Mobile Data Research Articles

Urban mobility and air pollution at the neighbourhood scale in the Megacity of São Paulo, Brazil

Developing countries’ megacities, characterised by dense populations and socioeconomic disparities, often face high levels of pollutants, mainly from vehicle emissions. Poor air quality can lead to a range of public health problems which in turn, need to be addressed by public policies. The main goal of this article is to examine the impact of public policies and the influence of transport modes on air pollution of three districts from the Metropolitan Region of São Paulo, in Brazil: Pinheiros, Parque Dom Pedro II and Taboão da Serra. The method was held through a comparative analysis, which in turn, took into account urban indicators, urban mobility data and pollutants levels (CO and NOx). These data were collected from the 2007 and 2017 São Paulo Metrô Origin and Destination Surveys and the Air Quality Database of the Environmental Company of the State of São Paulo. As key findings, the three study areas’ pollutant concentrations presented a downward trend from 2007 to 2017 as the same time there was an increase in the public transport and non-motorized transport modes. However, it is important to highlight the confluence of the state and federal public policies occuring at the same period such as PRONCOVE, Rodoanel and the Yellow Subway Line. The identified socio-environmental disparities in the urban realm highlight the importance of localised analyses in order to reveal problems and opportunities to get a better response in terms of urban mobility and air quality. Thus, the perpetuation of constant policy’s updates and interdisciplinary collaboration is crucial.

Comparing the impacts of built environment across different objective life neighborhoods on the out-of-home leisure activities of employed people using massive mobile phone data

An adequately-designed built environment can prompt willingness of employed people to carry out out-of-home leisure activities. However, since much of the literature relies on static or perceived residential neighborhoods which may have substantial deviations from realistic environmental exposure, some associations between out-of-home leisure activities and built environment may be biased. In this study, we develop three kinds of objective life neighborhood measurements (residential space, job-housing space, and activity space) using massive mobile phone data, to distinguish the differentiated impacts of built environment across different objective life neighborhoods on the out-of-home leisure activities. According to the results, the activity space makes the exploratory models far more reasonable and persuasive than other life neighborhoods, especially in urban development areas, and the benefits of activity space cannot be attained solely by expanding the distance parameters of residential space or job-housing space. Furthermore, the results demonstrate the relative importance of built environment for out-of-home leisure activities, especially the factors of population density, proportion of urban green space, and metro coverage rate, may be underestimated if the ordinary life neighborhoods are applied. This study provides some references for urban planners to formulate effective environmental intervention strategies to maintain active and healthy lifestyles of employed people.

Adaptive Caching Strategies for IoV Based on LTE Signal Quality Analysis

The Internet of Vehicles (IoV) enhances vehicular communication, promising better connectivity and user experiences. However, maintaining stable data delivery in dynamic vehicular environments is challenging. This paper introduces Mob-Cach, a novel mobility-aware cache management strategy that leverages real-time vehicular mobility information. Our preliminary study on LTE signal quality shows that mobility patterns significantly impact signal degradation. Based on these findings, we developed a dynamic caching framework that adapts to real-time mobility data, optimizing content delivery and network performance. Simulations demonstrate significant improvements in cache hit rates, reduced latency, and overall network efficiency. This research lays a foundation for future advancements in IoV networks, addressing the challenges of mobile and dynamic network topologies.

A nowcasting model of industrial production using alternative data and machine learning approaches

Recent years have seen a growing trend to utilize "alternative data" in addition to traditional statistical data in order to understand and assess economic conditions in real time. In this paper, we construct a nowcasting model for the Indices of Industrial Production (IIP), which measure production activity in the manufacturing sector in Japan. The model has the following characteristics: First, it uses alternative data (mobility data and electricity demand data) that is available in real-time and can nowcast the IIP one to two months before their official release. Second, the model employs machine learning techniques to improve the nowcasting accuracy by endogenously changing the mixing ratio of nowcast values based on traditional economic statistics (the Indices of Industrial Production Forecast) and nowcast values based on alternative data, depending on the economic situation. The estimation results show that by applying machine learning techniques to alternative data, production activity can be nowcasted with high accuracy, including when it went through large fluctuations during the spread of the COVID-19 pandemic.

Prediction model establishment for residential community occupancy considering urban environment

The accurate prediction of occupancy rates within the residential sector is pivotal for urban and energy planning and management. This study aims at addressing the often-neglected influence of regional variations within urban settings on occupancy rates by employing multi-source urban data and explainable machine learning algorithms. Utilizing mobile signaling data from January and July 2022 in Guangzhou, we developed prediction models using the CatBoost algorithm with hyperparameters finely tuned via Bayesian optimization. The SHapley Additive exPlanation was applied to elucidate the influence of different urban configurations. Our findings reveal significant seasonal variations and distinct differences between weekdays and weekends, showing higher occupancy rates in winter than in summer and elevated rates during weekends. Additionally, significant fluctuations were observed among different age groups on weekdays. The prediction models exhibited robust capabilities, achieving R2 values exceeding 0.95 for weekdays and 0.91 for weekends. Key findings highlight that the density of points of interest related to shopping and food, along with variability in building heights, significantly impact occupancy rates. The influence of diverse urban configurations varies across different hours of the day. By the comprehensive analysis of regional variations and the application of advanced machine learning techniques to predict and explain residential occupancy rates, this research offers tools to optimize the management and sustainability of built environments, enhances the understanding of urban operation patterns and contributes to advancing smart city technologies, providing valuable insights for urban and energy planning.

Unlocking Insights: The Imperative of Real-Time Data Collection and Mapping Systems

In today's rapidly evolving landscape, the need for timely and accurate data is paramount for informed decision-making across various sectors. However, traditional data collection methods often fall short in providing real-time insights necessary for addressing dynamic challenges. This study examines the critical problem of outdated data and the imperative for adopting real-time data collection and mapping systems to overcome it. The aim of the article is to highlight the transformative potential of such systems in unlocking actionable insights for decision-makers. By leveraging advanced technologies like geospatial mapping, sensor networks, and mobile data collection apps, organizations can gather, process, and analyze data in near real-time. This enables them to monitor trends, detect anomalies, and respond promptly to emerging issues. Ultimately, the adoption of real-time data systems offers a solution to the problem of outdated data, empowering decision-makers with the timely information needed to navigate complex scenarios and drive positive outcomes

Understanding the role of mobility in the recorded levels of violent crimes during COVID-19 pandemic: a case study of Tamil Nadu, India

Purpose/GoalThis research investigates the potential link between mobility and violent crimes in Tamil Nadu, India, using an empirical study centred on the COVID-19 pandemic waves (2020–2022). The goal is to understand how these events influenced crime, employing a counterfactual approach.MethodsThe study employs the XGBoost algorithm to forecast counterfactual events across different timeframes with varying levels of mobility. The mobility data sources include historical bus and passenger records spanning a decade, along with Google Community Mobility Reports added during the pandemic phases. The foundation for crime analysis is built upon the univariate time series of violent crimes reported as First Information Reports from 2010 to 2022.FindingsResults indicate a significant correlation between mobility and violent crimes when mobility drops below a specific threshold. However, no such correlation is observed when mobility is above this threshold during the non-pandemic periods. The COVID-19 pandemic had a major impact on people’s and vehicular mobility, especially during the complete lockdown periods of the first two waves, and also affected crime rates.ConclusionsThe decrease in recorded incidents could also be attributed to fewer criminal opportunities. Additionally, this could be due to unfavourable situational factors, such as victims’ limited access to appropriate health and law enforcement agencies to report crimes. Furthermore, frontline services were busy with pandemic-related commitments, which could have contributed to a lack of crime registration even when crimes were committed.

POI recommendation by deep neural matrix factorization integrated attention-aware meta-paths

With the continuous accumulation of massive amounts of mobile data, point-of-interest (POI) recommendation has become a vital task for location-based social networks. Deep neural networks or matrix factorization (MF) alone are challenging to effectively learn user–POI interaction functions. Moreover, the user–POI interaction matrix is sparse, and the heterogeneous characteristics of auxiliary information are underused. Therefore, we propose an innovative POI recommendation method that integrates attention-aware meta-paths based on deep neural matrix factorization (DNMF-AM). First, we develop a multi-relational heterogeneous information network of “user–POI–geographic region–POI category.” Multiple-weighted isomorphic information networks based on meta-paths are employed to obtain node-embedding vectors across different relationships. Attention networks are employed to aggregate node vectors across various relationships and serve as auxiliary information to mitigate the challenges of data sparsity. Subsequently, the internal embedding vectors of the users and POIs are extracted using feature embedding based on the user–POI interaction matrix. Second, these vectors are integrated with the embedding vectors obtained by aggregating the attention networks. Third, deep neural matrix factorization is used to learn linear and nonlinear user–POI interactions to mitigate the implicit feedback problem. This outcome is achieved using generalized matrix factorization and convolution-constrained multi-head self-attention mechanism deep neural networks. Extensive experiments conducted on two real-world datasets demonstrate that the DNMF-AM outperforms the optimal baseline NeuMF-CAA by 4.24% and 5.04% in terms of HR@10 and NDCG@10, respectively.

Synergizing intelligent signal processing with wavelength-division multiplexing for enhanced efficiency and speed in photonic network communications

Abstract The explosive growth of worldwide mobile data traffic seeks innovations in communication technology to cater to the mounting need for rapid connectivity, high-capacity connections. The mainstreaming of 5G technologies for communication is a dramatic step towards meeting the aforementioned goals, with the ability for reshaping IoT (Internet of things), D2D (device-to-Device) communications, and the smart grids. This work conveys an in-depth study of the fundamental innovations that underlie 5G, including full-duplex distribution, huge multiple-input-multiple-output, ultra-dense connections, the phenomenon of beamforming and millimeter-wave approaches. A special emphasis is focused on the integration of photonic technologies, or microwave photonics, which serves as a critical multidisciplinary study topic. Optical fibers, with their tremendous bandwidth and capacity, have been determined as the best medium for backhaul and fronthaul amenities, outpacing conventional copper cables to accommodate tiny cells and next-generation networks. The synergy between optical and wireless access technologies is analyzed with the emphasis on the central role of wavelength-division multiplexing (WDM) for improving network efficiency and speed. The investigation additionally explores the possibility of intelligent signal processing methods combined with WDM to optimize photonic network communications. The mingling of these technologies anticipates producing unrivaled levels of performance, rupturing the path for an additional intelligent, interconnected era.

Tensor Decomposition for Spatiotemporal Mobility Pattern Learning with Mobile Phone Data

Detecting urban mobility patterns is crucial for policymakers in urban and transport planning. Mobile phone data have been increasingly deployed to measure the spatiotemporal variations in human mobility. This work applied non-negative Tucker decomposition (NTD) to mobile phone-based origin–destination (O-D) matrices to explore mobility patterns’ latent spatial and temporal relationships in the province of Liège, Belgium. Four [Formula: see text] traffic tensors have been built for one regular weekday, one regular weekend day, one holiday weekday, and one holiday weekend day, respectively. The proposed method inferred spatial clusters and temporal patterns while interpreting the correlation between spatial clusters and temporal patterns through geographical visualization. As a result, we found the similarity of O-D and destination–origin (D-O) patterns and the symmetry for the trips of the temporal patterns with evening peak and morning peaks on the weekday. Moreover, we investigated the attraction of different spatial clusters with two temporal patterns on a regular weekday and validated the reconstructed demand using population counts and commuting matrices. Finally, the differences in spatial and temporal interactions have been addressed in detail.

Assessing the impact of summer heat on the movement of people in Tokyo based on mobile phone location data

Using human location data from mobile phones, we developed a method to detail transportation modes around Shinagawa Station in Tokyo and analyzed the impact of hot weather conditions on traffic behavior. Each user's movement history was collected based on the global positioning system (GPS) communication of applications. We estimated the modes of transportation (train, car, walking, and stationary) for each user. By conducting route searches based on each user's data, we calculated the traffic volume for each mode of transportation, linking the estimated population data to network data and confirming accuracy against official census data. In August 2019, we examined the relationship between the estimated transportation modes and the hourly wet-bulb globe temperature (WBGT), which was calculated by the Ministry of the Environment using values measured by the Tokyo Regional Meteorological Observatory. The results indicated that as WBGT increases, fewer pedestrians walk and more remain stationary. Sensitivity to WBGT was four times greater at 20:00 p.m. than at 08:00 a.m., despite similar WBGT levels. This may be because fewer people go out during the hottest part of the day, and the sum of those who go out during the day, i.e., those who return home, is greatly reduced. However, this trend was not observed on weekdays. The results suggest that the impact of summer heat is latent during weekdays due to the demands of work and errands but becomes more pronounced during free time, such as holidays and nighttime.

Towards a Framework to Evaluate Generative Time Series Models for Mobility Data Features

Towards a Framework to Evaluate Generative Time Series Models for Mobility Data Features

Developing a system for the real-time collection and analysis of mobile vehicle emission data

In this study, an integrated system for remotely monitoring air pollution data was developed. This system combines vehicle exhaust emission data with driving information. It comprises an exhaust extraction device (EED), an improved On-Board Diagnostic II (OBDII) device, a mobile app, and a backend server database. The EED detects exhaust gases such as NOx, COx, and PM2.5 and transmits this information to the OBDII device through 2.4G communication. The OBDII device gathers vehicle driving information through a CAN 2.0 interface, integrates this information with vehicle exhaust emission data, and transmits the integrated information to the app through Bluetooth. The app forwards the integrated data to the backend database through 4G/5G communication for storage. Overall, this system provides a platform for the comprehensive collection, analysis, and application of big data pertaining to vehicle pollution sources and driving behavior. Analysis of actual driving experiments involving three vehicles revealed a correlation of over 80 % between vehicles’ revolutions per minute and CO2 and PM2.5 emissions, indicating that driving behaviors affect vehicle exhaust emissions. In summary, the proposed system not only stores relevant driving information in real time but also provides a platform for the future analysis of driving behavior and vehicle exhaust emissions.

Green infrastructure and community health: exploring the characteristics of campus users in three university towns in Texas

ABSTRACT Green infrastructure in college campuses and their surrounding communities can bring public health benefits. Using green infrastructure measures in three university towns in Texas, United States, this study explores if and how green infrastructure can impact the physical and mental health status of university campus users in the community. Campus users were estimated as the number of visitors to campus using the SafeGraph mobility data, green infrastructure was measured with land cover and Landsat 8 satellite images, and health status was measured with the CDC health status data. The results of spatial error models suggest a significant moderating effect of tree canopy coverage indicating that green infrastructure could potentially benefit campus users’ health. This gives indirect evidence supporting the health-significant roles of the campus environment. For city planners and policymakers, increasing tree canopy coverage stands out as a promising green infrastructure planning and management strategy that can contribute to the health of campus users/communities.

Cost analysis of IPv6 distributed mobility management protocols in comparison with TFMIPv6.

The past decade has witnessed a significant evolution in the role of the Internet, transitioning from individual connectivity to an integral aspect of various domains. This shift has prompted a move in IP paradigms from hierarchical to distributed architectures characterized by decentralized structures. This transition empowers efficient data routing and management across diverse networks. However, traditional distributed mobility management (DMM) protocols, reliant on tunneling mechanisms, incur overheads, costs, and delays, exacerbating challenges in managing the exponential growth of mobile data traffic. This research proposes Tunnel-Free Mobility for IPv6 (TFMIPv6) as a solution to address the shortcomings of existing DMM protocols. TFMIPv6 eliminates the need for tunneling, simplifying routing processes and reducing latency. A comprehensive cost analysis and performance evaluation are conducted, comparing TFMIPv6 with traditional protocols such as MIPv6, PMIPv6, FMIPv6, and HMIPv6. The study reveals significant improvements with TFMIPv6. Signaling costs are reduced by 50%, packet delivery costs by 23%, and tunneling costs are completely eliminated (100%). Real-world network traffic datasets are used for simulation, providing statistical evidence of TFMIPv6's efficacy in supporting an uninterrupted movement of IPv6 data across networks.

Quantifying local mobility patterns in urban human mobility data

Understanding fine-scale dynamics of human mobility patterns is pivotal for effective urban planning, public health strategies, and retail analysis. This study introduces a novel mobility measure – the Local Mobility Index (LMI) – combining geometry-based mobility metrics and accessibility measures. The LMI can be considered a measure of ‘relative localness’ by integrating preferences into the assessment of local mobility patterns, offering a novel measure for understanding mobility behavior in urban contexts. The LMI improves upon existing measures as it captures individual choice for local destinations through measuring whether individuals select nearby destinations; accounting for the unequal spatial distribution of urban amenities. Our contribution is mainly methodological, advancing the field by introducing a metric that captures different aspects of mobility compared to conventional mobility metrics. Leveraging mobile-phone-based GPS data, we examine the LMI using 759 individuals across three cities in England. We found that the LMI captures a new and distinct dimension of urban mobility, as evidenced by its weak correlation with established metrics. Therefore, LMI's capacity to highlight previously undetected aspects of mobility behavior, underscores its importance for advancing research and urban planning.

Digital payments and relational embedding: Turning relations into data and data into relations

Research on the digital economy has highlighted the assetization of data. This article argues for expanding existing research on data and datafication processes by focusing on how relationships are made and unmade through and from data. We introduce a general analytic model of “relationing” and show how relationships between users, companies, and products are created in three different moments—entanglement, dissection, and matching—first in the digital economy, then in physical stores. We show how payments with mobile phones connect the digital to the brick-and-mortar economy. Applying our model, we illustrate how a mobile phone's various data streams, money's record-keeping function, and retailers’ loyalty programs produce qualitatively and quantitatively new relations between customers, retailers, banks, app providers, and payment intermediaries. We argue that “relational embedding” captures the inherent relationality between users, their data points, and other economic actors: algorithmically relating users’ data profiles to other users’ profiles yields personalized recommendations, ads, or rebates, continuing the relationship between retailers and customers.

A mobile data collection method for balancing energy consumption and delay in strip-shaped wireless sensor networks with branches

Strip-shaped Wireless Sensor Networks (WSNs) with branches are commonly used in various long and narrow applications, such as mines, factories, subways, and pipelines, and they face serious energy hole problems caused by multi-hop communication. The Mobile Data Collector (MDC) can alleviate the energy hole problem. Current solutions have two limitations: one is the balance between energy consumption and delay, and the other is the overly ideal network model, e.g., the square region or circular area. This paper focuses on strip-shaped networks and proposes a novel mobile data collection method to find a trade-off between energy preservation and data delivery delay. Firstly, the MDC path is planned by solving the diameter of the tree in the network, resulting in reduced delay. Secondly, the network energy consumption is further reduced by clustering and optimal transmission distance adjustment. Then, a network lifetime balancing mechanism is designed to balance network energy between backbone and branches. Finally, the performance of the algorithm proposed in this paper has been studied in four types of strip-shaped WSNs and compared with four existing MDC methods with evaluation metrics of maximum node energy consumption, network delay and weighted sum of both. The simulation results demonstrate that the proposed algorithm is applicable to different types of strip-shaped WSNs with branches and achieves excellent network performance, which can effectively balance network energy consumption and data collection delay.

Causal discovery reveals complex patterns of drought-induced displacement

The increasing frequency and severity of droughts present a significant risk to vulnerable regions of the globe, potentially leading to substantial human displacement in extreme situations. Drought-induced displacement is a complex and multifaceted issue that can perpetuate cycles of poverty, exacerbate food and water scarcity, and reinforce socio-economic inequalities. However, our understanding of human mobility in drought scenarios is currently limited, inhibiting accurate predictions and effective policy responses. Drought-induced displacement is driven by numerous factors and identifying its key drivers, causal-effect lags, and consequential effects is often challenging, typically relying on mechanistic models and qualitative assumptions. This paper presents a novel, data-driven methodology, grounded in causal discovery, to retrieve the drivers of drought-induced displacement within Somalia from 2016 to 2023. Our model exposes the intertwined vulnerabilities and the leading times that connect drought impacts, water and food security systems along with episodes of violent conflict, emphasizing that causal mechanisms change across districts. These findings pave the way for the development of algorithms with the ability to learn from human mobility data, enhancing anticipatory action, policy formulation, and humanitarian aid.

Understanding the Use of Mobility Data in Disasters: Exploratory Qualitative Study of COVID-19 User Feedback.

Human mobility data have been used as a potential novel data source to guide policies and response planning during the COVID-19 global pandemic. The COVID-19 Mobility Data Network (CMDN) facilitated the use of human mobility data around the world. Both researchers and policy makers assumed that mobility data would provide insights to help policy makers and response planners. However, evidence that human mobility data were operationally useful and provided added value for public health response planners remains largely unknown. This exploratory study focuses on advancing the understanding of the use of human mobility data during the early phase of the COVID-19 pandemic. The study explored how researchers and practitioners around the world used these data in response planning and policy making, focusing on processing data and human factors enabling or hindering use of the data. Our project was based on phenomenology and used an inductive approach to thematic analysis. Transcripts were open-coded to create the codebook that was then applied by 2 team members who blind-coded all transcripts. Consensus coding was used for coding discrepancies. Interviews were conducted with 45 individuals during the early period of the COVID-19 pandemic. Although some teams used mobility data for response planning, few were able to describe their uses in policy making, and there were no standardized ways that teams used mobility data. Mobility data played a larger role in providing situational awareness for government partners, helping to understand where people were moving in relation to the spread of COVID-19 variants and reactions to stay-at-home orders. Interviewees who felt they were more successful using mobility data often cited an individual who was able to answer general questions about mobility data; provide interactive feedback on results; and enable a 2-way communication exchange about data, meaning, value, and potential use. Human mobility data were used as a novel data source in the COVID-19 pandemic by a network of academic researchers and practitioners using privacy-preserving and anonymized mobility data. This study reflects the processes in analyzing and communicating human mobility data, as well as how these data were used in response planning and how the data were intended for use in policy making. The study reveals several valuable use cases. Ultimately, the role of a data translator was crucial in understanding the complexities of this novel data source. With this role, teams were able to adapt workflows, visualizations, and reports to align with end users and decision makers while communicating this information meaningfully to address the goals of responders and policy makers.