Previous approaches for computing duplicate-sensitive aggregates in sensor networks (<i>e.g.</i>, TAG) have used a tree topology, order to conserve energy and avoid double-counting readings. However, topology is not robust against node communication failures, which are common networks. In this paper, we present <i>synopsis diffusion</i>, general framework achieving signi.cantly more accurate reliable answers by combining energy-efficient multi-path routing schemes with techniques that...

We propose the first differentially private aggregation algorithm for distributed time-series data that offers good practical utility without any trusted server. This addresses two important challenges in participatory data-mining applications where (i) individual users collect temporally correlated (such as location traces, web history, personal health data), and (ii) an untrusted third-party aggregator wishes to run aggregate queries on data.

We discuss about the IrisNet (Internet-scale resource-intensive sensor network services) project at Intel Research, we design an architecture and build a system that enable easy deployment of such wide-area sensing services. aim to provide missing software components for realizing worldwide Web. Wide-area architectures pervasive new generation powerful distributed A Web, in which users can query, as single unit, vast quantities data from thousands or even millions widely distributed,...

Web 2.0 is an emerging paradigm for applications and user interactions. In this article, Aman Kansal, Suman Nath, Jie Liu, Feng Zhao from Microsoft Research discuss the development of SenseWeb, a peer-produced sensor network environment, used everyday life decisions.

Mobile app ecosystems have experienced tremendous growth in the last six years. This has triggered research on dynamic analysis of performance, security, and correctness properties mobile apps ecosystem. Exploration execution using automated UI actions emerged as an important tool for this research. However, existing largely developed analysis-specific automation techniques, wherein logic exploring is intertwined with analyzing properties. PUMA a programmable framework that separates these...

Existing energy-efficient approaches to in-network aggregation in sensor networks can be classified into two categories, tree-based and multi-path-based, with each having unique strengths weaknesses. In this paper, we introduce Tributary-Delta, a novel approach that combines the advantages of tree multi-path by running them simultaneously different regions network. We present schemes for adjusting response changes network conditions, show how many useful aggregates readily computed within...

FlashDB is a self-tuning database optimized for sensor networks using NAND flash storage. In practical systems used in different packages such as on-board chips, compact cards, secure digital cards and related formats. Our experiments reveal non-trivial differences their access costs. Furthermore, databases may be subject to types of workloads. We show that existing are not all devices or workloads performance thus suboptimal many systems. uses novel index dynamically adapts its storage...

Data Management in the Worldwide Sensor Web W ith rapidly increasing number of large-scale sensor network deployments, vision a worldwide web is close to becoming reality.Ranging from camera networks that monitor large wildlife reserves biological sensors implanted body vital signs, these deployments generate tremendous volumes priceless data.Simply put, data raison d'être any sensing exercise.Most researchers would probably agree we have placed too much attention on networking distributed...

We propose ACE (Acquisitional Context Engine), a middleware that supports continuous context-aware applications while mitigating sensing costs for inferring contexts. provides user's current context to running on it. In addition, it dynamically learns relationships among various attributes (e.g., whenever the user is Driving, he not AtHome). exploits these automatically learned two powerful optimizations. The first inference caching allows opportunistically infer one attribute (AtHome) from...

We consider the problem of computing all-pair correlations in a warehouse containing large number (e.g., tens thousands) time-series (or, signals). The arises automatic discovery patterns and anomalies data intensive applications such as center management, environmental monitoring, scientific experiments. However, with existing techniques, solving for stream is extremely expensive, due to problem's inherent quadratic I/O CPU complexities.

In today’s agile world, developers often rely on continuous integration pipelines to help build and validate their changes by executing tests in an efficient manner. One of the significant factors that hinder developers’ productivity is flaky tests—tests may pass fail with same version code. Since test failures are not deterministically reproducible, have spend hours only discover occasional nothing do changes. However, ignoring can be dangerous, since those represent real faults production...

Ad networks for mobile apps require inspection of the visual layout their ads to detect certain types placement frauds. Doing this manually is error prone, and does not scale sizes today's app stores. In paper, we design a system called DECAF automatically discover various frauds scalably effectively. uses automated navigation, together with optimizations scan through large number elements within limited time. It also includes framework efficiently detecting whether an violate extensible set...

The rise of smartphones equipped with various sensors has enabled personalization applications based on user contexts extracted from sensor readings. At the same time it raised serious concerns about privacy contexts.

This paper describes the design, implementation, and evaluation of VanarSena, an automated fault finder for mobile applications (``apps''). The techniques in VanarSena are driven by a study 25 million real-world crash reports Windows Phone apps reported 2012. Our analysis indicates that modest number root causes responsible many observed failures, but they occur wide range places app, requiring coverage possible execution paths. adopts ``greybox'' testing method, instrumenting app binary to...

Geocentric Web interfaces such as Microsoft Virtual Earth and Google Maps are useful for visualizing spatially geographically related data driving directions, directory entries, weather traffic conditions, to name a few. The desire add information these has led developers create custom applications that overlay housing prices, crime rates, bus locations, other on top of browsable maps. These possible due APIs publish location We envision new class relies real-time sensor its mash-up with the...

Webcams, microphones, pressure gauges and other sensors provide exciting new opportunities for querying monitoring the physical world. In this paper we focus on wide area sensor databases, containing (XML) data derived from spread over tens to thousands of miles. We present first scalable system executing XPATH queries such databases. The maintains logical view as a single XML document, while physically is fragmented across any number host nodes. For scalability, stored close sensors, but...

Previous approaches for computing duplicate-sensitive aggregates in wireless sensor networks have used a tree topology, order to conserve energy and avoid double-counting readings. However, topology is not robust against node communication failures, which are common networks. In this article, we present synopsis diffusion , general framework achieving significantly more accurate reliable answers by combining energy-efficient multipath routing schemes with techniques that double-counting....

Efficient thermal management is important in modern data centers as cooling consumes up to 50% of the total energy. Unlike previous work, we consider proactive management, whereby servers can predict potential overheating events due dynamics center configuration and workload, giving operators enough time react. However, such forecasting very challenging scales complexity. Moreover, a physical system influenced by cyber effects, including workload scheduling servers. We propose ThermoCast,...

Mobile advertising is an increasingly important driver in the Internet economy. We point out fundamental trade-offs between variables mobile advertisement ecosystem. In order to increase relevance, ad campaigns tend become more targeted and personalized by using context information extracted from user's interactions smartphone's sensors. This raises privacy concerns that are hard overcome due limited resources (energy bandwidth) available on phones. absence of a trusted third party, it...

Advertising is the primary source of revenue for many mobile apps. One important goal ad delivery process targeting users, based on criteria like users' geolocation, context, demographics, long-term behavior, etc. In this paper we report an in-depth study that broadly characterizes what information apps send to networks and how effectively, if at all, utilize users. Our a novel tool, called MadScope, can (1) quickly harvest ads from large collection apps, (2) systematically probe network...

Most current sensor network research explores the use of extremely simple sensors on small devices called motes and focuses over-coming resource constraints these devices. In contrast, our challenges multimedia is motivated by fact that devices, such as cameras, are rapidly becoming inexpensive, yet their in a presents number unique challenges. For example, data rates involved with orders magnitude greater than those for this cannot easily be processed traditional techniques focus scalar...