HUMAN AND SYSTEMS ENGINEERING:

1
HUMAN AND SYSTEMS ENGINEERING
Introduction to Particle Filtering
Sanjay Patil and Ryan Irwin Intelligent
Electronics Systems Human and Systems
Engineering Center for Advanced Vehicular
Systems URL www.cavs.msstate.edu/hse/ies/publica
tions/seminars/msstate/2005/particle_filtering/
2
Abstract

• The conventional techniques in speech recognition
  applications
• model speech as Gaussian mixtures
• lacks robustness to noise and mismatched channel
• Nonlinear techniques
• model speech as a time-varying and
  non-stationary signal
• Particle filtering
• a nonlinear method
• based on sequential Monte Carlo techniques
• a technique that can be used for prediction or
  filtering of signal
• works by approximating the target probability
  distribution (e.g. amplitude of speech signal)
• possible to increase the number of Gaussian
  mixtures to improve the prediction or filtering
  of signal.

3

• Drawing samples to represent a probability
  distribution function

Particles and their weights

• consider a pdf p(x) (blue line)
• generate random samples (red lines) which can
  represent this pdf (N of samples)
• Conclusion
• approximation depends on
• number (N) of samples
• amplitude (?x) of a sample (i) is its weight.
• each sample is called as Particle

4

• Particle filtering algorithm

Problem Statement find what x is at a given
time instant Observations known can be measured
(y1, y2, y3, y4 y5, y6, y7, yk-2, yk-1, yk,
...) States unknown (hence need to
calculated) (x0, x1, x2, x3, x4 x5, x6, x7,
xk-2, xk-1, xk, ) subscripts indicate time
index.
5

• Particle filtering algorithm continued

General two-stage framework (Prediction-Update
stages)

• Assume that pdf p(xk-1 y1k-1) is available at
  time k -1.
• Prediction stage
• This is a priori of the state at time k ( without
  the information on measurement). Thus, it is the
  probability of the state given only the previous
  measurements
• Update stage
• This is posterior pdf from predicted prior pdf
  and newly available measurement.

6

• Particle filtering algorithm step-by-step (1)

Initial set-up No observations available Known
parameters x0, p(x0), p(xkxk-1), p(ykxk),
noise statistics Draw samples to represent x0 by
its distribution p(x0)
time
Measurements / Observations
States (unknown / hidden) cannot be measured
N 5
(1.00, -1.176, 0.427, 0.906, 1.072)
7

• Particle filtering algorithm step-by-step (2)

Known parameters x0, p(x0), p(xkxk-1),
p(ykxk), noise statistics Still no observations
or measurements are available. Predict x1 using
equation
time
Measurements / Observations
States (unknown / hidden) cannot be measured
(0.5370, -0.9480, 0.63080, 1.51697, 0.39145 )
8

• Particle filtering algorithm step-by-step (3)

Known parameters x0, p(x0), p(xkxk-1),
p(ykxk), noise statistics First observation /
measurement is available. Update x1 using equation
time
Measurements / Observations
0.42
States (unknown / hidden) cannot be measured
(0.5370, 0.63080, 0.630, 0.630, 1.0 )
0.685
9

• Particle filtering algorithm step-by-step (4)

Known parameters x0, p(x0), p(xkxk-1),
p(ykxk), noise statistics Second observation /
measurement is NOT available. Predict x2 using
equation
time
Measurements / Observations
States (unknown / hidden) cannot be measured
(-1.651, 0.831, 1.888, 1.459, 2.540)
10

• Particle filtering algorithm step-by-step (5)

Known parameters x0, p(x0), p(xkxk-1),
p(ykxk), noise statistics Second observation /
measurement is available. update x2 using equation
time
-0.01
Measurements / Observations
States (unknown / hidden) cannot be measured
(-1.651, -1.651, 0.831, 0.831, 1.0 )
-0.12
11

• Particle filtering algorithm step-by-step (6)

Known parameters x0, p(x0), p(xkxk-1),
p(ykxk), noise statistics kth observation /
measurement is available. Predict and Update xk
using equation
12

• Particle filtering - visualization

Drawing samples
Predicting next state
Updating this state
What is THIS STEP???
Resampling.
13

• Applications

• Most of the applications involve tracking
• Visual Tracking e.g. human motion (body parts)
• Prediction of (financial) time series e.g.
  mapping gold price, stocks
• Quality control in semiconductor industry
• Military applications
• Target recognition from single or multiple images
• Guidance of missiles
• For IES NSF funded project, particle filtering
  has been used for
• Time series estimation for speech signal (Java
  demo)
• Speaker Verification (details on next slide)

14

• Speaker Verification

• Time series estimation of speech signal
• Speaker Verification
• Hypothesis particle filters approximate the
  probability distribution of a signal. If large
  number of particles are used, it approximates the
  pdf better. Only needed is the initial guess of
  the distribution.
• ! How are we going to achieve this..

15

• Pattern Recognition Applet

• Java applet that gives a visual of algorithms
  implemented at IES
• Classification of Signals
• PCA - Principal Component Analysis
• LDA - Linear Discrimination Analysis
• SVM - Support Vector Machines
• RVM - Relevance Vector Machines
• Tracking of Signals
• LP - Linear Prediction
• KF - Kalman Filtering
• PF Particle Filtering

URL http//www.cavs.msstate.edu/hse/ies/projects/
speech/software/demonstrations/applets/util/patter
n_recognition/current/index.html
16

• Classification Algorithms Best Case

• Data sets need to be differentiated
• Classifying distinguishes between sets of data
  without the samples
• Algorithms separate data sets with a line of
  discrimination
• To have zero error the line of discrimination
  should completely separate the classes
• These patterns are easy to classify

17

• Classification Algorithms Worst Case

• Toroidals are not classified easily with a
  straight line
• Error should be around 50 because half of each
  class is separated
• A proper line of discrimination of a toroidal
  would be a circle enclosing only the inside set
• The toroidal is not common in speech patterns

18

• Classification Algorithms Realistic Case

• A more realistic case of two mixed distributions
  using RVM
• This algorithm gives a more complex line of
  discrimination
• More involved computation for RVM yields better
  results than LDA and PCA
• Again, LDA, PCA, SVM, and RVM are pattern
  classification algorithms
• More information given online in tutorials about
  algorithms

19

• Signal Tracking Algorithms Kalman Filter

• Predicts the next state of the signal given prior
  information
• Signals must be time based or drawn from left to
  right
• X-axis represents time axis
• Algorithms interpolate data ensuring periodic
  sampling
• Kalman filter is shown here

20

• Signal Tracking Algorithms Particle Filter

• The model has realistic noise
• Gaussian noise is actually generated at each step
• Noise variances and number of particles can be
  customized
• Algorithm runs as previously described
• State prediction stage
• State update stage
• Each step gives a collection of possible next
  states of signal
• The collection is represented in the black
  particles
• Mean value of particles becomes the predicted
  state

21

• Summary

• Particle filtering promises to be one of the
  nonlinear techniques.
• More points to follow

22

• References

• S. Haykin and E. Moulines, "From Kalman to
  Particle Filters," IEEE International Conference
  on Acoustics, Speech, and Signal Processing,
  Philadelphia, Pennsylvania, USA, March 2005.
• M.W. Andrews, "Learning And Inference In
  Nonlinear State-Space Models," Gatsby Unit for
  Computational Neuroscience, University College,
  London, U.K., December 2004.
• P.M. Djuric, J.H. Kotecha, J. Zhang, Y. Huang, T.
  Ghirmai, M. Bugallo, and J. Miguez, "Particle
  Filtering," IEEE Magazine on Signal Processing,
  vol 20, no 5, pp. 19-38, September 2003.
• N. Arulampalam, S. Maskell, N. Gordan, and T.
  Clapp, "Tutorial On Particle Filters For Online
  Nonlinear/ Non-Gaussian Bayesian Tracking," IEEE
  Transactions on Signal Processing, vol. 50, no.
  2, pp. 174-188, February 2002.
• R. van der Merve, N. de Freitas, A. Doucet, and
  E. Wan, "The Unscented Particle Filter,"
  Technical Report CUED/F-INFENG/TR 380, Cambridge
  University Engineering Department, Cambridge
  University, U.K., August 2000.
• S. Gannot, and M. Moonen, "On The Application Of
  The Unscented Kalman Filter To Speech
  Processing," International Workshop on Acoustic
  Echo and Noise, Kyoto, Japan, pp 27-30, September
  2003.
• J.P. Norton, and G.V. Veres, "Improvement Of The
  Particle Filter By Better Choice Of The Predicted
  Sample Set," 15th IFAC Triennial World Congress,
  Barcelona, Spain, July 2002.
• J. Vermaak, C. Andrieu, A. Doucet, and S.J.
  Godsill, "Particle Methods For Bayesian Modeling
  And Enhancement Of Speech Signals," IEEE
  Transaction on Speech and Audio Processing, vol
  10, no. 3, pp 173-185, March 2002.