Tarek Abdelzaher

Tracking is one of the major applications of wireless sensor networks. EnviroSuite, as a programming paradigm, provides a comprehensive solution for programming tracking applications, wherein moving environmental targets are uniquely and identically mapped to logical objects to raise the level of programming abstraction. Such mapping is done through distributed group management algorithms, which organize nodes in the vicinity of targets into groups, and maintain the uniqueness and identity of target representation such that each target is given a consistent name. Challenged by tracking fast-moving targets, this paper explores, in a systematic way, various group management optimizations including semi-dynamic leader election, piggy-backed heartbeats, and implicit leader election. The resulting tracking protocol, Lightweight EnviroSuite, is integrated into a surveillance system. Empirical performance evaluation on a network of 200 XSM motes shows that, due to these optimizations, Lightweight EnviroSuite is able to track targets more than 3 times faster than the fastest targets trackable by the original EnviroSuite even when 20% of nodes fail.

... Shin, Chair Assistant Professor Peter Chen Associate Professor Farnam Jahanian AssistantProfessor Kimberly M ... Professor Farnam Jahanian for his friendly support, and for sharing interesting intuition ... Rich Friedrich, Gita Gopal, Tai Jin, David Mosberger, Anna Zara, Martin Arlitt, ...

Highly dynamic sensor networks, such as mobile robotic sensor networks, have been applied in various kinds of application scenarios such as real-time planet exploration and deep-ocean discovery. In these types of networks, mobility and energy management protocols change the connectivity among the neighboring nodes quickly. Traditional state-based protocols, designed for static and/or low-mobility networks, suffer excessive delay in updating their routing or neighborhood tables, leading to severe packet loss and communication delay in the highly dynamic situations. To provide robust and timely communication, we exploit the concept of Lazy-Binding to deal with the elevated network dynamics. Based on this concept and the knowledge of the node positions, we introduce Implicit Geographic Forwarding (IGF), a new protocol for highly dynamic sensor networks that is altogether state-free. We compare our work against several typical routing protocols in static, mobile and energy-conserving networks under a wide range of system and workload configurations. In the presence of mobility and other dynamics, IGF achieves as much as 10 times improvement in the delivery ratio and significant reduction in both the end-to-end delay and control overhead. In addition to extensive simulations, we also implement and evaluate the IGF protocol on the Berkeley mote platform.

... Jin Heo1, Dan Henriksson1, Xue Liu2, Tarek Abdelzaher1 1 Department of Computer Science, University of Illinois at Urbana-Champaign 2 School of Computer Science, McGill University Abstract ... Composing the two policies leads to an adverse interaction. ...

ABSTRACT The seven articles in this special issue focus on composable context aware services. The articles roughly fall into four areas: a survey of ambient networks, extensions to network protocols to facilitate service discovery and composition, networking middleware and software frameworks for integration of context awareness in service composition, and an example of creating an ecosystem to build context-aware mobile social networks.

Proliferation of QoS-sensitive client-server Internet applications such as high-quality audio, video-on-demand, e-commerce, and commercial Web hosting has generated an impetus to provide performance guarantees. These applications require a guaranteed minimum amount of resources to operate acceptably to the users, thus calling for QoS-provisioning mechanisms. One good place to locate such mechanisms is in server communication subsystems. Server-side communication subsystems manage an increasing number of connection end-points, thus readily controlling important bottleneck resources. We propose, implement, and evaluate a novel communication server architecture that maximizes the aggregate utility of QoS-sensitive connections for a community of clients even in the case of overload. A contribution of this architecture is that it manages QoS from the user space and is transparent to the application. It does not require modifications to the OS kernel, which improves portability and reduces development cost. Results from an experimental evaluation on a microkernel indicate that it achieves end-system overload protection and traffic prioritization, improves insulation between independent clients, adapts to offered load, and enhances aggregate service utility.

Abstract The sheer scale and rapid rise of Big Data mandates highly scalable, self-adaptive, and energy-conserving data-intensive compute clusters. Based on our analysis of the traces from a production Hadoop cluster at Yahoo!, we observe that file size, file lifespan, and file heat are statistically correlated and very strongly associated with the hierarchical directory structure (ie, absolute file path) in which the files are organized. Leveraging that observation, we present predictive GreenHDFS; an energy-conserving variant of the Hadoop ...

Advances in hardware miniaturization enable the mass-production of low-cost sensor devices equipped with microprocessors, memory and wireless communication. A key challenge in sensor networks is to limit network energy consumption, as it is impractical to replace batteries on thousands of deployed devices in a remote or dangerous environment. Moreover, the network must adapt to potentially harsh environmental conditions, such as temperature and radiation, in a way that addresses energy concerns, the challenge is to optimize system behavior and energy consumption under performance constraints.

Moore’s law, automation considerations, and the pervasive need for timely information lead to a next generation of distributed systems that are open, highly interconnected, and deeply embedded in the physical world by virtue of pervasive sensing and sensor-based decision-making. These systems offer new research challenges that stem from scale, composition of large numbers of components, and tight coupling between computation, communication, and distributed interaction with both physical and social contexts. These growing challenges span a large spectrum ranging from new models of computation for systems that live in physical and social spaces, to the enforcement of reliable, predictable, and timely end-to-end behavior in the face of high interactive complexity, increased uncertainty, and imperfect implementation. This chapter discusses the top challenges in composing large-scale sensing systems and conjectures on research directions of increasing interest in this realm.

Predictability and adaptability are two essential requirements of real-time systems. This paper presents a new real-time kernel whose design explicitly addresses the above requirements. It is intended for systems involving both static and dynamic tasks, and it employs a single scheduling algorithm to accommodate all task types. The algorithm utilizes the extent of known information about each type to achieve better management of time. It provides a priori guarantees for static tasks and on-line guarantees for dynamic ones. The kernel adapts to changes in the environment which may result in the dynamic generation of new tasks, on-line modification of tasks' timing constraints and large changes in task arrival rates

The value of real-time hydrologic data dissemination including river stage, streamflow, and precipitation for operational stormwater management efforts is particularly high for communities where flash flooding is common and costly. Ideally, such data would be presented within a watershed-scale geospatial context to portray a holistic view of the watershed. Local hydrologic sensor networks usually lack comprehensive integration with sensor networks managed by other agencies sharing the same watershed due to administrative, political, but mostly technical barriers. Recent efforts on providing unified access to hydrological data have concentrated on creating new SOAP-based web services and common data format (e.g. WaterML and Observation Data Model) for users to access the data (e.g. HIS and HydroSeek). Geospatial Web technology including OGC sensor web enablement (SWE), GeoRSS, Geo tags, Geospatial browsers such as Google Earth and Microsoft Virtual Earth and other location-based service tools provides possibilities for us to interact with a digital watershed in near-real-time. OGC SWE proposes a revolutionary concept towards a web-connected/controllable sensor networks. However, these efforts have not provided the capability to allow dynamic data integration/fusion among heterogeneous sources, data filtering and support for workflows or domain specific applications where both push and pull mode of retrieving data may be needed. We propose a light weight integration framework by extending SWE with open source Enterprise Service Bus (e.g., mule) as a backbone component to dynamically transform, transport, and integrate both heterogeneous sensor data sources and simulation model outputs. We will report our progress on building such framework where multi-agencies" sensor data and hydro-model outputs (with map layers) will be integrated and disseminated in a geospatial browser (e.g. Microsoft Virtual Earth). This is a collaborative project among NCSA, USGS Illinois Water Science Center, Computer Science Department at UIUC funded by the Adaptive Environmental Infrastructure Sensing and Information Systems initiative at UIUC.

Most currently deployed sensor networks use the same channel to communicate information among nodes. This is a source of great inefficiency as it poorly utilizes the available wireless spectrum. This paper takes advantage of radio capabilities of MicaZ motes that can communicate on multiple frequencies as specified in the 802.15.4 standard. We consider the case of a data collection sensor network where multiple base-stations are responsible for draining data from sensor nodes. A key question becomes how to assign nodes to wireless channels such that network throughput is maximized. The problem is reduced to one of load balancing. A control theoretical approach is used to design a self-regulating load-balancing algorithm that maximizes total network throughput. It is evaluated both in simulation and on an experimental testbed. The results demonstrate a significant performance improvement. It is shown that a control theory approach is indeed needed to guarantee stability in data collection networks and prevent undue oscillation of nodes among different wireless channels upon dynamic changes in load conditions.

This paper computes end-to-end delay bounds for prioritized data flows in disruption-tolerant networks (DTNs). DTNs suffer intermittent connectivity among nodes due to node mobility. When deployed in mission-critical applications, such as disaster response, an interesting question becomes to quantify end-to-end packet delays under assumptions on node mobility. In this paper, we answer this question for the special case of DTNs with recurrent mobility patterns. A recurrent pattern refers to one where nodes revisit the same locations repeatedly. We devise a suitable model for recurrent DTNs that captures their timing and mobility properties. We then apply the recently proposed delay composition algebra to the resulting network model in order to determine an upper bound on end-to-end communication delays of network flows. Evaluation results show that the upper bound is moderate in its pessimism and can be used for deployment planning purposes.

Abstract In this paper, we develop a cooperative mechanism, RELICS, to combat selfishness in DTNs. In DTNs, nodes belong to self-interested individuals. A node may be selfish in expending resources, such as energy, on forwarding messages from others, unless ...

We have developed an analysis-based design tool, ANDES, for modeling a wireless sensor network system and analyzing its performance before deployment ANDES enables designers to systematically develop a model for the system, refine it iteratively by tuning the system parameters based on existing analysis techniques, and resolve key design decisions according to the required system performance. We also present a real-time communication schedulability analysis for sensor networks based on exact characterization which utilizes information regarding network topology and workload characteristics to analyze the schedulability of a set of periodic streams with real-time constraints. We further demonstrate the use of ANDES for the designers through detailed case studies where we design wireless sensor network applications (for target detection and environmental monitoring) using ANDES and validate the results through simulations. Currently, ANDES supports communication schedulability analysis, target tracking analysis and real-time capacity analysis which work on system models with differing levels of detail. ANDES has been developed by extending the AADL/OSATE framework which has been used extensively for real-time and embedded systems. Based on key insights gained from the development of this analysis tool, we address issues in AADL for its use in the field of wireless sensor networks. We have developed a plug-in for ANDES, called ModelGeneration, which bridges the gap between the semantics needed for sensor networks and the syntax supported by AADL. This makes it easy for sensor network designers to build system models that are intuitive to them. Furthermore, ANDES is extensible and new analysis techniques can be easily incorporated into the toolset.

This paper first develops a multivariable discrete-time model reference adaptive control scheme with relaxed design conditions based on an LDU decomposition of the system high frequency gain matrix, and then applies the adaptive control scheme to a web cache system to provide proportional differentiation on average hit rates of different content classes of client requests. A key task fulfilled is

This paper presents the delay composition algebra: a set of simple operators for systematic transformation of distributed real-time task systems into single-resource task systems such that schedulability properties of the original system are preserved. The transformation allows performing schedulability analysis on distributed systems using uniprocessor theory and analysis tools. Reduction-based analyses techniques have been used in other contexts such as control theory and circuit theory, by defining rules to compose together components of the system and reducing them into equivalent single components that can be easily analyzed. This paper is the first to develop such reduction rules for distributed real-time systems. By successively applying operators such as PIPE and SPLIT on operands that represent workload on composed subsystems, we show how a distributed task system can be reduced to an equivalent single resource task set from which the end-to-end delay and schedulability of tasks can be inferred. We show through simulations that the proposed analysis framework is less pessimistic with increasing system scale compared to traditional approaches.

... Since our protocol stack for the physical motes will not use UDP/IP, which has 28 bytes header overhead, to adjust for these large overheads, compared to the small payload size, another change we made was that the ... For each routing algorithm we report the average ...