Pervasive Computing

Activity inference based on object use has received considerable recent attention. Such inference requires statistical models that map activities to the objects used in performing them. Proposed techniques for constructing these models (hand definition, learning from data, and web extraction) all share the problem of model incompleteness: it is difficult to either manually or automatically identify all the possible objects that may be used to perform an activity, or to accurately calculate the probability with which they will be used. In this paper, we show how to use auxiliary information, called an ontology, about the functional similarities between objects to mitigate the problem of model incompleteness. We show how to extract a large, relevant ontology automatically from WordNet, an online lexical reference system for the English language. We adapt a statistical smoothing technique, called shrinkage, to apply this similarity information to counter the incompleteness of our models. Our results highlight two advantages of performing shrinkage. First, overall activity recognition accuracy improves by 15.11% by including the ontology to re-estimate the parameters of models that are automatically mined from the web. Shrinkage can therefore serve as a technique for making web-mined activity models more attractive. Second, smoothing yields an increased recognition accuracy when objects not present in the incomplete models are used while performing an activity. When we replace 100% of the objects with other objects that are functionally similar, we get an accuracy drop of only 33% when using shrinkage as opposed to 91.66% (equivalent to random guessing) without shrinkage. If training data is available, shrinkage further improves classification accuracy.

Location-aware services can benefit from accurate and reliable indoor location tracking. The widespread adoption of 802.11x wireless LAN as the network infrastructure creates the opportunity to deploy WiFi-based location services with few additional hardware costs. While recent research has demonstrated adequate performance, localization error increases significantly in crowded and dynamic situations due to electromagnetic interferences. This paper proposes collaborative localization as an approach to enhance position estimation by leveraging more accurate location information from nearby neighbors within the same cluster. The current implementation utilizes ZigBee radio as the neighbor-detection sensor. This paper introduces the basic model and algorithm for collaborative localization. We also report experiments to evaluate its performance under a variety of clustering scenarios. Our results have shown 28.2-56% accuracy improvement over the baseline system Ekahau, a commercial WiFi localization system.

Many pervasive-computing applications depend on knowledge of user location. Because most current location-sensing techniques work only either indoors or outdoors, researchers have started using 802.11 beacon frames from access points (APs) to provide broader coverage. To use 802.11 beacons, they need to know AP locations. Because the actual locations are often unavailable, they use estimated locations from war driving. But these estimated locations may be different from actual locations. In this paper, we analyzed the errors in these estimates and the effect of these errors on other applications that depend on them. We found that the estimated AP locations have a median error of 32 meters. We considered the error in tracking user positions both indoors and outdoors. Using actual AP locations, we could improve the accuracy as much as 70% for indoors and 59% for outdoors. We also analyzed the effect of using estimated AP locations in computing AP coverage range and estimating interference among APs. The coverage range appeared to be shorter and the interference appeared to be more severe than in reality.

Pervasive applications rely on data captured from the physical world through sensor devices. Data provided by these devices, however, tend to be unreliable. The data must, therefore, be cleaned before an application can make use of them, leading to additional complexity for application development and deployment. Here we present Extensible Sensor stream Processing (ESP), a framework for building sensor data cleaning infrastructures for use in pervasive applications. ESP is designed as a pipeline using declarative cleaning mechanisms based on spatial and temporal characteristics of sensor data. We demonstrate ESP’s effectiveness and ease of use through three real-world scenarios.

People increasingly depend on the digital world to communicate with one another, but such communication is rarely secure. Users typically have no common administrative control to provide mutual authentication, and sales of certified public keys to individuals have made few inroads. The only remaining mechanism is key exchange. Because they are not authenticated, users must verify the exchanged keys through some out-of-band mechanism. Unfortunately, users appear willing to accept any key at face value, leaving communication vulnerable. This paper describes LoKey, a system that leverages the Short Message Service (SMS) to verify keys on users’ behalf. SMS messages are small, expensive, and slow, but they utilize a closed network, between devices—phones—that are nearly ubiquitous and authenticate with the network operator. Our evaluation shows LoKey can establish and verify a shared key in approximately 30 seconds, provided only that one correspondent knows the other’s phone number. By verifying keys asynchronously, two example applications—an instant messaging client and a secure email service—can provide assurances of message privacy, integrity, and source authentication while requiring only that users know the phone number of their correspondent.

Although numerous context-aware applications have been developed and there have been technological advances for acquiring contextual information, it is still difficult to develop and prototype interesting context-aware applications. This is largely due to the lack of programming support available to both programmers and end-users. This lack of support closes off the context-aware application design space to a larger group of users. We present iCAP, a system that allows end-users to visually design a wide variety of context-aware applications, including those based on if-then rules, temporal and spatial relationships and environment personalization. iCAP allows users to quickly prototype and test their applications without writing any code. We describe the study we conducted to understand end-users’ mental models of context-aware applications, how this impacted the design of our system and several applications that demonstrate iCAP’s richness and ease of use. We also describe a user study performed with 20 end-users, who were able to use iCAP to specify every application that they envisioned, illustrating iCAP’s expressiveness and usability.