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

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

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

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. 1997 Slide L1–6 . Moses. Stoica and R. Prentice Hall.Random Signals Random Signal random signal probabilistic statements about future variations t current observation time Here: X 1 t=.

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

Second Definition of PSD 1 N y(t)e. Stoica and R. Prentice Hall. t=1 y(t)e.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.i!t Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. 1997 Slide L1–8 .

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

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

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

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

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.i!t N t=1 Natural estimator N 2 Used by Schuster ( 1900) to determine “hidden periodicities” (hence the name). Lecture notes to accompany Introduction to Spectral Analysis by P.Periodogram Recall 2nd definition of (! ): 1 N y(t)e. Stoica and R. Prentice Hall. 1997 Slide L2–2 . Moses.

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

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

1 X = ^c(!) k=.Relationship Between ^p (! ) and ^c(! ) If: the biased ACS estimator r Then: ^(k) is used in ^c(!).(N . 1997 .i!t N t=1 X 2 = N . N ^p(!) = 1 y(t)e.i!k ^p(!) = ^c(!) Consequence: Both p ! and ^( ) ^c(!) can be analyzed simultaneously.1) r ^(k)e. Moses. Slide L2–5 Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. 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. ^p(!) and ^c(!) have poor performance. ( )] Lecture notes to accompany Introduction to Spectral Analysis by P. Thus. jk =0 there are “so many” that when summed in ! .1 are small. Moses. even for large N . Stoica and R. ( )g ^ ( ) . Prentice Hall. r(k) may be large for large jkj ^ Even if the errors r k r k Nj. Intuitive explanation: r(k) . the PSD error is large. p! f^( ) . 1997 Slide L2–6 .

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

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 . Moses. Prentice Hall. Stoica and R. (!) may differ quite a bit from (!).Bartlett Window WB (! ) 1 sin(!N=2) 2 WB (!) = N sin(!=2) " # 0 WB (!)=WB (0). Slide L2–8 Lecture notes to accompany Introduction to Spectral Analysis by P. WB 1=N . 1997 .

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 . ^ 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. 1997 Slide L2–9 . Prentice Hall.

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

(!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. 1997 Slide L2–11 . Stoica and R. (!1) ^p(!2) . Moses. dev = φ(ω) too - Resolvability properties depend on both bias and variance. Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.Periodogram Variance As N !1 E ^p(!1) .

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

1 for odd k Assume: X X X Let N For k ~ ~ = N=2 and W = W 2 = e.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 = . 2 = 2p: Y (2p) = For k X ~ N = 1 3 5 : : : N . y(t + N )]W tgW tp Lecture notes to accompany Introduction to Spectral Analysis by P. 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. Moses. Prentice Hall. 1997 . Stoica and R. ~ 4 = 0 2 4 : : : N .i2 =N . Slide L2–13 t=1 ~ ~ f y(t) .

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

1 N k=0 sampled. Moses.1 ! N k=0 ^ φ(ω) continuous curve Lecture notes to accompany Introduction to Spectral Analysis by P.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 . Prentice Hall. Stoica and R. 1997 Slide L2–15 . N=8 ω 2 k N .

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

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

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.Blackman-Tukey Method. Stoica and R. Prentice Hall. 1997 . Moses. con't ^BT (!) = 1 ^p( )W (! . )d 2 .

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

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

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.Window Examples Triangular Window. Prentice Hall. 1997 Slide L3–6 . M 0 = 25 −10 −20 dB −30 −40 −50 −60 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Rectangular Window. Moses. Stoica and R.

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

f ^p(!j )g are (nearly) uncorrelated random variables for 2 j N .Daniell Method By a previous result. Stoica and R. Moses. 1997 Slide L3–10 .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 . k +J 1 ^D(!k ) = ^p (!j ) 2J + 1 j=k.J X Lecture notes to accompany Introduction to Spectral Analysis by P. for N 1. Prentice Hall.

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

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

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

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

. a rational factored as (!) can be B (!) 2 2 (!) = A(!) A(z) = 1 + a1z.n B (z) = 1 + b1z.AR. 1997 .m and.1 + + anz. Prentice Hall. Moses.1 + + bmz. MA. Stoica and R.g. and ARMA Models By Spectral Factorization theorem. 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. e.

Moses. Prentice Hall. i) = X m j =0 bj e(t . k) and take E f g to give: X r (k ) + X 2 m bj hj . 1997 Slide L4–4 .k = j =0 = 0 for k > m B (q) = 1 h q. 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 .ARMA Covariance Structure ARMA signal model: y(t)+ X n i=1 n aiy(t . k)g Lecture notes to accompany Introduction to Spectral Analysis by P. j ) (b0 = 1) Multiply by y (t . j )y (t . Stoica and 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(. Moses. 2 1 a1 = 0 .1) : : : r(. r (. .1) . . . Stoica and R.n) . . Prentice Hall.AR Signals: Yule-Walker Equations AR: m = 0. . 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. 1997 Slide L4–5 . . r(1) r(0) . Lecture notes to accompany Introduction to Spectral Analysis by P. .

. . Prentice Hall.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) ^(. . Stoica and R. r ^(1) r(0) ^ ^.n) 1 r r ^2 . 1997 Slide L4–6 .1) : : : ^(. 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) . ^(. . .1 = 0 a . Moses. r .

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

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).. Direct Solution: For one given value of n: For k Rn+1 1 0 = 0 . order-recursive solution to YW equations 2 6 6 6 4 | 0 . n .n . . {z . .. 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.0 .1 1 . Prentice Hall. .. Moses.. Stoica and R. .1 .. .Levinson–Durbin Algorithm Fast.

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 Relevant Properties of R: Rx = y $ Rx = y.Levinson-Durbin Alg. Stoica and R. Moses. 1997 Slide L4–9 . 2 4 Rn+1 n+1 r ~n n+1 r ~n 0 3 5 2 r ~n = 6 4 . Prentice Hall. where x = xn : : : x1]T ~ ~ ~ Nested structure 2 Rn+2 = 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.Levinson-Durbin Alg. Lecture notes to accompany Introduction to Spectral Analysis by P. 2 kn = . 1997 . Prentice Hall. 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. Moses. n = n ) n+1 = 0 + kn 2+1 = n + kn n = n(1 .

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

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

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

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

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

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

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

Prentice Hall. 1997 Slide L4–18 .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. Stoica and R. Moses.

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

Examples: communications radar. 1997 . Stoica and R. !1 !2 !3 - An “Ideal” line spectrum Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. Prentice Hall.Line Spectra Many applications have signals with (near) sinusoidal components. 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 .

uniformly distributed on . ] (prevents model ambiguities) f'k g = independent rv's. Stoica and R. superimposed in white noise of k 2. 1997 . 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.Line Spectral Signal Model Signal Model: Sinusoidal components of frequencies !k and powers 2 . Prentice Hall. 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 .

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

Stoica and R. Prentice Hall. 1997 Slide L5–5 .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 . Moses. ^ ei!k t + residuals k Lecture notes to accompany Introduction to Spectral Analysis by P.

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

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

! Difficult Implementation: The NLS cost function F is multimodal. Prentice Hall. 1997 Slide L5–8 . Lecture notes to accompany Introduction to Spectral Analysis by P. it is difficult to avoid convergence to local minima.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. Moses. Stoica and R.5.

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

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

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 = . . 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. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. r(L + 2) . r(L + M . 1997 Slide L5–13 .High-Order and Overdetermined YW Equations ARMA covariance: r (k ) + X L i=1 bir(k . . . .

n) spurious zeros of B (q ) are located strictly inside the unit circle. The Minimum-Norm solution is b = . Stoica and R. if the Minimum-Norm solution b is used. y = . Prentice Hall.1U Important property: The additional (L . diagonal and nonsingular Thus.V . Moses.High-Order and Overdetermined YW Equations. 1997 Slide L5–14 . (U V )b = . Lecture notes to accompany Introduction to Spectral Analysis by P. 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).

V ^ . V from the Let U . this is the price paid for increased accuracy when L > n. Stoica and R. Prentice Hall.HOYW Equations. f^ g When the SNR is low. . SVD of . Moses. Compute ^ = . 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. Lecture notes to accompany Introduction to Spectral Analysis by P. this approach may give spurious frequency estimates when L > n. Practical Solution Let ^= ^ ^ ^ ^ but made from f^(k)g instead of fr(k)g.

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

Stoica and R. 1997 Slide L6–2 . 0 2 n 3 7 5 Lecture notes to accompany Introduction to Spectral Analysis by P.i(m.. Moses. m + 1) . 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)! ]T A = a(!1) : : : a(!n)] (m n) rank Note: Define (A) = n y(t) y(t . .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 .i! : : : e. 1) (for m 3 7 7 7 5 n) 2 46 y(t) = 6 ~ 6 4 y(t .. Prentice Hall.

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

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

A G= G=0 a (!n) ) fa(!k )gn=1 ? R(G) k 2 6 4 3 7 5 Thus. 1997 Slide L6–5 .MUSIC Method a (!1) . 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. f!kgn=1 are the unique solutions of k a (!)GG a(!) = 0. Moses. Prentice Hall. . Stoica and R.

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

Moses. Stoica and R. ) f!kgn=1 can be found from the roots of ^ k aT (z. If: =n+1 m=n+1 Then: ^ g G = ^1.1)^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. Prentice Hall. 1997 Slide L6–7 .Pisarenko Method Pisarenko is a special case of MUSIC with m (the minimum possible value).

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

Stoica and R. 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. Moses.Min-Norm Method.1) 1 ^ g " # that are closest to the unit circle. 1997 Slide L6–9 . Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.

g " # Min-Norm solution: As: ^ = .1 exists iff = k k2 6= 1 (This holds. Moses. Stoica and R.Let S ^= S g1 .S(S S ). for N 1.1 gm " # " # Min-Norm Method: Determining g ^ Then: 1 2 R(G) ) S 1 = 0 ^ ^ ^ ^ g g ) S ^= . k k2) = (S S ) ) (S S ). 1997 Slide L6–10 .S =(1 . Prentice Hall.1 = =(1 . at least. k k2) ) ^ = .1 g + S S . (S S ). k k2) g Lecture notes to accompany Introduction to Spectral Analysis by P.) ^^ I=S S= Multiplying the above equation by gives: (1 .

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

Stoica and R. Moses. 1997 Slide L6–12 . find S . 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. Prentice Hall.1S1S2 ^^ ^^ f!kgn=1 = . ^ ^ ^ ^ The frequency estimates are found by: where f^kgn=1 are the eigenvalues of k ^ = (S1S1).ESPRIT Implementation From the eigendecomposition of R. then S1 and S2 .

Moses. Stoica and R. 1997 . Prentice Hall.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.

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

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

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

1 k=0 X 1 ei!k k = 0 : : : N .i!k = 1 eiN (~.!) . Moses. Prentice Hall.!) .i!t 2 ~ ^p(~) = ! 4 N t=1 2 N 1 ~ = N y(t)ei!(N . k) 2 ! . 1 H (!) = hk e ! N ei(~. 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. Stoica and R.t) t=1 X X = N where ( X 1 k=0 hk y(N . 1997 .Filter Bank Interpretation of the Periodogram 1 N y(t)e.

! 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. !). Prentice Hall. 1997 Slide L7–5 . for N = 50. and: narrow filter passband.Filter Bank Interpretation of the Periodogram. con't jH (!)j as a function of (~ . Stoica and R. Moses. ^( ) Lecture notes to accompany Introduction to Spectral Analysis by P. power calculation from only 1 sample of filter output.

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

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.i! : : : e. 1997 Slide L7–7 .im! ]T hk e. Stoica and R. Moses. k) 2 6 4 | . Prentice Hall. yF (t) = X m k=0 | hk y(t . y(t .i!k = h a(!) Lecture notes to accompany Introduction to Spectral Analysis by P.Capon Method Idea: Data-dependent bandpass filter design. = h0 h1 : : : hm] .

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

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

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

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

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

. @. Sensor 2 Sensor m Problem: Detect and locate n radiating sources by using an array of m passive sensors.. @. sonar. . mechanical Receiving sensors: Hydrophones. Prentice Hall. . seismometers Applications: Radar.@ . Moses. . seismology. antennas.R . @.The Spatial Spectral Estimation Problem Source 1 vSource 2 B B B B N B cccc ccc c c c c c v @ . .. Stoica and R. 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. 1997 Slide L8–2 . communications. @. v Source n @.@ . @. Sensor 1 @. Emitted energy: Acoustic. electromagnetic.

Stoica and R. the array is calibrated. 1997 Slide L8–3 . AOA). The number of sources n is known.) 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. (We do not treat the detection problem) The sensors are linear dynamic elements with known transfer characteristics and known locations (That is. Moses. Prentice Hall.

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

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. Then: x(t) = (t) cos !ct + '(t)] where (t) '(t) vary “slowly enough” so that (t . k ) ' (t) cos !ct + '(t) . k) ' (t) '(t . !c k ] hk (t) x(t . k ) ' jHk (!c)j (t)cos !ct + '(t) . Moses. 1997 . Prentice Hall. !c k + argfHk (!c)g] where Hk function (!) = Ffhk(t)g is the kth sensor's transfer Hence.Narrowband Assumption Assume: The emitted signals are narrowband with known carrier frequency !c . the kth sensor output is yk (t) = jHk (!c)j (t) cos !ct + '(t) . Stoica and R.

i!ct ~ = yk(t) cos(!ct) .i !ct+ (t)] 2 Demodulate z (t) (translate to baseband): 2z(t)e.i 2!ct+ (t)] g n o | {z } | {z } Lowpass filter 2z(t)e. 1997 Slide L8–6 or .i!c k + ek (t) where s(t) is the complex envelope of x(t).!ct = (t)f ei (t) + e. 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. Stoica and R. by low-pass filtering and sampling the signal yk (t)=2 = yk(t)e. Prentice Hall.Complex Signal Representation The noise-free output has the form: z(t) = (t) cos !ct + (t)] = = (t) ei !ct+ (t)] + e. Lecture notes to accompany Introduction to Spectral Analysis by P.i!ct to obtain (t)ei (t) lowpass highpass Hence.i! + ek (t) | s(t) {z } | Hk (!c) {z k c } c k yk (t) = s(t)Hk (!c) e. Moses.

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

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 . Moses. 1997 Slide L8–10 . Stoica and R. Prentice Hall. 1) c Lecture notes to accompany Introduction to Spectral Analysis by P.

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

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

Stoica and R. 1997 Slide L9–2 .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. Moses. Lecture notes to accompany Introduction to Spectral Analysis by P.

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

Analogy between Temporal and Spatial Filtering Spatial Filter: fyk(t)gm=1 = k the “spatial samples” obtained with a sensor array.i!c 2 : : : e. 1997 Slide L9–4 . 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. Stoica and R.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. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.

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

Prentice Hall. h a 0 . Here. Moses.Spatial Filtering. Stoica and R. 1997 Slide L9–6 . 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. con't Example: The response magnitude h a of a spatial filter (or beamformer) for a 10-element ULA.

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

1997 . 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. ( )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. n o n o The (dominant) peaks of h DOAs of the sources. Stoica and R. Prentice Hall. Moses. 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.

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. y t can be considered as impinging on the array from all DOAs. () 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. 1997 Slide L9–9 . Moses. Stoica and R. Prentice Hall.

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

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

Then: R = E fy(t)y (t)g = APA + 2I Thus: The NLS. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L9–12 . MUSIC. Stoica and R. MIN-NORM and ESPRIT methods of frequency estimation can be used. for DOA estimation. Moses.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. almost without modification. Prentice Hall. YW.

Stoica and R.1A y(t) t = 1 : : : N s X Then 1 N k I . 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.1A ]Rg X f kg For N . f ^k g = arg max trf A(A A). this is precisely the form of the NLS method of frequency estimation. Moses.1A ]y(t)k2 f ( ^) = N s t=1 1 N y (t) I .Nonlinear Least Squares Method 1 N ky(t) . Prentice Hall. A(A A). 1997 Slide L9–13 =1 .1A ]y(t) = N t=1 ^ = trf I . A(A A).1A ]Rg ^ Thus. A(A A).

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

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

Prentice Hall.YW-MUSIC DOA Estimator f^kg = where the n largest peaks of 1=~ ( )V2V2 ~( ) a ^^a ~( ). Moses. Stoica and R. 1997 Slide L9–16 . using 1 N y(t)~ (t) ^ . is the “array transfer vector” for y(t) a ~ at DOA ^ V2 is defined similarly to V2.=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. (L 1).

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

D= 0 e. Moses.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. Prentice Hall. where e.. Stoica and R. 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. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L9–18 ( ) = d sin( )=c .i!c ( 1) 0 .

1 0]A = . Moses. the two subarrays are often the first m and last m array elements. 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.ESPRIT Method. Prentice Hall.1 . Stoica and R. 1997 Slide L9–19 .1 A1 = Im. so m m and .1]A Lecture notes to accompany Introduction to Spectral Analysis by P.1 A2 = 0 Im.

Sign up to vote on this title
UsefulNot useful