Scalable Information-Driven Sensor Querying and Routing for ad hoc Heterogeneous Sensor Networks Maurice Chu, Horst Haussecker and Feng Zhao Xerox Palo.

Slides:



Advertisements
Similar presentations
Dynamic Location Discovery in Ad-Hoc Networks
Advertisements

1 ECE 776 Project Information-theoretic Approaches for Sensor Selection and Placement in Sensor Networks for Target Localization and Tracking Renita Machado.
CLUSTERING IN WIRELESS SENSOR NETWORKS B Y K ALYAN S ASIDHAR.
Sensor Network 教育部資通訊科技人才培育先導型計畫. 1.Introduction General Purpose  A wireless sensor network (WSN) is a wireless network using sensors to cooperatively.
MiLAN: Middleware to Support Sensor Network Applications Wendi B. Heinzelman, Amy L. Murphy, Hervaldo S. Carvalho, Mark A. Perillo University of Rochester.
1 Scalable Information-Driven Sensor Querying and Routing for ad hoc Heterogeneous Sensor Networks Maurice Chu, Horst Haussecker and Feng Zhao Based upon.
Model base human pose tracking. Papers Real-Time Human Pose Tracking from Range Data Simultaneous Shape and Pose Adaption of Articulated Models using.
Chapter 4: Linear Models for Classification
Cooperative Multiple Input Multiple Output Communication in Wireless Sensor Network: An Error Correcting Code approach using LDPC Code Goutham Kumar Kandukuri.
Uncertainty Representation. Gaussian Distribution variance Standard deviation.
Routing in WSNs through analogies with electrostatics December 2005 L. Tzevelekas I. Stavrakakis.
Los Angeles September 27, 2006 MOBICOM Localization in Sparse Networks using Sweeps D. K. Goldenberg P. Bihler M. Cao J. Fang B. D. O. Anderson.
Visual Recognition Tutorial
Volkan Cevher, Marco F. Duarte, and Richard G. Baraniuk European Signal Processing Conference 2008.
Directed Diffusion: A Scalable and Robust Communication Paradigm for Sensor Networks.
A Beacon-Less Location Discovery Scheme for Wireless Sensor Networks Lei Fang (Syracuse) Wenliang (Kevin) Du (Syracuse) Peng Ning (North Carolina State)
Incorporating Language Modeling into the Inference Network Retrieval Framework Don Metzler.
Online Data Gathering for Maximizing Network Lifetime in Sensor Networks IEEE transactions on Mobile Computing Weifa Liang, YuZhen Liu.
Probability Grid: A Location Estimation Scheme for Wireless Sensor Networks Presented by cychen Date : 3/7 In Secon (Sensor and Ad Hoc Communications and.
Algorithm Evaluation and Error Analysis class 7 Multiple View Geometry Comp Marc Pollefeys.
Nonlinear Stochastic Programming by the Monte-Carlo method Lecture 4 Leonidas Sakalauskas Institute of Mathematics and Informatics Vilnius, Lithuania EURO.
Radial Basis Function Networks
Data Selection In Ad-Hoc Wireless Sensor Networks Olawoye Oyeyele 11/24/2003.
Integrated Social and Quality of Service Trust Management of Mobile Groups in Ad Hoc Networks Ing-Ray Chen, Jia Guo, Fenye Bao, Jin-Hee Cho Communications.
Distributed Quality-of-Service Routing of Best Constrained Shortest Paths. Abdelhamid MELLOUK, Said HOCEINI, Farid BAGUENINE, Mustapha CHEURFA Computers.
MiLAN: Middleware to Support Sensor Network Applications Jonghak Kim, Da Huo, Hyungik Oh.
Authors: Sheng-Po Kuo, Yu-Chee Tseng, Fang-Jing Wu, and Chun-Yu Lin
Energy Efficient Routing and Self-Configuring Networks Stephen B. Wicker Bart Selman Terrence L. Fine Carla Gomes Bhaskar KrishnamachariDepartment of CS.
Prakash Chockalingam Clemson University Non-Rigid Multi-Modal Object Tracking Using Gaussian Mixture Models Committee Members Dr Stan Birchfield (chair)
EM and expected complete log-likelihood Mixture of Experts
Optimal Power Control, Rate Adaptation and Scheduling for UWB-Based Wireless Networked Control Systems Sinem Coleri Ergen (joint with Yalcin Sadi) Wireless.
Sergios Theodoridis Konstantinos Koutroumbas Version 2
Particle Filters for Shape Correspondence Presenter: Jingting Zeng.
Bandwidth Constrained Energy Efficient Transmission Protocol in Wireless Sensor Networks Jain-Shing LIU and Chun-Hung Richard LIN,Nonmembers IEICE TRANS.
Value of Information (VOI) Theory Advisor: Dr Sushil K Prasad By: DM Rasanjalee Himali.
HQ U.S. Air Force Academy I n t e g r i t y - S e r v i c e - E x c e l l e n c e Improving the Performance of Out-of-Order Sigma-Point Kalman Filters.
1 E. Fatemizadeh Statistical Pattern Recognition.
Approximate Dynamic Programming Methods for Resource Constrained Sensor Management John W. Fisher III, Jason L. Williams and Alan S. Willsky MIT CSAIL.
An Energy Efficient Hierarchical Clustering Algorithm for Wireless Sensor Networks Seema Bandyopadhyay and Edward J. Coyle Presented by Yu Wang.
Probabilistic Coverage in Wireless Sensor Networks Authors : Nadeem Ahmed, Salil S. Kanhere, Sanjay Jha Presenter : Hyeon, Seung-Il.
A Passive Approach to Sensor Network Localization Rahul Biswas and Sebastian Thrun International Conference on Intelligent Robots and Systems 2004 Presented.
SCALABLE INFORMATION-DRIVEN SENSOR QUERYING AND ROUTING FOR AD HOC HETEROGENEOUS SENSOR NETWORKS Paper By: Maurice Chu, Horst Haussecker, Feng Zhao Presented.
Differential Ad Hoc Positioning Systems Presented By: Ramesh Tumati Feb 18, 2004.
Combs, Needles, Haystacks: Balancing Push and Pull for Discovery in Large Scale Sensor Networks Xin Liu Department of Computer Science University of California.
Dr. Sudharman K. Jayaweera and Amila Kariyapperuma ECE Department University of New Mexico Ankur Sharma Department of ECE Indian Institute of Technology,
1 Iterative Integer Programming Formulation for Robust Resource Allocation in Dynamic Real-Time Systems Sethavidh Gertphol and Viktor K. Prasanna University.
The Restricted Matched Filter for Distributed Detection Charles Sestok and Alan Oppenheim MIT DARPA SensIT PI Meeting Jan. 16, 2002.
SCALABLE INFORMATION-DRIVEN SENSOR QUERYING AND ROUTING FOR AD HOC HETEROGENEOUS SENSOR NETWORKS Paper By: Maurice Chu, Horst Haussecker, Feng Zhao Presented.
Performance of Adaptive Beam Nulling in Multihop Ad Hoc Networks Under Jamming Suman Bhunia, Vahid Behzadan, Paulo Alexandre Regis, Shamik Sengupta.
The Problem of Location Determination and Tracking in Networked Systems Weikuan Yu, Hui Cao, and Vineet Mittal The Ohio State University.
CS 484 Load Balancing. Goal: All processors working all the time Efficiency of 1 Distribute the load (work) to meet the goal Two types of load balancing.
Spectrum Sensing In Cognitive Radio Networks
Paging Area Optimization Based on Interval Estimation in Wireless Personal Communication Networks By Z. Lei, C. U. Saraydar and N. B. Mandayam.
ICIIS Peradeniya, Sri Lanka1 An Enhanced Top-Down Cluster and Cluster Tree Formation Algorithm for Wireless Sensor Networks H. M. N. Dilum Bandara,
Submitted by: Sounak Paul Computer Science & Engineering 4 th Year, 7 th semester Roll No:
TreeCast: A Stateless Addressing and Routing Architecture for Sensor Networks Santashil PalChaudhuri, Shu Du, Ami K. Saha, and David B. Johnson Department.
1 Information Content Tristan L’Ecuyer. 2 Degrees of Freedom Using the expression for the state vector that minimizes the cost function it is relatively.
Bayesian Brain Probabilistic Approaches to Neural Coding 1.1 A Probability Primer Bayesian Brain Probabilistic Approaches to Neural Coding 1.1 A Probability.
Part 3: Estimation of Parameters. Estimation of Parameters Most of the time, we have random samples but not the densities given. If the parametric form.
Biointelligence Laboratory, Seoul National University
Localization for Anisotropic Sensor Networks
Deep Feedforward Networks
Weikuan Yu, Hui Cao, and Vineet Mittal The Ohio State University
Distributed Energy Efficient Clustering (DEEC) Routing Protocol
Mikhail Bilenko, Sugato Basu, Raymond J. Mooney
Information Based Criteria for Design of Experiments
A Distributed Clustering Scheme For Underwater Sensor Networks
Overview: Chapter 4 (cont)
Overview: Chapter 2 Localization and Tracking
Probabilistic Surrogate Models
Presentation transcript:

Scalable Information-Driven Sensor Querying and Routing for ad hoc Heterogeneous Sensor Networks Maurice Chu, Horst Haussecker and Feng Zhao Xerox Palo Alto Research Center International Journal of High Performance Computing Applications, 2002

Motivation Advances in wireless networking, microfabrication (MEMS) and embedded microprocessors have enabled a new generation of large sensor networks.

Problem How to dynamically query sensors and route data in a network so that information gain is maximized while latency and bandwidth consumption is minimized.

Paper proposal - Information-driven sensor querying (IDSQ) to optimize sensor selection. - Constrained anisotropic diffusion routing (CADR) to direct data routing and incrementally combine sensor measurements so as to minimize an overall cost function.

What is new? The use of a general form of information utility that models the information content as well as the spatial configuration of a network in a distributed way.

Main Idea “Allows sensors to become activated when there are interesting events to report and only those parts of the network with the most useful information balanced by the communication cost need to be active”

Formal Formulation Sensing Model and Measure of Uncertainty General model: Zi = time dependent sensor characteristics λi(t) = sensors characteristics X(t) = parameters

Application of the formal formulation to acoustic sensors Stationary problem: sensors localization and sensors characteristics. Assumptions: –measure only the amplitude. –Signals propagate Isotropically.

Belief Representation of the current a posteriori distribution of x, given measurements Z1, … Zn. Estimation: minimum mean square estimation. The approximation of the residual uncertainty is determined by the covariance

Criteria to select the next best sensor In general, belief is applied to centralized approaches. To apply it, in the sensor networks, we have to consider communication. We cannot flood the network.

Sensor Selection Task: select an optimal subset of sensors and to decide to an optimal order of how to incorporate these measures in the belief update. Representation: the uncertainty is represented as a Gaussian (ellipsoid) distribution in a state space. Error: difference between distribution Gaussian area and the area after a sensor is added.

Utility function -All distribution probability in a space R. -Assign to each value of R its uncertainty. -Calculate a new distribution probability after adding a new sensor. -Compare sensors to find the best utility value.

Information Utility Measure Information content is inversely related to the “size” of the higher probability uncertainty region of the estimate of x.

Techniques 1.Covariance: square approximation of the ellipsoid. 2.Fisher Information matrix is based on the class of likelihood densities in a space: surface area of the high probability region. 3.Entropy of Estimation Uncertainty: entropy is the log of the volume of the set of typical values for a random variable. It should be used in the case that the distribution is not Gaussian.

Techniques 4.Volume of High Probability Region: related to the probability of values distributions. 5.Sensor Geometry Based Measures: function of the geometric location of the sensor.

Composite Objective Function Node is a leader. The leader holds the current belief. Information is sent to the leader from nodes of a certain distance. Objective function: a function of both information utility and cost of bandwidth and latency: utility function minus the cost (bandwidth plus latency) between the sensor j and sensor i.

Updating the Belief State Assuming that that x is uniformly distributed in a subset of R and that z is conditionally independent of x, the probability of every values z of x is determined by:

Information-Driven Sensor Query Assumptions: all sensors are synchronized; All node labeled (1 to n); clustering algorithm; Algorithm for choosing a leader. Communication between the leader and nodes in the cluster (leader knows characteristics of the nodes). Calculate the belief space. Start to track sensors that have being incorporated into the belief space. Evaluate Belief Quality test Incorporate new nodes until quality is good.

Constrained Anisotropic Diffusion Route 1.Choose the sensor that satisfies the constraints and maximizes the objective function in the neighborhood (in communication distance). 2.Choose the next routing node in the direction of the gradient of the objective function. 3.Propagate the belief evaluation and the necessary information to recalculate it.