Mobile Sensor Data Research Articles

Detecting Drinking Episodes in Young Adults Using Smartphone-based Sensors.

Alcohol use in young adults is common, with high rates of morbidity and mortality largely due to periodic, heavy drinking episodes (HDEs). Behavioral interventions delivered through electronic communication modalities (e.g., text messaging) can reduce the frequency of HDEs in young adults, but effects are small. One way to amplify these effects is to deliver support materials proximal to drinking occasions, but this requires knowledge of when they will occur. Mobile phones have built-in sensors that can potentially be useful in monitoring behavioral patterns associated with the initiation of drinking occasions. The objective of our work is to explore the detection of daily-life behavioral markers using mobile phone sensors and their utility in identifying drinking occasions. We utilized data from 30 young adults aged 21-28 with past hazardous drinking and collected mobile phone sensor data and daily Experience Sampling Method (ESM) of drinking for 28 consecutive days. We built a machine learning-based model that is 96.6% accurate at identifying non-drinking, drinking and heavy drinking episodes. We highlight the most important features for detecting drinking episodes and identify the amount of historical data needed for accurate detection. Our results suggest that mobile phone sensors can be used for automated, continuous monitoring of at-risk populations to detect drinking episodes and support the delivery of timely interventions.

HazeEst: Machine Learning Based Metropolitan Air Pollution Estimation From Fixed and Mobile Sensors

Metropolitan air pollution is a growing concern in both developing and developed countries. Fixed-station monitors, typically operated by governments, offer accurate but sparse data, and are increasingly being augmented by lower fidelity but denser measurements taken by mobile sensors carried by concerned citizens and researchers. In this paper, we introduce HazeEst—a machine learning model that combines sparse fixed-station data with dense mobile sensor data to estimate the air pollution surface for any given hour on any given day in Sydney. We assess our system using seven regression models and tenfold cross validation. The results show that estimation accuracy of support vector regression (SVR) is similar to decision tree regression and random forest regression, and higher than extreme gradient boosting, multi-layer perceptrons, linear regression, and adaptive boosting regression. The air pollution estimates from our models are validated via field trials, and results show that SVR not only yields high spatial resolution estimates that correspond well with the pollution surface obtained from fixed and mobile sensor monitoring systems, but also indicates boundaries of polluted area better than other regression models. Our results can be visualized using a Web-based application customized for metropolitan Sydney. We believe that the continuous estimates provided by our system can better inform air pollution exposure and its impact on human health.

Evaluation of a duty-cycled asynchronous X-MAC protocol for vehicular sensor networks

An asynchronous medium access control (MAC) duty-cycled protocols have higher energy efficiency and lower packet latency than synchronized ones due to reduced idle listening. Moreover, they provide efficient utilization of energy supplied to mobile sensors. They are considered very important in MAC protocols due to the adverse effects of hidden terminals which causes energy consumption in sensor networks. Therefore, in this paper, the impact of hidden terminals on the performance of an asynchronous duty-cycled MAC protocol X-MAC for vehicle-base sensor is investigated via analysis and simulations. We propose a Markov model to analyze the quality-of-service (QoS) parameters in terms of energy consumption, delay, and throughput. Our analytical model provides QoS parameter values that closely match the simulation results under various network conditions. Our model is more computationally efficient and provides accurate results quickly compared with simulations. More importantly, our model enables the designers to obtain a better understanding of the effects of different numbers of mobile sensor nodes and data arrival rates on the performance of an asynchronous MAC duty-cycled protocol.

Scalable Passive Sleep Monitoring Using Mobile Phones: Opportunities and Obstacles.

BackgroundSleep is a critical aspect of people’s well-being and as such assessing sleep is an important indicator of a person’s health. Traditional methods of sleep assessment are either time- and resource-intensive or suffer from self-reporting biases. Recently, researchers have started to use mobile phones to passively assess sleep in individuals’ daily lives. However, this work remains in its early stages, having only examined relatively small and homogeneous populations in carefully controlled contexts. Thus, it remains an open question as to how well mobile device-based sleep monitoring generalizes to larger populations in typical use cases.ObjectiveThe aim of this study was to assess the ability of machine learning algorithms to detect the sleep start and end times for the main sleep period in a 24-h cycle using mobile devices in a diverse sample.MethodsWe collected mobile phone sensor data as well as daily self-reported sleep start and end times from 208 individuals (171 females; 37 males), diverse in age (18−66 years; mean 39.3), education, and employment status, across the United States over 6 weeks. Sensor data consisted of geographic location, motion, light, sound, and in-phone activities. No specific instructions were given to the participants regarding phone placement. We used random forest classifiers to develop both personalized and global predictors of sleep state from the phone sensor data.ResultsUsing all available sensor features, the average accuracy of classifying whether a 10-min segment was reported as sleep was 88.8%. This is somewhat better than using the time of day alone, which gives an average accuracy of 86.9%. The accuracy of the model considerably varied across the participants, ranging from 65.1% to 97.3%. We found that low accuracy in some participants was due to two main factors: missing sensor data and misreports. After correcting for these, the average accuracy increased to 91.8%, corresponding to an average median absolute deviation (MAD) of 38 min for sleep start time detection and 36 min for sleep end time. These numbers are close to the range reported by previous research in more controlled situations.ConclusionsWe find that mobile phones provide adequate sleep monitoring in typical use cases, and that our methods generalize well to a broader population than has previously been studied. However, we also observe several types of data artifacts when collecting data in uncontrolled settings. Some of these can be resolved through corrections, but others likely impose a ceiling on the accuracy of sleep prediction for certain subjects. Future research will need to focus more on the understanding of people’s behavior in their natural settings in order to develop sleep monitoring tools that work reliably in all cases for all people.

Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K).

The Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K) is enabling the collection of high-frequency mobile sensor data for the development and validation of novel multisensory biomarkers and sensor-triggered interventions.

Mining Probabilistic Representative Gathering Patterns for Mobile Sensor Data

With the increasing availability of mobile IoT (Internet of Things) devices, it becomes feasible to collect user's trajectory data to discover valuable knowledge. Unfortunately, data uncertainty in trajectory data is inherent due to measurement errors or time-discretized sampling. Because of the nature of dealing with inexact data, new mining approaches are necessary to effectively account for ambiguities in the data. This paper formalizes the concept of probabilistic gathering pattern, which is to find the moving objects that travel together during a certain time period. In contrast to exact spatio-temporal trajectory, an uncertain trajectory is an ordered sequence of random variables. The distance between two uncertain series is also a random variable. Firstly, a new expected distance similarity function is designed to cluster the moving objects. Secondly, in order to avoid considerable redundancy in the complete set of gathering patterns, a new algorithm to only mining the representative set of probabilistic gathering patterns is presented, and some pruning strategies are proposed to greatly improve the mining efficiency. Finally, the efficiency of the approaches is validated by extensive experiments.

Modal Activity-Based Stochastic Model for Estimating Vehicle Trajectories from Sparse Mobile Sensor Data

Probe vehicles that measure position and speed have emerged as a promising tool for traffic data collection and performance measurement, but the sampling rates of most probe vehicle sensor data available today are low (ranging from 10 to 60 s per sample), and the data coverage is limited. Therefore, it is challenging to accurately estimate the vehicle dynamic states in both space and time based on these sparse mobile sensor data. In this paper, a stochastic model is proposed to estimate the second-by-second vehicle speed trajectories by examining all possible sequences of modal activities (i.e., acceleration, deceleration, cruising, and idling) between consecutive data points from sparse position and speed measurements. The likelihood of occurrence of each sequential pattern is first quantified by mode-specific a priori distributions. The vehicle dynamic state probability is then formulated as the product of probabilities for multiple independent events. Therefore, a detailed vehicle speed trajectory can be reconstructed using the optimal modal activity sequence, which maximizes the likelihood. The proposed model is calibrated and validated using the Next-Generation SIMulation dataset. The results show the substantial improvements on the accuracy of estimated vehicle trajectories compared with a baseline method based on linear interpolation. The proposed model is applied to a large-scale vehicle activity dataset to demonstrate the estimation of hourly traffic delay variation.

The relationship between mobile phone location sensor data and depressive symptom severity

BackgroundSmartphones offer the hope that depression can be detected using passively collected data from the phone sensors. The aim of this study was to replicate and extend previous work using geographic location (GPS) sensors to identify depressive symptom severity.MethodsWe used a dataset collected from 48 college students over a 10-week period, which included GPS phone sensor data and the Patient Health Questionnaire 9-item (PHQ-9) to evaluate depressive symptom severity at baseline and end-of-study. GPS features were calculated over the entire study, for weekdays and weekends, and in 2-week blocks.ResultsThe results of this study replicated our previous findings that a number of GPS features, including location variance, entropy, and circadian movement, were significantly correlated with PHQ-9 scores (r’s ranging from −0.43 to −0.46, p-values < .05). We also found that these relationships were stronger when GPS features were calculated from weekend, compared to weekday, data. Although the correlation between baseline PHQ-9 scores with 2-week GPS features diminished as we moved further from baseline, correlations with the end-of-study scores remained significant regardless of the time point used to calculate the features.DiscussionOur findings were consistent with past research demonstrating that GPS features may be an important and reliable predictor of depressive symptom severity. The varying strength of these relationships on weekends and weekdays suggests the role of weekend/weekday as a moderating variable. The finding that GPS features predict depressive symptom severity up to 10 weeks prior to assessment suggests that GPS features may have the potential as early warning signals of depression.

MCrave: Continuous Estimation of Craving During Smoking Cessation.

Craving usually precedes a lapse for impulsive behaviors such as overeating, drinking, smoking, and drug use. Passive estimation of craving from sensor data in the natural environment can be used to assist users in coping with craving. In this paper, we take the first steps towards developing a computational model to estimate cigarette craving (during smoking abstinence) at the minute-level using mobile sensor data. We use 2,012 hours of sensor data and 1,812 craving self-reports from 61 participants in a smoking cessation study. To estimate craving, we first obtain a continuous measure of stress from sensor data. We find that during hours of day when craving is high, stress associated with self-reported high craving is greater than stress associated with low craving. We use this and other insights to develop feature functions, and encode them as pattern detectors in a Conditional Random Field (CRF) based model to infer craving probabilities.

MSieve: Differential Behavioral Privacy in Time Series of Mobile Sensor Data.

Differential privacy concepts have been successfully used to protect anonymity of individuals in population-scale analysis. Sharing of mobile sensor data, especially physiological data, raise different privacy challenges, that of protecting private behaviors that can be revealed from time series of sensor data. Existing privacy mechanisms rely on noise addition and data perturbation. But the accuracy requirement on inferences drawn from physiological data, together with well-established limits within which these data values occur, render traditional privacy mechanisms inapplicable. In this work, we define a new behavioral privacy metric based on differential privacy and propose a novel data substitution mechanism to protect behavioral privacy. We evaluate the efficacy of our scheme using 660 hours of ECG, respiration, and activity data collected from 43 participants and demonstrate that it is possible to retain meaningful utility, in terms of inference accuracy (90%), while simultaneously preserving the privacy of sensitive behaviors.

"Kind and Grateful": A Context-Sensitive Smartphone App Utilizing Inspirational Content to Promote Gratitude.

BackgroundPrevious research has shown that gratitude positively influences psychological wellbeing and physical health. Grateful people are reported to feel more optimistic and happy, to better mitigate aversive experiences, and to have stronger interpersonal bonds. Gratitude interventions have been shown to result in improved sleep, more frequent exercise and stronger cardiovascular and immune systems. These findings call for the development of technologies that would inspire gratitude. This paper presents a novel system designed toward this end.MethodsWe leverage pervasive technologies to naturally embed inspiration to express gratitude in everyday life. Novel to this work, mobile sensor data is utilized to infer optimal moments for stimulating contextually relevant thankfulness and appreciation. Sporadic mood measurements are inventively obtained through the smartphone lock screen, investigating their interplay with grateful expressions. Both momentary thankful emotion and dispositional gratitude are measured. To evaluate our system, we ran two rounds of randomized control trials (RCT), including a pilot study (N = 15, 2 weeks) and a main study (N = 27, 5 weeks). Studies’ participants were provided with a newly developed smartphone app through which they were asked to express gratitude; the app displayed inspirational content to only the intervention group, while measuring contextual cues for all users.ResultsIn both rounds of the RCT, the intervention was associated with improved thankful behavior. Significant increase was observed in multiple facets of practicing gratitude in the intervention groups. The average frequency of practicing thankfulness increased by more than 120 %, comparing the baseline weeks with the intervention weeks of the main study. In contrast, the control group of the same study exhibited a decrease of 90 % in the frequency of thankful expressions. In the course of the study’s 5 weeks, increases in dispositional gratitude and in psychological wellbeing were also apparent. Analyzing the relation between mood and gratitude expressions, our data suggest that practicing gratitude increases the probability of going up in terms of emotional valence and down in terms of emotional arousal. The influences of inspirational content and contextual cues on promoting thankful behavior were also analyzed: We present data suggesting that the more successful times for eliciting expressions of gratitude tend to be shortly after a social experience, shortly after location change, and shortly after physical activity.ConclusionsThe results support our intervention as an impactful method to promote grateful affect and behavior. Moreover, they provide insights into design and evaluation of general behavioral intervention technologies.

Finding Significant Stress Episodes in a Discontinuous Time Series of Rapidly Varying Mobile Sensor Data.

Management of daily stress can be greatly improved by delivering sensor-triggered just-in-time interventions (JITIs) on mobile devices. The success of such JITIs critically depends on being able to mine the time series of noisy sensor data to find the most opportune moments. In this paper, we propose a time series pattern mining method to detect significant stress episodes in a time series of discontinuous and rapidly varying stress data. We apply our model to 4 weeks of physiological, GPS, and activity data collected from 38 users in their natural environment to discover patterns of stress in real-life. We find that the duration of a prior stress episode predicts the duration of the next stress episode and stress in mornings and evenings is lower than during the day. We then analyze the relationship between stress and objectively rated disorder in the surrounding neighborhood and develop a model to predict stressful episodes.

A Mobile Sensing Approach for Regional Surveillance of Fugitive Methane Emissions in Oil and Gas Production.

This paper addresses the need for surveillance of fugitive methane emissions over broad geographical regions. Most existing techniques suffer from being either extensive (but qualitative) or quantitative (but intensive with poor scalability). A total of two novel advancements are made here. First, a recursive Bayesian method is presented for probabilistically characterizing fugitive point-sources from mobile sensor data. This approach is made possible by a new cross-plume integrated dispersion formulation that overcomes much of the need for time-averaging concentration data. The method is tested here against a limited data set of controlled methane release and shown to perform well. We then present an information-theoretic approach to plan the paths of the sensor-equipped vehicle, where the path is chosen so as to maximize expected reduction in integrated target source rate uncertainty in the region, subject to given starting and ending positions and prevailing meteorological conditions. The information-driven sensor path planning algorithm is tested and shown to provide robust results across a wide range of conditions. An overall system concept is presented for optionally piggybacking of these techniques onto normal industry maintenance operations using sensor-equipped work trucks.

TouchSignatures: Identification of user touch actions and PINs based on mobile sensor data via JavaScript

Conforming to W3C specifications, mobile web browsers allow JavaScript code in a web page to access motion and orientation sensor data without the user's permission. The associated risks to user security and privacy are however not considered in W3C specifications. In this work, for the first time, we show how user security can be compromised using these sensor data via browser, despite that the data rate is 3–5 times slower than what is available in app. We examine multiple popular browsers on Android and iOS platforms and study their policies in granting permissions to JavaScript code with respect to access to motion and orientation sensor data. Based on our observations, we identify multiple vulnerabilities, and propose TouchSignatures which implements an attack where malicious JavaScript code on an attack tab listens to such sensor data measurements. Based on these streams, TouchSignatures is able to distinguish the user's touch actions (i.e., tap, scroll, hold, and zoom) and her PINs, allowing a remote website to learn the client-side user activities. We demonstrate the practicality of this attack by collecting data from real users and reporting high success rates using our proof-of-concept implementations. We also present a set of potential solutions to address the vulnerabilities. The W3C community and major mobile browser vendors including Mozilla, Google, Apple and Opera have acknowledged our work and are implementing some of our proposed countermeasures.

Dynamic Recognition of Driver’s Propensity Based on GPS Mobile Sensing Data and Privacy Protection

Driver’s propensity is a dynamic measurement of driver’s emotional preference characteristics in driving process. It is a core parameter to compute driver’s intention and consciousness in safety driving assist system, especially in vehicle collision warning system. It is also an important influence factor to achieve the Driver-Vehicle-Environment Collaborative Wisdom and Control macroscopically. In this paper, dynamic recognition model of driver’s propensity based on support vector machine is established taking the vehicle safety controlled technology and respecting and protecting the driver’s privacy as precondition. The experiment roads travel time obtained through GPS is taken as the characteristic parameter. The sensing information of Driver-Vehicle-Environment was obtained through psychological questionnaire tests, real vehicle experiments, and virtual driving experiments, and the information is used for parameter calibration and validation of the model. Results show that the established recognition model of driver’s propensity is reasonable and feasible, which can achieve the dynamic recognition of driver’s propensity to some extent. The recognition model provides reference and theoretical basis for personalized vehicle active safety systems taking people as center especially for the vehicle safety technology based on the networking.

An Automotive Distributed Mobile Sensor Data Collection with Machine Learning Based Data Fusion and Analysis on a Central Backend System

One of the most extensive examples for ubiquitous computing today is automotion. The equipment of sensors and independent computing devices in current vehicles is vast if not endless. Furthermore, traffic infrastructure is realized using global and local computing devices. Communication initiated by the car itself (e.g., to an emergency hotline) will be obligatory in some countries soon. And finally, by using a smart phone the driver brings an additional powerful computing device and sensor set to the vehicle. However, all these automotive sensors and computing devices are used just for fixed (and in most cases single) purposes. Data exchange between vehicles or vehicles and infrastructure is rarely done. And dynamic changes like compensating for a broken sensor with available other data, using old sensor equipment for new functions, or improving old driver assistance systems with new sensors is not possible, either.The objective of the collaborative research project Smart Adaptive Data Aggregation (SADA) is to develop technologies that enable linking data from distributed mobile on-board sensors (on vehicles) with data from previously unknown stationary (e.g., infrastructure) or mobile sensors (e.g., other vehicles, smart devices). One focus of the project is the dynamic and fully-automated switching between different sensors or sensor configurations, including the adaptation of data fusion processes. Technically, one important component for some of the SADA use cases is a central backend system that (1) collects sensor data of the vehicles and/or the infrastructure, (2) fuses these data, and (3) carries out machine learning (ML) based analysis of the data to generate new information for the drivers (sometimes refered to by the term “virtual sensors”).The article gives a short overview of the SADA project and describes in more detail the concept of the backend system architec- ture, the user interface, and the methods and processes needed for a demonstration use-case.

Energy-Efficient Location and Activity-Aware On-Demand Mobile Distributed Sensing Platform for Sensing as a Service in IoT Clouds

The Internet of Things (IoT) envisions billions of sensors deployed around us and connected to the Internet, where the mobile crowd sensing technologies are widely used to collect data in different contexts of the IoT paradigm. Due to the popularity of Big Data technologies, processing and storing large volumes of data has become easier than ever. However, large scale data management tasks still require significant amounts of resources that can be expensive regardless of whether they are purchased or rented (e.g. pay-as-you-go infrastructure). Further, not everyone is interested in such large scale data collection and analysis. More importantly, not everyone has the financial and computational resources to deal with such large volumes of data. Therefore, a timely need exists for a cloud-integrated mobile crowd sensing platform that is capable of capturing sensors data, on-demand, based on conditions enforced by the data consumers. In this paper, we propose a context-aware, specifically, location and activity-aware mobile sensing platform called C-MOSDEN ( Context-aware Mobile Sensor Data ENgine) for the IoT domain. We evaluated the proposed platform using three real-world scenarios that highlight the importance of 'selective sensing'. The computational effectiveness and efficiency of the proposed platform are investigated and is used to highlight the advantages of context-aware selective sensing.

Investigation of the Usability of Mobile Sensors for Weather Forecasting

Crowd sourcing is a&#x0D; popular method for providing data from people by the use of mobile sensor,&#x0D; internet and communication technologies. However efficient use of the raw data&#x0D; provided by the sensors with different characteristics in order to obtain&#x0D; accurate results is not investigated in detail.&amp;nbsp;&#x0D; This study aims to investigate the data collected by mobile sensors&#x0D; integrated in the smartphones for scientific purposes such as weather&#x0D; forecasting. In this context, accuracy of the data provided mobile humidity, pressure&#x0D; and temperature sensors was examined in this study. Data provided by 5 smart&#x0D; phones and 3 Bluetooth sensors were tested in this context. Accuracy assessment&#x0D; process was performed by calculating the Root Mean Square Errors of the data&#x0D; with respect to reference data collected by TST Sensor simultaneously. This&#x0D; study shows that accuracy of the data collected with the mobile sensors is&#x0D; affected by several external parameters such as climatic conditions, handling&#x0D; habits of the user, and etc. Although it is possible to calculate correction&#x0D; constant for each sensor separately, it is not possible to calculate a unique&#x0D; and universal correction constant in order to increase the accuracy of the raw&#x0D; data collected by the mobile sensors. Therefore further studies should be&#x0D; executed for improving the accuracy of the mobile sensor data for scientific&#x0D; purposes.

Device Nimbus: An Intelligent Middleware for Smarter Services for Health and Fitness

The world is experiencing an unprecedented explosion in the number of smart devices and mobile apps that are available to users. In the health and fitness domains, many of the devices and technologies in the market place are restricted to proprietary platforms, typically working in isolation with fixed hardware settings. Consequently, an important challenge is to investigate techniques for combining and analyzing data encapsulated by these ubiquitous technologies. In this paper, we introduce an intelligent middleware—Device Nimbus—to meet this challenge. The middleware supports the integration of distributed and heterogeneous mobile sensor data, enabling both context and predictive analysis. We describe a minimum viable product of Device Nimbus and report the results of preliminary tests spanning multiple data sources focused on fitness apps, illustrating the efficacy of the middleware.

Everyday space–time geographies: using mobile phone-based sensor data to monitor urban activity in Harbin, Paris, and Tallinn

This paper proposes a methodology for using mobile telephone-based sensor data for detecting spatial and temporal differences in everyday activities in cities. Mobile telephone-based sensor data has great applicability in developing urban monitoring tools and smart city solutions. The paper outlines methods for delineating indicator points of temporal events referenced as ‘midnight’, ‘morning start’, ‘midday’, and ‘duration of day’, which represent the mobile telephone usage of residents (what we call social time) rather than solar or standard time. Density maps by time quartiles were also utilized to test the versatility of this methodology and to analyze the spatial differences in cities. The methodology was tested with data from cities of Harbin (China), Paris (France), and Tallinn (Estonia). Results show that the developed methods have potential for measuring the distribution of temporal activities in cities and monitoring urban changes with georeferenced mobile phone data.