Search Results (8,481)

Accurately predicting the next location of urban individuals is a central issue in human mobility research. Human mobility exhibits diverse patterns, requiring the integration of spatiotemporal contexts for location prediction. In this context, multi-view learning has become a prominent method in location prediction. Despite notable advances, current methods still face challenges in effectively capturing non-spatial proximity of regional preferences, complex temporal periodicity, and the ambiguity of location semantics. To address these challenges, we propose a representation-enhanced multi-view learning network (ReMVL-Net) for location prediction. Specifically, we propose a community-enhanced spatial representation that transcends geographic proximity to capture latent mobility patterns. In addition, we introduce a multi-granular enhanced temporal representation to model the multi-level periodicity of human mobility and design a rule-based semantic recognition method to enrich location semantics. We evaluate the proposed model using mobile phone data from Fuzhou. Experimental results show a 2.94% improvement in prediction accuracy over the best-performing baseline. Further analysis reveals that community space plays a key role in narrowing the candidate location set. Moreover, we observe that prediction difficulty is strongly influenced by individual travel behaviors, with more regular activity patterns being easier to predict. Full article

In this paper, we develop a geometric formulation of datasets. The key novel idea is to formulate a dataset to be a fuzzy topological measure space as a global object and equip the space with an atlas of local charts using graphs of fuzzy linear logical functions. We call such a space a logifold. In applications, the charts are constructed by machine learning with neural network models. We implement the logifold formulation to find fuzzy domains of a dataset and to improve accuracy in data classification problems. Full article

With the increasing deployment of unmanned aerial vehicles (UAVs) in indoor environments, the demand for high-precision six-degrees-of-freedom (6-DoF) localization has grown significantly. Ultra-wideband (UWB) technology has emerged as a key enabler for indoor UAV navigation due to its robustness against multipath effects and high-accuracy ranging capabilities. However, conventional UWB-based systems primarily rely on range measurements, operate at low measurement frequencies, and are incapable of providing attitude information. This paper proposes a tightly coupled error-state extended Kalman filter (TC–ESKF)-based UWB/inertial measurement unit (IMU) fusion framework. To address the challenge of initial state acquisition, a weighted nonlinear least squares (WNLS)-based initialization algorithm is proposed to rapidly estimate the UAV’s initial position and attitude under static conditions. During dynamic navigation, the system integrates time-difference-of-arrival (TDOA) and angle-of-arrival (AOA) measurements obtained from the UWB module to refine the state estimates, thereby enhancing both positioning accuracy and attitude stability. The proposed system is evaluated through simulations and real-world indoor flight experiments. Experimental results show that the proposed algorithm outperforms representative fusion algorithms in 3D positioning and yaw estimation accuracy. Full article

Space-Air-Ground Integrated Network (SAGIN), which is considered a network architecture with great development potential, exhibits significant cross-domain collaboration characteristics at present. However, most of the existing works ignore the matching and adaptability of differential tasks and heterogeneous resources, resulting in significantly inefficient task execution and undesirable network performance. As a consequence, we formulate a category attribute-oriented resource allocation and task offloading optimization problem with the aim of minimizing the overall scheduling cost. We first introduce a task–resource matching matrix to facilitate optimal task offloading policies with computation resources. In addition, virtual queues are constructed to take the impacts of randomized task arrival into account. To solve the optimization objective which jointly considers bandwidth allocation, transmission power control and task offloading decision effectively, we proposed a deep reinforcement learning (DRL) algorithm framework considering type matching. Simulation experiments demonstrate the effectiveness of our proposed algorithm as well as superior performance compared to others. Full article

Network connectivity and area coverage are the most important aspects in the applications of wireless sensor networks (WSNs). The resource and energy constraints of sensor nodes, operational conditions, and network size pose challenges to the optimal coverage of targets in the region of interest (ROI). The main idea is to achieve maximum area coverage and connectivity with strategic deployment and the minimal number of sensor nodes. This work addresses the problem of network area coverage in randomly distributed WSNs and provides an efficient deployment strategy using an enhanced version of cuckoo search optimization (ECSO). The “sequential update evaluation” mechanism is used to mitigate the dependency among dimensions and provide highly accurate solutions, particularly during the local search phase. During the preference random walk phase of conventional CSO, particle swarm optimization (PSO) with adaptive inertia weights is defined to accelerate the local search capabilities. The “opposition-based learning (OBL)” strategy is applied to ensure high-quality initial solutions that help to enhance the balance between exploration and exploitation. By considering the opposite of current solutions to expand the search space, we achieve higher convergence speed and population diversity. The performance of ECSO-OBL is evaluated using eight benchmark functions, and the results of three cases are compared with the existing methods. The proposed method enhances network coverage with a non-uniform distribution of sensor nodes and attempts to cover the whole ROI with a minimal number of sensor nodes. In a WSN with a 100 m² area, we achieved a maximum coverage rate of 98.45% and algorithm convergence in 143 iterations, and the execution time was limited to 2.85 s. The simulation results of various cases prove the higher efficiency of the ECSO-OBL method in terms of network coverage and connectivity in WSNs compared with existing state-of-the-art works. Full article

Multimodal sentiment analysis (MSA) faces key challenges such as incomplete modality inputs, long-range temporal dependencies, and suboptimal fusion strategies. To address these, we propose MGMR-Net, a Mamba-guided multimodal reconstruction and fusion network that integrates modality-aware reconstruction with text-centric fusion within an efficient state-space modeling framework. MGMR-Net consists of two core components: the Mamba-collaborative fusion module, which utilizes a two-stage selective state-space mechanism for fine-grained cross-modal alignment and hierarchical temporal integration, and the Mamba-enhanced reconstruction module, which employs continuous-time recurrence and dynamic gating to accurately recover corrupted or missing modality features. The entire network is jointly optimized via a unified multi-task loss, enabling simultaneous learning of discriminative features for sentiment prediction and reconstructive features for modality recovery. Extensive experiments on CMU-MOSI, CMU-MOSEI, and CH-SIMS datasets demonstrate that MGMR-Net consistently outperforms several baseline methods under both complete and missing modality settings, achieving superior accuracy, robustness, and generalization. Full article

This study proposes an intelligent patrol planning framework that leverages reinforcement learning, spatiotemporal crime forecasting, and simulated sensor telemetry to optimize autonomous vehicle (AV) navigation in urban environments. Crime incidents from Washington DC (2024–2025) and Seattle (2008–2024) are modeled as a dynamic spatiotemporal graph, capturing the evolving intensity and distribution of criminal activity across neighborhoods and time windows. The agent’s state space incorporates synthetic AV sensor inputs—including fuel level, visual anomaly detection, and threat signals—to reflect real-world operational constraints. We evaluate and compare three learning strategies: Deep Q-Network (DQN), Double Deep Q-Network (DDQN), and Proximal Policy Optimization (PPO). Experimental results show that DDQN outperforms DQN in convergence speed and reward accumulation, while PPO demonstrates greater adaptability in sensor-rich, high-noise conditions. Real-map simulations and hourly risk heatmaps validate the effectiveness of our approach, highlighting its potential to inform scalable, data-driven patrol strategies in next-generation smart cities. Full article

Remote sensing images are often degraded by atmospheric haze, which not only reduces image quality but also complicates information extraction, particularly in high-level visual analysis tasks such as object detection and scene classification. State-space models (SSMs) have recently emerged as a powerful paradigm for vision tasks, showing great promise due to their computational efficiency and robust capacity to model global dependencies. However, most existing learning-based dehazing methods lack physical interpretability, leading to weak generalization. Furthermore, they typically rely on spatial features while neglecting crucial frequency domain information, resulting in incomplete feature representation. To address these challenges, we propose ScaleViM-PDD, a novel network that enhances an SSM backbone with two key innovations: a Multi-scale EfficientViM with Physical Decoupling (ScaleViM-P) module and a Dual-Domain Fusion (DD Fusion) module. The ScaleViM-P module synergistically integrates a Physical Decoupling block within a Multi-scale EfficientViM architecture. This design enables the network to mitigate haze interference in a physically grounded manner at each representational scale while simultaneously capturing global contextual information to adaptively handle complex haze distributions. To further address detail loss, the DD Fusion module replaces conventional skip connections by incorporating a novel Frequency Domain Module (FDM) alongside channel and position attention. This allows for a more effective fusion of spatial and frequency features, significantly improving the recovery of fine-grained details, including color and texture information. Extensive experiments on nine publicly available remote sensing datasets demonstrate that ScaleViM-PDD consistently surpasses state-of-the-art baselines in both qualitative and quantitative evaluations, highlighting its strong generalization ability. Full article

Financial market forecasting remains challenging due to complex nonlinear dynamics and regime-dependent behaviors that traditional models struggle to capture effectively. This research introduces the Adaptive Financial Reservoir Network with Hypernetwork Flow (AFRN–HyperFlow) framework, a novel ensemble architecture integrating Echo State Networks, temporal convolutional networks, mixture density networks, adaptive Hypernetworks, and deep state-space models for enhanced financial time-series prediction. Through comprehensive feature engineering incorporating technical indicators, spectral decomposition, reservoir-based representations, and flow dynamics characteristics, the framework achieves superior forecasting performance across diverse market conditions. Experimental validation on 26,817 balanced samples demonstrates exceptional results with an F1-score of 0.8947, representing a 12.3% improvement over State-of-the-Art baseline methods, while maintaining robust performance across asset classes from equities to cryptocurrencies. The adaptive Hypernetwork mechanism enables real-time regime-change detection with 2.3 days average lag and 95% accuracy, while systematic SHAP analysis provides comprehensive interpretability essential for regulatory compliance. Ablation studies reveal Echo State Networks contribute 9.47% performance improvement, validating the architectural design. The AFRN–HyperFlow framework addresses critical limitations in uncertainty quantification, regime adaptability, and interpretability, offering promising directions for next-generation financial forecasting systems incorporating quantum computing and federated learning approaches. Full article

Synthetic aperture radar (SAR), as an active microwave imaging system, has the capability of all-weather and all-time observation. In response to the challenges of aircraft detection in SAR images due to the complex background interference caused by the continuous scattering of airport buildings and the demand for real-time processing, this paper proposes a YOLOv7-MTI recognition model that combines the attention mechanism and involution. By integrating the MTCN module and involution, performance is enhanced. The Multi-TASP-Conv network (MTCN) module aims to effectively extract low-level semantic and spatial information using a shared lightweight attention gate structure to achieve cross-dimensional interaction between “channels and space” with very few parameters, capturing the dependencies among multiple dimensions and improving feature representation ability. Involution helps the model adaptively adjust the weights of spatial positions through dynamic parameterized convolution kernels, strengthening the discrete strong scattering points specific to aircraft and suppressing the continuous scattering of the background, thereby alleviating the interference of complex backgrounds. Experiments on the SAR-AIRcraft-1.0 dataset, which includes seven categories such as A220, A320/321, A330, ARJ21, Boeing737, Boeing787, and others, show that the mAP and mRecall of YOLOv7-MTI reach 93.51% and 96.45%, respectively, outperforming Faster R-CNN, SSD, YOLOv5, YOLOv7, and YOLOv8. Compared with the basic YOLOv7, mAP is improved by 1.47%, mRecall by 1.64%, and FPS by 8.27%, achieving an effective balance between accuracy and speed, providing research ideas for SAR aircraft recognition. Full article

Deep neural network (DNN) hardware (HW) accelerators have achieved great success in improving DNNs’ performance and efficiency. One key reason is the dataflow in executing a DNN layer, including on-chip data partitioning, computation parallelism, and scheduling policy, which have large impacts on latency and energy consumption. Unlike prior works that required considerable efforts from HW engineers to design suitable dataflows for different DNNs, this work proposes an efficient data-centric approach, named Dataflow Code Propagation (DCP), to automatically find the optimal dataflow for DNN layers in seconds without human effort. It has several attractive benefits that prior studies lack, including the following: (i) We translate the HW dataflow configuration into a code representation in a unified dataflow coding space, which can be optimized by back-propagating gradients given a DNN layer or network. (ii) DCP learns a neural predictor to efficiently update the dataflow codes towards the desired gradient directions to minimize various optimization objectives, e.g., latency and energy. (iii) It can be easily generalized to unseen HW configurations in a zero-shot or few-shot learning manner. For example, without using additional training data, Extensive experiments on several representative models such as MobileNet, ResNet, and ViT show that DCP outperforms its counterparts in various settings. Full article

Porous materials are characterized by the pore surface area (S) and volume (V) accessible to a confined fluid. For mesoporous materials NMR measurements of diffusion are used to assess the S/V ratio, because at short times, only the diffusivity of molecules in the adsorbed layer is affected by confinement and the fractional population of these molecules is proportional to the S/V ratio. For materials with sub-nanometer pores, this might not be true, as the adsorbed layer can encompass the entire pore volume. Here, using molecular simulations, we explore the role played by S and S/V in determining the dynamical behavior of two carbon-bearing fluids—CO₂ and ethane—confined in sub-nanometer pores of silica. S and V in a silicalite model representing a sub-nanometer porous material are varied by selectively blocking a part of the pore network by immobile methane molecules. Three classes of adsorbents were thus obtained with either all of the straight (labeled ‘S-major’) or zigzag channels (‘Z-major’) remaining open or a mix of a fraction of both types of channel blocked, resulting in half of the total pore volume being blocked (‘Half’). While the adsorption layers from opposite surfaces overlap, encompassing the entire pore volume for all pores except the intersections, the diffusion coefficient is still found to be reduced at high S/V, especially for CO₂, albeit not so strongly as would be expected in the case of wider pores. This is because of the presence of channel intersections that provide a wider pore space with non-overlapping adsorption layers. Full article

With the diversification of space applications, the tracking of maneuvering targets has gradually gained attention. Issues such as their wide range of movement and observation outliers caused by human operation are worthy of in-depth discussion. This paper presents a novel distributed diffusion multi-noise Interacting Multiple Model (IMM) filter for maneuvering target tracking in heavy-tailed noise. The proposed approach leverages parallel Gaussian and Student-t filters to enhance robustness against non-Gaussian process and measurement noise. This hybrid filter is implemented as a node within a distributed network, where the diffusion algorithm leads to the global state asymptotically reaching consensus as the filtering time progresses. Furthermore, a fusion of multiple motion models within the IMM algorithm enables robust tracking of maneuvering targets across the distributed network and process outlier caused by maneuver compared to previous studies. Simulation results demonstrate the effectiveness of the proposed filter in tracking maneuvering targets. Full article

Monitoring pH and acidity during yoghurt fermentation is essential for product quality and process efficiency. Conventional measurement methods, however, are invasive and labour-intensive. This study developed artificial neural network (ANN) models to predict pH and titratable acidity during yoghurt fermentation using CIELAB colour parameters (L, a*, b*). Reconstituted milk powder with 12% total solids was prepared with varying protein levels (4.2–4.8%), inoculum concentrations (1–3%), and fermentation temperatures (36–44 °C). Data were collected every 10 min until pH 4.6 was reached. Forty models were trained for each output variable, using 90% of the data for training and 10% for validation. The first two phases of the fermentation process were clearly distinguishable, lasting between 4.5 and 7 h and exceeding 0.6% lactic acid in all treatments evaluated. The best pH model used two hidden layers with 28 neurons (R² = 0.969; RMSE = 0.007), while the optimal acidity model had four hidden layers with 32 neurons (R² = 0.868; RMSE = 0.002). The strong correlation between colour and physicochemical changes confirms the feasibility of this non-destructive approach. Integrating ANN models and colourimetry offers a practical solution for real-time monitoring, helping improve process control in industrial yoghurt production. Full article

We introduce a waypoint-based flow capture location model (WbFCLM) for siting facilities on networks with the objective of maximally capturing flows. The advantages of the waypoint-based formulation are that (1) demand is modeled along observed trajectories rather than assumed travel paths, and (2) demand is allowed to vary across geographic space. We demonstrate the WbFCLM using a test dataset derived from vehicle tracking data. Though solving the WbFCLM can require considerable computing power, it can be used to site facilities on networks where both flows and demand vary spatially. Full article