Client-side network control and configuration

In this paper, we propose and develop a platform to support data collection for mobile crowdsensing from mobile device sensors that is under-pinned by real-time mobile data stream mining. We experimentally show that mobile data mining provides an efficient and scalable approach for data collection for mobile crowdsensing. Our approach results in reducing the amount of data sent, as well as the energy usage on the mobile phone, while providing comparable levels of accuracy to traditional models of intermittent/continuous sensing and sending. We have implemented our Context-Aware Real-time Open Mobile Miner (CAROMM) to facilitate data collection from mobile users for crowdsensing applications. CAROMM also collects and correlates this real-time sensory information with social media data from both Twitter and Facebook. CAROMM supports delivering real-time information to mobile users for queries that pertain to specific locations of interest. We have evaluated our framework by collecting real-time data over a period of days from mobile users and experimentally demonstrated that mobile data mining is an effective and efficient strategy for mobile crowdsensing Read More ›

Radio Frequency (RF) fingerprinting, based onWiFi or cellular signals, has been a popular approach to indoor localization. However, its adoption in the real world has been stymied by the need for sitespecific calibration, i.e., the creation of a training data set comprising WiFi measurements at known locations in the space of interest. While efforts have been made to reduce this calibration effort using modeling, the need for measurements from known locations still remains a bottleneck. In this paper, we present Zee -- a system that makes the calibration zero-effort, by enabling training data to be crowdsourced without any explicit effort on the part of users. Zee leverages the inertial sensors (e.g., accelerometer, compass, gyroscope) present in the mobile devices such as smartphones carried by users, to track them as they traverse an indoor environment, while simultaneously performing WiFi scans. Zee is designed to run in the background on a device without requiring any explicit user participation. The only site-specific input that Zee depends on is a map showing the pathways (e.g., hallways) and barriers (e.g., walls). A significant challenge that Zee surmounts is to track users without any a priori, user-specific knowledge such as the user's initial location, stride-length, or phone placement. Zee employs a suite of novel techniques to infer location over time: (a) placement-independent step counting and orientation estimation, (b) augmented particle filtering to simultaneously estimate location and user-specific walk characteristics such as the stride length,(c) back propagation to go back and improve the accuracy of ocalization in the past, and (d) WiFi-based particle initialization to enable faster convergence. We present an evaluation of Zee in a large office building. Read More ›

Here we present our ubiquitous computing vision, Eywa. Eywa is an open publish-subscribe system that employs crowdsourcing for tasking and social networks & machine learning for identifying relevance. We argue that crowdsourcing (and the social networks and machine learning that enable it) should be a first class citizen in ubiquitous computing. We also observe that cloud computing is a natural platform to host such future ubiquitous computing systems. We discuss about some applications enabled by Eywa, and focus on our CuratedLiving application (which emphasizes “less choice more relevance” approach) as a case study. Read More ›

Crowd-enabled place-centric systems gather and reason over large mobile sensor datasets and target everyday user locations (such as stores, workplaces, and restaurants). Such systems are transforming various consumer services (for example, local search) and data-driven organizations (city planning). As the demand for these systems increases, our understanding of how to design and deploy successful crowdsensing systems must improve. In this paper, we present a systematic study of the coverage and scaling properties of place-centric crowdsensing. During a two-month deployment, we collected smartphone sensor data from 85 participants using a representative crowdsensing system that captures 48,000 different place visits. Our analysis of this dataset examines issues of core interest to place-centric crowdsensing, including place-temporal coverage, the relationship between the user population and coverage, privacy concerns, and the characterization of the collected data. Collectively, our findings provide valuable insights to guide the building of future place-centric crowdsensing systems and applications. Read More ›

Crowd-enabled place-centric systems gather and reason over large mobile sensor datasets and target everyday user locations (such as stores, workplaces, and restaurants). Such systems are transforming various consumer services (for example, local search) and data-driven organizations (city planning). As the demand for these systems increases, our understanding of how to design and deploy successful crowdsensing systems must improve. In this paper, we present a systematic study of the coverage and scaling properties of place-centric crowdsensing. During a two-month deployment, we collected smartphone sensor data from 85 participants using a representative crowdsensing system that captures 48,000 different place visits. Our analysis of this dataset examines issues of core interest to place-centric crowdsensing, including place-temporal coverage, the relationship between the user population and coverage, privacy concerns, and the characterization of the collected data. Collectively, our findings provide valuable insights to guide the building of future place-centric crowdsensing systems and applications. Read More ›

Mobile crowdsensing is becoming a vital technique for environment monitoring, infrastructure management, and social computing. However, deploying mobile crowdsensing applications in large-scale environments is not a trivial task. It creates a tremendous burden on application developers as well as mobile users. In this paper we try to reveal the barriers hampering the scale-up of mobile crowdsensing applications, and to offer our initial thoughts on the potential solutions to lowering the barriers. Read More ›

To minimize battery drain due to background communication in cellular-connected devices such as smartphones, the duration for which the cellular radio is kept active should be minimized. This, in turn, calls for scheduling the background communication so as to maximize the throughput. It has been recognized in prior work that a key determinant of throughput is the wireless link quality. However, as we show here, another key factor is the load in the cell, arising from the communication of other nodes. Unlike link quality, the only way, thus far, for a cellular client to obtain a measure of load has been to perform active probing, which defeats the goal of minimizing the active duration of the radio. In this paper, we address the above dilemma by making the following contributions. First, we show experimentally that to obtain good throughput, considering link quality alone is insufficient, and that cellular load must also be factored in. Second, we present a novel technique called LoadSense for a cellular client to obtain a measure of the cellular load, locally and passively, that allows the client to determine the ideal times for communication when available throughput to the client is likely to be high. Finally, we present the Peek-n-Sneak protocol, which enables a cellular client to 'peek' into the channel and 'sneak' in with its background communication when the conditions are suitable. When multiple clients in a cell perform Peen-n-Sneak, it enables them to coordinate their communications, implicitly and in an entirely distributed manner, akin to CSMA in wireless LANs, helping improve throughput (and reduce energy drain) for all. Our experimental evaluation shows overall device energy savings of 20-60% even when Peek-n-Sneak is deployed incrementally Read More ›

This paper describes a multitiered architecture for realizing an inference management firewall (IMF) that employs context-aware information masking techniques for systematic management of risk-vs-value trade-off of sensor data. Previously we have demonstrated an initial implementation of the IMF running as messaging services on the Information Fabric, which is a middleware asset developed under the International Technology Alliance (ITA) research program. Furthermore, we have presented an additional asset, recently implemented on a commercially-available mobile device running the Android operating system, which is intended to operate as an information source and first-line inference management capability at the edge of the network. The low-cost and widespread use of Android-based mobile devices offers a popular platform for crowdsourced participatory sensing. The focus of our current work is on the integration of these two technology assets in support of policy-managed, sensor-driven workflows in coalition scenarios. Read More ›

Mobile phones or smartphones are rapidly becoming the central computer and communication device in people's lives. Application delivery channels such as the Apple AppStore are transforming mobile phones into App Phones, capable of downloading a myriad of applications in an instant. Importantly, today's smartphones are programmable and come with a growing set of cheap powerful embedded sensors, such as an accelerometer, digital compass, gyroscope, GPS, microphone, and camera, which are enabling the emergence of personal, group, and communityscale sensing applications. We believe that sensor-equipped mobile phones will revolutionize many sectors of our economy, including business, healthcare, social networks, environmental monitoring, and transportation. In this article we survey existing mobile phone sensing algorithms, applications, and systems. We discuss the emerging sensing paradigms, and formulate an architectural framework for discussing a number of the open issues and challenges emerging in the new area of mobile phone sensing research. Read More ›

We present Darwin, an enabling technology for mobile phone sensing that combines collaborative sensing and classification techniques to reason about human behavior and context on mobile phones. Darwin advances mobile phone sensing through the deployment of efficient but sophisticated machine learning techniques specifically designed to run directly on sensor-enabled mobile phones (i.e., smartphones). Darwin tackles three key sensing and inference challenges that are barriers to mass-scale adoption of mobile phone sensing applications: (i) the human-burden of training classifiers, (ii) the ability to perform reliably in different environments (e.g., indoor, outdoor) and (iii) the ability to scale to a large number of phones without jeopardizing the 'phone experience' (e.g., usability and battery lifetime). Darwin is a collaborative reasoning framework built on three concepts: classifier/model evolution, model pooling, and collaborative inference. To the best of our knowledge Darwin is the first system that applies distributed machine learning techniques and collaborative inference concepts to mobile phones. We implement the Darwin system on the Nokia N97 and Apple iPhone. While Darwin represents a general framework applicable to a wide variety of emerging mobile sensing applications, we implement a speaker recognition application and an augmented reality application to evaluate the benefits of Darwin. We show experimental results from eight individuals carrying Nokia N97s and demonstrate that Darwin improves the reliability and scalability of the proof-of-concept speaker recognition application without additional burden to users. Read More ›

An emerging category of devices at the edge of the Internet are consumer-centric mobile sensing and computing devices, such as smartphones, music players, and in-vehicle sensors. These devices will fuel the evolution of the Internet of Things as they feed sensor data to the Internet at a societal scale. In this article, we examine a category of applications that we term mobile crowdsensing, where individuals with sensing and computing devices collectively share data and extract information to measure and map phenomena of common interest. We present a brief overview of existing mobile crowdsensing applications, explain their unique characteristics, illustrate various research challenges, and discuss possible solutions. Finally, we argue the need for a unified architecture and envision the requirements it must satisfy. Read More ›

A number of novel wireless networked services, ranging from participatory sensing to social networking, leverage the increasing capabilities of mobile devices and the movements of the individuals carrying them. For many of these systems, their effectiveness fundamentally depends on coverage and the particular mobility patterns of the participants. Given the strong spatial and temporal regularity of human mobility, the needed coverage can typically only be attained through a large participant base. In this paper we explore an alternative approach to attain coverage without scale -- (soft) controlling the movement of participants. We present Crowd Soft Control (CSC), an approach to exert limited control over the temporal and spatial movements of mobile users by leveraging the built-in incentives of location-based gaming and social applications. By pairing network services with these location-based apps, CSC allows researchers to use an application's incentives (e.g. game objectives) to control the movement of participating users, increasing the effectiveness and efficiency of the associated network service. After outlining the case for Crowd Soft Control, we present an initial prototype of our ideas and discuss potential benefits and costs in the context of two case studies. Read More ›

Automated and scalable approaches for understanding the semantics of places are critical to improving both existing and emerging mobile services. In this paper, we present CrowdSense@Place (CSP), a framework that exploits a previously untapped resource -- opportunistically captured images and audio clips from smartphones -- to link place visits with place categories (e.g., store, restaurant). CSP combines signals based on location and user trajectories (using WiFi/GPS) along with various visual and audio place 'hints' mined from opportunistic sensor data. Place hints include words spoken by people, text written on signs or objects recognized in the environment. We evaluate CSP with a seven-week, 36-user experiment involving 1,241 places in five locations around the world. Our results show that CSP can classify places into a variety of categories with an overall accuracy of 69%, outperforming currently available alternative solutions. Read More ›

We are becoming increasingly aware that the effectiveness of mobile crowdsourcing systems critically depends on the whims of their human participants, impacting everything from participant engagement to their compliance with the crowdsourced tasks. In response, a number of such systems have started to incorporate different incentive features aimed at a wide range of goals that span from improving participation levels, to extending the systems' coverage, and enhancing the quality of the collected data. Despite the many related efforts, the inclusion of incentives in crowdsourced systems has so far been mostly ad-hoc, treating incentives as a wild-card response fitted for any occasion and goal. Using data from a large, 2-day experiment with 96 participants at a corporate conference, we present an analysis of the impact of two incentive structures on the recruitment, compliance and user effort of a basic mobile crowdsourced service. We build on these preliminary results to argue for a principled approach for selecting incentive and incentive structures to match the variety of requirements of mobile crowdsourcing applications and discuss key issues in working toward that goal. Read More ›