Map-free visual relocalization computes camera pose using only a query image and a reference image. Therefore, it is hindered by challenges in feature-point matching and the absence of scale information in monocular images. These issues may cause significant rotational and metric errors, leading to localization failures. To address these challenges, we propose a map-free visual relocalization method enhanced with instance knowledge and depth knowledge. By utilizing instance-based matching, our approach improves the robustness of feature-point matching by focusing on relevant regions across scenes. Additionally, our depth estimation techniques provide reliable depth knowledge from a single image, improving scale recovery and reducing translation errors. Our method surpasses the previous state-of-the-art by 1.071m and 13.593° on the translation error and rotation error, respectively. Furthermore, we are also one of the winners in the Map-free Workshop & Challenge (ECCV2024), underscoring its superiority compared to concurrent approaches.

We introduce Crowd-Sourced Splatting (CSS), a novel 3D Gaussian Splatting (3DGS) pipeline designed to overcome the challenges of pose-free scene reconstruction using crowd-sourced imagery. The dream of reconstructing historically significant but inaccessible scenes from collections of photographs has long captivated researchers. However, traditional 3D techniques struggle with missing camera poses, limited viewpoints, and inconsistent lighting. CSS addresses these challenges through robust geometric priors and advanced illumination modeling, enabling high-quality novel view synthesis under complex, real-world conditions. Our method demonstrates clear improvements over existing approaches, paving the way for more accurate and flexible applications in AR, VR, and large-scale 3D reconstruction.

Human Activity Recognition (HAR) is crucial for applications like urban traffic management and health monitoring but faces challenges in handling complex patterns and missing sensor data. In this work, we propose a conditional Kolmogorov-Arnold network diffusion (CKAD) framework for HAR, which separates the prediction process into two stages: sensor data recovery and HAR classification. In data recovery progress, we employ the diffusion generation framework and propose a KAN-denosing model to enhance the ability for sensor feature construction. Moreover, we design the Conditional Information Interpolation mechanism to obtain the robust motion sensor representation. Experiments on three public datasets demonstrate that our model outperforms existing state-of-the-art methods, especially in scenarios with incomplete sensor data, proving its robustness and efficacy.

Intelligent on-demand delivery relies on precise indoor positioning to enhance courier experiences in complex environments. In multi-floor buildings, inefficient navigation causes delays for new couriers, highlighting the need for accurate indoor floor plans. Current on-site surveys are resourceintensive, and crowdsourced methods often rely on shallow models, struggling with accuracy. Existing studies lack sufficient diverse data for large-scale commercial applications, limiting practicality for nationwide malls. Leveraging extensive crowdsourced data collected by couriers, instant delivery platforms offer promising options for indoor topology mining, presenting an opportunity for innovation in efficient navigation and delivery optimization. Challenges in indoor floor plan mining arise from diverse smartphone usage patterns and complex courier activities hindering accurate trajectory information provision. Couriers reporting previous arrival data to prevent delays complicates trajectory accuracy enhancement. Crowdsourced tracks lack precise alignment, diminishing accuracy post traditional trajectory fusion. To tackle these issues, our proposed WeakMap system employs weak-supervised learning using couriers' inaccurate arrival data. Employing a trajectory representation learning model extracts spatio-temporal features, mitigating noisy data through hierarchical semantic feature clustering. Further, a Heterogeneity trajectory embedding method and multi-scale cross-attention fusion enhance accuracy. Two weak-supervised tasks contribute to trajectory representation learning pre-training, presenting a novel solution for accurate indoor floor plan generation. The WeakMap system utilizes crowdsourced data, providing a cost-effective solution without manual mapping or professional equipment. In our assessment, using extensive data from China's largest instant delivery platform, WeakMap achieved an RMSE of 5.7 m and 79.3 % average accuracy, surpassing state-of-the-art methods. Leveraging WeakMap's floor plan information, we optimized downstream tasks, resulting in a 42.0 % reduction in prediction errors. This innovation extends to weak-supervised crowdsourcing applications, enhancing courier experiences and order scheduling in intelligent instant delivery services. Moreover, WeakMap offers a practical solution for generating indoor floor plans, seamlessly integrating with various positioning systems for widespread adoption.

Visual dense SLAM can facilitate pose estimation and map reconstruction for sensor carriers in unknown environments. However, in uncontrolled environments such as offices, shopping malls, and train stations, frequent occurrences of people walking back and forth or temporary movement of objects within the scene are common. Most existing visual dense SLAM systems do not account for these dynamic factors, leading to localization drift and map distortion. In this paper, we propose DGS-SLAM, a system capable of achieving robust localization and high-fidelity static map reconstruction in dynamic environments. We utilize semantic 3D Gaussians for scene representation, effectively eliminating interference from dynamic objects and refining the reconstruction of static background. We enhance the tracking accuracy and mapping quality of dense SLAM by using a distance distribution-based Gaussian pruning algorithm and implementing a coarse-to-fine tracking strategy with bundle adjustment and differentiable rendering. We perform qualitative and quantitative evaluations on two publicly available dynamic environment datasets. The results indicate that our method effectively reduces the interference caused by dynamic objects, enabling visual dense SLAM to maintain competitive tracking accuracy and mapping performance in dynamic environments.

The success of 3D Gaussian splatting in 3D reconstruction has recently led to efforts to integrate it with SLAM systems. However, most existing research has focused on indoor tracking and mapping, while outdoor Gaussian SLAM methods still heavily rely expensive LiDAR sensor. To address these challenges, we propose LOG-SLAM, a novel method for large-scale outdoor tracking and mapping using Gaussian Splatting. Our approach supports tracking through monocular or visual-inertial input, progressively constructing the 3D Gaussian map from depth and pose estimates obtained during the tracking process. Additionally, we introduce a submap-based strategy for managing large-scale maps, enabling the reconstruction of kilometer-scale environments. A loop closure detection module is also incorporated to reduce accumulated errors. Furthermore, we present a novel dynamic object removal method based on rendering loss that mitigates the interference of dynamic objects on the reconstruction. Our experiments on KITTI and KITTI-360 demonstrate that our method achieves localization performance comparable to traditional SLAM systems, while outperforming recent GS/NeRF-based SLAM approaches in terms of mapping and rendering quality.

Single-camera-training person reidentification (SCT re-ID) aims to train a reidentification (re-ID) model using single-camera-training (SCT) datasets where each person appears in only one camera. The main challenge of SCT re-ID is to learn camera-invariant feature representations without cross-camera same-person (CCSP) data as supervision. Previous methods address it by assuming that the most similar person should be found in another camera. However, this assumption is not guaranteed to be correct. In this article, we propose a novel solution: the camera-invariant meta-learning network (CIMN) for SCT re-ID. CIMN operates under the premise that camera-invariant feature representations should remain robust despite changes in camera settings. To achieve this, we partition the training data into a meta-train set and a meta-test set based on camera IDs. We then conduct a cross-camera simulation (CCS) using a meta-learning strategy, aiming to enforce the feature representations learned from the meta-train set to be robust when applied to the meta-test set. We further introduce three specific loss functions to leverage potential identity relations between the meta-train set and the meta-test set. Through the CCS and the introduced loss functions, CIMN can extract feature representations that are both camera-invariant and identity-discriminative even in the absence of CCSP data. Our experimental results demonstrate that CIMN can extract feature representations that are both camera-invariant and identity-discriminative, even in the absence of CCSP data. our method achieves comparable performance with and without the use of CCSP data, and outperforms state-of-the-art methods on three SCT re-ID benchmarks.

Accurately estimating the position and attitude of vehicles is essential for intelligent transportation systems. The GNSS/INS integrated system offers precise navigation information. However, the system’s positioning errors may accumulate rapidly in challenging GNSS signal conditions. Odometer (ODO) and nonholonomic constraints (NHC) are commonly employed to mitigate the rapid accumulation of INS errors. Compensating for the mounting angles of INS and the scale factor of the odometer is necessary to fully exploit the potential of ODO/NHC. However, many studies employ Kalman filters with small mounting angle assumption, which limits their applicability for large mounting angles in practice. To accurately estimate the mounting angle of INS with any installation attitude, we propose a new algorithm called Fast Estimation of Mounting Angle and Scale Factor (FEMASF). FEMASF employs Singular Value Decomposition (SVD) to obtain a closed-form solution for the parameters. It also incorporates an enhanced Sage-Husa scheme, enhancing overall estimation accuracy by reducing the weight of outlier data through a forgetting factor. Extensive simulation experimental results demonstrate that our proposed FEMASF algorithm outperforms filter-based methods in terms of accuracy and convergence speed for large mounting angles. Specifically, for the −90° mounting angle, FEMASF achieves 0.45° angle error, while the velocity-based Kalman filter (VKF) fails to converge and the position-based Kalman filter (PKF) yields about 4° error. Furthermore, neither VKF nor PKF converges for the 120° mounting angle, whereas FEMASF exhibits only about 3.2° estimation error.

Transportation mode-aware trajectory recovery is the fundamental for individual oriented downstream tasks in intelligent transportation systems. Different from vehicle based trajectory recovery, it suffers from the heterogeneity and sparsity issues arising from the insufficient data labelled for distinct transportation modes (e.g., obtaining limited individual trajectories from modes like walking or cycling due to privacy concerns while obtaining rich vehicle trajectories from the mode like driving). To alleviate this, we develop a novel Cross-trAnsportation-mode Knowledge transfEr method with meta learning, coined as Cake, to first learn generalized parameters from source modes (i.e., the relatively dense modes) and then share the meta knowledge with the targets (i.e., more sparse modes), which significantly improve the recovery performance for the sparse. To achieve this, we first develop an efficient fine-GRAined Personalized trajectory rEcovery model called Grape, to incorporate the cross-granularity features with coarse-centered and fine-centered subgraph learning and learn the intrinsic characteristics of transportation modes with auto-correlation efficiently. Then we design a personalized memory to store distinct parameters for diverse interests of individual groups and read the memory for predictor’s input features according to the previous learnt features. At last, we employ the feature reuse strategy based on meta learning to iteratively make adaptions from the source to the target. Extensive experimental results on real-world dataset demonstrate that our proposed method significantly outperforms the state-of-arts for the sparse modes.

We focus on the transportation-aware trajectory recovery problem, which is distinct from the conventional vehicle-based trajectory recovery, facing three major challenges: heterogeneity, personalization and efficiency. For the heterogeneity, the velocity of the mobile object is intrinsically correlated with the specific transportation mode, containing inherent heterogeneity. For the personalization, the trajectory data is complicated by substantial variations in users, which are different in personalized behaviors. For the efficiency, previous works mostly employ sequence-to-sequence framework which limits their efficiency due to the auto-regressive inference pattern. To address these challenges, we design a novel efficient and effective multi-modal deep model, coined as PTrajRec, for transportation-aware trajectory recovery. Specifically, we initially embed location, behavior, and transportation mode modalities in distinct channels, which not only reflect spatio-temporal information encapsulated in location sequences but also introduce the heterogeneity and personalization characteristics associated with mode and behavior sequences. For further modeling these modalities, we employ the auto-correlation mechanism to learn periodic dependencies on the temporal dimension and the graph attention mechanism to learn road network dependencies on the spatial dimension. At last, we propose a dual-view constraint mechanism to assist the efficient trajectory recovery framework and design three auxiliary tasks to address the inherent heterogeneity and efficiency design. Extensive experimental results on two real-world datasets demonstrate the superiority of our proposed method compared to state-of-the-art baselines with reduced computation cost and excellent performance.

Neural implicit representation has recently achieved significant advancements, especially in the field of SLAM(Simultaneous Localization and Mapping). Previous NeRF-based SLAM methods have difficulties with object-level localization and reconstruction and struggle in dynamic and illumination-varied environments. We propose ONeK-SLAM, a robust object-level SLAM system that effectively combines feature points and neural radiance fields. ONeK-SLAM uses the joint information at the object level to improve localization accuracy and enhance reconstruction details. Moreover, our approach detects and eliminates dynamic objects based on the joint errors, while also harnessing the illumination invariance offered by feature points. Consequently, ONeK-SLAM achieves high-precision localization and detailed object-level mapping, even in dynamic and illumination-varying environments. Our evaluations, conducted on three public datasets that include both dynamic and variable lighting sequences, demonstrate that our method outperforms recent NeRF-based SLAM method in both localization and reconstruction.

When there exists a mismatch between the system model or inaccuracies in the statistical characteristics of the noise, it can directly impact the accuracy of current state estimation and potentially lead to filter divergence. In the domain of dynamic positioning and navigation, the Kalman filter has been proposed as a solution to address these issues by incorporating fading factors. Unlike conventional methods for estimating fading factors, this study introduces the RGAM (Resnet Global Attention Mechanism) algorithm model to estimate multiple fading factors associated with state prediction covariance. The RGAM algorithm model leverages the robust estimation capability of network models to accurately estimate the fading factors corresponding to relevant epochs. To validate the effectiveness of the proposed algorithm, experiments were conducted using an actual road data set within the context of a GNSS/INS (Global Navigation Satellite System/ Internal Navigation System) integrated system. Through these experiments, it was observed that employing a filter with accurate fading factors facilitates the avoidance of filter divergence and achieves stable filtering performance in the integrated system.

In large-scale outdoor environments, vehicles often encounter situations like retracing their path or turning around, leading to many reverse loop closures where the vehicles traverse previously covered paths from opposite viewpoints. Existing monocular SLAM methods, due to insufficient utilization of semantic information and neglect of leveraging reverse loop closures, result in significant scale drift and pose drift when confronted with such scenarios. In this paper, we introduce SMORE-SLAM, a semantic monocular SLAM with scale correction and reverse loop closure module. We constrain scale drift by harnessing semantic information across a wide spatial extent. Furthermore, we detect and correct reverse loop closures using semantic point cloud to reduce pose drift. Experimental results on the KITTI odometry dataset and the Oxford RobotCar dataset demonstrate the capability of our research in scale correction and reverse loop closure detection, enabling a reduction in trajectory errors of monocular SLAM.

GNSS (global navigation satellite systems) technology enables high-precision single-point positioning (SPP) in open environments. However, the accuracy of GNSS positioning is significantly compromised in complex urban canyons due to signal obstructions and non-line-of-sight propagation errors. To address this challenge, we propose a GNSS displacement estimation algorithm. This method learns nonlinear dependencies between GNSS raw measurements and corresponding position changes, capturing dynamic and layered features in GNSS measurement data for displacement estimation. We introduce a denoising auto-encoder (DAE) to preprocess raw GNSS observations, reducing the impact of noise. The model simultaneously outputs estimated displacement and model confidence. The fusion process dynamically combines positioning results from the SPP algorithm and the D-Tran model, adaptively blending them to achieve accurate and optimal positioning estimation. This approach optimizes the accuracy of estimated positioning results while maintaining confidence in the estimation. Experimental results show a 61% reduction in root mean square error (RMSE) and 100% availability in urban canyon environments compared to traditional single-point positioning techniques.

Traffic forecasting belongs to intelligent transportation systems and is helpful for public property and life safety. Therefore, to forecast traffic accurately, researchers pay great attention to dealing with complex problems by mining intricate spatial and temporal dependencies of the traffic. However, some challenges still hold back traffic forecasting: 1) Most studies mainly focus on modeling correlations of traffic time series of close distances on the road network and ignore correlations of remote but similar traffic time series; 2) Previous static graph-based methods failed to reflect the dynamic changed spatial relations of multiple time series in the evolving traffic system. To tackle the above issues, we design a new dynamic multi-graph spatio-temporal convolution network (DMGSTCN) in this paper, which utilizes the gated causal convolution with the dynamic multi-graph convolution network (DMGCN) to simultaneously extract spatial and temporal information. Specifically, DMGCN uses not only distance-based graphs but also structure-based graphs to obtain spatial information from nearby and remote but similar traffic time series, respectively. Moreover, to dynamically model spatial correlations, DMGCN first splits neighbors of each traffic time series into different regions according to relative position relationships. Then DMGCN assigns different weights to different regions at different time slices. Empirical evaluations on four traffic forecasting benchmarks reveal that DMGSTCN outperforms existing methods.

In the context of the Internet of Things, utilizing the inherent inertial sensors in smartphones for human activity recognition (HAR) has garnered considerable attention owing to its wide-ranging applications. However, prevailing HAR approaches primarily treat activity identification as a single-label classification task, focusing solely on discerning pedestrian motion modes or device usage modes, while disregarding their interrelatedness. Additionally, HAR methods employing sliding windows encounter challenges associated with the multiclass window problem, wherein certain sample labels differ from the label assigned to the window. This paper aims to address these issues. This paper presents a novel approach for simultaneously recognizing pedestrian motion and device usage modes by utilizing the segmentation transformer. The proposed joint recognition framework effectively annotates sensor data at each timestamp and achieves dense prediction of time-series data through the encoding and decoding of the annotated data. To optimize the utilization of information extracted from each Transformer layer, a global up-sampling decoder based on the pyramid attention module is introduced, enabling dense decoding of features obtained from each Transformer layer. We performed experiments on two publicly available datasets to comprehensively assess the effectiveness of the proposed methodology. The results demonstrate that our approach achieves an accuracy of 99.79% and a weighted F-score of 99.77%, surpassing the performance of existing state-of-the-art methods. Furthermore, we constructed heterogeneous datasets to validate the robustness of our method. The extensive experimental findings indicate that the joint recognition framework effectively uncovers the inherent correlations between pedestrian motion and device usage modes, leading to enhanced accuracy in recognition and addressing the challenges posed by the multiclass window problem.

Location-based service (LBS) is the core of internet of things (IoTs), which serves tracking, navigation and monitoring. The ubiquitous magnetic signals are temporally stable and spatially distinguishable, and can achieve high-precision and ubiquitous positioning results without additional infrastructure, which is favored by researchers and has become a major research hotspot. Although there has been extensive research in the field of indoor magnetic positioning, there is still room for optimization in terms of positioning accuracy and robustness. Aiming at the problem that the magnetometer is offset and susceptible to environmental interference, we propose an online magnetometer calibration algorithm without user perception. Aiming at the inconsistency of magnetic data spatial scale problem caused by differences in device sampling frequency and user walking speed, we leverage different scales to segment the magnetic data, extract the magnetic sequence features of the corresponding scales through Transformer, utilize the attention mechanism to score the weights of the different scale features, and finally fuse the multiple scale features for positioning. We conduct extensive and well-designed experiments on public datasets and self-collected datasets. The experimental results indicate that the proposed method effectively solves the magnetic spatial scale problem and improves indoor magnetic positioning accuracy.

Tightly integrated navigation based on Extended Kalman filter (EKF) has become ubiquitous across domains such as autonomous driving.The accuracy of the innovation matrix plays a crucial role in the filtering process. However, the variance in data quality across different satellites due to various errors, as well as the nonlinear errors introduced by the linearization of the measurement equation and the errors resulting from the Gaussian noise assumption, can lead to inaccuracies in the innovation matrix. We address these aforementioned issues and propose an adaptive solution relying on Resnet- Efficient Multi-Head Self-Attention (Res-EMSA) to adjust the innovation matrix. Specifically, the network model extracts the features of Global Navigation Satellite System (GNSS) and Inertial Navigation System (INS) data through the residual network, and then the features are fused with different weights. After that, the EMSA network is utilized for feature mapping, and ultimately, the fully connected layer is used to perform weight matrix regression. Experimental results reveal that Res-EMSA model exhibits a significant enhancement in positioning accuracy, with a 39% increase compared to the traditional EKF model.

This letter addresses challenges stemming from sensor manufacturing processes and technological constraints, such as nonlinear stochastic noise, leading to rapid INS positioning error divergence. To enhance the performance of GNSS/INS (Global Navigation Satellite System/Integrated Navigation Systems) integrated navigation methods, we propose an AI-based adaptive error compensation method. We introduce deep learning to correct INS computational errors, leveraging its precision modeling without strict noise assumptions. Our approach integrates filtering methods and deep learning approaches, avoiding the uncontrollable introduction of positioning errors inherent in end-to-end models' black-box mode. We design a novel deep model structure to improve generalization while reducing parameters and computational complexity. Validation is conducted using a vehicular navigation data acquisition platform, simulating scenarios of GNSS signal loss. Experimental results demonstrate a 77.70% improvement in recognition rates across different road segments compared to traditional methods based on Extended Kalman Filter, indicating significant practical value.

The IEEE 802.11-2016 standard enables devices to gather precise ranging information through the time-of-flight evaluation, facilitating the development of accurate indoor location-based services. Researchers have indicated that the protocol's most effective performance is in scenarios with direct line-of-sight, despite providing meter-level ranging accuracy. In real indoor environments, the accuracy diminishes considerably due to random errors caused by interference such as multipath effects and non-line-of-sight signal propagation. Therefore, it is essential to accurately evaluate the reliability of each ranging measurement and effectively leverage neighboring high-quality observations to improve positioning accuracy. This study presents a novel optimization algorithm that relies on the motion observation series by incorporating adjacent ranging observations and a priori motion knowledge into a factor graph model, resulting in a unified optimization objective. Consequently, our system can dynamically estimate the confidence of fine time measurements ranging measurements. It optimizes the position estimation of the current user by maximizing the probability of not only the current ranging measurements but also the adjacent historical measurements and a priori motion. Additionally, to enable real-time positioning, a fast-solving procedure employing an adaptive gradient is proposed, capable of providing evaluations in under 10ms. The system has been tested in real indoor environments, showing improved performance compared to existing methods. It achieves meter-level real-time positioning accuracy at 1 sigma without requiring a specific device pose, additional sensor, or expensive site survey. This makes our proposal highly applicable for wide adoption and readiness for the market.

Travel time estimation(TTE) is a critical component of intelligent transportation systems. To achieve efficient and accurate trajectory-based travel time estimation, it is essential to design a streamlined model that reduces computation and memory costs. However, this is challenging as traditional deep neural networks are limited in their calculation capabilities and can be cumbersome due to the high number of model parameters. To overcome these challenges, we propose a novel approach to travel time estimation, utilizing a well-designed deep neural network model called Knowledge Distillation for Travel Time Estimation (KDTTE). By implementing knowledge distillation techniques, the model’s computational and memory requirements are reduced, while simultaneously improving its accuracy. The student model leverages the knowledge of the Teacher model to learn features it would not have been able to on its own, thereby enhancing the overall accuracy of the model. Our approach, referred to as KDTTE, was tested on two real-world datasets and showed improved accuracy compared to nine state-of-the-art baselines, demonstrating a 34.0% and 86.8% increase in accuracy on the Chengdu and Porto datasets, respectively.

Traffic forecasting is crucial for public safety and resource optimization, yet is very challenging due to the temporal changes and the dynamic spatial correlations. To capture these intricate dependencies, spatio-temporal networks, such as recurrent neural networks with graph convolution networks, are applied. However, traffic forecasting is still a non-trivial task because of three major challenges: 1) Previous spatio-temporal networks are based on end-to-end training and thus fail to handle the distribution shift in the non-stationary traffic time series. 2) Existing methods always utilize the one-hour input to forecast future traffic and the long-term historical trend knowledge is ignored. 3) The efficient and effective algorithm for modeling multi-scale spatial correlations is still lacking in prior networks. Therefore, in this paper, rather than proposing yet another end-to-end model, we provide a novel disentangle-fusion framework STWave+to mitigate the distribution shift issue. The framework first decouples the complex one-hour traffic data into stable trends and fluctuating events, followed by a dual-channel spatio-temporal network to model trends and events, respectively. Moreover, long-term trends are used as a self-supervised signal in STWave+to teach overall temporal information into one-hour trends through a contrastive loss. Finally, reasonable future traffic can be predicted through the adaptive fusion of one-hour trends and events. Additionally, we incorporate a novel query sampling strategy and multi-scale graph wavelet positional encoding into the full graph attention network to efficiently and effectively model dynamic hierarchical spatial correlations. Extensive experiments on four traffic datasets show the superiority of our approach,i.e., the higher forecasting accuracy with lower computational cost.

With the proliferation of intelligent sensors integrated into mobile devices, fine-grained human activity recognition (HAR) based on lightweight sensors has emerged as a useful tool for personalized applications. Although shallow and deep learning algorithms have been proposed for HAR problems in the past decades, these methods have limited capability to exploit semantic features from multiple sensor types. To address this limitation, we propose a novel HAR framework, DiamondNet, which can create heterogeneous multisensor modalities, denoise, extract, and fuse features from a fresh perspective. In DiamondNet, we leverage multiple 1-D convolutional denoising autoencoders (1-D-CDAEs) to extract robust encoder features. We further introduce an attention-based graph convolutional network to construct new heterogeneous multisensor modalities, which adaptively exploit the potential relationship between different sensors. Moreover, the proposed attentive fusion subnet, which jointly employs a global-attention mechanism and shallow features, effectively calibrates different-level features of multiple sensor modalities. This approach amplifies informative features and provides a comprehensive and robust perception for HAR. The efficacy of the DiamondNet framework is validated on three public datasets. The experimental results demonstrate that our proposed DiamondNet outperforms other state-of-the-art baselines, achieving remarkable and consistent accuracy improvements. Overall, our work introduces a new perspective on HAR, leveraging the power of multiple sensor modalities and attention mechanisms to significantly improve the performance.

Sensor-rich smartphones have facilitated a lot of services and applications. Indoor/Outdoor (IO) status serves as a critical foundation for various upstream tasks, including seamless pedestrian navigation, power management, and activity recognition. Nevertheless, achieving robust, efficient, and accurate IO detection remains challenging due to environmental complexities and device heterogeneity. To tackle this challenge, some researchers have turned to deep learning for IO detection, which can deal with complex scenarios and achieve high detection accuracy. However, deep learning methods are often blamed for their expensive computational cost. Therefore, in this paper, we introduce a novel efficient IO detection method-DeepSIO, which can detect IO status accurately and efficiently. Specifically, different from existing IO detection methods, DeepSIO is developed based on spiking neural networks (SNN) that are more biologically plausible and computationally efficient than other deep neural networks. To better capture useful features, we propose to utilize dense connections between SNN layers. Extensive experiments are conducted in three typical scenarios, and experimental results demonstrate that DeepSIO outperforms state-of-the-art methods, achieving an accuracy of about 99.7%. Moreover, it has better generalization ability and can adapt well to new environments and devices.

Traffic forecasting is crucial for public safety and resource optimization, yet is very challenging due to the temporal changes and the dynamic spatial correlations of the traffic data. To capture these intricate dependencies, spatio-temporal networks, such as recurrent neural networks with graph convolution networks, graph convolution networks with temporal convolution networks, and temporal attention networks with full graph attention networks, are applied. However, previous spatio-temporal networks are based on end-to-end training and thus fail to handle the distribution shift in the non-stationary traffic time series. On the other hand, the efficient and effective algorithm for modeling spatial correlations is still lacking in prior networks.In this paper, rather than proposing yet another end-to-end model, we aim to provide a novel disentangle-fusion framework STWave to mitigate the distribution shift issue. The framework first decouples the complex traffic data into stable trends and fluctuating events, followed by a dual-channel spatio-temporal network to model trends and events, respectively. Finally, reasonable future traffic can be predicted through the fusion of trends and events. Besides, we incorporate a novel query sampling strategy and graph wavelet-based graph positional encoding into the full graph attention network to efficiently and effectively model dynamic spatial correlations. Extensive experiments on six traffic datasets show the superiority of our approach, i.e., the higher forecasting accuracy with lower computational cost.

The demand for high-accuracy continuous positioning of vehicles in complex environments has become more and more urgent. Although existing tightly-coupled navigation based on GNSS/INS can provide continuous positioning in these scenes with bad GNSS signals or short-term GNSS outages, it still cannot meet the application requirements. To tackle the aforementioned problems, we propose a novel algorithm to predict GNSS pseudo-measurement when the GNSS signals are blocked. The algorithm introduces a hybrid multi-head attention neural network to learn the dynamic nonlinear relationship of the raw GNSS observations. Three attention neural networks are used to capture richer features from the pseudorange, pseudorange rate, satellite position and carrier position as well as the satellite velocity and the carrier velocity, respectively, which can improve the generalization ability and avoid overfitting. Extensive experiments on dataset show that the proposed algorithm can obtain at least $20 \sim 30$% positioning accuracy improvement compared with other tightly-coupled navigation methods.

Human activity recognition (HAR) based on wearable sensors has been a prosperous research topic in recent years. Considering that the sensor may be worn at diverse body positions, obtaining enough labeled human activity data for each body-worn position is usually expensive and labor-intensive. Furthermore, the variability and diversity of data distribution induced by different body-worn positions make the HAR model trained on data collected from one body position perform poorly for other body-worn positions. To achieve accurate HAR with low labeling cost, which we call unsupervised cross-position HAR, in this article, we propose a deep multiscale transfer learning (DMSTL) model. In the model, we first introduce an unsupervised source selection method to select the most similar source domain for transferring domain knowledge. Then, we develop a multiscale spatial–temporal Net (MSSTNet) to learn comprehensive multimodal representations from multiple feature subspaces. Finally, we design the category-level adaptation module and the domain-level adversarial module for learning domain-invariant features. We conduct extensive experiments on three public HAR data sets and demonstrate the reasonable generalization performance of the DMSTL, which remarkably outperforms other state-of-the-art baselines.

Next Point-of-Interest (POI) recommendation is a pivotal issue for researchers in the field of location-based social networks. While many recent efforts show the effectiveness of recurrent neural network-based next POI recommendation algorithms, several important challenges have not been well addressed yet: (i) The majority of previous models only consider the dependence of consecutive visits, while ignoring the intricate dependencies of POIs in traces; (ii) The nature of hierarchical and the matching of sub-sequence in POI sequences are hardly model in prior methods; (iii) Most of the existing solutions neglect the interactions between two modals of POI and the density category. To tackle the above challenges, we propose an auto-correlation enhanced multi-modal Transformer network (AutoMTN) for the next POI recommendation. Particularly, AutoMTN uses the Transformer network to explicitly exploits connections of all the POIs along the trace. Besides, to discover the dependencies at the sub-sequence level and attend to cross-modal interactions between POI and category sequences, we replace self-attention in Transformer with the auto-correlation mechanism and design a multi-modal network. Experiments results on two real-world datasets demonstrate the ascendancy of AutoMTN contra state-of-the-art methods in the next POI recommendation.

In recent years, traffic forecasting has gradually attracted attention in data mining because of the increasing availability of large-scale traffic data. However, it faces substantial challenges of complex temporal-spatial correlations in traffic. Recent studies mainly focus on modeling the local spatial correlations by utilizing graph neural networks and neglect the influence of long-distance spatial correlations. Besides, most existing works utilize recurrent neural networks-based encoder–decoder architecture to forecast multistep traffic volume and suffer from accumulative errors in recurrent neural networks. To deal with these issues, we propose the memory attention (MA) enhanced graph convolution long short-term memory network (MAEGCLSTM), a novel deep learning model for traffic forecasting. Specifically, MAEGCLSTM combines the MA and the vanilla graph convolution long short-term memory to capture global and local spatio-temporal dependencies, respectively. Then MAEGCLSTM utilizes a simplified GCLSTM to effectively fuse the global and local information. Moreover, we integrate the MAEGCLSTM into an encoder–decoder architecture to forecast multistep traffic volume. Besides MAEGCLSTM, we add the convolution neural network and encoder–decoder attention into the decoder to ease accumulative errors caused by iterative prediction and gain whole historical information from the encoder. Experiments on four real-world traffic data sets show that our model significantly outperforms by up to 6.07 % $6.07 \% $ improvement in L 1 $L1$ measure over 14 baselines.

Most traffic accidents are caused by bad driving habits. Online monitoring of the abnormal driving behaviors of drivers can help reduce traffic accidents. Recently, abnormal driving behavior recognition based on the sensors' data embedded in commodity smartphones has attracted much attention. Though much progress has been made about driving behavior recognition, the existing works cannot achieve high recognition accuracy and show poor robustness. To improve the driving behaviors recognition accuracy and robustness, we propose an algorithm based on Soft Thresholding and Temporal Convolutional Network (S-TCN) for driving behavior recognition. In this algorithm, we first introduce a soft attention mechanism to learn the importance of different sensors. The TCN has the advantages of small memory requirement and high computational efficiency. And the soft thresholding can further filter the redundant features and extract the main features. So, we fuse the TCN and soft thresholding to improve the model's stability and accuracy. Our proposed model is extensively evaluated on four real public data sets. The experimental results show that our proposed model outperforms best state-of-the-art baselines by 2.24%.

Traditional Wi-Fi-based floor identification methods mainly have been tested in small experimental scenarios, and generally, their accuracies drop significantly when applied in real large and multistorey environments. The main challenge emerges when the complexity of Wi-Fi signals on the same floor exceeds the complexity between the floors along the vertical direction, leading to a reduced floor distinguishability. A second challenge regards the complexity of Wi-Fi features in environments with atrium, hollow areas, mezzanines, intermediate floors, and crowded signal channels. In this article, we propose an adaptive Wi-Fi-based floor identification algorithm to achieve accurate floor identification also in these environments. Our algorithm, based on the Wi-Fi received signal strength indicator and spatial similarity, first identifies autonomous blocks parcelling the whole environment. Then, local floor identification is performed through the proposed Wi-Fi models to fully harness the Wi-Fi features. Finally, floors are estimated through the joint optimization of the autonomous blocks and the local floor models. We have conducted extensive experiments in three real large and multistorey buildings greater than 140 000 m$^2$using 19 different devices. Finally, we show a comparison between our proposal and other state-of-the-art algorithms. Experimental results confirm that our proposal performs better than other methods, and it exhibits an average accuracy of 97.24%.

Accurate location data of ground vehicles is very important for various intelligent transportation applications. As one of the most commonly-used navigation solutions at present, the GNSS/INS integrated navigation still cannot meet the accuracy and stability command of current applications. This paper presents a National Marine Electronics Association (NMEA) protocol data-assisted high-accuracy navigation algorithm based on GNSS position error estimation using multi-task learning (MT-e&R), which can accurately estimate the GNSS position error and GNSS measurement noise covariance matrix with the assistance of protocol data and a multi-task learning model. Extensive experimental results on practical navigation data collected in various urban environments of Beijing demonstrate that our proposed approach can improve the performance of integrated navigation system. The positioning errors of integrated navigation equipped with single-frequency receiver are reduced by 36.17% and 39.58% for double-frequency receiver, which confirms the reasonable environmental adaptability of our proposed MT-e&R algorithm.

Forecasting traffic flow and speed in the urban is important for many applications, ranging from the intelligent navigation of map applications to congestion relief of city management systems. Therefore, mining the complex spatio-temporal correlations in the traffic data to accurately predict traffic is essential for the community. However, previous studies that combined the graph convolution network or self-attention mechanism with deep time series models (e.g., the recurrent neural network) can only capture spatial dependencies in each time slot and temporal dependencies in each sensor, ignoring the spatial and temporal correlations across different time slots and sensors. Besides, the state-of-the-art Transformer architecture used in previous methods is insensitive to local spatio-temporal contexts, which is hard to suit with traffic forecasting. To solve the above two issues, we propose a novel deep learning model for traffic forecasting, named Locality-aware spatio-temporal joint Transformer (Lastjormer), which elaborately designs a spatio-temporal joint attention in the Transformer architecture to capture all dynamic dependencies in the traffic data. Specifically, our model utilizes the dot-product self-attention on sensors across many time slots to extract correlations among them and introduces the linear and convolution self-attention mechanism to reduce the computation needs and incorporate local spatio-temporal information. Experiments on three real-world traffic datasets, England, METR-LA, and PEMS-BAY, demonstrate that our Lastjormer achieves state-of-the-art performances on a variety of challenging traffic forecasting benchmarks.

Travel Time Estimation (TTE) is indispensable in intelligent transportation system (ITS). It is significant to achieve the fine-grained Trajectory-based Travel Time Estimation (TTTE) for multi-city scenarios, namely to accurately estimate travel time of the given trajectory for multiple city scenarios. However, it faces great challenges due to complex factors including dynamic temporal dependencies and fine-grained spatial dependencies. To tackle these challenges, we propose a meta learning based framework, MetaTTE, to continuously provide accurate travel time estimation over time by leveraging well-designed deep neural network model called DED, which consists of Data preprocessing module and Encoder-Decoder network module. By introducing meta learning techniques, the generalization ability of MetaTTE is enhanced using small amount of examples, which opens up new opportunities to increase the potential of achieving consistent performance on TTTE when traffic conditions and road networks change over time in the future. The DED model adopts an encoder-decoder network to capture fine-grained spatial and temporal representations. Extensive experiments on two real-world datasets are conducted to confirm that our MetaTTE outperforms nine state-of-art baselines, and improve 29.35% and 25.93% accuracy than the best baseline on Chengdu and Porto datasets, respectively.

With the ever-increasing intelligent perception capability of mobile terminals, the demand for fine-grained human activity recognition has become urgent. Transportation mode recognition, as a special branch of human activity recognition, plays a vital role in various mobile smart services. Some studies have been conducted on attitude-independence transportation mode recognition from various perspectives. However, it is challenging to recognize cross-place transportation mode due to the diversity of human activity and carrier deployment place. Towards this end, we propose a robust cross-place transportation mode recognition algorithm, which consists of three parts: Multi-Sensor Neural Network model, a variant bootstrap (ensemble learning) method, and data augmentation. The Multi-Sensor Neural Network model leverages multiple SF-SEDNets to extract spatial-temporal and spectral fusion features from different sensor combinations. The proposed data augmentation method addresses the unbalance data problem without introducing additional data collection costs, and the variant bootstrap method improves the robustness of our proposed algorithm. We evaluate our proposed algorithm on the Sussex-Huawei Locomotion-Transportation recognition challenge 2019 dataset. Extensive experimental results indicate that our algorithm achieves 81.45% macro-F1 score on the test dataset, which is 3.03% higher than that provided by the winning entry of this competition and is 14.85% higher than that of the official baseline.

State estimation with sensors is essential for mobile robots. Due to different performance of sensors in different environments, how to fuse measurements of various sensors is a problem. In this paper, we propose a tightly coupled multi-sensor fusion framework, Lvio-Fusion, which fuses stereo camera, Lidar, IMU, and GPS based on the graph optimization. Especially for urban traffic scenes, we introduce a segmented global pose graph optimization with GPS and loop-closure, which can eliminate accumulated drifts. Additionally, we creatively use a actor-critic method in reinforcement learning to adaptively adjust sensors’ weight. After training, actor-critic agent can provide the system better and dynamic sensors’ weight. We evaluate the performance of our system on public datasets and compare it with other state-of-the-art methods, which shows that the proposed method achieves high estimation accuracy and robustness to various environments. And our implementations are open source and highly scalable.

With the explosive development of pervasive computing and the Internet of Things (IoT), indoor positioning and navigation have attracted immense attention over recent years. Pedestrian dead reckoning (PDR) is a potential autonomous localization technology that obtains position estimation employing built-in sensors. However, most existing PDR methods assume that the smartphone is held horizontally and points to the walking direction. To solve reckoning errors caused by inconsistency of headings between walking heading and pointing of smartphone, we design an accurate and robust PDR method based on walking patterns, which is identified by multihead convolutional neural networks. In addition to adaptively adjust the threshold of step detection and select the most suitable step length model according to the results of walking pattern recognition, a novel heading estimation approach independent of device orientation is proposed. To mitigate accumulative errors, we proposed an online trajectory calibration method based on forward and backward magnetic fingerprint trajectory matching. We conduct extensive and well-designed experiments in typical scenarios, and the experimental results indicate that the 75th percentile localization accuracy of the three scenarios is 1.06, 1.08, and 1.22 m, respectively, using the commercial smartphone embedded sensor without any dedicated infrastructures or training data. Despite the intricate pedestrian locomotion, the proposed PDR method has great potential in pedestrian positioning.

In recent years, with the widespread popularity of smart mobile devices, seamless location-based service (LBS) between indoor and outdoor (IO) space has become an emerging requirement. As the direct indicator of location and contextual status, accurate indoor/outdoor status guarantees stability and continuity for human localization and activity recognition. Rich-sensor smartphones have made possible various context sensing. However, it is challenging to realize fast and accurate IO switching detection in dynamic and complex environments with variable sensory signals. Toward this end, we propose a multisensor deep learning model to predict the IO state, which consists of four parts: 1) sensor vectors; 2) 1-D sensor DenseNet; 3) LSTM; and 4) MLP. Unlike the traditional feature engineering and fixed threshold strategy, we utilize 1-D sensor DenseNet to extract higher level features for different sensors, and then utilize LSTM to capture signal change pattern in time series. Finally, we leverage the MLP to classify indoor/outdoor scenes. We evaluated our proposed architecture on two data sets, one for only differentiating IO scenes and one for detecting the indoor/outdoor switching scenes. Extensive experimental results show that our proposed method outperforms the fixed threshold method, traditional machine learning-based methods, and other existing neural network-based methods in terms of switching delay and recognition accuracy.

Person re-identification (ReID) suffers from camera view variants. Existing works, which typically learn a feature for each image, share a limitation that the learned features are single-view: each feature only contains information in one camera view. Thus, view bias occurs when matching pedestrians across camera views. In this paper, we seek to mitigate the view bias by generating multi-view features (fusion of features from a fixed number of cameras). To this end, we define the complementary-view features (complementary features to generate multi-view features with single-view features) and perform in-depth analysis. Based on this insight, we alleviate the view bias in testing and training, respectively. In testing, we present Multi-view Message Passing (MVMP), which generates multi-view features by aggregating single-view features from the neighborhood. In training, we propose Multi-view Feature Fusion Network (MFFN), which involves the single-view feature extractor and the complementary-view feature aggregator. MFFN makes the network sensitive to view-specific cues by adding constraints on multi-view features rather than single-view features. In addition, MVMP and MFFN have two key advantages: (1) They are parameter-free. (2) They can be applied to any Convolutional Neural Networks (CNNs) readily without extra supervision. Extensive experiments are conducted to validate the superiority of our method for person ReID over state-of-the-art methods on four benchmark datasets (Market-1501, DukeMTMC-reID, CUHK03, and MSMT17). The code is available at https://github.com/Yinsongxu/MVMP_MFFN.

In recent years, with the rapid development of public transportation, the ways people travel has become more diversified and complicated. Transportation mode detection, as a significant branch of human activity recognition (HAR), is of great importance in analyzing human travel patterns, traffic prediction and planning. Though many works have been devoted to transportation mode detection, there remains challenge for accurate and robust transportation pattern identification. In this paper, we propose a residual and LSTM recurrent networks-based transportation mode detection algorithm using multiple light-weight sensors integrated in commodity smartphones. Feature representation learning is adopted separately on multiple preprocessed sensor data using deep residual and LSTM network, which can enhance the identification accuracy and support one or more sensors. Residual units are introduced to accelerate the learning speed and enhance the accuracy of transportation mode detection. Furthermore, we also leverage the attention model to learn the significance of different features and different timesteps to enhance the recognition accuracy. Extensive experimental results on three datasets indicate that using our proposed model can achieve the best recognition accuracy for eight transportation modes including being stationary, walking, running, cycling, taking a car, taking a bus, taking a subway and taking a train, which outperforms other benchmark algorithms.

The IEEE 802.11mc-2016 protocol enables certified devices to obtain precise ranging information using time-of-flight-based techniques. The ranging error increases in indoor environments due to the multipath effect. Traditional methods utilize only the ranging measurements of the current location, thus limiting the abilities to reduce the influence of multipath problems. This article introduces a robust positioning method that leverages the constraints of multiple positioning nodes at different positions. We transfer a sequence of temporal ranging measurements into multiple virtual positioning clients (VPCs) in the spatial domain by considering their spatial constraints. Defining an objective function and the spatial constraints of the VPCs as Karush-Kuhn-Tucker conditions, we solve the positioning estimation with nonconvex optimization. We propose an iterative weight estimation method for the time of flight ranging and the VPC to optimize the positioning model. An extensive experimental campaign demonstrates that our proposal can remarkably improve the positioning accuracy in complex indoor environments.

The ability to accurately estimate a user's stride length plays a great important role in various applications. For a new target pedestrian or device, their heterogeneity dramatically reduces the performance of the current stride-length estimation (SLE) methods. To address the issue of heterogeneity, in this article, we propose an SLE method based on a long short-term memory (LSTM) network and denoising autoencoders (DAEs). The LSTM network is used to mine temporal dependencies and extract significant eigenvectors from the corrupted inertial sensor observations. Then, DAEs are adopted to automatically eliminate the inherent noise in eigenvectors and obtain denoised eigenvectors. Finally, a regression module maps the denoised eigenvectors to the resulting stride length. To mitigate the heterogeneity, we propose an unperceived model updating framework based on active online learning to establish a personalized model for a given target pedestrian or device. The proposed framework utilizes a magnetism-aided map-matching approach to automatically generate personalized training data and utilizes online learning technologies to evolve the stride-length model. The extensive experimental results demonstrate that the proposed method outperforms other state-of-the-art algorithms and achieves a promising accuracy with a stride-length error rate of 4.59% at a confidence level of 80%.

Indoor positioning methods have appeared to fulfill indoor location-based systems requirements, it is still a great challenge to obtain high precision results of indoor positioning. For example, fingerprint-based methods reach high performances but have a high cost for to survey the environment in order to collect sample and to maintain location fingerprints. Systems based on log-distance path loss model suffer from the multi-path problem and the adjustment of Wi-Fi station powers, and achieve low accuracy in complex environments. The appearance of fine time measurement protocol supported Wi-Fi access points provide a novel way to develop accurate indoor positioning algorithms. Considering the influence of the indoor multi-path effect to the fine time measurement ranging accuracy, we propose a multi-dimensional scaling based positioning algorithm to reduce the impact of ranging errors. We leverage the multidimensional scaling algorithm to estimate the rough position of positioning clients. Successively, adjusting the weight of fine time measurement ranging, we optimize the positioning results with the application of a SMACOF strategy. Through experiments conducted in a complex real-world scenario, we demonstrate that the system proposed reach an accuracy below the 2.5 meters at 80% of the cases.

Pedometer is an enabling technique for smartphone- based pedestrian positioning systems. Because the sensor drifts, these algorithms can only estimate moving distances from step counts. In order to detect step events, researchers have tried to leverage the peak detection and the periodicity attribute of step acceleration signals. However, many human behaviors are having acceleration peaks and periodic, causing traditional detectors error- prone when the phone is shaken periodically leading state-of-the-art system to high false positive ratio and consequently to big mistake of distance estimations. Based on the acceleration feature analysis of step events, we present a deep convolution neural network based step detection scheme to improve the pedometer robustness. Finally, the proposed step detection algorithm is tested in a realistic situation, showing a high anti periodic negative-step movement capability.

Smartphone based indoor positioning has become a hot topic in pervasive computing, because of the need to improve indoor location-based services. In order to strengthen positioning accuracy, researchers have tried to leverage high-resolution magnetic fingerprint with particle filter and dynamic time warping (DTW). These approaches are computation-hungry, which increases hardware cost for positioning companies. By analyzing magnetic features for pedestrian users, we present a mass-centered weight update scheme to decrease calculation overheads. Finally, the proposed positioning algorithm is tested in a realistic situation, showing high-quality localization capability.

With the increasing sensing ability of smartphone, both recognizing and understanding a user's activity using sensor data have become a popular topic of ubiquitous computing systems. Individual transportation mode identification can provide essential data for road planning and traffic management. In this paper, we present a Convolutional Neural Networks (CNN) based method to extract expressive and discriminative features automatically for transportation mode identification. The signal preprocessing in the time and frequency domain is performed before the sensor data is fed into the deep learning framework. We optimize various important hyper-parameters such as learning rate, kernel size and number of convolutional layers to adapt the characteristics of multiple sensor signals. Extensive experimental results indicate that the proposed CNN based transportation mode identification algorithm can achieve 98% accuracy to distinguish between car, bus, train and metro, which outperforms the Support Vector Machines and Adaboost based transportation identification with better robustness and generalization.

Identifying different floors in multistory buildings is a very important task for precise indoor localization in industrial and commercial applications. The accuracy from existing studies is rather low, especially in multistory buildings with irregular structures such as hollow areas, which is common in various industrial and commercial sites. As a better solution, this paper proposes a hybrid floor identification (HYFI) algorithm, which exploits wireless access point (AP) distribution and barometric pressure information. It first extracts the distribution probability of APs scanned in different floors from offline training fingerprints and adopts Bayesian classification to accurately identify floor in well-partitioned zones without hollow areas. The floor information obtained from wireless AP distribution is then used to initialize and calibrate barometric pressure-based floor identification to compensate variable environmental effects. Extensive experiments confirm that the HYFI approach significantly outperforms purely wireless fingerprinting-based or purely barometric pressure-based floor identification approaches. In our field tests in multistory facilities with irregular hollow areas, it can identify the floor level with more than 96.1% accuracy.

With the rapid increase of location based services, various indoor positioning technologies have emerged. The existing indoor positioning technologies have different characteristics in localization accuracy, real-time performance, coverage and cost. To meet the requirements of high accuracy, low cost, and broad coverage in complex indoor scenes, we design an indoor localization system WiMag: Multimode Fusion Localization System based on Magnetic/WiFi/PDR. Based on the particle filter framework, the system optimally selects the fusion strategy to combine the location results according to the identified smartphone status. The experimental results demonstrate that the proposed WiMag outperforms all single indoor positioning technologies with higher accuracy (1.6m average localization error), wider localization coverage and better robustness.

Personal dead reckoning (PDR) localization technology can provide effective and critical assistance for public security, such as emergency rescue or anti-terror training in the indoor or underground environment without the need of deploying additional positioning infrastructure. However, the PDR suffers from the severe position error accumulation with time due to the inaccurate step length and moving direction estimation. To improve the self-positioning accuracy, this paper proposed a novel indoor self-localization algorithm using two kinds of automatic calibration methods, i.e., opportunistic magnetic trajectory matching and indoor landmark identification. Extensive experiments performed in two representative indoor environments, including an office building and a supermarket, demonstrate that the proposed self-localization algorithm can obtain an 80 percentile localization accuracy of 1.4m and 2m in the two representative indoor environments, respectively, which outperforms the art-of-the-state PDR algorithms.

In this article, we address the problem of how to make the wireless device-to-device (D2D) video multicast systems have better quality provision with consideration of internet-of-things (IoT) applications. We propose an opportunistic transmission and fair resource allocation framework, including joint application-layer and physical-layer transmission and optimization. First, we use a parallel subchannels structure by concatenating the Fountain codes and diversity-embedded space-time block codes to provide reliable and flexible transmission in heterogeneous circumstances. Second, we exploit the quality of heterogeneous user experience (quality of experience) metric under D2D video multicast systems, with consideration of various channel states, device capability, video content urgency, and the number of demanding users. Third, we formulate reliable multiple video streams broadcasting to heterogeneous devices as an aggregate maximum utility achieving problem, and we use opportunistic scheduling to select suitable users in each transmission interval to improve the broadcasting utility. Fourth, we use the utility fair scheme to guide rate allocation among multicontent video multicast. Extensive performance comparison and analysis are presented to demonstrate efficiency of the proposed solution.

Presently existing lightweight indoor/outdoor detection schemes on phones acquire accuracy by sensing variations of ambient physical environmental properties with inherent sensors on mobile phones, with which, however, the detection scheme cannot work well in some ambient environments, where the variations are not very observable. This detection scheme is with very high dependency on light. The I/O detector does not work well in poor lighting or fast changing lighting settings, therefore the I/O detection is very much challenged at times like dawn, dusk, or night. The target of this paper is finding a pervasive detection scheme independent of physical environments. In this paper, we present MobiIO, an lightweight indoor and outdoor detection scheme based on analyses of human activities. By recording human indoor and outdoor motion activities with sensors, typical features of their activities are extracted. We compare assorted combinations or groupings of various properties with SVM classifier. We classify indoor/outdoor settings through classifiers like SVM, Bayes, decision trees, HMM and compare the effects of classification in between various classifying algorithms.

The presentation of Bluetooth Low Energy (BLE; e.g., Bluetooth 4.0) makes Bluetooth based indoor positioning have extremely broad application prospects. In this paper, we propose a received signal strength indication (RSSI) based Bluetooth positioning method. There are two phases in the procedure of our positioning: offline training and online locating. In the phase of offline training, we use piecewise fitting based on the lognormal distribution model to train the propagation model of RSSI for every BLE reference nodes, respectively, in order to reduce the influence of the positioning accuracy because of different locations of BLE reference nodes. Here we design a Gaussian filter to pre-process the receiving signals in different sampling points. In the phase of online locating, we use weighted sliding window to reduce fluctuations of the real-time signals. In addition, we propose a distance weighted filter based on triangle trilateral relations theorem, which can reduce the influence of positioning accuracy due to abnormal RSSI and improve the location accuracy effectively. Besides, in order to reduce the errors of targets coordinates caused by ordinary least squares method, we propose a collaborative localization algorithm based on Taylor series expansion. Another important feature of our method is the active learning ability of BLE reference nodes. Every reference node adjusts its pre-trained model according to the received signals from detecting nodes actively and periodically, which improve the accuracy of positioning greatly. Experiments show that the probability of locating error less than 1.5 meter is higher than 80% using our positioning method.

Crowdsourcing-based localization has attracted wide research concern to the metropolitan-scale positioning. However, crowdsourcing-based fingerprints collection with assorted mobile smart devices brings fingerprint confusion, which significantly degrades the localization accuracy. To solve the device diversity problem, many solutions have been raised like the Device-Clustering algorithm. Based on macro Device-Cluster (DC) rather than natural device, DC algorithm maintains less device types and slight calibration overhead. Despite high positioning accuracy, the selection of suitable clustering algorithms in DC system becomes another puzzle. In this paper, we reshape the novel Device-Clustering algorithm to enhance the indoor positioning by comparing the application of different clustering algorithms. The experimental result indicates the reliability of DC strategy in broad clustering scheme as well as the suitable locating process corresponding to distinct environment.

This paper presents a LBS middleware software, which contains application and positioning modules. The first module is designed to improve the granularity of the pushed information in current LBS systems and to conveniently add new applications. Taking the advantage of PUB/SUB system's asynchronous, loosely-coupled and multiplex communication mechanism, the middleware pushes LBS information with its granularity refined. By combining position information and the PUB/SUB paradigm, users can easily focus on his points of interest. Positioning module enables the system of handling both indoor and outdoor scenario. And the system can easily add new indoor positioning server and dispatch work load to them judging by their current work load. Experimental results showed that the middleware can offer an efficient service to users and system performance can be improved under massive requests.

We present a novel approach to indoor wireless localization using label propagation based on semi-supervised learning. Our aim is to reduce the effort of collecting labeled data in the offline training phrase, which are expensive to obtain. This learning algorithm combines labeled and unlabeled data in learning process to fully realize a global consistency assumption: similar data should have similar labels, which has intimate connections with random walks to propagate label through the dataset along high density areas defined by unlabeled data. We test our algorithm in 802.11 wireless LAN environments, and demonstrate the advantage of our approach in both accuracy and its ability to utilize a much smaller set of labeled training data.

Most existing mobile beacon-assisted localization algorithms do not make effective use of the node distribution information and let the mobile landmark traverse the entire network, which causes large path length and low utilization rate of beacon messages. In order to reduce the path length and use the mobile beacon more effectively, a novel mobile beacon-assisted localization algorithm based on network-density clustering (MBL(ndc)) for wireless sensor networks is presented, which combines node clustering, incremental localization and mobile beacon assisting together. Simulation results demonstrate that the proposed MBL(ndc) algorithm offers comparable localization accuracy as the mobile beacon-assisted localization algorithm with HILBERT trajectory, but with less than 50% path length of the later, which shortens the period of positioning the whole network.

For the rate control of H. 264/AVC, one of the most important things is to get the statistics of the current frame accurately. To achieve this, a novel adaptive coding characteristics prediction scheme is presented to improve the accuracy of R-D modeling, by exploiting spatio-temporal correlations. With the proposed prediction scheme, we present a novel rate function and a linear distortion model, and then deduce a simple close-form solution to the problem of optimum bit allocation, just in a TMN-8-alike way. Extensive experiments show that improvements with gains up to 0.92dB per frame over JVT-G012, the current standardized rate control scheme, are achieved by the proposed scheme for a variety of test sequences with less demanding bandwidth.