Signal interactions Cross correlation, cross spectral coupling and significance testing Centre for Doctoral Training in Healthcare Innovation

Similar documents
Introduction to time-frequency analysis Centre for Doctoral Training in Healthcare Innovation

Up/down-sampling & interpolation Centre for Doctoral Training in Healthcare Innovation

Probability and Statistics for Final Year Engineering Students

Stochastic Processes. M. Sami Fadali Professor of Electrical Engineering University of Nevada, Reno

Applied Probability and Stochastic Processes

ELEG 3143 Probability & Stochastic Process Ch. 6 Stochastic Process

Correlation, discrete Fourier transforms and the power spectral density

Wavelet entropy as a measure of solar cycle complexity

Wavelet analysis of the parameters of edge plasma fluctuations in the L-2M stellarator

L29: Fourier analysis

2. SPECTRAL ANALYSIS APPLIED TO STOCHASTIC PROCESSES

13.42 READING 6: SPECTRUM OF A RANDOM PROCESS 1. STATIONARY AND ERGODIC RANDOM PROCESSES

2A1H Time-Frequency Analysis II

3. ESTIMATION OF SIGNALS USING A LEAST SQUARES TECHNIQUE

Application of the cross wavelet transform and wavelet coherence to geophysical time series

Introduction to Signal Processing

Spectral Analysis of Random Processes

Problem Sheet 1 Examples of Random Processes

ECE302 Spring 2006 Practice Final Exam Solution May 4, Name: Score: /100

2A1H Time-Frequency Analysis II Bugs/queries to HT 2011 For hints and answers visit dwm/courses/2tf

Elec4621 Advanced Digital Signal Processing Chapter 11: Time-Frequency Analysis

EEM 409. Random Signals. Problem Set-2: (Power Spectral Density, LTI Systems with Random Inputs) Problem 1: Problem 2:

L6: Short-time Fourier analysis and synthesis

Correlator I. Basics. Chapter Introduction. 8.2 Digitization Sampling. D. Anish Roshi

Stochastic Processes: I. consider bowl of worms model for oscilloscope experiment:

Effects of data windows on the methods of surrogate data

Assignment #09 - Solution Manual

Statistics of Stochastic Processes

Advanced Digital Signal Processing -Introduction

Introduction to Biomedical Engineering

HST.582J/6.555J/16.456J

6.435, System Identification

Chapter 6. Random Processes

Notes on Wavelets- Sandra Chapman (MPAGS: Time series analysis) # $ ( ) = G f. y t

ECE 450 Homework #3. 1. Given the joint density function f XY (x,y) = 0.5 1<x<2, 2<y< <x<4, 2<y<3 0 else

Course content (will be adapted to the background knowledge of the class):

Adaptive Filtering. Squares. Alexander D. Poularikas. Fundamentals of. Least Mean. with MATLABR. University of Alabama, Huntsville, AL.

HST.582J / 6.555J / J Biomedical Signal and Image Processing Spring 2007

SEISMIC WAVE PROPAGATION. Lecture 2: Fourier Analysis

Stochastic Processes. A stochastic process is a function of two variables:

SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY UNIT 3 RANDOM PROCESS TWO MARK QUESTIONS

NONLINEAR TIME SERIES ANALYSIS, WITH APPLICATIONS TO MEDICINE

Classic Time Series Analysis

Lecture 15. Theory of random processes Part III: Poisson random processes. Harrison H. Barrett University of Arizona

Basics on 2-D 2 D Random Signal

ECE Homework Set 3

: The coordinate origin dependence of the phase distribution. Fig B(t) OBS OBS PRS PCS

MATSUYAMA CITY RAINFALL DATA ANALYSIS USING

Signal Modeling Techniques in Speech Recognition. Hassan A. Kingravi

EA2.3 - Electronics 2 1

Fourier Transform in Image Processing. CS/BIOEN 6640 U of Utah Guido Gerig (slides modified from Marcel Prastawa 2012)

Prof. Dr.-Ing. Armin Dekorsy Department of Communications Engineering. Stochastic Processes and Linear Algebra Recap Slides

WAVEPAL. A Python software for the frequency and wavelet analyses of irregularly sampled time series. Guillaume Lenoir

ECE-340, Spring 2015 Review Questions

MATSUYAMA CITY RAINFALL DATA ANALYSIS USING WAVELET TRANSFORM

TAKEHOME FINAL EXAM e iω e 2iω e iω e 2iω

EL1820 Modeling of Dynamical Systems

System Modeling and Identification CHBE 702 Korea University Prof. Dae Ryook Yang

EE/CpE 345. Modeling and Simulation. Fall Class 9

Random signals II. ÚPGM FIT VUT Brno,

Signal Processing With Wavelets

Fourier Analysis and Power Spectral Density

( nonlinear constraints)

ECG782: Multidimensional Digital Signal Processing

Jean Morlet and the Continuous Wavelet Transform

A6523 Signal Modeling, Statistical Inference and Data Mining in Astrophysics Spring 2013

State Space Representation of Gaussian Processes

A6523 Modeling, Inference, and Mining Jim Cordes, Cornell University. False Positives in Fourier Spectra. For N = DFT length: Lecture 5 Reading

Characteristic Behaviors of Wavelet and Fourier Spectral Coherences ABSTRACT

Lecture 30. DATA 8 Summer Regression Inference

Waves, the Wave Equation, and Phase Velocity. We ll start with optics. The one-dimensional wave equation. What is a wave? Optional optics texts: f(x)

Lecture Wigner-Ville Distributions

EEG- Signal Processing

A computationally efficient approach to generate large ensembles of coherent climate data for GCAM

Statistical Analysis of fmrl Data

Frequency Based Fatigue

Time Series. Anthony Davison. c

MATLAB Signal Processing Toolbox. Greg Reese, Ph.D Research Computing Support Group Academic Technology Services Miami University

Fundamentals of Noise

Gaussian Processes in Machine Learning

Theoretical and Simulation-guided Exploration of the AR(1) Model

PART 1. Review of DSP. f (t)e iωt dt. F(ω) = f (t) = 1 2π. F(ω)e iωt dω. f (t) F (ω) The Fourier Transform. Fourier Transform.

3. Lecture. Fourier Transformation Sampling

If we want to analyze experimental or simulated data we might encounter the following tasks:

Lecture 8: Signal Detection and Noise Assumption

IV. Covariance Analysis

CS 6604: Data Mining Large Networks and Time-series. B. Aditya Prakash Lecture #12: Time Series Mining

Fourier Transform for Continuous Functions

Package biwavelet. September 29, 2012

D.S.G. POLLOCK: BRIEF NOTES

Review of Fourier Transform

SNR Calculation and Spectral Estimation [S&T Appendix A]

Periodogram of a sinusoid + spike Single high value is sum of cosine curves all in phase at time t 0 :

Information and Communications Security: Encryption and Information Hiding

Gaussian Processes for Audio Feature Extraction

The General Linear Model (GLM)

Lecture 4 - Spectral Estimation

E 4101/5101 Lecture 6: Spectral analysis

EE/CpE 345. Modeling and Simulation. Fall Class 10 November 18, 2002

EE482: Digital Signal Processing Applications

Transcription:

Signal interactions Cross correlation, cross spectral coupling and significance testing Centre for Doctoral Training in Healthcare Innovation Dr. Gari D. Clifford, University Lecturer & Director, Centre for Doctoral Training in Healthcare Innovation, Institute of Biomedical Engineering, University of Oxford

Cross correlation Time domain Cross spectral coupling Fourier Wavelet Detecting significance in the time domain Parametric Non-parametric Detecting significance in the frequency domain Parametric Non-parametric

A time domain statistic A measure of similarity of two time series as a function of a time-lag applied to one of them Also known as cross-covariance, a sliding dot product or inner-product. Commonly used to search a long duration signal for a shorter, known feature. (* indicates complex conjugate)

Similar in nature to the convolution of two functions. Convolution involves reversing a signal, then shifting it and multiplying by another signal Correlation only involves shifting it and multiplying (no reversing) What s the correlation function between sine & cosine?

Cross correlation function Cos & Sin are the same, but π/2 out of phase So C has max/min (=±1) at ±π/2 (± 100 samples) Completely (anti-) correlated at this point Another max every 200 sample shift (with alternating sign) Function dies away at edges (less samples) Completely symmetric Theoretically equal to correlation coefficient r=max(c ) f=0.5;t=[1:1000]/200; x=sin(2*pi*f*t); y=cos(2*pi*f*t); [c lags] = xcorr(x,y,'coeff');

Cross correlation of sine with itself

FT(C) is the PSD! (Wiener Khinchin theorem, Wiener Khintchine theorem, Wiener Khinchin Einstein theorem or the Khinchin Kolmogorov theorem) http://mathworld.wolfram.com/wiener-khinchintheorem.html

Take two random time series y(1)=randn(1,1);for(i=2:1000); y(i)=y(i-1)/2+randn(1,1);end x(1)=randn(1,1);for(i=2:1000); x(i)=x(i-1)/2+randn(1,1);end A correlation >0 doesn t mean there is really any significant correlation But - size of correlation doesn t imply anything regarding significance

Parametric: Bartlett s (1935) correction (Orcutt & James, Biometrika (1948) Vol. 35, No. 3-4, 397-413) Nonparametric: Method of surrogates / bootstrap test (Politis, Statistical Science (2003) Vol. 18, No. 2, 219 230)

1. Measure correlation, Cr 2. Take one time series and shuffle order of samples (removes all temporal information) 3. Check correlation, Cn 4. Repeat step 1 N times (n=1:n) 5. Significance, p = length(cn<cr)/n (Proportion of times that we see a higher correlation from a random time series!)

What does the shuffling do? Preserves all statistics except autocorrelation Same mean, variance, skew, kurtosis Non-parametric technique Does not assume any distribution

Recall respiration from the ECG (EDR), HR (RSA) PPG, and IP: Test significance between correlation

Coherence analysis, or cross-spectral analysis, may be used to identify variations which have similar spectral properties (high power in the same spectral frequency bands) Similar to FFT results with real and imaginary coefficients The cross-spectrum is defined from the covariance function Cxy: Complex function: the power is: and the phase information is: The coherence spectrum is analogous to the conventional correlation coefficient and is defined as:

Two signals One single freq One dual freq Share a common freq

Coherence only at common frequency Normalized See: mschohere.m

Example with EDR and RSA c.f. STFT

Non-stationary frequency coupling W is the wavelet transform of x at scale a, and translation (time shift),. Let s consider the Morlet wavelet

Cross wavelet transform of two time series x(t) and y(t) is given by: Cross wavelet power: (common power in both time series) Wavelet Coherency: where <> represents a smoothing operator achieved by a convolution in time and scale:

COI Black arrows indicate the phase at a given time & frequency (point right for in-phase, left for anti-phase, down for X leading Y by 90 and up for Y leading X by 90)

Are my peaks real? Parametric tests False alarm probability when compared to the amplitude you would expect from a background noise (such as white noise) Non-parametric tests Bootstrap or surrogate methods phase randomisation

If the PSD, P(ω), is normalized Scargle shows that the distribution of P(ω) is exponential So the probability that P(ω) will be between some positive z and z + dz is exp( z)dz Therefore, if we scan some M independent frequencies, the probability that none give values larger than z is (1 e -z ) M. So P(> z) 1 (1 e -z ) M is the false alarm probability of the null hypothesis (that the data values are independent Gaussian random values) i.e. the significance level of any peak in P(ω) that we do see. A small value for the false-alarm probability indicates a highly significant periodic signal.

A small value for the false-alarm probability indicates a highly significant periodic signal

Instead of shuffling time locations, shuffle phases in Fourier domain Test cross spectral coherence of surrogates is > real coherence. If larger over many bootstrap iterations, we have significance Similar to time series bootstrap method

COI Black arrows indicate the phase at a given time & frequency (point right for in-phase, left for anti-phase, down for X leading Y by 90 and up for Y leading X by 90)

The MatLab wavelet coherence package: wtc-r16.zip http://www.pol.ac.uk/home/research/waveletcoherence/download.html Grinsted, A., S. Jevrejeva, J. Moore, "Application of the cross wavelet transform and wavelet coherence to geophysical time series." submitted to a special issue of Nonlinear Proc. Geophys., 'Nonlinear analysis of multivariate geoscientific data - advanced methods, theory and application', 2004 [pdf] Torrence, C., and G.P. Compo, A practical guide to wavelet analysis, Bull. Am. Meteorol. Soc., 79, 61-78, 1998.

Spectral estimation of unevenly sampled data without resampling Variable integration step size Equivalent to least squares fitting of sines to data!