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

1 y(t)e. Stoica and R.Deterministic Signals fy(t)g1 . Moses.1 = t= If: discrete-time deterministic data sequence X 1 t=. Prentice Hall.1 jy(t)j2 < 1 X Then: Y (! ) = 1 t=. 1997 Slide L1–4 .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)y (t . Moses.1 X (k)e. 1997 .1 X Z 1 where 4 S (!) = jY (!)j2 = Energy Spectral Density We can write S (!) = where X 1 k=. S (!)d! t=. Stoica and R. k) Slide L1–5 Lecture notes to accompany Introduction to Spectral Analysis by P.Energy Spectral Density Parseval's Equality: jy(t)j2 = 21 . Prentice Hall.i!k (k) = 1 t=.

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

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

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

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

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

Prentice Hall. Stoica and R. 1997 Slide L1–11 . 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 . ]g Find an estimate of Two Main Approaches : Nonparametric: – Derived from the PSD definitions. Moses.

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

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

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

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

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

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

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

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

1997 Slide L2–9 . Moses. Prentice Hall.Smearing and Leakage Main Lobe Width: smearing or smoothing Details in resolvable. Stoica and R. ^ 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. 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 .

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

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

Moses.i 2 .Discrete Fourier Transform (DFT) Finite DTFT: YN (!) = X N t=1 y(t)e. Stoica and R. 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 . Prentice Hall. 1997 Slide L2–12 . 1 Direct computation of O N 2 flops ( ) .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.

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

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

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

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

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

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

w(k) is a psd sequence) ^BT (!) 0 (which is desirable) Then: Time-Bandwidth Product M . 1997 Slide L3–4 . Prentice Hall.1) Ne = k=.1 X .(M (0) w Z w (k ) = equiv time width 1 2 . 0 Z | {z } If W (!) 0 (. Stoica and R. Moses.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. )d 2 .

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

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

k +J 1 ^D(!k ) = ^p (!j ) 2J + 1 j=k.Daniell Method By a previous result.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 . f ^p(!j )g are (nearly) uncorrelated random variables for 2 j N . Moses. 1997 Slide L3–10 . Prentice Hall.J X Lecture notes to accompany Introduction to Spectral Analysis by P. 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.Daniell Method. Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Stoica and R. for N 1. Moses. 1997 Slide L3–11 . Then.

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

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

1997 Slide L4–2 ( ) . ! can approximate arbitrarily well any continuous PSD. Note.i! . provided m and n are chosen sufficiently large.θ) φ(ω) possibly revise assumption on φ(ω) Rational Spectra jkj m k e. Stoica and R. P By Weierstrass theorem.i!k (!) is a rational function in e.i!k (!) = P jkj n k e. Prentice Hall.Basic Idea of Parametric Spectral Estimation Observed Data Assumed functional form of 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) Estimate parameters in φ(ω.θ) ^ θ Estimate PSD ^ ^ =φ(ω. Moses.

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

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

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

(Y Y ). 2) Lecture notes to accompany Introduction to Spectral Analysis by P. . y(N . . 3 7 5 Y =4 y(n) y(n . . ^ = . 1997 Slide L4–7 . 1) : : : y (t . 1) y(N . Prentice Hall. Find i=1 aiy(t .1(Y y) where 3 7 5 2 6 t=n+1 je(t)j2 y(1) y(2) 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 . n)]T . Moses. . 1) y(n + 1) y(n) . 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 ) . . . Stoica and R. n) .

Levinson–Durbin Algorithm Fast. order-recursive solution to YW equations 2 6 6 6 4 | 0 . . n . .. .n .1 1 ..0 . Moses. {z . 1997 Slide L4–8 . . Direct Solution: For one given value of n: For k Rn+1 1 0 = 0 . Prentice Hall.1 ... 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. . . 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).. Stoica and R.

con't Relevant Properties of R: Rx = y $ Rx = y. 1997 Slide L4–9 . . Stoica and R. Moses. 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. Prentice Hall. 2 4 Rn+1 n+1 r ~n n+1 r ~n 0 3 5 2 r ~n = 6 4 .Levinson-Durbin Alg. 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. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. 1997 . 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. n = n ) n+1 = 0 + kn 2+1 = n + kn n = n(1 .Levinson-Durbin Alg. Moses. 2 kn = .

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

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

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

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

Moses. K t=K +1 X Lecture notes to accompany Introduction to Spectral Analysis by P. f kgK=1 by using AR k 1b) Estimate the noise sequence ^(t) = y(t) + ^1y(t . 1) + + K y(t .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 . K ) 1a) Estimate the coefficients modelling techniques. 1997 Slide L4–15 . 1) + 2y(t . K ) ^2 = N 1 j^(t)j2 e N . for some large K e(t) ' y(t) + 1y(t . Prentice Hall. Stoica and R. 2) + = AR(1) with j kj ! 0 as k ! 1 Step 1: Approximate.

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

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

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

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

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

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

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

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

1997 Slide L5–6 . . 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.Nonlinear Least Squares (NLS) Method min' g f! Let: X N k k k t=1 | y(t) . Prentice Hall. . 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 . Moses. .

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

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

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

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

. 1997 Slide L5–13 . r(L + M . r(L + 2) . Prentice Hall. .High-Order and Overdetermined YW Equations ARMA covariance: r (k ) + X L i=1 bir(k . Moses. Stoica and R. . 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. 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. .

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

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

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

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

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

Moses..Eigendecomposition of R and Its Properties. 0 n 3 7 5 with C (since rank S ..1) = AC S = A(PA S ( ) = rank(A) = n). Therefore. Prentice Hall. 2 . 1997 . Stoica and R. = = nonsingular 0 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 . since S G j j 6= 0 4 . .. con't As rank (APA ) = n: k > 2 k = 1 ::: n k = 2 k = n + 1 ::: m 0 1.

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

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

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. 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. Moses. 1997 Slide L6–7 . Prentice Hall.Pisarenko Method Pisarenko is a special case of MUSIC with m (the minimum possible value).

and reduce risk of false frequency estimates. Moses. g # Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L6–8 . Uses m n (as in MUSIC). that has minimum Euclidean norm. Prentice Hall. 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).

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

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

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

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

Stoica and R. 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. Prentice Hall.

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

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

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

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

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

Prentice Hall. Both approaches compromise bias for variance. 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. Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P. more data points for the power calculation stage. 2. and bandpass filter each subsample. Each filter covers a small band centered on ! . Moses. This “multiwindow approach” is similar to the Daniell method. Use several bandpass filters on the whole sample.Possible Improvements to the Filter Bank Approach 1. This approach leads to Bartlett and Welch methods. Split the available sample. 1997 Slide L7–6 .

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

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

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

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

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

Stoica and R. Prentice Hall.Spatial Methods — Part 1 Lecture 8 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L8–1 . 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.R . Prentice Hall. 1997 Slide L8–2 . .The Spatial Spectral Estimation Problem Source 1 vSource 2 B B B B N B cccc ccc c c c c c v @ . mechanical Receiving sensors: Hydrophones. Emitted energy: Acoustic. @. seismology.@ . @. antennas. Sensor 1 @. electromagnetic. . @. Moses. @. sonar. seismometers Applications: Radar. communications. Sensor 2 Sensor m Problem: Detect and locate n radiating sources by using an array of m passive sensors.. Stoica and R. . .. @.@ . v Source n @.

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

Then the output of sensor k is yk (t) = hk (t) x(t . thermal noise in sensor electronics.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. Lecture notes to accompany Introduction to Spectral Analysis by P.. Basic Problem: Estimate the time delays known but x t unknown. 1997 Slide L8–4 .. background noise. () f kg with hk(t) This is a time-delay estimation problem in the unknown input case. 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. k) + ek (t) ( = convolution operator). etc.g. Prentice Hall. Stoica and R.g.

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

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

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

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

Slide L8–11 d = spatial sampling period Lecture notes to accompany Introduction to Spectral Analysis by P. 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. 1997 .i!s : : : e. !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. Moses.i(m.1)!s ]T By direct analogy with the vector a(! ) made from uniform samples of a sinusoidal time series. Stoica and R.

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

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

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

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

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

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

hence filtering out interferences located outside the filter's beam (but possibly having the same temporal characteristics as the signal). 1997 Slide L9–7 . Stoica and R. To locate an emitter in the field of view. Lecture notes to accompany Introduction to Spectral Analysis by P. 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. Prentice Hall. Moses.

Stoica and R. 1997 . Prentice Hall.Nonparametric Spatial Methods A Filter Bank Approach to DOA estimation. ( )Rh( ) give the Slide L9–8 Lecture notes to accompany Introduction to Spectral Analysis by P. 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. n o n o The (dominant) peaks of h DOAs of the sources. Moses. 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.

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

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

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

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

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

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

is . and the SVD of . Moses. Prentice Hall.) = n. 1997 Slide L9–15 . Stoica and R. = U1 U2 ] 0 n n 0 V1 g n .= y (M L) Also assume: M > n L > n ( ) m = M + L > 2n) rank ~ (A) = rank(A) = 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 ~ .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 .

(L 1). using 1 N y(t)~ (t) ^ .YW-MUSIC DOA Estimator f^kg = where the n largest peaks of 1=~ ( )V2V2 ~( ) a ^^a ~( ). Prentice Hall. 1997 Slide L9–16 . Stoica and R. 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. Moses.

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

ESPRIT Method Assumption: The array is made from two identical subarrays separated by a known displacement vector. Prentice Hall. Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P. D= 0 e. Moses..i!c ( 1) 0 . 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. 1997 Slide L9–18 ( ) = d sin( )=c .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. where e..

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

Sign up to vote on this title
UsefulNot useful