Feng Zhao

Scalable Information-Driven Sensor Querying and Routing for Ad Hoc Heterogeneous Sensor Networks

This paper describes two novel techniques, information-driven sensor querying (IDSQ) and constrained anisotropic diffusion routing (CADR), for energy-efficient data querying and routing in ad hoc sensor networks for a range of collaborative signal processing tasks. The key idea is to introduce an information utility measure to select which sensors to query and to dynamically guide data routing. This allows us to maximize information gain while minimizing detection latency and bandwidth consumption for tasks such as localization and tracking. Our simulation results have demonstrated that the information-driven querying and routing techniques are more energy efficient, have lower detection latency, and provide anytime algorithms to mitigate risks of link/node failures.

Collaborative signal and information processing: an information-directed approach

This paper describes information-based approaches to processing and organizing spatially distributed, multimodal sensor data in a sensor network. Energy-constrained networked sensing systems must rely on collaborative signal and information processing (CSIP) to dynamically allocate resources, maintain multiple sensing foci, and attend to new stimuli of interest, all based on task requirements and resource constraints. Target tracking is an essential capability for sensor networks and is used as a canonical problem for studying information organization problems in CSIP. After formulating a CSIP tracking problem in a distributed constrained optimization framework, the paper describes information-driven sensor query and other techniques for tracking individual targets as well as combinatorial tracking problems such as counting targets. Results from simulations and experimental implementations have demonstrated that these information-based approaches are scalable and make efficient use of scarce sensing and communication resources.

Collaborative in-network processing for target tracking

This paper presents a class of signal processing techniques for collaborative signal processing in ad hoc sensor networks, focusing on a vehicle tracking application. In particular, we study two types of commonly used sensors—acoustic-amplitude sensors for target distance estimation and direction-of-arrival sensors for bearing estimation—and investigate how networks of such sensors can collaborate to extract useful information with minimal resource usage. The information-driven sensor collaboration has several advantages: tracking is distributed, and the network is energy-efficient, activated only on a when-needed basis. We demonstrate the effectiveness of the approach to target tracking using both simulation and field data.

Lightweight sensing and communication protocols for target enumeration and aggregation

The development of lightweight sensing andcommunication protocols is a key requirement for designing resource constrained sensor networks. This paper introduces a set of efficient protocols and algorithms, DAM, EBAM, and EMLAM, for constructing and maintaining sensor aggregates that collectively monitor target activity in the environment. A sensor aggregate comprises those nodes in a network that satisfy a grouping predicate for a collaborative processing task. The parameters of the predicate depend on the task and its resource requirements. Since the foremost purpose of a sensor network is to selectively gather information about the environment, the formation of appropriate sensor aggregates is crucial for optimally allocating resources to sensing and communication tasks.This paper makes minimal assumptions about node onboard processing and communication capabilities so as to allow possible implementations on resource-constrained hardware. Factors affecting protocol performance are discussed. The paper presents simulation results showing how the protocol performance varies as key network and task parameters are varied. It also provides probabilistic analyses of network behavior consistent with the simulation results. The protocols have been experimentally validated on a sensor network testbed comprising 25 Berkeley MICA sensor motes.

Distributed group management for track initiation and maintenance in target localization applications

The tradeoff between performance and scalability is a fundamental issue in distributed sensor networks. In this paper, we propose a novel scheme to efficiently organize and utilize network resources for target localization. Motivated by the essential role of geographic proximity in sensing, sensors are organized into geographically local collaborative groups. In a target tracking context, we present a dynamic group management method to initiate and maintain multiple tracks in a distributed manner. Collaborative groups are formed, each responsible for tracking a single target. The sensor nodes within a group coordinate their behavior using geographically-limited message passing. Mechanisms such as these for managing local collaborations are essential building blocks for scalable sensor network applications.

A dual-space approach to tracking and sensor management in wireless sensor networks

Wireless ad hoc sensor networks have the advantage of spanning a large geographical region and being able to collaboratively detect and track non-local spatio-temporal events. This paper presents a dual-space approach to event tracking and sensor resource management in sensor networks. The dual-space transformation maps a non-local phenomenon, e.g., the edge of a half-plane shadow, to a single point in the dual space, and maps locations of distributed sensor nodes to a set of lines that partitions the dual space. The detection problem becomes finding and tracking the cell that contains the point in the arrangement defined by these lines. This mechanism can be effectively used for power management of the sensor network - nodes that will not be immediately visited by an event can be turned off to save energy required for sensing, processing, and communication. The approach has been successfully demonstrated on a laboratory testbed built using the UC Berkeley motes sensors. An implemented application of detecting and tracking light shadow edges moving over a sensor field is described.

A distributed algorithm for managing multi-target identities in wireless ad-hoc sensor networks

This paper presents a scalable distributed algorithm for computing and maintaining multi-target identity information. The algorithm builds on a novel representational framework, Identity-Mass Flow, to overcome the problem of exponential computational complexity in managing multi-target identity explicitly. The algorithm uses local information to efficiently update the global multi-target identity information represented as a doubly stochastic matrix, and can be efficiently mapped to nodes in a wireless ad hoc sensor network. The paper describes a distributed implementation of the algorithm in sensor networks. Simulation results have validated the Identity-Mass Flow framework and demonstrated the feasibility of the algorithm.

State-centric programming for sensor-actuator network systems

Distributed embedded systems such as wireless sensor and actuator networks require new programming models and software tools to support the rapid design and prototyping of sensing and control applications. Unlike centralized platforms and Web-based distributed systems, these distributed sensor-actuator network (DSAN) systems are characterized by a massive number of potentially failing nodes, limited energy and bandwidth resources, and the need to rapidly respond to sensor input. We describe a state-centric, agent-based design methodology to mediate between a system developers mental model of physical phenomena and the distributed execution of DSAN applications. Building on the ideas of data-centric networking, sensor databases, and proximity-based group formation, we introduce the notion of collaboration groups, which abstracts common patterns in application-specific communication and resource allocation. Using a distributed tracking application with sensor networks, well demonstrate how state-centric programming can raise the abstraction level for application developers.

Distributed state representation for tracking problems in sensor networks

This paper investigates the problem of designing decentralized representations to support monitoring and inferences in sensor networks. State-space models of physical phenomena such as those arising from tracking multiple interacting targets, while commonly used in signal processing and control, suffer from the curse of dimensionality as the number of phenomena of interest increases. Furthermore, mapping an inference algorithm onto a distributed sensor network must appropriately allocate scarce sensing and communication resources. We address the state-space explosion problem by developing a distributed state-space model that switches between factored and joint state spaces as appropriate. We develop a collaborative group abstraction as a mechanism to effectively support the information ow within and across subspaces of the state-space model, which can be efficiently supported in a communication-constrained network. The approach has been implemented and demonstrated in a simulation of tracking multiple interacting targets.

Estimation of distributed hybrid systems Using particle filtering methods

Networked embedded systems are composed of a large number of components that interact with the physical world via a set of sensors and actuators, have their own computational capabilities, and communicate with each other via a wired or wireless network. Such systems are best modeled by distributed hybrid systems that capture the interaction between the physical and computational components. Monitoring and diagnosis of any dynamical system depend crucially on the ability to estimate the system state given the observations. Estimation for distributed hybrid systems is particularly challenging because it requires keeping track of multiple models and the transitions between them. This paper presents a particle filtering based estimation algorithm for a class of distributed hybrid systems. The hybrid estimation methodology is demonstrated on a cryogenic propulsion system.