deep neural networks and the high accuracy of PDE solvers. Because NNFWI does not require extra training data and optimization loops, it provides an attractive and straightforward alternative to conventional FWI.

physics.geo-ph↗
arXiv2021

An End-to-End Earthquake Detection Method for Joint Phase Picking and Association using Deep Learning

Earthquake monitoring by seismic networks typically involves a workflow consisting of phase detection/picking, association, and location tasks. In recent years, the accuracy of these individual stages has been improved through the use of machine learning techniques. In this study, we introduce a new, end-to-end approach that improves overall earthquake detection accuracy by jointly optimizing each stage of the detection pipeline. We propose a neural network architecture for the task of multi-station processing of seismic waveforms recorded over a seismic network. This end-to-end architecture consists of three sub-networks: a backbone network that extracts features from raw waveforms, a phase picking sub-network that picks P- and S-wave arrivals based on these features, and an event detection sub-network that aggregates the features from multiple stations and detects earthquakes. We use these sub-networks in conjunction with a shift-and-stack module based on back-projection that introduces kinematic constraints on arrival times, allowing the model to generalize to different velocity models and to variable station geometry in seismic networks. We evaluate our proposed method on the STanford EArthquake Dataset (STEAD) and on the 2019 Ridgecrest, CA earthquake sequence. The results demonstrate that our end-to-end approach can effectively pick P- and S-wave arrivals and achieve earthquake detection accuracy rivaling that of other state-of-the-art approaches.

physics.geo-ph↗
arXiv2021

Earthquake Phase Association using a Bayesian Gaussian Mixture Model

Earthquake phase association algorithms aggregate picked seismic phases from a network of seismometers into individual earthquakes and play an important role in earthquake monitoring. Dense seismic networks and improved phase picking methods produce massive earthquake phase data sets, particularly for earthquake swarms and aftershocks occurring closely in time and space, making phase association a challenging problem. We present a new association method, the Gaussian Mixture Model Association (GaMMA), that combines the Gaussian mixture model for phase measurements (both time and amplitude), with earthquake location, origin time, and magnitude estimation. We treat earthquake phase association as an unsupervised clustering problem in a probabilistic framework, where each earthquake corresponds to a cluster of P and S phases with hyperbolic moveout of arrival times and a decay of amplitude with distance. We use a multivariate Gaussian distribution to model the collection of phase picks for an event, the mean of which is given by the predicted arrival time and amplitude from the causative event. We carry out the pick assignment for each earthquake and determine earthquake parameters (i.e., earthquake location, origin time, and magnitude) under the maximum likelihood criterion using the Expectation-Maximization (EM) algorithm. The GaMMA method does not require the typical association steps of other algorithms, such as grid-search or supervised training. The results on both synthetic test and the 2019 Ridgecrest earthquake sequence show that GaMMA effectively associates phases from a temporally and spatially dense earthquake sequence while producing useful estimates of earthquake location and magnitude.

physics.geo-ph↗