Outline I.Stochastic Modeling for Pollution Tracking II.Simulations III.Filtering in a Random Environment IV.Overview of Center.

Slides:



Advertisements
Similar presentations
Random Processes Introduction (2)
Advertisements

State Estimation and Kalman Filtering CS B659 Spring 2013 Kris Hauser.
Tracking Unknown Dynamics - Combined State and Parameter Estimation Tracking Unknown Dynamics - Combined State and Parameter Estimation Presenters: Hongwei.
Dynamic Bayesian Networks (DBNs)
CHAPTER 8 A NNEALING- T YPE A LGORITHMS Organization of chapter in ISSO –Introduction to simulated annealing –Simulated annealing algorithm Basic algorithm.
STAT 497 APPLIED TIME SERIES ANALYSIS
Observers and Kalman Filters
Workshop on Stochastic Differential Equations and Statistical Inference for Markov Processes Day 1: January 19 th, Day 2: January 28 th Lahore University.
Bayesian Model Selection and Multi-target Tracking Presenters: Xingqiu Zhao and Nikki Hu Joint work with M. A. Kouritzin, H. Long, J. McCrosky, W. Sun.
Brief Introduction to the Alberta Group of MITACS-PINTS Center I. Groups: (Project Leader : Prof. Mike Kouritzin) University of Alberta (base), H.E.C.
Artificial Learning Approaches for Multi-target Tracking Jesse McCrosky Nikki Hu.
Prediction in Interacting Systems: Applications & Simulations Jarett Hailes November 1, 2002 dX t = μ(X t )dt + σ(X t )dB t dx = this- >mu()*dt + …
Outline Formulation of Filtering Problem General Conditions for Filtering Equation Filtering Model for Reflecting Diffusions Wong-Zakai Approximation.
Manifold Filtering Problem Lockheed Martin Jarett Hailes Jonathan Wiersma Richard VanWeelden July 21, 2003.
ODE and Discrete Simulation or Mean Field Methods for Computer and Communication Systems Jean-Yves Le Boudec EPFL MLQA, Aachen, September
Group problem solutions 1.(a) (b). 2. In order to be reversible we need or equivalently Now divide by h and let h go to Assuming (as in Holgate,
Transport Equations and Flux Laws Basic concepts and ideas 1.A concept model of Diffusion 2.The transient Diffusion Equation 3.Examples of Diffusion Fluxes.
Today Introduction to MCMC Particle filters and MCMC
Amos Storkey, School of Informatics. Density Traversal Clustering and Generative Kernels a generative framework for spectral clustering Amos Storkey, Tom.
© 2003 by Davi GeigerComputer Vision November 2003 L1.1 Tracking We are given a contour   with coordinates   ={x 1, x 2, …, x N } at the initial frame.
The Monte Carlo Method: an Introduction Detlev Reiter Research Centre Jülich (FZJ) D Jülich
Monte Carlo Methods in Partial Differential Equations.
Physics of fusion power
Adaptive Signal Processing Class Project Adaptive Interacting Multiple Model Technique for Tracking Maneuvering Targets Viji Paul, Sahay Shishir Brijendra,
MITACS-PINTS Prediction In Interacting Systems Project Leader : Michael Kouriztin.
Muhammad Moeen YaqoobPage 1 Moment-Matching Trackers for Difficult Targets Muhammad Moeen Yaqoob Supervisor: Professor Richard Vinter.
第四章 Brown运动和Ito公式.
Particle Filtering in Network Tomography
Mean Field Methods for Computer and Communication Systems Jean-Yves Le Boudec EPFL ACCESS Distinguished Lecture Series, Stockholm, May 28,
Dr. R. Nagarajan Professor Dept of Chemical Engineering IIT Madras Advanced Transport Phenomena Module 2 Lecture 4 Conservation Principles: Mass Conservation.
BsysE595 Lecture Basic modeling approaches for engineering systems – Summary and Review Shulin Chen January 10, 2013.
Object Tracking using Particle Filter
SIS Sequential Importance Sampling Advanced Methods In Simulation Winter 2009 Presented by: Chen Bukay, Ella Pemov, Amit Dvash.
What is a model Some notations –Independent variables: Time variable: t, n Space variable: x in one dimension (1D), (x,y) in 2D or (x,y,z) in 3D –State.
PINTS Network. Multiple Target Tracking Nonlinear Filtering Used for detection, tracking, and prediction of a target in a noisy environment Based entirely.
Presentation: Random walk models in biology E.A.Codling et al. Journal of The Royal Society Interface R EVIEW March 2008 Random walk models in biology.
BROWNIAN MOTION A tutorial Krzysztof Burdzy University of Washington.
Forward-Scan Sonar Tomographic Reconstruction PHD Filter Multiple Target Tracking Bayesian Multiple Target Tracking in Forward Scan Sonar.
EECS 274 Computer Vision Segmentation by Clustering II.
The Logistic Growth SDE. Motivation  In population biology the logistic growth model is one of the simplest models of population dynamics.  To begin.
4. Atmospheric chemical transport models 4.1 Introduction 4.2 Box model 4.3 Three dimensional atmospheric chemical transport model.
Mobile Robot Localization (ch. 7)
Week 21 Stochastic Process - Introduction Stochastic processes are processes that proceed randomly in time. Rather than consider fixed random variables.
Tracers for Flow and Mass Transport
Lab for Remote Sensing Hydrology and Spatial Modeling Dept of Bioenvironmental Systems Engineering National Taiwan University 1/45 GEOSTATISTICS INTRODUCTION.
Fokker-Planck Equation and its Related Topics
1 EE571 PART 3 Random Processes Huseyin Bilgekul Eeng571 Probability and astochastic Processes Department of Electrical and Electronic Engineering Eastern.
Paging Area Optimization Based on Interval Estimation in Wireless Personal Communication Networks By Z. Lei, C. U. Saraydar and N. B. Mandayam.
The Unscented Particle Filter 2000/09/29 이 시은. Introduction Filtering –estimate the states(parameters or hidden variable) as a set of observations becomes.
Stochastic Hydrology Random Field Simulation Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University.
Colorado Center for Astrodynamics Research The University of Colorado 1 STATISTICAL ORBIT DETERMINATION Kalman Filter with Process Noise Gauss- Markov.
A Probabilistic Appraoch to Nonlinear Diffusion: Feature-Driven Filtering Hamid Krim ECE Dept. NCSU, Raleigh, NC
Mean Field Methods for Computer and Communication Systems Jean-Yves Le Boudec EPFL Network Science Workshop Hong Kong July
Geometrically motivated, hyperbolic gauge conditions for Numerical Relativity Carlos Palenzuela Luque 15 December
Modeling Chemical Reactions Project Proposal
Diffusion over potential barriers with colored noise
水分子不時受撞,跳格子(c.p. 車行) 投骰子 (最穩定) 股票 (價格是不穏定,但報酬過程略穩定) 地震的次數 (不穩定)
Appendix A: Probability Theory
Outline Introduction Signal, random variable, random process and spectra Analog modulation Analog to digital conversion Digital transmission through baseband.
Assoc. Prof. Dr. Peerapol Yuvapoositanon
Mathematical Finance An Introduction
Introduction to particle filter
Brownian Motion for Financial Engineers
Introduction to particle filter
A Tutorial on Bayesian Speech Feature Enhancement
STOCHASTIC HYDROLOGY Random Processes
Brownian Motion & Itô Formula
Biointelligence Laboratory, Seoul National University
Lesson 4: Application to transport distributions
Presentation transcript:

Outline I.Stochastic Modeling for Pollution Tracking II.Simulations III.Filtering in a Random Environment IV.Overview of Center

1. Description Of Water Pollution Problem

n Factories along river or groundwater system n Undesired chemicals or bacteria released by each factory at random times n Chemicals initially distributed according to some proportional function n Contaminants react, drift, and disperse through water sheet n Quality of water ? Location of chief polluters ? How to predict the transport of the contaminants?

2. The Mathematical Model

... where : n u(t,x) - concentration of contaminants n D > 0 - dispersion rate n V - water velocity or drift rate n R - birth and death of bacteria or adsorption of chemicals n A i j - random contaminant deposits n  i j - random release times n u 0 (x) - initial contaminant concentration n b.c. - dispersive flux across the boundary = contaminant concentration does not change at boundary

3. Filtering Problem for Signal Process u(t,x) n Observations - sample pollution at discrete well sites or average of lake: Y k = h(u tk, V k ) n Optimal filter: i.e best guess E [f(u(t k ))|Y k ] Y k = σ {Y 1,...,Y k } n Find u N approximating u n Calculate the approximate filter: n E [f(u N (t k ))|Y k ]

4. Construction Of Markov Chain n Divide the region [0,L 1 ] x [0,L 2 ] into L 1 N x L 2 N cells n Construct discretized operators  N and  N n n k (t) - the number of particles in cell k at time t n {n k (t)} is modeled as a Markov chain n Particles evolve according to births and deaths from reactions, random walks from diffusion and drift, area dependent births from Poisson noise source n l -1 is mass of each particle

The Approximate Markov Process n is given by : n Semi-group and martingale theories can be used to analyze the mathematical structure of the Markov chain

5. The Law Of Large Numbers For both quenched and annealed approaches: u N (t) converges to u(t) and  N u N (t) converges to  u(t) in distribution sense as N   Quenched approach: evolving Markov chains for each fixed path of the driving Poisson source Annealed approach: considering the Poisson source as a random medium for the Markov chains Results in filtering theory state that we can construct approximate filters from u N The estimation of  u(t, x) can help locate the polluters

6. Model Development 1.Thomas Kurtz introduced and studied this type of Markov chain approximation for ODE’s. (1971) 2.Arnold and Theodosopulu extended model to the partial differential equations of chemical reactions. (1980) 3.Peter Kotelenez established high density limits for model. (1988) 4.Douglas Blount established a general technique for establishing crucial estimates in these models. (1991, 1994) 5.Kouritzin and Long revised model to i) speed up the implementation, ii) allow convection, iii) allow more general nonlinearities, and iv) Poisson-measure driving noise. New analysis methods were required.

THE DINGHY PROBLEM n A dinghy is lost at sea n The dinghy moves randomly n We know the underlying stochastic model n We only have very noisy observations from a high-altitude sensor n We want to track the location and the orientation of the dinghy n The goal  rescue the dinghy’s occupants

Nonlinear Filtering for Diffusions in Random Environments 1.Background for Signal Process Motivation: tracking problem of a dinghy lost at sea. Formal SDE for the motion of the dinghy: dX t = - 1 a (X t )  W (X t )dt + b(X t )dt +  (X t )dB t, 2 where W is a real-valued random field on R d that can be nowhere differentiable d a =   T and b = (b 1,…,b d ) with b i =   j a ij. j =1 B,W are independent random sources Example: dX t = b(X t ) dt +  (X t ) dB t models motion of dighy itself –½ a(X t )  W(X t )dt brings in effect of the waves.

Find Solution (X w t, P w t )via Dirichlet form theory,  w (u,v) =  R d e -w(x) dx/2, R on the Hilbert space H =L 2 (R d ; e -w (x) dx) with domain D(  w ) = {u  H: | a 1/2  u|  H}. P RR RP (W,  (W),Q) models the random environment. Consider the law P w f of X w t with initial law f(x)dx as a probability measure on C (R +, R d ). The mapping (W,f)  W  L 1 + (R d )  P w f is measurable. · u, v  D(  w ) Example: W(x) = W((0,x]) is a Brownian sheet = a zero mean Gaussian field with W (A), W (B) independent when A  B = , W (A  B) = W (A) + W (B) and E(W(A)) 2 =  2 (A).

Construction of diffusion in random environments: - Let f be our initial random probability measure on R d. RRP P We define a new probability measure on C (R +, R d ) by P w f (A) = E Q [P w f (A)]. P Then (X t, t  ) is a diffusion in the random medium W with initial law f if its law is P w f. - (X t, t  ) is not Markov! P - Can we come up with an equation for P w f (X [0,t]  dx| observations up to t)? i.e. Is there a useful filtering equation?

2. Filtering Model The signal process is the “diffusion” in a random medium defined as above. The observation model: Y t =  t 0 h(X s )ds + V t. To calculate the optimal filter E [  (X [0,t] )|  t ],  t =  {Y s, s  t}]

3. Filtering for Historical Processes: Quenched Approach R E Fix W  W, consider the historical process X w [0,t] (s)= X w t  s, s  R+ and calculate the optimal pathspace filter E [  (X w [0,t] )|  w t ]. Let P 0 be a new probability. measure that turns Y into a Brownian motion and  w [0,t] (  ) = E 0 [  (X w [0,t] ) A t |  w t ], A t = exp{  t 0 h(X s )dY s – ½  t 0 |h(X s ) | 2 ds} Obtain an Zakai equation for the unnormalized pathspace measure valued filtering process.

Solution is given in terms of multiple Wiener-Ito’s integrals.  w [0,t] (  ) =  R d T t w, 0  (x)f(x)dx+  t 0 [  R d T t 1 w, 0 (U w,t 1 t-t 1  )(x)f(x)dx]dY t 1 +  t 0  t 1 0 [  R d T t 2 w, 0 (U w,t 2 t 1 -t 2 (U w,t 1 t-t 1  )) (x) f(x)dx] dY t 2 dY t 1 + … Here T t w, 0 is the evolution operator for X under P 0 and U is an associated operator. Measurability of the joint distribution of (X w, Y w ) with respect to W. 3. Filtering for Historical Processes: Quenched Approach (continued)

4. Filtering for Diffusions in Random Environments: Annealed Approach. There is no known stochastic evolution equation for the filtering process associated with the diffusion in random medium based on the noisy observation. The random environments are not accessible, which must be averaged. out (i.e annealed approach):  [0,t] (  ) =   w [0,t] (  )Q(dW) T 0 t g =  T w,0 t g Q(dW) From the measurability we get the chaos expansion for the filtering process associated to our filtering model:  [0,t] (  ) =  R d T 0 t  (x) f(x)dx +  t 0 [  R d T 0 t1 (U w,t 1 t-t 1  ) (x) f(x)dx] dY t 1 +  t 0  t 1 0 [  R d T 0 t 2 (U w,t2 t 1 -t 2 (U w,t 1 t -t 1  ))(x)f(x)dx]dY t 2 dY t 1 + … Truncate expansion and approximate the stochastic integrals  Implementation method

5. Mathematical Background. 1)Brox and Schumacher independently introduced the one- dimensional diffusion in random medium using Ito-McKean time change techniques. (1986) 2)Tanaka showed this model is recurrent. (1993) 3)Many authors have established properties like self-similarity. 4)Mathieu introduced the multidimensional model using Dirichlet forms and studies the behaviour as the amplitude of B goes to zero. (1994) 5)Kouritzin, Long, and Sun estimate the paths of such processes based on corrupted, distorted, partial observations.

Brief Introduction to the Alberta Branch of MITACS-PINTS Center 1.Postdoctoral Fellow and Graduate Students Dr. Hongwei Long (PIms-MITACS Industrial PDF) Dr. Wei Sun (PIms-MITACS Industrial PDF) David Ballantyne (Graduate Student) Calvin Chan (Undergraduate Student) Hubert Chan (Graduate Student) Michelle Prefontaine (Graduate Student) Paul Wiebe (Graduate Student)

2.Simulation Front (i)Branching particle filtering; path-space filter, combination of filtering and parameter estimations, application to advanced historical tracking of dinghy lost at sea. (By David Ballantyne, Hubert Chan and Michael Kouritzin). (ii)Convolutional filters: application to a filtering model for mean reverting stochastic volatility using Levy driven prices as observation. (By Paul Wiebe, Michelle Prefontaine and Calvin Chan) (iii)Markov chain approximation: application to water pollution model which is characterized by a stochastic reaction- diffusion equation. (By David Ballantyne and Hubert Chan).

3.Theoretical Front (i)Uniqueness and weak convergence of solutions of the nonlinear filter equations (by Andrew Heunis and Vladimir Lucic, U. Waterloo): study the distributional uniqueness and weak convergence of the (normalized) filter equations. (ii)Convergence of Markov chain approximation to stochastic reaction diffusion equations. (by Michael Kouritzin and Hongwei Long). (iii)Nonlinear filtering for diffusions in random environments (by Michael A. Kouritzin, Hongwei Long and Wei Sun).

3. Theoretical Front (continued) (iv) Holder continuity for spatial and path processes via spectral analysis. (by D. Blount and Michael Kouritzin). (v) On a class of discrete generation interacting particle systems.(by P. Del Moral, Micheal Kouritzin and L. Miclo). (vi) A pathspace branching particle filter. (by Michael Kouritzin). (vii) Rates for branching particle approximation of continuous- discrete filters. (by D. Blount an Michael Kouritzin). (viii) Convolutional filters. (by Micheal Kouritzin and Paul Wiebe).

4.Sponsors and their interests (i) Lockheed Martin Naval Electronics and Surveillance System: surveillance and tracking, search and rescue, anti-narcotic smuggling, air traffic management, and global positioning. (ii) Lockheed Martin Canada, Montreal: same interests as above. (iii) VisionSmart, Edmonton: Quality control of industrial processes such as real time analysis of oriented strand board (OSB) density variations using thermography techniques and pattern recognition of naturally occurring substances etc. (iv) Acoustic Positioning Research Inc., Edmonton: Track stage performers using acoustic techniques and adjust lighting, sound effects. Create performer-movement- controlled. (v) Stantec (future): environmental monitoring, pollution tracking.

5. Ideas for the future (i) Filtering by Markov chain approximation. (ii) Tracking and estimation of bacteria and other species (iii) Filtering when observations in random environments. (iv) Implement chaos method and try out with random environments.