Locomotion activity recognition (LAR) is important for a number of applications, such as indoor localization, fitness tracking, and aged care. Existing methods usually use handcrafted features, which requires expert knowledge laborious, the achieved result might still be suboptimal. To relieve burden designing selecting we propose deep learning method LAR by using data from multiple sensors available on most smart devices. Experimental results show that proposed method, learns useful...

Robust and accurate step counting is important for indoor localization algorithms that rely on smartphone inertial sensors. Existing solutions do not consider users' false walking state (e.g., when a user in still uses her phone texting, playing games, watching movies), which results the over-counting problem. In this paper, we propose robust algorithm to solve overcounting problem caused by walking. Experimental show proposed outperforms commonly-used peak detection-based method can improve...

Pedestrian dead reckoning (PDR) is a popular indoor localization method due to its independence of additional infrastructures and the wide availability smart devices. Step length estimation key component PDR, which has an important influence on performance PDR. Existing step models suffer from various limitations such as requiring knowledge user's height, lack consideration varying phone carrying ways, dependence spatial constraints. To solve these problems, we propose deep learning-based...

Logs that record system abnormal states (anomaly logs) can be regarded as outliers, and the k-Nearest Neighbor (kNN) algorithm has relatively high accuracy in outlier detection methods. Therefore, we use kNN to detect anomalies log data. However, there are some problems when using anomalies, three of which are: excessive vector dimension leads inefficient algorithm, unlabeled data cannot support imbalance number distorts classification decision algorithm. In order solve these problems,...

Location estimation is significant in mobile and ubiquitous computing systems. The complexity smaller scale of the indoor environment impose a great impact on location estimation. key lies representation fusion uncertain information from multiple sources. improvement complicated comprehensive issue. A lot research has been done to address this However, existing typically focuses certain aspects problem specific methods. This paper reviews mainstream schemes improving levels perspectives by...

The utility and adoption of indoor localization applications have been limited due to the complex nature physical environment combined with an increasing requirement for more robust performance. Existing solutions this problem are either too expensive or dependent on infrastructure such as Wi-Fi access points. To address problem, we propose APFiLoc-a low cost, smartphone-based framework localization. key idea behind is obtain landmarks within use augmented particle filter fuse them...

Indoor localization techniques using Wi-Fi fingerprints have become prevalent in recent years because of their cost-effectiveness and high accuracy. The most common algorithm adopted for fingerprinting is weighted K-nearest neighbors (WKNN), which calculates neighboring points to a mobile user. However, existing WKNN cannot effectively address the problems that there difference observed AP sets during offline online stages also not all K are physically close In this paper, similarity...

Coarse registration of 3D point clouds plays an indispensable role for parametric, semantically rich, and realistic digital twin buildings (DTBs) in the practice GIScience, manufacturing, robotics, architecture, engineering, construction. However, existing methods have prominently been challenged by (i) high cost data collection numerous (ii) computational complexity from self-similar layout patterns. This paper studies two low-cost sets, i.e., colorful captured smartphones 2D CAD drawings,...

The traditional construction industry is characterized by high energy consumption and significant carbon emissions, primarily due to its reliance on on-site manual labor wet operations, which are not only low in mechanization but also result material efficiency substantial waste. Prefabricated offers a new solution with efficient production methods, significantly enhancing utilization efficiency. This paper focuses the scheduling optimization of prefabricated components. directly affects...

The recognition of locomotion activities (e.g., walking, running, still) is important for a wide range applications like indoor positioning, navigation, location-based services, and health monitoring. Recently, there has been growing interest in activity using accelerometer data. However, when utilizing only acceleration-based features, it difficult to differentiate varying vertical motion states from horizontal especially conducting user-independent classification. In this paper, we also...

Indoor localization is important for a variety of applications, such as location-based services, mobile social networks, and emergency response. Fusing spatial information an effective way to achieve accurate indoor with little or no need extra hardware. However, the existing methods that make use are either computationally expensive sensitive completeness landmarks. In this article, we propose novel, low-cost, high-accuracy method based on landmark graph. The experimental results show...

Wi-Fi fingerprinting is widely used in indoor localization due to the ubiquitous availability of infrastructure environments. The basic assumption that received signal strength indicator (RSSI) distance consistent with location distance. However, fluctuation signals environments, nearest neighbors selected using RSSI may not be those whose corresponding locations are target, which could lead a large error. In this paper, we propose novel method for by transforming raw into features learned...

A large proportion of indoor spatial data is generated by parsing floor plans. However, a mature and automatic solution for generating high-quality building elements (e.g., walls doors) space partitions rooms) still lacking. In this study, we present two-stage approach to mapping modeling (IMM) from plan images. The first stage vectorizes the on images second repairs topological inconsistencies between elements, separates spaces, generates maps models. To reduce shape complexity boundary...

In cloud computing, how to reasonably allocate computing resources for batch jobs ensure the load balance of dynamic clusters and meet user requests is an important challenging task. Most existing studies are based on deep Q network, which utilizes neural networks estimate expected value cumulative return in scheduling process. The value-based DQN algorithms ignore complete information contained distribution lack strong adaptability time-varying cluster resources. Therefore, capture inherent...

According to the European Union, buildings account for 40% of overall energy use and 36% CO2 emissions, with existing energy-inefficient main source losses. Efforts enhance thermal comfortability users in can result overheating if not appropriately designed, which turn could lead health issues as well continual emission greenhouse gases. To provide further insights into this dilemma contribute improving energy-efficient building designs, we undertake a new modeling technique encompassing...

Indoor localization has become a hot topic in recent years because of its wide applications. Map matching is popular method used to improve the accuracy without adding hardware. However, existing map methods are usually computationally expensive, leading unsuitability running on resource-limited devices such as smartphones. In this paper, we present an efficient system for indoor localization, called HTrack, which uses hidden Markov model, considering user's heading and spatial information....

Abstract Semantically rich maps are the foundation of indoor location‐based services. Many map providers such as OpenStreetMap and automatic mapping solutions focus on representation detection geometric information (e.g., shape room) a few semantics stairs furniture) but neglect room usage. To mitigate issue, this work proposes general tagging method for public buildings, which can benefit both existing by inferring missing usage based maps. Two kinds statistical learning‐based methods...

Technologies and systems for indoor positioning, mapping, navigation (IPMN) have rapidly developed over the latest decade due to advanced radio light communications, internet of things, intelligent smart devices, big data, so forth. Thus, a group surveys IPMN technologies, systems, standards, solutions can be found in literature. However, currently there is no proposed solution that satisfy all application requirements; one biggest challenges lack standardization, even though several...

Indoor localization is important for a variety of applications such as emergency response, shopping guide, and location-based services. Localization based on smartphone inertial sensors one the most widely-used indoor techniques since it can provide continuous real-time locations without requiring additional infrastructure. However, suffers from accumulated error problem, which be addressed by using sensory landmarks. In this paper, we first introduce concept landmarks, then show how...