User-centric context inference for mobile crowdsensing

2020

Our work aims to make opportunistic crowdsensing a reliable means of detecting urban phenomena, as a component of smart city development. We believe that the optimal method for achieving this is by enforcing the cost-effective collection of high quality data. We then investigate a supporting middleware solution that reduces both the network traffic and computation at the cloud. To this end, our research focuses on defining a set of protocols that together implement “context-aware in-network collaborative mobile crowdsensing” by combining: (i) The inference of the crowdsensors' physical context so as to characterize the gathered data; (ii) The context-aware grouping of crowdsensors to share the workload and filter out low quality data; and (iii) Data aggregation at the edge to enhance the knowledge transferred to the cloud.

—Mobile sensing has been gaining ground due to the increasing capabilities of mobile, personal devices that are carried around by citizens, giving access to a large variety of data and services, all based on the way humans interact. Nevertheless, pervasive platforms used to capture and to infer human interaction are still designed in a simplistic way, aspect which prevents them to adequately scale. This paper contributes by bringing awareness into the challenges faced by mobile sensing platforms that capture and perform learning based on human interaction; how current solutions perform activity recognition, which classification models they consider, and which type of behavior inference can be seamlessly provided as of today. The paper contributes to raise awareness to current challenges faced by these platforms, and provides a set of guidelines towards a better functional design.

IEEE Communications Surveys & Tutorials, 2016

The evolution of smartphones together with increasing computational power have empowered developers to create innovative context-aware applications for recognizing user related social and cognitive activities in any situation and at any location. The existence and awareness of the context provides the capability of being conscious of physical environments or situations around mobile device users. This allows network services to respond proactively and intelligently based on such awareness. The key idea behind context-aware applications is to encourage users to collect, analyze and share local sensory knowledge in the purpose for a large scale community use by creating a smart network. The desired network is capable of making autonomous logical decisions to actuate environmental objects, and also assist individuals. However, many open challenges remain, which are mostly arisen due to the middleware services provided in mobile devices have limited resources in terms of power, memory and bandwidth. Thus, it becomes critically important to study how the drawbacks can be elaborated and resolved, and at the same time better understand the opportunities for the research community to contribute to the context-awareness. To this end, this paper surveys the literature over the period of 1991-2014 from the emerging concepts to applications of context-awareness in mobile platforms by providing up-to-date research and future research directions. Moreover, it points out the challenges faced in this regard and enlighten them by proposing possible solutions.

Springer eBooks, 2013

We use mobile sensor data to predict a mobile phone user's semantic place, e.g. at home, at work, in a restaurant etc. Such information can be used to feed context-aware systems, that adapt for instance mobile phone settings like energy saving, connection to Internet, volume of ringtones etc. We consider the task of semantic place prediction as classification problem. In this paper we exploit five feature groups: (i) daily patterns, (ii) weekly patterns, (iii) WLAN information, (iv) battery charging state and (v) accelerometer data. We compare the performance of a Random Forest algorithm and two Support Vector Machines, one with an RBF kernel and one with a Pearson VII function based kernel, on a labelled dataset, and analyse the separate performances of the feature groups as well as promising combinations of feature groups. The winning combination of feature groups achieves an accuracy of 0.871 using a Random Forest algorithm on daily patterns and accelerometer data. A detailed analysis reveals that daily patterns are the most discriminative feature group for the given semantic place labels. Combining daily patterns with WLAN information, battery charging state or accelerometer data further improves the performance. The classifiers using these selected combinations perform better than the classifiers using all feature groups. This is especially encouraging for mobile computing, as fewer features mean that less computational power is required for classification.

IEEE International Conference on Systems, Man and Cybernetics, 2016

Due to the increasing spread of location-aware applications, developers interest in user location estimation has grown in recent years. As users spend the majority of their time in few meaningful places (i.e., groups of near locations that can be considered as a unique place, such as home, school or the workplace), this paper presents a new energy efficient method to estimate user presence in a meaningful place. Specifically, instead of using commonly used but energy hungry methods such as GPS and network positioning techniques, the proposed method applies a Machine Learning algorithm based on Decision Trees, to predict the user presence in a meaningful place by collecting and analyzing: a) user activity, b) information from received notifications (receipt time, generating service, sender-receiver relationship), and c) device status (battery level and ringtone mode). The results demonstrate that, using 20 days of training data and testing the system with data coming from 14 persons, the accuracy (percentage of correct predictions) is 89.40% (standard deviation: 8.27%) with a precision of 89.04% and a recall of 89.40%. Furthermore, the paper analyzes the importance of each considered feature, by comparing the prediction accuracy obtained with different combinations of features.

Lecture Notes in Computer Science, 2009

Mobile phones and accompanying network layers provide a platform to capture and share location, image, and acoustic data. This substrate enables participatory sensing: coordinated data gathering by individuals and communities to explore the world around them. Realizing such widespread and participatory sensing poses difficult challenges. In this paper, we discuss one particular challenge: creating a recruitment service to enable sensing organizers to select well-suited participants. Our approach concentrates on finding participants based on geographic and temporal coverage, as determined by context-annotated mobility profiles that model transportation mode, location, and time. We outline a threestage recruitment framework designed to be parsimonious so as to limit risk to participants by reducing the location and context information revealed to the system. Finally, we illustrate the utility of the framework, along with corresponding modeling technique for mobility information, by analyzing data from a pilot mobility study consisting of ten users.

Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct Publication - UbiComp '14 Adjunct, 2014

Mobile phones include a variety of sensors that can be used to develop context-aware applications and gather data about the user's behavior, including the places he visits, his level of activity and how frequently and with whom he socializes. The collection and analysis of these data has been the focus of recent attention in ubiquitous computing, giving rise to the field known as mobile sensing. In this work, we present a collaborative extension to InCense, a toolkit to facilitate behavioral data gathering from populations of mobile phone users.

2010

Mobile phones have recently been used to collect large-scale continuous data about human behavior. In a paradigm known as people centric sensing, users are not only the carriers of sensing devices, but also the sources and consumers of sensed events. This paper describes a data collection campaign wherein Nokia N95 phones are allocated to a heterogeneous sample of nearly 170 participants from Lausanne, a mid-tier city in Switzerland, to be used over a period of one year. The data collection software runs on the background of the phones in a non-intrusive manner, yielding data on modalities such as social interaction and spatial behavior. The main motivations for organizing a new campaign on top of the ones that have been successfully conducted in the past are the following: First, in comparison to the Reality Mining data, generated in 2004-2005, the present data set is expected to provide a richer means to study location attributes, in particular, because today's mobile phones are more powerful and equipped with more sensors. Second, we aim to recruit a heterogeneous set of participants, comprising family and leisure related social networks in addition to organizationally driven ones. This paper provides a methodological description of the project and shows the potential of the resulting data set in terms of illuminating multiple aspects of human behavior.

Proceedings of the The 51st Annual Design Automation Conference on Design Automation Conference - DAC '14, 2014

This paper presents a scalable and collaborative mobile crowdsensing framework for efficient collective understanding of users, contexts, and their environments. Collaborative mobile crowdsensing enables information to be gathered and shared by users who are directly involved (participatory sensing) or integrated seamlessly as needed (opportunistic sensing) through user mobile platforms. To address the scalability needs of the mobile ecosystem, we additionally employ compressive sensing techniques for approximate gathering and processing of sensor data -this requires new mechanisms for sensor data collection, tunable approximate processing, and mobile networking architecture, to create a compressive collaborative mobile crowdsensing platform called SenseDroid. The proposed framework is build using a multi-tired hierarchical architecture to sense spatial variations of a parameter of interest, perceive spatio-temporal fields, and enable energy efficient local mobile sensing with a small number of measurements. This approximate, yet tunable approach combines different sensing approaches opportunistically while trading scalability (and coverage) for data accuracy (and energy efficiency). In this paper we propose and discuss the framework and the challenges associated with compressive and collaborative mobile sensing for multi-tired hierarchical mobile network architecture for emerging mobile collaborative applications.

Pervasive and Mobile Computing, 2018

Location-Based Services (LBSs) provide users especially in smartphones with information and services based on their location and interests. Predicting people's next location could help in improving the quality of such services and, in turn, boosting people's confidence on those services. Any successful LBS algorithm or model should target three major goals: Location prediction accuracy, high throughput or fast response and efficiency in terms of utilizing smartphone resources. This paper proposes a new approach to discover and predict people's next location based on their mobility patterns, while being computationally efficient. The approach starts by discovering Regions-of-Interest (RoIs) in people's historical trajectories (which denote the locations where users have been previously, frequently). A new model based on Markov Chain (MC) is proposed to overcome the drawback of classical MC. Our proposed model considers both space and time contexts where the space represents a specific location that has been visited by a user while the time represents the location visiting time. We show how classical MC model can be extended to include movement times and how time will improve prediction accuracy. One Unique finding in our research is related to the value of integrating users' mobility location/space with time context. In particular, time context is formulated in a way to add extra information to the space context. For better abstraction during building the model, a general transformation function is used to transform the n-order MC into first order. Results show that the proposed approach provides significant improvements in predicting people's next location compared with the stateof-the-art models when applied on two large real-life datasets, GeoLife and