Jonas Michel

Gander: Personalizing Search of the Here and Now

In Personalized Networked Spaces (PNets), people and devices are integrated with the environment and demand fluid interactions to enable connectivity to information, services, and people. PNet applications exhibit significant spatiotemporal demands in which connectivity to resources and information is personalized and focused on the here and now. We introduce Gander, a personalized search engine for the here and now. We examine how search expectations are affected when users and applications interact directly with the physical environment. We define a formal conceptual model of search in PNets that provides a clear definition of the framework and ultimately enables reasoning about relationships between search processing and the relevance of results. We assess our model by evaluating sophisticated Gander queries in a simulated PNet.

Gander: Mobile, Pervasive Search of the Here and Now in the Here and Now

The vision of the Internet of Things (IoT) will enable networked environments populated with vast amounts of data that can be exploited by humans. The volume of digitally available data in such emerging computing spaces presents an imminent need for search mechanisms that enable humans and applications to find relevant information within their digitally accessible physical surroundings. This paper presents Gander, a search engine for these pervasive computing spaces enabled by the IoT and characterized by large volumes of highly transient data. Gander is founded on a novel conceptual model of search that resolves queries about a users here and now by leveraging proximally available resources in the here and now. We formally describe the model underlying Gander, describe the networking protocols that enable Ganders search, and provide a realization of Gander via an extensible framework. Employing this Gander framework, we describe a concrete middleware implementation for wirelessly networked environments. We evaluate this implementation of Gander through a user study that examines the perceived utility of myGander, a real-world mobile application enabled by the Gander middleware, and we benchmark the performance of Gander in large pervasive computing spaces through network simulation.

Real-Time Optimization of Video Transmission in a Network of AAVs

Mobile cyberphysical systems have received considerable attention over the last decade, as communication, computing and control come together on a common platform. Understanding the complex interactions that govern the behavior of large complex cyberphysical systems is not an easy task. The goal of this paper is to address this challenge in the particular context of multimedia delivery over an autonomous aerial vehicle (AAV) network. Bandwidth requirements and stringent delay constraints of real-time video streaming, paired with limitations on computational complexity and power consumptions imposed by the underlying implementation platform, make cross-layer and cross-domain co-design approaches a necessity. In this paper, we propose a novel, low-complexity rate-distortion optimized (RDO) protocol specifically targeted at video streaming over mobile embedded networks. We test the performance of our RDO algorithm on simulation models developed for aerial mobility of multiple wirelessly communicating AAVs. Results show that our optimized streaming leads to 47% and 39% less video distortion with very little computational overhead compared to regular ACKed and non-ACKed transmission, respectively.

A Cloudlet-Based Proximal Discovery Service for Machine-to-Machine Applications

Many of today’s applications attempt to connect mobile users with resources available in their immediate surroundings. Existing approaches for discovering available resources are either centralized, providing a single point of lookup somewhere in the cloud or ad hoc, requiring mobile devices to directly connect to other nearby devices. In this paper, we explore an approach based on cloudlets, marrying these two approaches to reflect both the proximity requirements of the applications and the dynamic nature of the resources. We present the design and implementation of a cloudlet-based proximal discovery service, solving key technical challenges along the way. We then use real world data traces to demonstrate, evaluate, and benchmark our service and compare it to a completely centralized approach. We find that, in supporting highly localized queries, our service outperforms the centralized approach without significantly affecting the quality of the discovery results.

A spatiotemporal model for ephemeral data in pervasive computing networks

Pervasive computing evokes a vision of digitally-accessible environments with which applications and users interact in localized ways. In this vision, information is ephemeral: it is created, moved, stored, and deleted on-demand at rapid rates. Without a formal data model that enables the data itself to speak about its spatial and temporal bearings, it is difficult to build support for accessing an information-rich digital world in a general-purpose way. In this paper, we demonstrate the need for an expressive data model of the inherently ephemeral data in pervasive computing and propose the beginnings of such a model that explicitly tags information with spatial and temporal semantics. Our model is founded on spatiotemporal trajectories, which capture the spatial and temporal semantics of data and the phenomenon it represents. We further demonstrate both the need for and potential impact of a general-purpose expressive spatiotemporal data model using several use cases.

myGander: A mobile interface and distributed search engine for pervasive computing

Motivated by the growing amount of digitally-accessible information in our physical surroundings and the ephemeral nature of that information, there is a profound need to efficiently search information with spatiotemporal underpinnings without a priori indexing. Human users in Personalized Network Spaces (PNetS), pervasive computing environments connected by opportunistic peer-to-peer connections, need information that is immediate and localized. This tight integration of the user with his immediate surroundings introduces novel search requirements. The requisite support for performing search of the here and now in the here and now calls for a new paradigm of search that explicitly separates search from advanced indexing of data. In light of this vision, this paper presents Gander, a scalable search engine for PNetS, along with myGander, a prototype mobile interface for Gander. The utility and usability of myGander as supported by Gander is demonstrated through a practical real-world pervasive computing scenario.

Mobilizing search of the here and now

The escalating number of digitally-accessible devices pervading our everyday environments gives rise to the availability of tremendous amounts of human- and device-generated data. This data possesses strong spatial and temporal semantics, it captures phenomena and states of the environment, and is extremely volatile, being created, moved, stored, and deleted on-demand at rapid rates. The requisite support for general-purpose expressive search of the “here” and “now” has eluded realization due to the complexities of indexing, storing, and retrieving relevant information within a vast collection of highly ephemeral data. We aim to address this gap in the research (i) through search mechanisms that are sensitive to the bearings of space and time on the relevance of search results and (ii) more fundamentally, through the design of a general-purpose data model that facilitates information availability in pervasive computing environments and exposes spatial and temporal relationships between digital data and physical phenomena. Finally, this paper outlines simulation and deployment evaluations of the proposed research.