This paper examines a sensor scheduling problem for target tracking using a network of distributed passive sensors. It considers how to schedule sensor activation to minimise bandwidth and battery usage while maintain...
详细信息
This paper examines a sensor scheduling problem for target tracking using a network of distributed passive sensors. It considers how to schedule sensor activation to minimise bandwidth and battery usage while maintaining track accuracy. The sensor network is made up of simple acoustic, binary proximity sensors. Four different scheduling methods are considered. Two of these methods are heuristic while the other two are optimal, one-step ahead schedulers for differing objectives. The performance of all four algorithms is compared using simulations. The results show that it is possible to obtain adequate performance with only a subset of the sensor field active at any one time.
We propose a distributed coalition formation strategy for collaborative sensing tasks in camera sensor networks. The proposed model supports taskdependent node selection and aggregation through an announcement/bidding...
详细信息
ISBN:
(纸本)9783540730897
We propose a distributed coalition formation strategy for collaborative sensing tasks in camera sensor networks. The proposed model supports taskdependent node selection and aggregation through an announcement/bidding/ selection strategy. It resolves node assignment conflicts by solving an equivalent constraint satisfaction problem. Our technique is scalable, as it lacks any central controller, and it is robust to node failures and imperfect communication. Another unique aspect of our work is that we advocate visually and behaviorally realistic virtual environments as a simulation tool in support of research on largescale camera sensor networks. Specifically, our visual sensor network comprises uncalibrated static and active simulated video surveillance cameras deployed in a virtual train station populated by autonomously self-animating pedestrians. The readily reconfigurable virtual cameras generate synthetic video feeds that emulate those generated by real surveillance cameras monitoring public spaces. Our simulation approach, which runs on high-end commodity PCs, has proven to be beneficial because this type of research would be difficult to carry out in the real world in view of the impediments to deploying and experimenting with an appropriately complex camera network in extensive public spaces.
The problem we address in this paper is how to detect an intruder moving through a polygonal space that is equipped with a camera sensor network. We propose a probabilistic sensor tasking algorithm in which cameras se...
ISBN:
(纸本)9783540730897
The problem we address in this paper is how to detect an intruder moving through a polygonal space that is equipped with a camera sensor network. We propose a probabilistic sensor tasking algorithm in which cameras sense the environment independently of one another, thus reducing the communication overhead. Since constant monitoring is prohibitively expensive with complex sensors such as cameras, the amount of sensing done is also minimized. To be effective, a minimum detection probability must be guaranteed by the system over all possible paths through the space. The straightforward approach of enumerating all such paths is intractable, since there is generally an infinite number of potential paths. Using a geometric decomposition of the space, we lowerbound the detection probability over all paths using a small number of linear constraints. The camera tasking is computed for set of example layouts and shows large performance gains with our probabilistic scheme over both constant monitoring as well as over a deterministic heuristic.
With the proliferation of various kinds of sensor networks, we will see large amounts of heterogeneous data. They have different characteristics such as data content, formats, modality and quality. Existing research h...
ISBN:
(纸本)9783540730897
With the proliferation of various kinds of sensor networks, we will see large amounts of heterogeneous data. They have different characteristics such as data content, formats, modality and quality. Existing research has largely focused on issues related to individual sensor networks; how to make use of diverse data beyond the individual network level is largely unaddressed. In this paper, we propose a semantics-based approach for this problem and describe a system that constructs applications that utilize many sources of data simultaneously. We propose models to formally describe the semantics of data sources, and processing modules that perform various kinds of operations on data. Based on such formal semantics, our system composes data sources and processing modules together in response to users queries. The semantics provides a common ground such that data sources and processing modules from various parties can be shared and reused among applications. We describe our system architecture, illustrate application deployment, and share our experiences in the semantic approach.
With the field of wireless sensor networks rapidly maturing, the focus shifts from "easy" deployments, like remote monitoring, to more difficult domains where applications impose strict, real-time constraint...
ISBN:
(纸本)9783540730897
With the field of wireless sensor networks rapidly maturing, the focus shifts from "easy" deployments, like remote monitoring, to more difficult domains where applications impose strict, real-time constraints on performance. One such class of applications is safety critical systems, like fire and burglar alarms, where events detected by sensor nodes have to be reported reliably and timely to a sink node. A complicating factor is that systems must operate for years without manual intervention, which puts very strong demands on the energy efficiency of protocols running on current sensor-node *** we are not aware of a solution that meets all requirements of safety-critical systems, i.e. provides reliable data delivery and low latency and low energy consumption, we present Dwarf, an energy-efficient, robust and dependable forwarding algorithm. The core idea is to use unicast-based partial flooding along with a delay-aware node selection strategy. Our analysis and extensive simulations of real-world scenarios show that Dwarf tolerates large fractions of link and node failures, yet is energy efficient enough to allow for an operational lifetime of several years.
The exposure of a path p is a measure of the likelihood that an object traveling along p is detected by a network of sensors and it is formally defined as an integral over all points x of p of the sensibility (the str...
ISBN:
(纸本)9783540730897
The exposure of a path p is a measure of the likelihood that an object traveling along p is detected by a network of sensors and it is formally defined as an integral over all points x of p of the sensibility (the strength of the signal coming from x) times the element of path length. The minimum exposure path (MEP) problem is, given a pair of points x and y inside a sensor field, find a path between x and y of a minimum exposure. In this paper we introduce the first rigorous treatment of the problem, designing an approximation algorithm for the MEP problem with guaranteed performance characteristics. Given a convex polygon P of size n with O(n) sensors inside it and any real number Ɛ > 0, our algorithm finds a path in P whose exposure is within an 1 + Ɛ factor of the exposure of the MEP, in time O(n/Ɛ2ψ), where ψ is a topological characteristic of the field. We also describe a framework for a faster implementation of our algorithm, which reduces the time by a factor of approximately Θ(1/Ɛ), by keeping the same approximation ratio.
Many-to-one transport results in congestion in wireless sensor networks. This is due to the sudden impulse of information flow to a single destination. Greater the number of event reporting nodes greater will be the d...
详细信息
Many-to-one transport results in congestion in wireless sensor networks. This is due to the sudden impulse of information flow to a single destination. Greater the number of event reporting nodes greater will be the degree of congestion; considering fixed transmission power for nodes. Therefore an important factor in the design of a congestion control scheme is the density of the network. In this paper we present a new congestion control scheme based on hop-by-hop packet delivery time and buffer size. The new congestion control scheme is at the transport layer and uses a TDMA-like mechanism to optimally adjust the reporting rate of events. Detailed simulation analysis confirm that the proposed congestion control scheme decreases packet drops and provide high packet delivery ratio (above 90%) from even very dense event reporting regions.
Activity recognition from on-body sensors is affected by sensor degradation, interconnections failures, and jitter in sensor placement and orientation. We investigate how this may be balanced by exploiting redundant s...
详细信息
Activity recognition from on-body sensors is affected by sensor degradation, interconnections failures, and jitter in sensor placement and orientation. We investigate how this may be balanced by exploiting redundant sensors distributed on the body. We recognize activities by a meta-classifier that fuses the information of simple classifiers operating on individual sensors. We investigate the robustness to faults and sensor scalability which follows from classifier fusion. We compare a reference majority voting and a naive Bayesian fusion scheme. We validate this approach by recognizing a set of 10 activities carried out by workers in the quality assurance checkpoint of a car assembly line. Results show that classification accuracy greatly increases with additional sensors (50% with 1 sensor, 80% and 98% with 3 and 57 sensors), and that sensor fusion implicitly allows to compensate for typical faults up to high fault rates. These results highlight the benefit of large on-body sensor network rather than a minimum set of sensors for activity recognition and prompts further investigation.
The widespread availability of digital EEG equipment and inexpensive computational resources make feasible the extraction of quantitative new information from the EEG such as the temporal distribution of interictal sp...
详细信息
The widespread availability of digital EEG equipment and inexpensive computational resources make feasible the extraction of quantitative new information from the EEG such as the temporal distribution of interictal spikes in patients with epilepsy. Unfortunately, there is no theoretical basis for analyzing the detected spikes to determine long-term seizure risk in this population. We have developed potentially useful algorithms for spike detection as well as analyses that are based on in vitro studies of the interaction between synapses and network function. These tools may provide new insights into the nature of epileptic networks and their evolution over time.
One of the challenge in developing smart sensor networks is the minimization of network delay or at the very least be able to have upper and lower boundaries of network delay when sensor nodes respond to higher level ...
详细信息
One of the challenge in developing smart sensor networks is the minimization of network delay or at the very least be able to have upper and lower boundaries of network delay when sensor nodes respond to higher level applications. In this paper, we present a highly efficient task scheduling method based on linear programming that integrates both sensing and networking communication delay. The objective is to minimize the total response time and global power consumption of the network with respect to the total number of sensor nodes in the network. Simulation results based on closed-form solutions for the task scheduling problem are presented for two scenarios with homogeneous and six scenarios with heterogeneous sensor nodes using single level tree-network topology. Specifically, for the heterogeneous scenarios, responding sequence that results in global optimum total respond time has also been found.
暂无评论