Dispersive processing of borehole acoustic logging data is essential for accurate evaluation of subsurface formation velocities, which directly impact reservoir characterization and geophysical interpretation. Traditional coherence-based techniques fail to address the dispersive nature of guided waves, leading to underestimated velocities, while model-based inversion requires strict parameter assumptions that limit applicability in complex formations. To overcome these challenges, we propose a physics-informed multi-task deep learning method that integrates an analytical exponential function, Gaussian noise modeling, and a hybrid CNN–BiLSTM architecture. Large-scale synthetic datasets are generated separately for fast and and slow formation scenarios, enabling the network to capture both local dispersion curve details and long-range sequential dependencies. A composite physics-informed loss function jointly constrains parameter accuracy and dispersion curve consistency, ensuring physically meaningful predictions. Numerical experiments on simulated dipole waveforms demonstrate that the proposed method can accurately reconstruct clean dispersion curves and invert reliable formation shear velocities. Furthermore, field tests on LWD quadrupole wave data validate its robustness and generalization capability, confirming its practical potential for borehole acoustic logging applications.

With the wide applications of the Internet of Things (IoT) technologies in intelligent manufacturing, production safety, smart citie, and other fields, target location awareness has been the primary issue as it can be used to personnel and material positioning, electronic fence settings, daily attendance statistics, and so on. Wireless sensor networks (WSNs) positioning can become an indispensable part of these location-based IoT applications that benefit from the ubiquitous sensing and communication ability. For addressing the large positioning errors caused by uncertain WSNs, this article proposes an improved particle filter-based hybrid optimization (IPFHO) algorithm. After mapping the noisy wireless signal to uncertain target location, the preliminary positioning of mobile target is realized by improved particle filter, whose coarse accuracy over time can be further optimized with use of iterative search method. We evaluate the proposed algorithm under different noise levels, process noise variances and computation times in extensive simulations. The results indicate that the positioning accuracy of proposed algorithm can be effectively improved in the presence of wireless ranging errors and anchor node calibration errors. Compared with the positioning error 0.27 m estimated by pure filtering, the positioning error of proposed algorithm can be reduced to 0.19 m by combining the filtering and search methods. The platform experimental results, which are consistent with the trend of the simulation results, validate the superior accuracy of proposed algorithm compared with relevant positioning algorithms.

Underwater nodes are prone to drift under the water flow action, which makes the topological structure of underwater wireless sensor networks (UWSNs) have great uncertainty. It is bound to bring about the occurrence of node nonuniform distribution over time. Given the obstacles and boundaries constrained areas, the intention of this article is to redeploy the drifted underwater nodes for regaining higher coverage rate. Hence, we propose a 3-D virtual force coverage algorithm (3D-VFCA). Inspired by the physical water flow action, the position evolution model of drifted underwater nodes is derived under the continuous gravity, buoyancy, propulsion, and resistance forces, which can reveal the mechanism of UWSNs deformation and coverage holes caused by the underwater node drift. Then, the coverage problem of dense and sparse node distributions is transformed into the optimization problem of weighted distance between node positions and clustering centers in the precoverage process, which can reduce large moving distances caused by node blind movements and solve the problem of inaccurate clustering centers caused by outliers. Furthermore, the improved virtual force algorithm is designed in consideration of underwater nodes, obstacles, and boundaries, which can drive the precovered underwater nodes to more optimal positions based on the adaptive moving distance per step. Finally, coverage performance evaluations of drifted underwater nodes are performed under different coverage algorithms, node numbers, and obstacles. The experimental results indicate that the proposed 3D-VFCA can improve the coverage rates in UWSNs.