Julien Eberle

From big smartphone data to worldwide research: The Mobile Data Challenge

This paper presents an overview of the Mobile Data Challenge (MDC), a large-scale research initiative aimed at generating innovations around smartphone-based research, as well as community-based evaluation of mobile data analysis methodologies. First, we review the Lausanne Data Collection Campaign (LDCC), an initiative to collect unique longitudinal smartphone dataset for the MDC. Then, we introduce the Open and Dedicated Tracks of the MDC, describe the specific datasets used in each of them, discuss the key design and implementation aspects introduced in order to generate privacy-preserving and scientifically relevant mobile data resources for wider use by the research community, and summarize the main research trends found among the 100+ challenge submissions. We finalize by discussing the main lessons learned from the participation of several hundred researchers worldwide in the MDC Tracks.

ExposureSense: Integrating daily activities with air quality using mobile participatory sensing

With an increasing number of rich embedded sensors, like accelerometer and GPS, smartphone becomes a pervasive people-centric sensing platform for inferring users daily activities and social contexts. Alternatively, wireless sensor network offers a comprehensive platform for capturing the surrounding environmental information using mobile sensing nodes, e.g., the OpenSense project [2] in Switzerland deploying air quality sensors like CO on public transports like buses and trams. The two sensing platforms are typically isolated from each other. In this paper, we build ExposureSense, a rich mobile participatory sensing infrastructure that integrates the two independent sensing paradigms. ExposureSense is able to monitor peoples daily activities as well to compute a reasonable estimation of pollution exposure in their daily life. Besides using external sensor networks, ExposureSense also supports pluggable sensors (e.g., O3) to further enrich air quality data using mobile participatory sensing with smartphones.

Symbolic representation of smart meter data

Currently smart meter data analytics has received enormous attention because it allows utility companies to analyze customer consumption behavior in real time. However, the amount of data generated by these sensors is very large. As a result, analytics performed on top of it become very expensive. Furthermore, smart meter data contains very detailed energy consumption measurement which can lead to customer privacy breach and all risks associated with it. In this work, we address the problem on how to reduce smart meter data numerosity and its detailed measurement while maintaining its analytics accuracy. We convert the data into symbolic representation and allow various machine learning algorithms to be performed on top of it. In addition, our symbolic representation admit an additional advantage to allow also algorithms which usually work on nominal and string to be run on top of smart meter data. We provide an experiment for classification and forecasting tasks using real-world data. And finally, we illustrate several directions to extend our work further.

An energy-aware method for the joint recognition of activities and gestures using wearable sensors

This paper presents an energy-aware method for recognizing time series acceleration data containing both activities and gestures using a wearable device coupled with a smartphone. In our method, we use a small wearable device to collect accelerometer data from a users wrist, recognizing each data segment using a minimal feature set chosen automatically for that segment. For each collected data segment, if our model finds that recognizing the segment requires high-cost features that the wearable device cannot extract, such as dynamic time warping for gesture recognition, then the segment is transmitted to the smartphone where the high-cost features are extracted and recognition is performed. Otherwise, only the minimum required set of low-cost features are extracted from the segment on the wearable device and only the recognition result, i.e., label, is transmitted to the smartphone in place of the raw data, reducing transmission costs. Our method automatically constructs this adaptive processing pipeline solely from training data.

Energy measurements campaign for positioning methods on State-of-the-Art smartphones

Mobile phones have been undergoing a breathtaking evolution over the last years, starting from simple devices with only voice services, nowadays they are referred as smartphones, powerful devices able to offer appealing services. One of the services that has recently become popular is referred as Location Based Services that use contextual and location information to enhance mobile user experience. As a basic assumption, this kind of services needs to be continuously aware of the phones location. The challenge emerges from the trade-off needed between the location precision and the energy consumption of the positioning methods, the most precise being also the energy-hungriest. This paper presents the results of an extensive energy measurements campaign that was conducted on different sensors and positioning methods on a smartphone. Results will help researchers and developers to design energy efficient and more accurate continuous location tracking algorithms to support Location Based Services.

Mobile observatory: an exploratory study of mobile air quality monitoring application

We present Mobile Observatory, a mobile air quality monitoring application that provides evaluation of air quality of the city of Zurich, Switzerland. As Mobile Observatory utilizes air quality data gathered by sensors mounted on around 10 trams in Zurich, it is able to provide neighborhood-level air quality information within the city. In this paper, we introduce a mobile air quality monitoring application Mobile Observatory. Also, we describe a user study with 10 participants and show our preliminary results in hopes of yielding insights toward improving civic and urban engagement on air quality.

Toward Self-monitoring Smart Cities: the OpenSense2 Approach

The sustained growth of urban settlements in the last years has had an inherent impact on the environment and the quality of life of their inhabitants. In order to support sustainability and improve quality of life in this context, we advocate the fostering of ICT-empowered initiatives that allow citizens to self-monitor their environment and assess the quality of the resources in their surroundings. More concretely, we present the case of such a self-monitoring Smart City platform for estimating the air quality in urban environments at high resolution and large scale. Our approach is a combination of mobile and human sensing that exploits both dedicated and participatory monitoring. We identify the main challenges in such a crowdsensing scenario for Smart Cities, and in particular we analyze issues related to scalability, accuracy, accessibility, privacy, and discoverability, among others. Moreover, we show that our approach has the potential to empower citizens to diagnose their environment using mobile and portable sensing devices, combining their personal data with a public higher accuracy air quality network.

Tree-structured classifier for acceleration-based activity and gesture recognition on smartwatches

This paper proposes a new method for recognizing both activities and gestures by using acceleration data collected on a smartwatch. While both activity recognition techniques and gesture recognition techniques employ acceleration data, these techniques are studied independently due to the large difference between the characteristics of activity sensor data and gesture sensor data. In this study, we combine their recognition using a tree structured classifier that combines features that are widely used to recognize activities with dynamic time warping-based k-nearest neighbor classifiers. Our method can recognize both activities and gestures with low computational cost by executing only the minimal set of feature extraction and classification processes that are required to recognize an input sensor-data segment. An experiment on 30 sessions of sensor data shows that our method can recognize both activities and gestures simultaneously with 95.8% accuracy while reducing computation costs by 97.3% when compared with a baseline method.

Online unsupervised state recognition in sensor data

Smart sensors, such as smart meters or smart phones, are nowadays ubiquitous. To be “smart”, however, they need to process their input data with limited storage and computational resources. In this paper, we convert the stream of sensor data into a stream of symbols, and further, to higher level symbols in such a way that common analytical tasks such as anomaly detection, forecasting or state recognition, can still be carried out on the transformed data with almost no loss of accuracy, and using far fewer resources. We identify states of a monitored system and convert them into symbols (thus, reducing data size), while keeping “interesting” events, such as anomalies or transition between states, as it is. Our algorithm is able to find states of various length in an online and unsupervised way, which is crucial since behavior of the system is not known beforehand. We show the effectiveness of our approach using real-world datasets and various application scenarios.

A model-based back-end for air quality data management

In this paper we present a hybrid model for real-time query processing over data stream collected by mobile air quality sensors. First, we introduce a novel indexing scheme for representing air quality and use it for generating and evaluating a static model over a yearly dataset. Then, this model is combined with a dynamic nearest-neighbor approach for real-time updates, and implemented into the Global Sensor Network (GSN) middleware, with added support for model queries.