Detecting event frontline or boundary sensors in a complex sensor network environment is one of the critical problems for sensor network applications. In this paper, we propose a novel algorithm for event frontline sensor detection based on statistical mixture methods with model selection (Akaike, 1973). A boundary sensor is considered as being associated with a multimodal local neighborhood of (univariate or multivariate) sensing readings, and each non-boundary sensor is treated as being with a unimodal sensor reading neighborhood. Furthermore, the set of sensor readings within each sensor's spatial neighborhood is formulated using Gaussian mixture model (McLachlan and Peel, 2000). Two classes of boundary and non-boundary sensors can be effectively classified using the model selection techniques for finite mixture models. Our extensive experimental results demonstrate that our algorithm effectively detects the event boundary with a high accuracy under moderate noise levels.

In this paper, we present a Gaussian mixture model based approach to capture the spatial characteristics of any target signal in a sensor network, and further propose a temporally-adaptive variant of the approach for dynamic multiple target tracking under changing environments, with the presence of both significant background event noises and a large portion of outlying sensor readings. The target position is estimated by adopting the mean-shift optimization to discriminate the target signals from the background noises. Our mixture model based algorithm is capable of fusing multivariate real-valued sensor measurements and its probability nature shows fault tolerance and robustness in noisy sensing environments. This consideration is practical as in real world applications, sensor readings are multi-modal and may contain errors. The simulation study validates our design and the results indicate that our mixture model based algorithm is an effective and capable approach for the two most typical target signal models under consideration. Desirable quantitative target tracking results are also achieved through extensive evaluations under challenging background conditions.

This paper targets the identification of outlying sensors (that is, outlying reading sensors) and the detection of the reach of events in sensor networks. Typical applications include the detection of the transportation front line of some vegetation or animalcule's growth over a certain geographical region. We propose and analyze two novel algorithms for outlying sensor identification and event boundary detection. These algorithms are purely localized and, thus, scale well to large sensor networks. Their computational overhead is low, since only simple numerical operations are involved. Simulation results indicate that these algorithms can clearly detect the event boundary and can identify outlying sensors with a high accuracy and a low false alarm rate when as many as 20 percent sensors report outlying readings. Our work is exploratory in that the proposed algorithms can accept any kind of scalar values as inputs-a dramatic improvement over existing work, which takes only 0/1 decision predicates. Therefore, our algorithms are generic. They can be applied as long as "events" can be modeled by numerical numbers. Though designed for sensor networks, our algorithms can be applied to the outlier detection and regional data analysis in spatial data mining.

Abstract Recent research points out that the flooding mechanism for topology update or route request in existing ad hoc routing protocols greatly degrades the network capacity. If we restrict the broadcast of control packets within a small subset of hosts in the network, the protocol overhead can be substantially reduced. This motivates our research of constructing a virtual backbone by computing a connected dominating set (CDS) in unit‐disk graphs. In this paper, we propose two distributed algorithms to approximate a minimum CDS. These algorithms take linear time. Their performance is verified by a complete theoretical analysis. Copyright © 2006 John Wiley & Sons, Ltd.

This paper targets the identification of faulty sensors and detection of the reach of events in sensor networks with faulty sensors. Typical applications include the detection of the transportation front line of a contamination and the diagnosis of network health. We propose and analyze two novel algorithms for faulty sensor identification and fault-tolerant event boundary detection. These algorithms are purely localized and thus scale well to large sensor networks. Their computational overhead is low, since only simple numerical operations are involved. Simulation results indicate that these algorithms can clearly detect the event boundary and can identify faulty sensors with a high accuracy and a low false alarm rate when as many as 20% sensors become faulty. Our work is exploratory in that the proposed algorithms can accept any kind of scalar values as inputs, a dramatic improvement over existing works that take only 0/1 decision predicates. Therefore, our algorithms are generic. They can be applied as long as the "events" can be modelled by numerical numbers. Though designed for sensor networks, our algorithms can be applied to the outlier detection and regional data analysis in spatial data mining.

Fault-tolerant target detection and localization is a challenging task in collaborative sensor networks. The paper introduces our exploratory work toward identifying a stationary target in sensor networks with faulty sensors. We explore both spatial and temporal dimensions for data aggregation to decrease the false alarm rate and improve the target position accuracy. To filter out extreme measurements, the median of all readings in the close neighborhood is used to approximate the local observation to the target. The sensor whose observation is a local maximum computes a position estimate at each epoch. Results from multiple epochs are combined to decrease the false alarm rate further and improve the target localization accuracy. Our algorithms have low computation and communication overheads. A simulation study demonstrates the validity and efficiency of our design.

We propose a novel vision for highway safety warning based on sensor networks; the system aims to provide user-friendly zero-delay safety warnings to motorists. Our idea leverages advanced sensing, networking and storage technologies. Highway sensors detect events and store event records at multiple designated locations such that passing drivers can be alerted to potential dangers or traffic delays through wireless communication among sensors deployed at the highway and in vehicles. We have designed a location-centric storage protocol which manages the propagation and storage of event records based on the time needed to clear the event in the highway. This protocol ensures that event records can be fairly distributed among highway sensors when event arrival rate is Poisson distributed with the same mean, as verified by our simulation study and theoretical performance analysis.