Basic Definitions and The Spectral Estimation Problem

Lecture 1

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L1–1

Informal Definition of Spectral Estimation

Given: A finite record of a signal. Determine: The distribution of signal power over frequency.
signal t ω t=1, 2, ... ω ω+∆ω π spectral density

! = (angular) frequency in radians/(sampling interval) f = !=2 = frequency in cycles/(sampling interval)

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L1–2

Applications Temporal Spectral Analysis Vibration monitoring and fault detection Hidden periodicity finding Speech processing and audio devices Medical diagnosis Seismology and ground movement study Control systems design Radar, Sonar

Spatial Spectral Analysis Source location using sensor arrays
Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997 Slide L1–3

i!t exists and is called the Discrete-Time Fourier Transform (DTFT) Lecture notes to accompany Introduction to Spectral Analysis by P.1 = t= If: discrete-time deterministic data sequence X 1 t=. Moses. Stoica and R. 1997 Slide L1–4 .Deterministic Signals fy(t)g1 .1 y(t)e. Prentice Hall.1 jy(t)j2 < 1 X Then: Y (! ) = 1 t=.

1 y(t)y (t .1 X Z 1 where 4 S (!) = jY (!)j2 = Energy Spectral Density We can write S (!) = where X 1 k=. Stoica and R.1 X (k)e. 1997 . Prentice Hall. k) Slide L1–5 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.Energy Spectral Density Parseval's Equality: jy(t)j2 = 21 .i!k (k) = 1 t=. S (!)d! t=.

Random Signals Random Signal random signal probabilistic statements about future variations t current observation time Here: X 1 t=. Stoica and R. Moses. Prentice Hall. 1997 Slide L1–6 .1 n jy(t)j2 = 1 o But: E jy(t)j2 < 1 E f g = Expectation over the ensemble of realizations E jy(t)j2 n o = Average power in y (t) PSD = (Average) power spectral density Lecture notes to accompany Introduction to Spectral Analysis by P.

First Definition of PSD (! ) = where r X 1 k=. k)g r(k) = r (.k) r(0) jr(k)j Z Note that r(k) = 21 . Prentice Hall. Moses. (!)ei!k d! n o (Inverse DTFT) Interpretation: so r(0) = E jy(t)j2 = 21 . Stoica and R.i!k (k) is the autocovariance sequence (ACS) r(k) = E fy(t)y (t . 1997 Slide L1–7 . (!)d! Z (!)d! = infinitesimal signal power in the band ! d! 2 Lecture notes to accompany Introduction to Spectral Analysis by P.1 r(k)e.

1997 Slide L1–8 .i!t (!) = Nlim E N !1 t=1 X > : 8 > < 2 9 > = > Note that 1 jY (!)j2 (!) = Nlim E N N !1 YN (!) = X where N is the finite DTFT of fy(t)g. t=1 y(t)e.Second Definition of PSD 1 N y(t)e.i!t Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. Stoica and R. Prentice Hall.

Prentice Hall. 1997 Slide L1–9 . we can restrict attention to !2 .Properties of the PSD P1: (!) = (! + 2 ) for all !.! ) Otherwise: (! ) 6= (. Then: (! ) = (. P2: ] () f 2 . Stoica and R.! ) P3: If y Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.1=2 1=2] (!) 0 (t) is real. Thus.

k = unit delay operator: q.1y(t) = y(t .Transfer of PSD Through Linear Systems System Function: where q . 1997 Slide L1–10 . Prentice Hall. Moses.i!k H (! ) = X 1 y (!) = jH (!)j2 e(!) k=0 Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R.1 H (q) = X 1 k=0 hk q. 1) H (q ) y (t) 2 y (! ) = jH (! )j e( e(t) e (! ) Then !) y(t) = X 1 k=0 hk e(t . k) hk e.

]g Find an estimate of Two Main Approaches : Nonparametric: – Derived from the PSD definitions. Moses. Parametric: – Assumes a parameterized functional form of the PSD Lecture notes to accompany Introduction to Spectral Analysis by P.The Spectral Estimation Problem The Problem: From a sample fy(1) : : : y(N )g (!): f ^(!) ! 2 . Prentice Hall. 1997 Slide L1–11 . Stoica and R.

1997 Slide L2–1 .Periodogram and Correlogram Methods Lecture 2 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Moses. Stoica and R.

i!t (!) = Nlim E N !1 t=1 X > : 8 > < 2 9 > = > Given : Drop “ fy(t)gN t=1 lim ” and “E N !1 f g” to get X ^p(!) = 1 y(t)e.Periodogram Recall 2nd definition of (! ): 1 N y(t)e. Stoica and R. Prentice Hall.i!t N t=1 Natural estimator N 2 Used by Schuster ( 1900) to determine “hidden periodicities” (hence the name). Moses. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L2–2 .

1 X r(k)e.i!k (k)” by “^(k)”: r r ^(k)e. 1997 Slide L2–3 .1) Lecture notes to accompany Introduction to Spectral Analysis by P.1 N . Prentice Hall.i!k ^c(!) = k=.(N . Moses.Correlogram Recall 1st definition of (! ): (! ) = P X 1 Truncate the “ ” and replace “r k=. Stoica and R.

k) k < 0 r Lecture notes to accompany Introduction to Spectral Analysis by P.Covariance Estimators (or Sample Covariances) Standard unbiased estimate: 1 N y(t)y (t . k) k 0 r(k) = N ^ t=k+1 X For both estimators: r ^(k) = ^ (. k t=k+1 X Standard biased estimate: 1 N y(t)y (t . 1997 Slide L2–4 . k) k 0 ^ r(k) = N . Moses. Stoica and R. Prentice Hall.

Prentice Hall. Moses. N ^p(!) = 1 y(t)e. Stoica and R.i!t N t=1 X 2 = N .1 X = ^c(!) k=.Relationship Between ^p (! ) and ^c(! ) If: the biased ACS estimator r Then: ^(k) is used in ^c(!).1) r ^(k)e. 1997 .i!k ^p(!) = ^c(!) Consequence: Both p ! and ^( ) ^c(!) can be analyzed simultaneously. Slide L2–5 Lecture notes to accompany Introduction to Spectral Analysis by P.(N .

even for large N . p! f^( ) . 1997 Slide L2–6 . Prentice Hall.Statistical Performance of ^p(! ) and ^c(! ) Summary: Both are asymptotically (for large N ) unbiased: E ^p(!) ! (!) as N ! 1 n o Both have “large” variance. Thus.1 are small. ( )g ^ ( ) . Moses. Stoica and R. ^p(!) and ^c(!) have poor performance. r(k) may be large for large jkj ^ Even if the errors r k r k Nj. ( )] Lecture notes to accompany Introduction to Spectral Analysis by P. the PSD error is large. jk =0 there are “so many” that when summed in ! . Intuitive explanation: r(k) .

1 jkj r(k)e. n o 0 k 1 . Stoica and R.1 n o n o X ! X N . 1 jkj N Bartlett.(N . ) d Z Ideally: WB (!) = Dirac impulse (!). Moses.1 X wB (k) = 8 < : = Thus.1) N .(N . ( )WB (! . N k=. 1997 .i!k r k=. jNj jkj N . window E ^p(!) = 21 .i!k k=. Slide L2–7 Lecture notes to accompany Introduction to Spectral Analysis by P. or triangular.i!k = 1. Prentice Hall.1) 1 = wB (k)r(k)e.Bias Analysis of the Periodogram E ^p(!) = E ^c(!) = E f^(k)g e.

WB 1=N .Bartlett Window WB (! ) 1 sin(!N=2) 2 WB (!) = N sin(!=2) " # 0 WB (!)=WB (0). Stoica and R. 1997 . (!) may differ quite a bit from (!). Moses. for N = 25 −10 −20 dB −30 −40 −50 −60 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Main lobe 3dB width For “small” N . Slide L2–8 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.

1997 Slide L2–9 . φ(ω) smearing ω ω (!) separated in f by less than 1=N are not ^ φ(ω) <1/Ν Thus: Periodogram resolution limit = 1=N . Prentice Hall. ^ φ(ω)≅W (ω) B Sidelobe Level: leakage φ(ω)≅δ(ω) leakage ω ω Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses.Smearing and Leakage Main Lobe Width: smearing or smoothing Details in resolvable.

Periodogram Bias Properties Summary of Periodogram Bias Properties: For “small” N . so ! is asymptotically unbiased. severe bias As N . 1997 Slide L2–10 . Moses. !1 ^( ) ( )! ( ) Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Stoica and R. WB ! !.

Moses. (!1) ^p(!2) . Prentice Hall. (!2) 2(!1) !1 = !2 = 0 !1 6= !2 nh ih ( io Inconsistent estimate Erratic behavior ^ φ(ω) ^ φ(ω) asymptotic mean = φ(ω) ω + 1 st. Stoica and R.Periodogram Variance As N !1 E ^p(!1) . Lecture notes to accompany Introduction to Spectral Analysis by P. dev = φ(ω) too - Resolvability properties depend on both bias and variance. 1997 Slide L2–11 .

N Let ! = N Then YN ( 2 k) is the Discrete Fourier Transform (DFT): N X Y (k) = N t=1 y(t)W tk k = 0 ::: N . Moses. 1 Direct computation of O N 2 flops ( ) .i 2 .Discrete Fourier Transform (DFT) Finite DTFT: YN (!) = X N t=1 y(t)e.i!t 2 k and W = e.1 fY (k)gN=0 from fy(t)gN : t=1 k Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. 1997 Slide L2–12 . Stoica and R.

1 = 2p + 1: X t=1 ~ ~ y(t) + y(t + N )]W tp Y (2p + 1) = Each is a N ~ N ~ = N=2-point DFT computation. 2 = 2p: Y (2p) = For k X ~ N = 1 3 5 : : : N . 1997 . ~ 4 = 0 2 4 : : : N .1 for odd k Assume: X X X Let N For k ~ ~ = N=2 and W = W 2 = e.i2 =N . Slide L2–13 t=1 ~ ~ f y(t) . Moses. y(t + N )]W tgW tp Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R.Radix–2 Fast Fourier Transform (FFT) N = 2m N=2 N Y (k) = y(t)W tk + y(t)W tk t=1 t=N=2+1 N=2 2 = y(t) + y(t + N=2)W Nk ]W tk t=1 Nk +1 for even k with W 2 = . Prentice Hall.

FFT Computation Count Let ck Then = number of flops for N = 2k point FFT. Moses.1 k2k ) ck = 2 ck = 1 N log2 N 2 Thus. Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P. 2k + 2c ck = 2 k. Prentice Hall. 1997 Slide L2–14 .

1 ! N k=0 ^ φ(ω) continuous curve Lecture notes to accompany Introduction to Spectral Analysis by P.1 N k=0 sampled.Zero Padding Append the given data by zeros prior to computing DFT (or FFT): fy(1) : : : y(N ) 0 : : : 0g | N {z } Goals: Apply a radix-2 FFT (so N Finer sampling of = power of 2) ^(!): 2 k N . Moses. 1997 Slide L2–15 . N=8 ω 2 k N . Prentice Hall. Stoica and R.

Improved Periodogram-Based Methods Lecture 3 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. 1997 Slide L3–1 . Prentice Hall. Stoica and R.

^( ) jj ^BT (!) = M . 1997 Slide L3–2 . Moses.1) w(k)^(k)e.Blackman-Tukey Method Basic Idea: Weighted correlogram. Stoica and R. Prentice Hall.i!k r fw(k)g = Lag Window w(k) 1 -M M k Lecture notes to accompany Introduction to Spectral Analysis by P.1 X k=.(M . with small weight applied to covariances r k with “large” k .

Blackman-Tukey Method. Prentice Hall. Z W (!) = = Conclusion: Effect: DTFT fw(k)g Spectral Window ^BT (!) = “locally” smoothed periodogram Variance decreases substantially Bias increases slightly By proper choice of M : MSE = var + bias2 ! 0 as N ! 1 Slide L3–3 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 . )d 2 . Stoica and R. Moses. con't ^BT (!) = 1 ^p( )W (! .

Prentice Hall.(M (0) w Z w (k ) = equiv time width 1 2 . Moses.1) Ne = k=. 1997 Slide L3–4 .1 X . Stoica and R. )d 2 . w(k) is a psd sequence) ^BT (!) 0 (which is desirable) Then: Time-Bandwidth Product M . 0 Z | {z } If W (!) 0 (.Window Design Considerations Nonnegativeness: ^BT (!) = 1 ^p( ) W (! . W (!)d! = equiv bandwidth e= W (0) Ne e = 1 Lecture notes to accompany Introduction to Spectral Analysis by P.

) Choose M as a tradeoff between variance and bias. con't e = 1=Ne = 0(1=M ) is the BT resolution threshold. Moses. Prentice Hall. Stoica and R. Ne (and hence fixed. 1997 Slide L3–5 . As M increases. Once M is given. Lecture notes to accompany Introduction to Spectral Analysis by P. e) is essentially ) Choose window shape to compromise between smearing (main lobe width) and leakage (sidelobe level).Window Design. The energy in the main lobe and in the sidelobes cannot be reduced simultaneously. bias decreases and variance increases. once M is given.

M 0 −10 = 25 −20 dB −30 −40 −50 −60 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. M 0 = 25 −10 −20 dB −30 −40 −50 −60 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Rectangular Window. Prentice Hall. Stoica and R.Window Examples Triangular Window. 1997 Slide L3–6 .

Bartlett Method Basic Idea:
1 Μ 2Μ ... Ν

^1(ω) ^2(ω) φ φ

... average

^L(ω) φ

^B(ω) φ
Mathematically:

yj (t) = y((j ; 1)M + t) t = 1 : : : M = the j th subsequence 4 (j = 1 : : : L = N=M ])

^j (!) = 1 yj (t)e;i!t M t=1
X

M

2

1 L ^ (!) ^B (!) = L j =1 j
X

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L3–7

Comparison of Bartlett and Blackman-Tukey Estimates

1 L ^B (!) = L j =1
X

8 > < > :

M ;1 X

=

'
Thus:

k=;(M ;1) M ;1 X k=;(M ;1)

k=;(M ;1) 8 M ;1 < 1 X L X
:

r ^j (k)e;i!k
9 =

9 > = >

L j =1

r ^j (k) e;i!k

r ^(k)e;i!k

^B (!) ' ^BT (!)
Since B ! implicitly uses method has

with a rectangular lag window wR k

()

^( )

fwR(k)g, the Bartlett

High resolution (little smearing) Large leakage and relatively large variance
Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997 Slide L3–8

Welch Method Similar to Bartlett method, but allow overlap of subsequences (gives more subsequences, and thus “better” averaging) use data window for each periodogram; gives mainlobe-sidelobe tradeoff capability

1

2 . . . subseq subseq #1 #2 subseq #S

N

Let S # of subsequences of length M . (Overlapping means S > N=M “better averaging”.)

=

])

Additional flexibility: The data in each subsequence are weighted by a temporal window Welch is approximately equal to non-rectangular lag window.

^BT (!) with a
Slide L3–9

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

1 ! = N j =0 Idea: “Local averaging” of (2J + 1) samples in the j (2 + 1) frequency domain should reduce the variance by about J . Moses. 1997 Slide L3–10 .J X Lecture notes to accompany Introduction to Spectral Analysis by P.Daniell Method By a previous result. f ^p(!j )g are (nearly) uncorrelated random variables for 2 j N . for N 1. Stoica and R. Prentice Hall. k +J 1 ^D(!k ) = ^p (!j ) 2J + 1 j=k.

Prentice Hall.Daniell Method. Lecture notes to accompany Introduction to Spectral Analysis by P. Then. 1997 Slide L3–11 . for N 1. Stoica and R. ^D(!) ' 1 ^p(!)d! 2 . con't As J increases: Bias increases (more smoothing) Variance decreases (more averaging) = 2J=N . Hence: ^D (! ) ' ^BT (! ) with a rectangular spectral Let Z window. Moses.

^ () – Implemented by averaging subsample periodograms. ^p(!) by a suitably selected ^(k) using a lag – Implemented by truncating and weighting r window in c ! ^( ) Bartlett. Moses. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. ^BT (!) with a rectangular Slide L3–12 – Implemented by local averaging of periodogram values. more general for Welch). 1997 . Welch periodograms – Approximate interpretation: BT ! with a suitable lag window (rectangular for Bartlett. Daniell Periodogram – Approximate interpretation: spectral window.Summary of Periodogram Methods Unwindowed periodogram – reasonable bias – unacceptable variance Modified periodograms – Attempt to reduce the variance at the expense of (slightly) increasing the bias. BT periodogram – Local smoothing/averaging of spectral window.

Parametric Methods for Rational Spectra Lecture 4 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L4–1 . Stoica and R. Prentice Hall. Moses.

Prentice Hall. ! can approximate arbitrarily well any continuous PSD. Stoica and R.θ) Estimate parameters in φ(ω.Basic Idea of Parametric Spectral Estimation Observed Data Assumed functional form of φ(ω. provided m and n are chosen sufficiently large.i!k (!) = P jkj n k e. 1997 Slide L4–2 ( ) . Note. P By Weierstrass theorem.i! .θ) ^ θ Estimate PSD ^ ^ =φ(ω. however: choice of m and n is not simple some PSDs are not continuous Lecture notes to accompany Introduction to Spectral Analysis by P.θ) φ(ω) possibly revise assumption on φ(ω) Rational Spectra jkj m k e. Moses.i!k (!) is a rational function in e.

a rational factored as (!) can be B (!) 2 2 (!) = A(!) A(z) = 1 + a1z.n B (z) = 1 + b1z. Stoica and R.1 + + anz.m and.1 + + bmz.. A(! ) = A(z )jz =ei! Signal Modeling Interpretation: e(t) e (! ) = 2 - B (q) A(q) white noise B (!) 2 2 A(!) ltered white noise y(t)y (! ) = ARMA: AR: MA: A(q)y(t) = B (q)e(t) A(q)y(t) = e(t) y(t) = B (q)e(t) Slide L4–3 Lecture notes to accompany Introduction to Spectral Analysis by P.g.AR. MA. Prentice Hall. and ARMA Models By Spectral Factorization theorem. 1997 . Moses. e.

Stoica and R. Prentice Hall. k) and take E f g to give: X r (k ) + X 2 m bj hj . 1997 Slide L4–4 . j ) (b0 = 1) Multiply by y (t . k)g Lecture notes to accompany Introduction to Spectral Analysis by P.ARMA Covariance Structure ARMA signal model: y(t)+ X n i=1 n aiy(t . Moses. i) = m j =0 bj E fe(t .k (h = 1) where H (q ) = A(q ) 0 k k=0 X X i=1 air(k . j )y (t . i) = X m j =0 bj e(t .k = j =0 = 0 for k > m B (q) = 1 h q.

1) . 2 1 a1 = 0 . Lecture notes to accompany Introduction to Spectral Analysis by P. .AR Signals: Yule-Walker Equations AR: m = 0. . . Moses. 1997 Slide L4–5 . r(1) r(0) . . .n) . r (. . . Stoica and R.1) : : : r(. an 0 3 7 7 7 5 2 6 6 6 4 Writing covariance equation in matrix form for k : : : n: =1 r(0) r(. r(n) : : : r(0) 1 = 2 R 0 2 6 6 6 4 7 7 7 5 " # " 32 6 6 6 4 3 7 7 7 5 # These are the Yule–Walker (YW) Equations. Prentice Hall.

1) : : : ^(. 1997 Slide L4–6 .AR Spectral Estimation: YW Method Yule-Walker Method: Replace r 2 6 6 6 4 (k) by r(k) and solve for f^ig and ^2: ^ a r ^(0) ^(. . . ^(.n) 1 r r ^2 . r ^(n) : : : r(0) ^ ^n a 0 32 7 7 7 5 6 6 6 4 3 7 7 7 5 2 6 6 6 4 3 7 7 7 5 Then the PSD estimate is ^2 ^(!) = ^ 2 jA(!)j Lecture notes to accompany Introduction to Spectral Analysis by P. .1 = 0 a . Prentice Hall. Moses. . r ^(1) r(0) ^ ^. Stoica and R.1) . . r . .

i) = y(t) + 'T (t) = a1 : : : an]T to minimize f( ) = X N This gives 2 y=6 4 y(n + 1) y(n + 2) y(N ) . . 2) Lecture notes to accompany Introduction to Spectral Analysis by P. n) . Prentice Hall.AR Spectral Estimation: LS Method Least Squares Method: e(t) = y(t) + X n 4 = y(t) + y(t) ^ where '(t) = y (t . 1) y(N . 1) y(n + 1) y(n) . . Find i=1 aiy(t . 3 7 5 Y =4 y(n) y(n .(Y Y ). Moses. . n)]T . ^ = . . y(N . 1997 Slide L4–7 .1(Y y) where 3 7 5 2 6 t=n+1 je(t)j2 y(1) y(2) y(N . Stoica and R. 1) : : : y (t . . .

Prentice Hall. .n . . .. 1997 Slide L4–8 . 32 7 7 7 5 } 6 6 6 4 1 n 3 7 7 7 5 2 6 6 6 4 ^ k = either r(k) or r(k).1 .0 .Levinson–Durbin Algorithm Fast..1 1 . Direct Solution: For one given value of n: For k Rn+1 1 0 = 0 . Stoica and R. . Moses... . n . {z . .. order-recursive solution to YW equations 2 6 6 6 4 | 0 . 0 2 n 3 7 7 7 5 O(n3) flops = 1 : : : n: O(n4) flops Levinson–Durbin Algorithm: Exploits the Toeplitz form of Rn+1 to obtain the solutions for k : : : n in O n2 flops! =1 ( ) Lecture notes to accompany Introduction to Spectral Analysis by P.

con't Relevant Properties of R: Rx = y $ Rx = y. Moses. 2 4 Rn+1 n+1 r ~n n+1 r ~n 0 3 5 2 r ~n = 6 4 .Levinson-Durbin Alg. Stoica and R. where x = xn : : : x1]T ~ ~ ~ Nested structure 2 Rn+2 = Thus. Prentice Hall. 1997 Slide L4–9 . n 3 7 5 1 Rn+2 4 1 n = 0 3 5 2 4 Rn+1 n+1 r ~n n+1 r ~n 0 32 54 1 n n = 0 0 n 3 5 2 4 2 3 5 where r n = n+1 + ~n n Lecture notes to accompany Introduction to Spectral Analysis by P. .

con't 2 Rn+2 4 82 < 4 +2 : 1 n n = 0 0 n 3 5 2 4 2 3 5 2 Rn+2 4 39 = 5 2 0 ~n = 0n 2 1 n 3 5 2 4 3 5 2 3 5 Combining these gives: Rn 0 1 n + kn ~n 0 1 3 5 2 4 = 4 n + kn n 2 0 n + kn 2 n = 4 n 2 +1 3 5 0 0 Thus. jknj2) 2 2 n " n # " ~n 1 # Computation count: k flops for the step k 2 ) n2 flops !k+1 2 to determine f k k gn=1 k Slide L4–10 This is O (n2) times faster than the direct solution. Stoica and R. 1997 . n = n ) n+1 = 0 + kn 2+1 = n + kn n = n(1 . Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. 2 kn = .Levinson-Durbin Alg. Moses.

Moses. 1) + + bme(t .m X Lecture notes to accompany Introduction to Spectral Analysis by P. r(k) = 0 for jkj > m and 2 2 = m r(k)e. m) Thus. Stoica and R. 1997 Slide L4–11 . Prentice Hall.i!k (!) = jB(!)j k=.MA Signals MA: n =0 y(t) = B (q)e(t) = e(t) + b1e(t .

1997 Slide L4–12 . with AR model order n . Insert sample covariances (! ) = This is length X m f^(k)g in: r r(k)e.MA Spectrum Estimation Two main ways to Estimate (! ): 1. Estimate fbkg and 2 and insert them in (!) = jB(!)j2 2 nonlinear estimation problem ^(!) is guaranteed to be 0 2. ^(!) is not guaranteed to be 0 =0 Both methods are special cases of ARMA methods described below.m ^BT (!) with a rectangular lag window of 2m + 1. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P.i!k k=. Stoica and R. Moses.

k)] g = E f A(q)y(t)] A(q)y(t . Stoica and R. Prentice Hall.m k e. A(q)y(t) = B (q)e(t) B (!) 2 = m=. k)] g = X n X n j =0 p=0 aj ap r(k + p .ARMA Signals ARMA models can represent spectra with both peaks (AR part) and valleys (MA part).i!k k (!) = 2 A(!) jA(!)j2 P where k = E f B(q)e(t)] B(q)e(t . j ) Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. 1997 Slide L4–13 .

Estimate fai r(k)g in (!) = P m .ARMA Spectrum Estimation Two Methods: 2g in (!) = 2 B(!) 2 1. Estimate fai bj A(!) nonlinear estimation problem. ^(!) is not guaranteed to be 0 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.m k e jA(!)j2 linear estimation problem (the Modified Yule-Walker method). Prentice Hall. Stoica and R. can use an approximate linear two-stage least squares method ^(!) is guaranteed to be 0 2.i!k k=. 1997 Slide L4–14 .

Prentice Hall.Two-Stage Least-Squares Method Assumption: The ARMA model is invertible: A e(t) = B(q) y(t) (q ) = y(t) + 1y(t . K ) ^2 = N 1 j^(t)j2 e N . K t=K +1 X Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L4–15 . f kgK=1 by using AR k 1b) Estimate the noise sequence ^(t) = y(t) + ^1y(t . 1) + 2y(t . for some large K e(t) ' y(t) + 1y(t . 2) + = AR(1) with j kj ! 0 as k ! 1 Step 1: Approximate. 1) + e and its variance + ^K y(t . K ) 1a) Estimate the coefficients modelling techniques. 1) + + K y(t . Stoica and R. Moses.

1997 Slide L4–16 . ^(t) ' . 1) : : : . con't fe(t)g by ^(t) in the ARMA equation. Moses. m)] e e = a1 : : : an b1 : : : bm]T Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R.y(t . Prentice Hall. e A(q)y(t) ' B (q)^(t) e and obtain estimates of fai bj g by applying least squares Step 2: Replace techniques. 1) : : : ^(t . Note that the ai and bj coefficients enter linearly in the above equation: y(t) .Two-Stage Least-Squares Method. y(t . n) e ^(t .

1997 Slide L4–17 . : : : r(m . r(m + M ) r(m + 1) r(m + 2) 3 7 If M n. . r f^ g . . . i) = 0 k > m 3 7 5 " r(m) r(m + 1) .6 4 Replace fr(k)g by f^(k)g and solve for faig. 1) : : : r(m . . . Prentice Hall. = m + 1 ::: m + M a1 an . n + 1) r(m . fast Levinson-type algorithms exist for obtaining ai . 5 . . If M > n overdetermined squares solution for ai . .. Stoica and R. . # 2 = r(m + M . Moses. the accuracy of ai is often better for M > n Lecture notes to accompany Introduction to Spectral Analysis by P. n + 2) . n + M ) .. least f^ g Note: For narrowband ARMA signals. = f^ g Y W system of equations.Modified Yule-Walker Method ARMA Covariance Equation: r(k) + 2 6 4 X n In matrix form for k i=1 air(k .

Moses. Stoica and R.Summary of Parametric Methods for Rational Spectra Method AR: YW or LS low low-medium medium-high medium-high medium low-medium No No Yes Computational Burden low Accuracy medium Guarantee ^(! ) 0 ? Yes MA: BT ARMA: MYW ARMA: 2-Stage LS Use for Spectra with (narrow) peaks but no valley Broadband spectra possibly with valleys but no peaks Spectra with both peaks and (not too deep) valleys As above Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. 1997 Slide L4–18 .

1997 Slide L5–1 . Moses.Parametric Methods for Line Spectra — Part 1 Lecture 5 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Stoica and R.

Prentice Hall. sonar geophysical seismology ARMA model is a poor approximation Better approximation by Discrete/Line Spectrum Models 2 6(2 1) 2 6 (!) 6 (2 2) 2 6 (2 2) 3 ! Slide L5–2 . !1 !2 !3 - An “Ideal” line spectrum Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 . Moses. Examples: communications radar. Stoica and R.Line Spectra Many applications have signals with (near) sinusoidal components.

Stoica and R. 1997 . uniformly distributed on . power f g f g y(t) = x(t) + e(t) t = 1 2 : : : n x(t) = k ei(!k t+ k ) k=1 xk (t) X | {z } Assumptions: A1: A2: k>0 !k 2 . Prentice Hall. ] (prevents model ambiguities) f'k g = independent rv's.Line Spectral Signal Model Signal Model: Sinusoidal components of frequencies !k and powers 2 . superimposed in white noise of k 2. Moses. ] (realistic and mathematically convenient) A3: e(t) = circular white noise with variance 2 E fe(t)e (s)g = 2 t s E fe(t)e(s)g = 0 (can be achieved by “slow” sampling) Slide L5–3 Lecture notes to accompany Introduction to Spectral Analysis by P.

r(k) = E fy(t)y (t . for p = j e i'p e. !p) + 2 p Slide L5–4 Lecture notes to accompany Introduction to Spectral Analysis by P.i'j o e Z n 1 i' d' 2 = 0. for p 6= j =2 . 1997 .i'j o = 1. e E xp(t)xj (t .Covariance Function and PSD Note that: E E n i'p e. k)g = n=1 2ei!pk + 2 k 0 p p P and (! ) = 2 X n p=1 2 (! . Stoica and R. Prentice Hall.i'j o = E nei'p o E ne. Moses. k) = 2 ei!pk p j p n o Hence.

1997 Slide L5–5 . ^ ei!k t + residuals k Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Moses. Stoica and R.Parameter Estimation Estimate either: f!k k 'k gn=1 k 2 (Signal Model) f!k 2gn=1 2 k k Once (PSD Model) ^ Major Estimation Problem: f!k g f!kg are determined: ^ X f ^2g can be obtained by a least squares method from k n 2ei!pk + residuals ^ ^ r(k) = p p=1 OR: Both k and 'k can be derived by a least squares method from f^ g f^ g X y(t) = with n k= k=1 k ei'k .

B = eiN!1 eiN!n k 2 6 4 3 7 5 Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. . 1997 Slide L5–6 . .Nonlinear Least Squares (NLS) Method min' g f! Let: X N k k k t=1 | y(t) . Moses. . X n k=1 {z F (! ') k ei(!k t+'k ) 2 } = kei'k = 1 : : : n]T Y = y(1) : : : y(N )]T ei!1 ei!n . Prentice Hall.

Prentice Hall. B ) = kY .1B Y This gives: ^ = (B B).1B Y ^ ! Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. con't Then: F = (Y . (B B).1B Y !=^ ! and ! = arg max Y B (B B ).Nonlinear Least Squares (NLS) Method. Stoica and R.1B Y ] B B] . B k2 = . Y B(B B). 1997 Slide L5–7 . (B B).1B Y ] +Y Y . B ) (Y .

Stoica and R.5. 1997 Slide L5–8 . Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. it is difficult to avoid convergence to local minima. Moses.NLS Properties Excellent Accuracy: 6 2 (for N var (^k ) = 3 2 ! N k Example: N = 300 SNRk = 2= 2 = 30 dB k q 1) Then var (^k ) 10. ! Difficult Implementation: The NLS cost function F is multimodal.

Unwindowed Periodogram as an Approximate NLS Method

For a single (complex) sinusoid, the maximum of the unwindowed periodogram is the NLS frequency estimate: Assume: Then:

n=1
X

B B=N
N t=1

BY=

y(t)e;i!t = Y (!)

(finite DTFT)

1 Y B (B B );1B Y = N jY (!)j2 = ^p(!) = (Unwindowed Periodogram)
So, with no approximation,

! = arg max ^p(!) ^ !

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L5–9

Unwindowed Periodogram as an Approximate NLS Method, con't

Assume: Then:

n>1

f!kgn=1 ' the ^ k
provided that

locations of the peaks of p !

^( )

n

largest

inf j!k ; !pj > 2 =N
which is the periodogram resolution limit. If better resolution desired then use a High/Super Resolution method.

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L5–10

High-Order Yule-Walker Method Recall:

y(t) = x(t) + e(t) =

k ei(!k t+'k ) +e(t) | {z } k=1 xk (t)
X

n

“Degenerate” ARMA equation for y (t):
n

(1 ; ei!k q;1)xk (t) = k ei(!kt+'k) ; ei!k ei !k(t;1)+'k] = 0
o

Let

B (q) = 1 +

4 bk q;k = A(q)A(q) k=1 A(q) = (1 ; ei!1 q;1) (1 ; ei!n q;1) A(q) = arbitrary
X

L

Then B

(q)x(t) 0 ) B (q)y(t) = B (q)e(t)
Slide L5–11

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

f!kgn=1. Prentice Hall. 1997 Slide L5–12 . Stoica and R. n ^ k Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. con't Estimation Procedure: Estimate f^igL=1 using an ARMA MYW technique b i ^( ) Roots of B q give “spurious” roots.High-Order Yule-Walker Method. along with L .

r(L + M . Prentice Hall. . Stoica and R. 1) : : : r(M ) r(L + M ) 4 4 = = {z } | {z r(L) r(L + 1) This is a high-order (if L > n) and overdetermined (if M > L) system of YW equations.High-Order and Overdetermined YW Equations ARMA covariance: r (k ) + X L i=1 bir(k . . 1997 Slide L5–13 . r(L + 2) . . i) = 0 k > L In matrix form for k 2 6 6 6 4 | = L + 1 ::: L + M 3 7 7 7 5 2 6 6 6 4 3 7 7 7 5 } : : : r(1) r(L + 1) : : : r(2) b = . Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. . .

Stoica and R.High-Order and Overdetermined YW Equations. 1997 Slide L5–14 . The Minimum-Norm solution is b = . Moses. Prentice Hall. (U V )b = . diagonal and nonsingular Thus. y = . Lecture notes to accompany Introduction to Spectral Analysis by P. if the Minimum-Norm solution b is used.V .1U Important property: The additional (L . n) spurious zeros of B (q ) are located strictly inside the unit circle. con't Fact: SVD of rank : ( )=n =U V U = (M n) with U U = In V = (n L) with V V = In = (n n).

Moses. Prentice Hall. Stoica and R. this is the price paid for increased accuracy when L > n. f^ g When the SNR is low.HOYW Equations. this approach may give spurious frequency estimates when L > n. 1997 Slide L5–15 . r . V be defined similarly to U . .1U ^ ^ ^ b ^(q) that Then !k n=1 are found from the n zeroes of B k are closest to the unit circle. Practical Solution Let ^= ^ ^ ^ ^ but made from f^(k)g instead of fr(k)g. Lecture notes to accompany Introduction to Spectral Analysis by P. Compute ^ = .V ^ . SVD of . V from the Let U .

1997 Slide L6–1 . Stoica and R.Parametric Methods for Line Spectra — Part 2 Lecture 6 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. Prentice Hall.

i! : : : e.1)! ]T A = a(!1) : : : a(!n)] (m n) rank Note: Define (A) = n y(t) y(t .. Moses.i(m. m + 1)]T e 4 R = E fy(t)~ (t)g = APA + 2I ~ y Then with 2 P = E fx(t)~ (t)g = ~ x 6 4 2 1 0 . 1) (for m 3 7 7 7 5 n) 2 46 y(t) = 6 ~ 6 4 y(t . 0 2 n 3 7 5 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L6–2 . Stoica and R. Prentice Hall. .The Covariance Matrix Equation Let: a(!) = 1 e. = Ax(t) + ~(t) ~ e where x(t) = x1(t) : : : xn(t)]T ~ ~(t) = e(t) : : : e(t .. m + 1) .

. 2 R = S G] 6 4 1 3 " . Stoica and R.ng: orthonormal eigenvectors associated with f n+1 : : : m g S = s1 : : : sn] (m n) G = g1 : : : gm. Prentice Hall.. Moses.n] (m (m . n)) Thus.Eigendecomposition of R and Its Properties R = APA + 2I (m > n) Let: 1 2 ::: m: eigenvalues of R fs1 : : : sng: orthonormal eigenvectors associated with f 1 : : : ng fg1 : : : gm. 1997 Slide L6–3 . m 7 5 S G # Lecture notes to accompany Introduction to Spectral Analysis by P.

= = nonsingular 0 n. ..1) = AC S = A(PA S ( ) = rank(A) = n). con't As rank (APA ) = n: k > 2 k = 1 ::: n k = 2 k = n + 1 ::: m 0 1. =0 A G=0 Slide L6–4 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 . 2 2 6 4 3 7 5 2 6 4 Note: RS = APA S + 2S = S 1 0 . Moses.. 0 n 3 7 5 with C (since rank S . since S G j j 6= 0 4 .Eigendecomposition of R and Its Properties. Stoica and R. Prentice Hall. 2 .. Therefore.

Moses.MUSIC Method a (!1) . Stoica and R. . f!kgn=1 are the unique solutions of k a (!)GG a(!) = 0. Prentice Hall. A G= G=0 a (!n) ) fa(!k )gn=1 ? R(G) k 2 6 4 3 7 5 Thus. Let: 1 N y(t)~ (t) ^ ~ y R = N t=m ^^ S G = S G made from the ^ eigenvectors of R X Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L6–5 .

Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. Here.Spectral and Root MUSIC Methods Spectral MUSIC Method: f!kgn=1 = the locations of the n highest peaks of the ^ k “pseudo-spectrum” function: 1 ^G a(!) ! 2 .1 : : : z.1)GG a(z) = 0 that are closest to the unit circle.1)]T Note: Both variants of MUSIC may produce spurious frequency estimates. a (!)G ^ Root MUSIC Method: ] f!kgn=1 = the angular positions of the n roots of: ^ k ^^ aT (z. Moses. a(z) = 1 z. Stoica and R. 1997 Slide L6–6 .(m.

Stoica and R. Prentice Hall. If: =n+1 m=n+1 Then: ^ g G = ^1. ) f!kgn=1 can be found from the roots of ^ k aT (z. 1997 Slide L6–7 .Pisarenko Method Pisarenko is a special case of MUSIC with m (the minimum possible value). Moses.1)^1 = 0 g no problem with spurious frequency estimates computationally simple (much) less accurate than MUSIC with m n+1 Lecture notes to accompany Introduction to Spectral Analysis by P.

Uses m n (as in MUSIC). Moses.Min-Norm Method Goals: Reduce computational burden. and reduce risk of false frequency estimates. Stoica and R. Prentice Hall. but only one vector in G (as in Pisarenko). 1997 Slide L6–8 . g # Lecture notes to accompany Introduction to Spectral Analysis by P. with first element equal ^ ^ to one. that has minimum Euclidean norm. R( ) Let " 1 = the vector in R(G).

Prentice Hall. Stoica and R. 1997 Slide L6–9 .Min-Norm Method.1) 1 ^ g " # that are closest to the unit circle. Moses. Lecture notes to accompany Introduction to Spectral Analysis by P. con't Spectral Min-Norm f!gn=1 = ^k the locations of the n highest peaks in the “pseudo-spectrum” 2 1 = a (!) 1 ^ g " # Root Min-Norm f!gn=1 = ^k the angular positions of the n roots of the polynomial aT (z.

1 gm " # " # Min-Norm Method: Determining g ^ Then: 1 2 R(G) ) S 1 = 0 ^ ^ ^ ^ g g ) S ^= . Stoica and R. 1997 Slide L6–10 .) ^^ I=S S= Multiplying the above equation by gives: (1 .Let S ^= S g1 . k k2) ) ^ = . (S S ). g " # Min-Norm solution: As: ^ = . at least. Moses. k k2) = (S S ) ) (S S ).1 g + S S . k k2) g Lecture notes to accompany Introduction to Spectral Analysis by P. for N 1. Prentice Hall.S =(1 .1 = =(1 .1 exists iff = k k2 6= 1 (This holds.S(S S ).

.i!n S1 = Im. where e.1 0]A A2 = 0 Im.i!1 0 . Prentice Hall. Stoica and R.1]S 2 6 4 | 3 7 5 Recall S = AC with jC j 6= 0. Then S2 = A2C = A1DC = S1 C .ESPRIT Method Let A1 = Im. D= 0 e.1 0]S S2 = 0 Im.1]A Then A2 Also. let = A1D. 1997 Slide L6–11 . determined as is uniquely = (S1S1)..1DC {z } So has the same eigenvalues as D . Moses.1S1S2 Lecture notes to accompany Introduction to Spectral Analysis by P.

Stoica and R. ^ ^ ^ ^ The frequency estimates are found by: where f^kgn=1 are the eigenvalues of k ^ = (S1S1). then S1 and S2 .1S1S2 ^^ ^^ f!kgn=1 = . 1997 Slide L6–12 . Moses. arg(^k ) ^ k ESPRIT Advantages: computationally simple no extraneous frequency estimates (unlike in MUSIC or Min–Norm) accurate frequency estimates Lecture notes to accompany Introduction to Spectral Analysis by P.ESPRIT Implementation From the eigendecomposition of R. Prentice Hall. find S .

Stoica and R. Prentice Hall. Moses. 1997 .Summary of Frequency Estimation Methods Method Periodogram Nonlinear LS Yule-Walker Pisarenko MUSIC Min-Norm ESPRIT Computational Accuracy / Risk for False Burden Resolution Freq Estimates small medium-high medium very high very high very high medium high medium small low none high high medium medium high small medium very high none Recommendation: Use Periodogram for medium-resolution applications Use ESPRIT for high-resolution applications Slide L6–13 Lecture notes to accompany Introduction to Spectral Analysis by P.

1997 Slide L7–1 . Moses. Stoica and R. Prentice Hall.Filter Bank Methods Lecture 7 Lecture notes to accompany Introduction to Spectral Analysis by P.

!+ ] N >1 1 2 is more general than 1. ( =2 2 =1 N > 1 leads to the variability / resolution compromise associated with all nonparametric methods. by assuming that ! is nearly constant over the bands f ( )g ( ) !. but 2 requires to ensure that the number of estimated values = = ) is < N . Nonparametric Approach: Implicitly smooth ! !=.Basic Ideas Two main PSD estimation approaches: 1. 1997 Slide L7–2 . Parametric Approach: Parameterize finite-dimensional model. Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. (!) by a 2. Stoica and R. Moses.

are conflicting. Prentice Hall. Z 6 ~ (! ) @ -! (b) Z !c ^FB (!) ' 1 2 (a) . as well as (b) and (c). c ( )d = (') (!) (a) consistent power calculation (b) Ideal passband filter with bandwidth (c) ( ) constant on 2 ! . !c @Q Q -! . Moses. 1997 . Stoica and R. Slide L7–3 Lecture notes to accompany Introduction to Spectral Analysis by P.with varying ! Power Calculation Bandpass Filter c and xed bandwidth - Power in the band - Division by lter bandwidth ^(!c ) - 6 y y (! ) . 2 ! + 2 ] Note that assumptions (a) and (b). jH ( )j2 ( )d = ' 21 !+ !.Filter Bank Approach to Spectral Estimation H 6 (!) 1 !c -! Filtered Signal y (t) .

Filter Bank Interpretation of the Periodogram 1 N y(t)e. 1 k=0 X 1 ei!k k = 0 : : : N .i!k = 1 eiN (~. Stoica and R.!) . 1 ~ hk = N otherwise 0 1 center frequency of H 3dB bandwidth of H (!) = ! ~ (!) ' 1=N Slide L7–4 Lecture notes to accompany Introduction to Spectral Analysis by P.i!t 2 ~ ^p(~) = ! 4 N t=1 2 N 1 ~ = N y(t)ei!(N .!) . Moses. k) 2 ! . 1 H (!) = hk e ! N ei(~. Prentice Hall. 1997 .t) t=1 X X = N where ( X 1 k=0 hk y(N .

and: narrow filter passband. for N = 50. Prentice Hall. ^( ) Lecture notes to accompany Introduction to Spectral Analysis by P. ! 0 −5 −10 −15 dB −20 −25 −30 −35 −40 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Conclusion: The periodogram p ! is a filter bank PSD estimator with bandpass filter as given above. 1997 Slide L7–5 . Moses. !). power calculation from only 1 sample of filter output.Filter Bank Interpretation of the Periodogram. Stoica and R. con't jH (!)j as a function of (~ .

Each filter covers a small band centered on ! . 2. Both approaches compromise bias for variance. Prentice Hall. 1997 Slide L7–6 . This approach leads to Bartlett and Welch methods. Use several bandpass filters on the whole sample. and in fact are quite related to each other: splitting the data sample can be interpreted as a special form of windowing or filtering. Split the available sample. Stoica and R. more data points for the power calculation stage.Possible Improvements to the Filter Bank Approach 1. ~ provides several samples for power calculation. and bandpass filter each subsample. This “multiwindow approach” is similar to the Daniell method. Moses. Lecture notes to accompany Introduction to Spectral Analysis by P.

m) h y(t) ~ E jyF (t)j2 = h Rh R = E fy(t)~ (t)g ~ y {z } 7 5 {z } n o y(t) 3 H (! ) = where a X m (! ) = 1 k=0 e. = h0 h1 : : : hm] . Prentice Hall. y(t .im! ]T hk e. yF (t) = X m k=0 | hk y(t . Stoica and R.Capon Method Idea: Data-dependent bandpass filter design.i!k = h a(!) Lecture notes to accompany Introduction to Spectral Analysis by P.i! : : : e. k) 2 6 4 | . 1997 Slide L7–7 . Moses.

Moses.1a(!) which should be the power of y (t) in a passband centered on ! .1 a (!)R a(!) with N 1 ^ y(t)~ (t) ~ y R = N . Stoica and R. con't Capon Filter Design Problem: min(h Rh) h Solution: subject to h a(!) = 1 h0 = R.Capon Method.1a o The power at the filter output is: E jyF (t)j2 = h0Rh0 = 1=a (!)R.m t=m+1 X Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. 1997 Slide L7–8 . n The Bandwidth 1 ' m+1 = (filter 1 length) Conclusion Estimate PSD as: +1 ^(!) = m^.1a=a R.

Prentice Hall. . but it splits the N -data point sample into (N m) subsequences of length m with maximum overlap. Moses. Stoica and R. bias decreases and variance increases. Capon uses one bandpass filter only. 1997 Slide L7–9 . Lecture notes to accompany Introduction to Spectral Analysis by P.Capon Properties m is the user parameter that controls the compromise between bias and variance: – as m increases.

i!k ^ k=. Moses.s) = m+1 r t=0 s=0 ^ a (!)Ra(!) R = ^(i . jkj ^BT (!) = r(k)e. s)e. Prentice Hall.i!(t.Relation between Capon and Blackman-Tukey Methods Consider ^BT (!) with Bartlett window: m m + 1 .m m + 1 1 m m ^(t . Stoica and R. 1997 Slide L7–10 .1a(!) Lecture notes to accompany Introduction to Spectral Analysis by P. j )] ^ r = m+1 X X X Then we have ^ a (!)Ra(!) ^BT (!) = m+1 m+1 ^C (!) = ^ a (!)R.

Prentice Hall. Moses. AR Then ^C (!) = 1 1 m 1= ^AR(!) m + 1 k=0 k X Consequences: Due to the average over k. 1997 Slide L7–11 . C ! generally has worse resolution and bias properties than the AR method. C ! generally has less statistical variability than the AR PSD estimator. Due to the low-order AR terms in the average. Stoica and R. ^( ) ^( ) Lecture notes to accompany Introduction to Spectral Analysis by P.Relation between Capon and AR Methods Let 2 ^k ^k (!) = ^ 2 jAk(!)j be the kth order AR PSD estimate of y (t).

Stoica and R. Prentice Hall. 1997 Slide L8–1 .Spatial Methods — Part 1 Lecture 8 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.

underwater surveillance Basic Approach: Determine energy distribution over space (thus the name “spatial spectral analysis”) Lecture notes to accompany Introduction to Spectral Analysis by P. seismology. communications. mechanical Receiving sensors: Hydrophones. antennas. electromagnetic. @. @. Emitted energy: Acoustic.@ . sonar. Moses. @.The Spatial Spectral Estimation Problem Source 1 vSource 2 B B B B N B cccc ccc c c c c c v @ . Sensor 1 @. ..@ .R . v Source n @.. .. . @. seismometers Applications: Radar. Stoica and R. . Prentice Hall. @. Sensor 2 Sensor m Problem: Detect and locate n radiating sources by using an array of m passive sensors. 1997 Slide L8–2 .

(We do not treat the detection problem) The sensors are linear dynamic elements with known transfer characteristics and known locations (That is. the array is calibrated.) Lecture notes to accompany Introduction to Spectral Analysis by P.Simplifying Assumptions Far-field sources in the same plane as the array of sensors Non-dispersive wave propagation Hence: The waves are planar and the only location parameter is direction of arrival (DOA) (or angle of arrival. AOA). Prentice Hall. 1997 Slide L8–3 . Stoica and R. Moses. The number of sources n is known.

1997 Slide L8–4 .Array Model — Single Emitter Case x(t) = k= the signal waveform as measured at a reference point (e. background noise.) hk (t) = ek (t) = Note: t 2 R (continuous-time signals). k) + ek (t) ( = convolution operator). Moses.. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. at the “first” sensor) the delay between the reference point and the kth sensor the impulse response (weighting function) of sensor k “noise” at the kth sensor (e.g. () f kg with hk(t) This is a time-delay estimation problem in the unknown input case. Basic Problem: Estimate the time delays known but x t unknown. Then the output of sensor k is yk (t) = hk (t) x(t . thermal noise in sensor electronics.g. etc.. Stoica and R.

the kth sensor output is yk (t) = jHk (!c)j (t) cos !ct + '(t) . k ) ' jHk (!c)j (t)cos !ct + '(t) . Moses. !c k ] hk (t) x(t .Narrowband Assumption Assume: The emitted signals are narrowband with known carrier frequency !c . Stoica and R. Then: x(t) = (t) cos !ct + '(t)] where (t) '(t) vary “slowly enough” so that (t . k ) ' '(t) Time delay is now ' to a phase shift !c k : x(t . !c k + arg Hk (!c)] + ek(t) Slide L8–5 Lecture notes to accompany Introduction to Spectral Analysis by P. !c k + argfHk (!c)g] where Hk function (!) = Ffhk(t)g is the kth sensor's transfer Hence. 1997 . Prentice Hall. k ) ' (t) cos !ct + '(t) . k) ' (t) '(t .

i! + ek (t) | s(t) {z } | Hk (!c) {z k c } c k yk (t) = s(t)Hk (!c) e. Prentice Hall. Stoica and R. by low-pass filtering and sampling the signal yk (t)=2 = yk(t)e.i !ct+ (t)] 2 Demodulate z (t) (translate to baseband): 2z(t)e. Moses.i 2!ct+ (t)] g n o | {z } | {z } Lowpass filter 2z(t)e.!ct = (t)f ei (t) + e.i!c k + ek (t) where s(t) is the complex envelope of x(t).i!ct to obtain (t)ei (t) lowpass highpass Hence.i!ct ~ = yk(t) cos(!ct) . iyk(t) sin(!ct) we get the complex representation: (for t 2 Z ) yk (t) = (t) ei'(t) jHk (!c)j ei arg H (! )] e. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L8–6 or .Complex Signal Representation The noise-free output has the form: z(t) = (t) cos !ct + (t)] = = (t) ei !ct+ (t)] + e.

Vector Representation for a Narrowband Source

Let

= m =
a( ) =
2 6 4

the emitter DOA the number of sensors
2 6 4

H1(!c) e;i!c
. .

1

3 7 5

Hm(!c) e;i!c m y1(t) e1(t) . . y(t) = e(t) = . . ym(t) em(t)
3 7 5 2 6 4

3 7 5

Then

y(t) = a( )s(t) + e(t)
NOTE: enters a via both For omnidirectional sensors the on .

()

f kg and fHk(!c)g. fHk(!c)g do not depend

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L8–7

Analog Processing Block Diagram

Analog processing for each receiving array element
SENSOR DEMODULATION

A/D CONVERSION

-

-

Internal Noise External Noise XXXXX ? XX ll z l - Cabling and , Filtering : ,, Incoming Transducer Signal x(t ; k )

Re yk (t)] Lowpass Re yk (t)] - !! ~ Filter A/D cos(!ct) 6 y k (t) Oscillator -sin(!ct) ? ~ - Im yk (t)] Lowpass Im yk (t)] - !! Filter A/D

-

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997

Slide L8–8

Multiple Emitter Case Given n emitters with received signals: DOAs:

fsk(t)gn=1 k

k

Linear sensors

) y(t) = a( 1)s1(t) +

+ a( n)sn(t) + e(t)

Let

A = a( 1) : : : a( n)] (m n) s(t) = s1(t) : : : sn(t)]T (n 1)
Then, the array equation is:

y(t) = As(t) + e(t)
Use the planar wave assumption to find the dependence of k on .
Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997 Slide L8–9

Prentice Hall.Uniform Linear Arrays Source h JJ J J J J JJ J ^ J J JJ JJ+ JJ JJ 2 ] J d sin J ^ JJ J JJ 3 J J v 4 1 v ]? v d - v ppp m vSensor ULA Geometry Sensor #1 = time delay reference Time Delay for sensor k: where c = wave propagation speed d sin k = (k . 1997 Slide L8–10 . Stoica and R. 1) c Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.

Stoica and R.1)!s ]T By direct analogy with the vector a(! ) made from uniform samples of a sinusoidal time series.i!s : : : e. Moses.i(m.Spatial Frequency Let: sin 4 !s = !c d sin = 2 dc=f = 2 d sin c c = c=fc = signal wavelength a( ) = 1 e. 1997 . !s = spatial frequency The function !s 7! a( ) is one-to-one for j!sj $ dj sin j 1 d =2 As =2 d =2 is a spatial Shannon sampling theorem. Slide L8–11 d = spatial sampling period Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.

Prentice Hall.Spatial Methods — Part 2 Lecture 9 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L9–1 . Stoica and R. Moses.

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses.Spatial Filtering Spatial filtering useful for DOA discrimination (similar to frequency discrimination of time-series filtering) Nonparametric DOA estimation There is a strong analogy between temporal filtering and spatial filtering. Prentice Hall. 1997 Slide L9–2 .

1] y(t) = u(t) : : : u(t . Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L9–3 .Analogy between Temporal and Spatial Filtering Temporal FIR Filter: yF (t) = hk u(t . k) = h y(t) k=0 h = ho : : : hm.i! : : : e.i(m.1)! ]T We can select h to enhance or attenuate signals with different frequencies ! . m + 1)]T If u m. Stoica and R. Moses. Prentice Hall.1 X (t) = ei!t then yF (t) = h a(!)] u(t) | {z } filter transfer function a(!) = 1 e.

i!c m ]T yF (t) = h a( )] s(t) | {z } The corresponding filter output is filter transfer function We can select h to enhance or attenuate signals coming from different DOAs.Analogy between Temporal and Spatial Filtering Spatial Filter: fyk(t)gm=1 = k the “spatial samples” obtained with a sensor array. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. Stoica and R. Prentice Hall.i!c 2 : : : e. Spatial FIR Filter output: yF (t) = X m k=1 hk yk (t) = h y(t) Narrowband Wavefront: The array's (noise-free) response to a narrowband ( sinusoidal) wavefront with complex envelope s t is: () y(t) = a( )s(t) a( ) = 1 e. 1997 Slide L9–4 .

B . ? qq 1 ! (Spatial sampling) (b) Spatial filter Lecture notes to accompany Introduction to Spectral Analysis by P. B B @ .i! Be 2 B !! B B B B ..? B -P C B C B B B . | e {z } ? a(!) 1 ? 0 qq ( 1) 1 (Temporal sampling) (a) Temporal filter narrowband source with DOA= Z z Z Z Z ~ aa 1reference !! h 0 0 1 B aa 2.@ 0 B 1 C h @@ C B B e.1 ? - 1 . C C B C B C B C B C B A @ . ! hm. Stoica and R..Analogy between Temporal and Spatial Filtering u(t) = ei!t z 1 s q. Prentice Hall. Cu(t) C B . B . Moses.i! C @ h a(!)]u(t) R @ C . s qq h - ? 1 ? - m.i m.1 q . 1997 Slide L9–5 .i! m | e {z a( ) aa m! point ( ) qq ( ) 1 C C C C C Cs(t) C C C C C A } - ? @ ? @ @ h - 1 @ R @ - P h a( )]s(t) - hm.

Spatial Filtering. 1997 Slide L9–6 . Here. Prentice Hall. h a 0 . where 0 j ( )j = ( ) = 25 0 30 −30 Th et a (d eg ) 60 −60 90 0 2 4 6 Magnitude 8 10 −90 Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. con't Example: The response magnitude h a of a spatial filter (or beamformer) for a 10-element ULA. Moses.

Prentice Hall. To locate an emitter in the field of view. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L9–7 . hence filtering out interferences located outside the filter's beam (but possibly having the same temporal characteristics as the signal). Stoica and R. Moses. by sweeping the filter through the DOA range of interest (“goniometer”).Spatial Filtering Uses Spatial Filters can be used To pass the signal of interest only.

( )Rh( ) give the Slide L9–8 Lecture notes to accompany Introduction to Spectral Analysis by P.Nonparametric Spatial Methods A Filter Bank Approach to DOA estimation. Moses. Prentice Hall. 1997 . n o n o The (dominant) peaks of h DOAs of the sources. Basic Ideas Design a filter h ( ) such that for each – It passes undistorted the signal with DOA = – It attenuates all DOAs 6= Sweep the filter through the DOA range of interest. and evaluate the powers of the filtered signals: E jyF (t)j2 = E jh ( )y(t)j2 = h ( )Rh( ) with R = E fy (t)y (t)g. Stoica and R.

y t can be considered as impinging on the array from all DOAs. Stoica and R.Beamforming Method Assume the array output is spatially white: Then: R = E fy(t)y (t)g = I E jyF (t)j2 = h h n o Hence: In direct analogy with the temporally white assumption for filter bank methods. 1997 Slide L9–9 . Prentice Hall. () Filter Design: min (h h) h Solution: subject to h a( ) = 1 h = a( )=a ( )a( ) = a( )=m E jyF (t)j2 = a ( )Ra( )=m2 n o Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.

Moses. pj > = wavelength array length array beamwidth Inconsistency problem: Beamforming DOA estimates are consistent if n inconsistent if n > . 1997 . ( )^ ( ) This is the direct spatial analog of the Blackman-Tukey periodogram. Stoica and R. Prentice Hall. Resolution Threshold: inf j k . but Slide L9–10 Lecture notes to accompany Introduction to Spectral Analysis by P.Implementation of Beamforming 1 N y(t)y (t) ^ R=N t=1 X The beamforming DOA estimates are: f^kg = the locations of the n largest peaks of a Ra . 1 = 1.

Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L9–11 . They do not make assumptions on the covariance properties of the data (and hence do not depend on them).1a( ) n o Implementation: f^kg = the locations of the n largest peaks of ^ 1=a ( )R.1a( ): Performance: Slightly superior to Beamforming. Stoica and R.1a( ) E jyF (t)j2 = 1=a ( )R.Capon Method Filter design: min(h Rh) subject to h a( ) = 1 h Solution: h = R. Moses.1a( )=a ( )R. Prentice Hall. Both Beamforming and Capon are nonparametric approaches.

Prentice Hall. MUSIC. Then: R = E fy(t)y (t)g = APA + 2I Thus: The NLS. almost without modification. MIN-NORM and ESPRIT methods of frequency estimation can be used. Stoica and R. 1997 Slide L9–12 .Parametric Methods Assumptions: The array is described by the equation: y(t) = As(t) + e(t) The noise is spatially white and has the same power in all sensors: E fe(t)e (t)g = 2I The signal covariance matrix P = E fs(t)s (t)g is nonsingular. Moses. Lecture notes to accompany Introduction to Spectral Analysis by P. YW. for DOA estimation.

this is precisely the form of the NLS method of frequency estimation. Stoica and R.1A ]y(t) = N t=1 ^ = trf I .Nonlinear Least Squares Method 1 N ky(t) . f ^k g = arg max trf A(A A). Moses.1A ]Rg ^ Thus. 1997 Slide L9–13 =1 . A(A A).1A ]Rg X f kg For N . A(A A). As(t)k2 min(t)g N f k g fs t=1 X | Minimizing f over s gives f ( s) {z } ^(t) = (A A). Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. A(A A).1A y(t) t = 1 : : : N s X Then 1 N k I .1A ]y(t)k2 f ( ^) = N s t=1 1 N y (t) I .

Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.Nonlinear Least Squares Method Properties of NLS: Performance: high Computational complexity: high Main drawback: need for multidimensional search. 1997 Slide L9–14 . Stoica and R. Moses.

Yule-Walker Method e y(t) = y(t) = A s(t) + ~(t) ~ y(t) ~ e(t) A " # " # " # Assume: Then: E fe(t)~ (t)g = 0 e 4 E fy(t)~ (t)g = AP A ~ . Moses. and the SVD of . Prentice Hall. 1997 Slide L9–15 .n ~ ~ Properties: A V2 = 0 V1 2 R(A) Then: rank |{z} |{z} Lecture notes to accompany Introduction to Spectral Analysis by P.n 0 V2 g L n M .) = n. is . = U1 U2 ] 0 n n 0 V1 g n . Stoica and R.= y (M L) Also assume: M > n L > n ( ) m = M + L > 2n) rank ~ (A) = rank(A) = n " # " # (.

1997 Slide L9–16 . is the “array transfer vector” for y(t) a ~ at DOA ^ V2 is defined similarly to V2. Moses. Prentice Hall.=N y t=1 X Properties: Computational complexity: medium Performance: satisfactory if m 2n Main advantages: – weak assumption on e t – the subarray A need not be calibrated f ( )g Lecture notes to accompany Introduction to Spectral Analysis by P.YW-MUSIC DOA Estimator f^kg = where the n largest peaks of 1=~ ( )V2V2 ~( ) a ^^a ~( ). using 1 N y(t)~ (t) ^ . (L 1). Stoica and R.

Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. 1997 Slide L9–17 . that is. Moses. Prentice Hall.MUSIC and Min-Norm Methods Both MUSIC and Min-Norm methods for frequency estimation apply with only minor modifications to the DOA estimation problem. Spectral forms of MUSIC and Min-Norm can be used for arbitrary arrays Root forms can be used only with ULAs MUSIC and Min-Norm break down if the source signals are coherent. if rank (P ) = rank(E fs(t)s (t)g) < n Modifications that apply in the coherent case exist.

i!c ( 1) 0 .. Prentice Hall.i!c ( n) ( ) = the time delay from subarray 1 to 2 6 4 3 7 5 subarray 2 for a signal with DOA = : where d is the subarray separation and is measured from the perpendicular to the subarray displacement vector. D= 0 e. Let m = # sensors in each subarray A1 = Im 0]A (transfer matrix of subarray 1) A2 = 0 Im]A (transfer matrix of subarray 2) Then A2 = A1D.ESPRIT Method Assumption: The array is made from two identical subarrays separated by a known displacement vector. where e. Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. 1997 Slide L9–18 ( ) = d sin( )=c ..

1 A2 = 0 Im. con't ESPRIT Scenario source θ known subarray 1 displac ement v ector subarray 2 Properties: Requires special array geometry Computationally efficient No risk of spurious DOA estimates Does not require array calibration Note: For a ULA.1 0]A = . Stoica and R. so m m and .1 .ESPRIT Method. Prentice Hall.1 A1 = Im.1]A Lecture notes to accompany Introduction to Spectral Analysis by P. the two subarrays are often the first m and last m array elements. 1997 Slide L9–19 . Moses.

Sign up to vote on this title
UsefulNot useful