This action might not be possible to undo. Are you sure you want to continue?
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 = t= If: discretetime deterministic data sequence X 1 t=.i!t exists and is called the DiscreteTime Fourier Transform (DTFT) Lecture notes to accompany Introduction to Spectral Analysis by P.1 y(t)e. 1997 Slide L1–4 .1 jy(t)j2 < 1 X Then: Y (! ) = 1 t=. Stoica and R. Moses.Deterministic Signals fy(t)g1 . Prentice Hall.
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=. 1997 . k) Slide L1–5 Lecture notes to accompany Introduction to Spectral Analysis by P.i!k (k) = 1 t=. Stoica and R.1 X (k)e.Energy Spectral Density Parseval's Equality: jy(t)j2 = 21 . Moses. Prentice Hall. S (!)d! t=.
Stoica and R. 1997 Slide L1–6 .1 n jy(t)j2 = 1 o But: E jy(t)j2 < 1 E f g = Expectation over the ensemble of realizations E jy(t)j2 n o = Average power in y (t) PSD = (Average) power spectral density Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Moses.Random Signals Random Signal random signal probabilistic statements about future variations t current observation time Here: X 1 t=.
1997 Slide L1–7 . Moses. 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. Stoica and R.i!k (k) is the autocovariance sequence (ACS) r(k) = E fy(t)y (t .k) r(0) jr(k)j Z Note that r(k) = 21 . (!)ei!k d! n o (Inverse DTFT) Interpretation: so r(0) = E jy(t)j2 = 21 .First Deﬁnition of PSD (! ) = where r X 1 k=.1 r(k)e. Prentice Hall.
Moses.i!t Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L1–8 . Stoica and R. t=1 y(t)e. Prentice Hall.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.Second Deﬁnition of PSD 1 N y(t)e.
Stoica and R.! ) Otherwise: (! ) 6= (. we can restrict attention to !2 .Properties of the PSD P1: (!) = (! + 2 ) for all !. Then: (! ) = (. P2: ] () f 2 . 1997 Slide L1–9 . Moses.! ) P3: If y Lecture notes to accompany Introduction to Spectral Analysis by P. Thus. Prentice Hall.1=2 1=2] (!) 0 (t) is real.
1y(t) = y(t .k = unit delay operator: q.Transfer of PSD Through Linear Systems System Function: where q . Moses.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) hk e. Prentice Hall. 1997 Slide L1–10 . Stoica and R. 1) H (q ) y (t) 2 y (! ) = jH (! )j e( e(t) e (! ) Then !) y(t) = X 1 k=0 hk e(t .
]g Find an estimate of Two Main Approaches : Nonparametric: – Derived from the PSD deﬁnitions. 1997 Slide L1–11 . Stoica and R. Prentice Hall.The Spectral Estimation Problem The Problem: From a sample fy(1) : : : y(N )g (!): f ^(!) ! 2 . Parametric: – Assumes a parameterized functional form of the PSD Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.
Stoica and R. 1997 Slide L2–1 . Moses. Prentice Hall.Periodogram and Correlogram Methods Lecture 2 Lecture notes to accompany Introduction to Spectral Analysis by P.
1997 Slide L2–2 .Periodogram Recall 2nd deﬁnition of (! ): 1 N y(t)e. Moses. Prentice Hall.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). Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P.
Correlogram Recall 1st deﬁnition of (! ): (! ) = P X 1 Truncate the “ ” and replace “r k=.1) Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.1 N . Moses.1 X r(k)e. Stoica and R.i!k ^c(!) = k=. 1997 Slide L2–3 .(N .i!k (k)” by “^(k)”: r r ^(k)e.
k) k 0 r(k) = N ^ t=k+1 X For both estimators: r ^(k) = ^ (. k) k 0 ^ r(k) = N . 1997 Slide L2–4 .Covariance Estimators (or Sample Covariances) Standard unbiased estimate: 1 N y(t)y (t .k) k < 0 r Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. k t=k+1 X Standard biased estimate: 1 N y(t)y (t . Prentice Hall. Stoica and R.
(N . Prentice Hall.i!k ^p(!) = ^c(!) Consequence: Both p ! and ^( ) ^c(!) can be analyzed simultaneously. Stoica and R. 1997 .1 X = ^c(!) k=. N ^p(!) = 1 y(t)e. Slide L2–5 Lecture notes to accompany Introduction to Spectral Analysis by P.Relationship Between ^p (! ) and ^c(! ) If: the biased ACS estimator r Then: ^(k) is used in ^c(!).i!t N t=1 X 2 = N . Moses.1) r ^(k)e.
r(k) may be large for large jkj ^ Even if the errors r k r k Nj. ( )g ^ ( ) . ^p(!) and ^c(!) have poor performance. Thus. 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. Intuitive explanation: r(k) . even for large N . ( )] Lecture notes to accompany Introduction to Spectral Analysis by P. p! f^( ) . the PSD error is large. Prentice Hall.1 are small. 1997 Slide L2–6 . Stoica and R. Moses.
i!k = 1.1 X wB (k) = 8 < : = Thus.(N . Prentice Hall.1 n o n o X ! X N . Stoica and R. ) d Z Ideally: WB (!) = Dirac impulse (!). n o 0 k 1 . Moses. 1997 .1) N .1) 1 = wB (k)r(k)e. jNj jkj N . window E ^p(!) = 21 . N k=.Bias Analysis of the Periodogram E ^p(!) = E ^c(!) = E f^(k)g e. ( )WB (! .i!k r k=.i!k k=. Slide L2–7 Lecture notes to accompany Introduction to Spectral Analysis by P.(N . or triangular.1 jkj r(k)e. 1 jkj N Bartlett.
Moses. Prentice Hall. Stoica and R. (!) may differ quite a bit from (!). 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 .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. 1997 . WB 1=N .
Prentice Hall. Stoica and R. Moses. 1997 Slide L2–9 .Smearing and Leakage Main Lobe Width: smearing or smoothing Details in resolvable. ^ 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 .
1997 Slide L2–10 . Prentice Hall. Stoica and R. !1 ^( ) ( )! ( ) Lecture notes to accompany Introduction to Spectral Analysis by P. WB ! !. Moses. severe bias As N .Periodogram Bias Properties Summary of Periodogram Bias Properties: For “small” N . so ! is asymptotically unbiased.
(!1) ^p(!2) . dev = φ(ω) too  Resolvability properties depend on both bias and variance. Lecture notes to accompany Introduction to Spectral Analysis by P.Periodogram Variance As N !1 E ^p(!1) . Stoica and R. 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. 1997 Slide L2–11 . Moses.
1 Direct computation of O N 2 ﬂops ( ) . Stoica and R. 1997 Slide L2–12 .1 fY (k)gN=0 from fy(t)gN : t=1 k Lecture notes to accompany Introduction to Spectral Analysis by P. N Let ! = N Then YN ( 2 k) is the Discrete Fourier Transform (DFT): N X Y (k) = N t=1 y(t)W tk k = 0 ::: N . Moses.i 2 .Discrete Fourier Transform (DFT) Finite DTFT: YN (!) = X N t=1 y(t)e.i!t 2 k and W = e. Prentice Hall.
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.i2 =N . ~ 4 = 0 2 4 : : : N . 1 = 2p + 1: X t=1 ~ ~ y(t) + y(t + N )]W tp Y (2p + 1) = Each is a N ~ N ~ = N=2point DFT computation. Moses. Stoica and R.1 for odd k Assume: X X X Let N For k ~ ~ = N=2 and W = W 2 = e. y(t + N )]W tgW tp Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 . 2 = 2p: Y (2p) = For k X ~ N = 1 3 5 : : : N . Slide L2–13 t=1 ~ ~ f y(t) .
2k + 2c ck = 2 k. Prentice Hall. 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. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L2–14 .
N=8 ω 2 k N .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 radix2 FFT (so N Finer sampling of = power of 2) ^(!): 2 k N . Moses. 1997 Slide L2–15 .1 ! N k=0 ^ φ(ω) continuous curve Lecture notes to accompany Introduction to Spectral Analysis by P.1 N k=0 sampled. Prentice Hall. Stoica and R.
Moses. 1997 Slide L3–1 .Improved PeriodogramBased Methods Lecture 3 Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Prentice Hall.
^( ) jj ^BT (!) = M .(M .BlackmanTukey Method Basic Idea: Weighted correlogram. 1997 Slide L3–2 . Moses.1 X k=. Stoica and R. Prentice Hall.i!k r fw(k)g = Lag Window w(k) 1 M M k Lecture notes to accompany Introduction to Spectral Analysis by P. with small weight applied to covariances r k with “large” k .1) w(k)^(k)e.
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. Stoica and R. Moses.BlackmanTukey Method. )d 2 . con't ^BT (!) = 1 ^p( )W (! . 1997 . Prentice Hall.
Moses. )d 2 . 1997 Slide L3–4 . Stoica and R. w(k) is a psd sequence) ^BT (!) 0 (which is desirable) Then: TimeBandwidth Product M .1 X .(M (0) w Z w (k ) = equiv time width 1 2 . 0 Z  {z } If W (!) 0 (.1) Ne = k=. W (!)d! = equiv bandwidth e= W (0) Ne e = 1 Lecture notes to accompany Introduction to Spectral Analysis by P.Window Design Considerations Nonnegativeness: ^BT (!) = 1 ^p( ) W (! . Prentice Hall.
As M increases. once M is given. e) is essentially ) Choose window shape to compromise between smearing (main lobe width) and leakage (sidelobe level). The energy in the main lobe and in the sidelobes cannot be reduced simultaneously. con't e = 1=Ne = 0(1=M ) is the BT resolution threshold. bias decreases and variance increases. Stoica and R.Window Design. Moses. Lecture notes to accompany Introduction to Spectral Analysis by P. ) Choose M as a tradeoff between variance and bias. Ne (and hence ﬁxed. 1997 Slide L3–5 . Prentice Hall. Once M is given.
M 0 −10 = 25 −20 dB −30 −40 −50 −60 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.Window Examples Triangular Window. Stoica and R. Moses. M 0 = 25 −10 −20 dB −30 −40 −50 −60 −3 −2 −1 0 1 ANGULAR FREQUENCY 2 3 Rectangular Window. 1997 Slide L3–6 .
Bartlett Method Basic Idea:
1 Μ 2Μ ... Ν
^1(ω) ^2(ω) φ φ
... average
^L(ω) φ
^B(ω) φ
Mathematically:
yj (t) = y((j ; 1)M + t) t = 1 : : : M = the j th subsequence 4 (j = 1 : : : L = N=M ])
^j (!) = 1 yj (t)e;i!t M t=1
X
M
2
1 L ^ (!) ^B (!) = L j =1 j
X
Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Moses, Prentice Hall, 1997
Slide L3–7
Comparison of Bartlett and BlackmanTukey 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 mainlobesidelobe 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 nonrectangular 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 . k +J 1 ^D(!k ) = ^p (!j ) 2J + 1 j=k. Prentice Hall.J X Lecture notes to accompany Introduction to Spectral Analysis by P.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 . for N 1.Daniell Method By a previous result. Stoica and R. Moses. 1997 Slide L3–10 .
Hence: ^D (! ) ' ^BT (! ) with a rectangular spectral Let Z window. ^D(!) ' 1 ^p(!)d! 2 .Daniell Method. Then. Stoica and R. con't As J increases: Bias increases (more smoothing) Variance decreases (more averaging) = 2J=N . for N 1. Prentice Hall. Moses. Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L3–11 .
^ () – Implemented by averaging subsample periodograms. BT periodogram – Local smoothing/averaging of spectral window. Daniell Periodogram – Approximate interpretation: spectral window. Lecture notes to accompany Introduction to Spectral Analysis by P.Summary of Periodogram Methods Unwindowed periodogram – reasonable bias – unacceptable variance Modiﬁed periodograms – Attempt to reduce the variance at the expense of (slightly) increasing the bias. ^BT (!) with a rectangular Slide L3–12 – Implemented by local averaging of periodogram values. Moses. Welch periodograms – Approximate interpretation: BT ! with a suitable lag window (rectangular for Bartlett. more general for Welch). Prentice Hall. ^p(!) by a suitably selected ^(k) using a lag – Implemented by truncating and weighting r window in c ! ^( ) Bartlett. 1997 . Stoica and R.
Prentice Hall. Moses.Parametric Methods for Rational Spectra Lecture 4 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L4–1 . Stoica and R.
! can approximate arbitrarily well any continuous PSD. 1997 Slide L4–2 ( ) .i!k (!) is a rational function in e. provided m and n are chosen sufﬁciently large. Prentice Hall.θ) ^ θ Estimate PSD ^ ^ =φ(ω. however: choice of m and n is not simple some PSDs are not continuous Lecture notes to accompany Introduction to Spectral Analysis by P.θ) Estimate parameters in φ(ω.i! .θ) φ(ω) possibly revise assumption on φ(ω) Rational Spectra jkj m k e. Note. Moses. P By Weierstrass theorem.Basic Idea of Parametric Spectral Estimation Observed Data Assumed functional form of φ(ω.i!k (!) = P jkj n k e. Stoica and R.
Stoica and R. Moses.n B (z) = 1 + b1z. 1997 . and ARMA Models By Spectral Factorization theorem. a rational factored as (!) can be B (!) 2 2 (!) = A(!) A(z) = 1 + a1z.AR. e..m and. MA. Prentice Hall. 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.1 + + anz.1 + + bmz.g.
k) and take E f g to give: X r (k ) + X 2 m bj hj .k = j =0 = 0 for k > m B (q) = 1 h q. j )y (t . i) = m j =0 bj E fe(t . j ) (b0 = 1) Multiply by y (t . i) = X m j =0 bj e(t . Stoica and R. Prentice Hall. k)g Lecture notes to accompany Introduction to Spectral Analysis by P.k (h = 1) where H (q ) = A(q ) 0 k k=0 X X i=1 air(k . Moses.ARMA Covariance Structure ARMA signal model: y(t)+ X n i=1 n aiy(t . 1997 Slide L4–4 .
. . Moses. . Lecture notes to accompany Introduction to Spectral Analysis by P.AR Signals: YuleWalker 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. . r(1) r(0) . 1997 Slide L4–5 . Stoica and R. 2 1 a1 = 0 . 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(.n) . .1) : : : r(. Prentice Hall. .1) . .
.1) .1) : : : ^(.1 = 0 a . . r .n) 1 r r ^2 . r ^(1) r(0) ^ ^. 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. . Stoica and R. .AR Spectral Estimation: YW Method YuleWalker Method: Replace r 2 6 6 6 4 (k) by r(k) and solve for f^ig and ^2: ^ a r ^(0) ^(. . . Prentice Hall. Moses. ^(. 1997 Slide L4–6 .
. 1) : : : y (t . 2) Lecture notes to accompany Introduction to Spectral Analysis by P. n)]T . ^ = . . 1) y(n + 1) y(n) . Find i=1 aiy(t . Prentice Hall. . . Stoica and R. n) .AR Spectral Estimation: LS Method Least Squares Method: e(t) = y(t) + X n 4 = y(t) + y(t) ^ where '(t) = y (t . 3 7 5 Y =4 y(n) y(n . y(N .1(Y y) where 3 7 5 2 6 t=n+1 je(t)j2 y(1) y(2) y(N . 1997 Slide L4–7 . 1) y(N . Moses. 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 ) . .(Y Y ). .
.Levinson–Durbin Algorithm Fast. Stoica and R. .. .. 1997 Slide L4–8 .. . Direct Solution: For one given value of n: For k Rn+1 1 0 = 0 . orderrecursive solution to YW equations 2 6 6 6 4  0 . n . .1 .0 .. {z . .n .. Moses.1 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). Prentice Hall.
. Stoica and R. Prentice Hall. 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. where x = xn : : : x1]T ~ ~ ~ Nested structure 2 Rn+2 = Thus. Moses. 2 4 Rn+1 n+1 r ~n n+1 r ~n 0 3 5 2 r ~n = 6 4 .LevinsonDurbin Alg. 1997 Slide L4–9 . con't Relevant Properties of R: Rx = y $ Rx = y.
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. 1997 . Prentice Hall. 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. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. n = n ) n+1 = 0 + kn 2+1 = n + kn n = n(1 .LevinsonDurbin Alg. Stoica and R. 2 kn = .
r(k) = 0 for jkj > m and 2 2 = m r(k)e.m X Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L4–11 . m) Thus. 1) + + bme(t . Moses. Prentice Hall.i!k (!) = jB(!)j k=.MA Signals MA: n =0 y(t) = B (q)e(t) = e(t) + b1e(t . Stoica and R.
Estimate fbkg and 2 and insert them in (!) = jB(!)j2 2 nonlinear estimation problem ^(!) is guaranteed to be 0 2. Moses. 1997 Slide L4–12 . with AR model order n . ^(!) is not guaranteed to be 0 =0 Both methods are special cases of ARMA methods described below.MA Spectrum Estimation Two main ways to Estimate (! ): 1. Stoica and R. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P.m ^BT (!) with a rectangular lag window of 2m + 1.i!k k=. Insert sample covariances (! ) = This is length X m f^(k)g in: r r(k)e.
ARMA Signals ARMA models can represent spectra with both peaks (AR part) and valleys (MA part). Stoica and R. A(q)y(t) = B (q)e(t) B (!) 2 = m=.m k e. Moses.i!k k (!) = 2 A(!) jA(!)j2 P where k = E f B(q)e(t)] B(q)e(t . Prentice Hall. 1997 Slide L4–13 . j ) Lecture notes to accompany Introduction to Spectral Analysis by P. k)] g = E f A(q)y(t)] A(q)y(t . k)] g = X n X n j =0 p=0 aj ap r(k + p .
Prentice Hall. Estimate fai bj A(!) nonlinear estimation problem. ^(!) is not guaranteed to be 0 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. can use an approximate linear twostage least squares method ^(!) is guaranteed to be 0 2.m k e jA(!)j2 linear estimation problem (the Modiﬁed YuleWalker method).ARMA Spectrum Estimation Two Methods: 2g in (!) = 2 B(!) 2 1. Estimate fai r(k)g in (!) = P m . 1997 Slide L4–14 .i!k k=. Stoica and R.
2) + = AR(1) with j kj ! 0 as k ! 1 Step 1: Approximate. for some large K e(t) ' y(t) + 1y(t . Moses. f kgK=1 by using AR k 1b) Estimate the noise sequence ^(t) = y(t) + ^1y(t . K ) ^2 = N 1 j^(t)j2 e N . 1997 Slide L4–15 .TwoStage LeastSquares Method Assumption: The ARMA model is invertible: A e(t) = B(q) y(t) (q ) = y(t) + 1y(t . 1) + 2y(t . Stoica and R. 1) + e and its variance + ^K y(t . 1) + + K y(t . Prentice Hall. K t=K +1 X Lecture notes to accompany Introduction to Spectral Analysis by P. K ) 1a) Estimate the coefﬁcients modelling techniques.
1) : : : . 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. Moses. Stoica and R. Note that the ai and bj coefﬁcients enter linearly in the above equation: y(t) . Prentice Hall.TwoStage LeastSquares Method. ^(t) ' . con't fe(t)g by ^(t) in the ARMA equation.y(t . 1997 Slide L4–16 . y(t . n) e ^(t . m)] e e = a1 : : : an b1 : : : bm]T Lecture notes to accompany Introduction to Spectral Analysis by P.
. n + M ) . 1997 Slide L4–17 . . If M > n overdetermined squares solution for ai . Stoica and R. : : : r(m . the accuracy of ai is often better for M > n Lecture notes to accompany Introduction to Spectral Analysis by P.. . = f^ g Y W system of equations. n + 1) r(m . = m + 1 ::: m + M a1 an .. r f^ g . Moses. 5 .6 4 Replace fr(k)g by f^(k)g and solve for faig. n + 2) . i) = 0 k > m 3 7 5 " r(m) r(m + 1) . . .Modiﬁed YuleWalker Method ARMA Covariance Equation: r(k) + 2 6 4 X n In matrix form for k i=1 air(k . 1) : : : r(m . . fast Levinsontype algorithms exist for obtaining ai . # 2 = r(m + M . Prentice Hall. . least f^ g Note: For narrowband ARMA signals. r(m + M ) r(m + 1) r(m + 2) 3 7 If M n. .
Stoica and R. 1997 Slide L4–18 . Moses. Prentice Hall.Summary of Parametric Methods for Rational Spectra Method AR: YW or LS low lowmedium mediumhigh mediumhigh medium lowmedium No No Yes Computational Burden low Accuracy medium Guarantee ^(! ) 0 ? Yes MA: BT ARMA: MYW ARMA: 2Stage 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 L5–1 . Stoica and R.Parametric Methods for Line Spectra — Part 1 Lecture 5 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Moses.
!1 !2 !3  An “Ideal” line spectrum Lecture notes to accompany Introduction to Spectral Analysis by P. Moses.Line Spectra Many applications have signals with (near) sinusoidal components. Stoica and R. 1997 . Examples: communications radar. Prentice Hall. sonar geophysical seismology ARMA model is a poor approximation Better approximation by Discrete/Line Spectrum Models 2 6(2 1) 2 6 (!) 6 (2 2) 2 6 (2 2) 3 ! Slide L5–2 .
] (prevents model ambiguities) f'k g = independent rv's. ] (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.Line Spectral Signal Model Signal Model: Sinusoidal components of frequencies !k and powers 2 . 1997 . uniformly distributed on . Moses. Stoica and R. 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 .
!p) + 2 p Slide L5–4 Lecture notes to accompany Introduction to Spectral Analysis by P.i'j o e Z n 1 i' d' 2 = 0. k) = 2 ei!pk p j p n o Hence.i'j o = 1. e E xp(t)xj (t .Covariance Function and PSD Note that: E E n i'p e. for p = j e i'p e. r(k) = E fy(t)y (t . for p 6= j =2 .i'j o = E nei'p o E ne. Stoica and R. Moses. 1997 . k)g = n=1 2ei!pk + 2 k 0 p p P and (! ) = 2 X n p=1 2 (! . Prentice Hall.
^ ei!k t + residuals k Lecture notes to accompany Introduction to Spectral Analysis by P. 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 . Stoica and R. Prentice Hall. Moses.
Nonlinear Least Squares (NLS) Method min' g f! Let: X N k k k t=1  y(t) . . Prentice Hall. . Moses. 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. 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 . .
1B Y !=^ ! and ! = arg max Y B (B B ). 1997 Slide L5–7 . B ) = kY .1B Y ^ ! Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Prentice Hall.Nonlinear Least Squares (NLS) Method.1B Y ] B B] . con't Then: F = (Y .1B Y This gives: ^ = (B B). (B B). B ) (Y . Moses. B k2 = . Y B(B B).1B Y ] +Y Y . (B B).
! Difﬁcult Implementation: The NLS cost function F is multimodal. Stoica and R. 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. 1997 Slide L5–8 . Moses.5. Prentice Hall.
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
HighOrder YuleWalker 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
HighOrder YuleWalker Method. con't Estimation Procedure: Estimate f^igL=1 using an ARMA MYW technique b i ^( ) Roots of B q give “spurious” roots. f!kgn=1. Stoica and R. Moses. Prentice Hall. 1997 Slide L5–12 . along with L . n ^ k Lecture notes to accompany Introduction to Spectral Analysis by P.
r(L + 2) .HighOrder and Overdetermined YW Equations ARMA covariance: r (k ) + X L i=1 bir(k . . . 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 = . Stoica and R. Prentice Hall. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. . 1) : : : r(M ) r(L + M ) 4 4 = = {z }  {z r(L) r(L + 1) This is a highorder (if L > n) and overdetermined (if M > L) system of YW equations. 1997 Slide L5–13 . . . r(L + M .
(U V )b = .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). The MinimumNorm solution is b = . 1997 Slide L5–14 . Moses.1U Important property: The additional (L . Lecture notes to accompany Introduction to Spectral Analysis by P. y = . diagonal and nonsingular Thus. n) spurious zeros of B (q ) are located strictly inside the unit circle.HighOrder and Overdetermined YW Equations. Prentice Hall. if the MinimumNorm solution b is used. Stoica and R.
Moses. Practical Solution Let ^= ^ ^ ^ ^ but made from f^(k)g instead of fr(k)g. f^ g When the SNR is low. Lecture notes to accompany Introduction to Spectral Analysis by P. this approach may give spurious frequency estimates when L > n. . SVD of . V be deﬁned similarly to U . Prentice Hall. r . Stoica and R. 1997 Slide L5–15 . Compute ^ = . V from the Let U .1U ^ ^ ^ b ^(q) that Then !k n=1 are found from the n zeroes of B k are closest to the unit circle. this is the price paid for increased accuracy when L > n.V ^ .HOYW Equations.
Stoica and R.Parametric Methods for Line Spectra — Part 2 Lecture 6 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. Prentice Hall. 1997 Slide L6–1 .
1)! ]T A = a(!1) : : : a(!n)] (m n) rank Note: Deﬁne (A) = n y(t) y(t ... .i(m. Prentice Hall. 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 . 0 2 n 3 7 5 Lecture notes to accompany Introduction to Spectral Analysis by P.The Covariance Matrix Equation Let: a(!) = 1 e. m + 1) .i! : : : e. 1) (for m 3 7 7 7 5 n) 2 46 y(t) = 6 ~ 6 4 y(t . Stoica and R. Moses. 1997 Slide L6–2 .
1997 Slide L6–3 .n] (m (m .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.. m 7 5 S G # Lecture notes to accompany Introduction to Spectral Analysis by P.ng: orthonormal eigenvectors associated with f n+1 : : : m g S = s1 : : : sn] (m n) G = g1 : : : gm. Stoica and R. 2 R = S G] 6 4 1 3 " . Prentice Hall. n)) Thus.. Moses.
. Moses.1) = AC S = A(PA S ( ) = rank(A) = n). since S G j j 6= 0 4 . 1997 .. Therefore.Eigendecomposition of R and Its Properties. 0 n 3 7 5 with C (since rank S . = = nonsingular 0 n. . =0 A G=0 Slide L6–4 Lecture notes to accompany Introduction to Spectral Analysis by P. con't As rank (APA ) = n: k > 2 k = 1 ::: n k = 2 k = n + 1 ::: m 0 1. 2 . Stoica and R.. 2 2 6 4 3 7 5 2 6 4 Note: RS = APA S + 2S = S 1 0 . Prentice Hall.
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. Stoica and R. Prentice Hall. 1997 Slide L6–5 .MUSIC Method a (!1) . Moses. A G= G=0 a (!n) ) fa(!k )gn=1 ? R(G) k 2 6 4 3 7 5 Thus. .
1997 Slide L6–6 . a (!)G ^ Root MUSIC Method: ] f!kgn=1 = the angular positions of the n roots of: ^ k ^^ aT (z. Here. Moses.(m.1)GG a(z) = 0 that are closest to the unit circle. Prentice Hall.1)]T Note: Both variants of MUSIC may produce spurious frequency estimates. a(z) = 1 z.1 : : : z. Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P.Spectral and Root MUSIC Methods Spectral MUSIC Method: f!kgn=1 = the locations of the n highest peaks of the ^ k “pseudospectrum” function: 1 ^G a(!) ! 2 .
) f!kgn=1 can be found from the roots of ^ k aT (z. 1997 Slide L6–7 . Prentice Hall.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. Moses. Stoica and R. If: =n+1 m=n+1 Then: ^ g G = ^1.Pisarenko Method Pisarenko is a special case of MUSIC with m (the minimum possible value).
1997 Slide L6–8 . Moses.MinNorm Method Goals: Reduce computational burden. with ﬁrst element equal ^ ^ to one. Uses m n (as in MUSIC). and reduce risk of false frequency estimates. g # Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. R( ) Let " 1 = the vector in R(G). Stoica and R. but only one vector in G (as in Pisarenko). that has minimum Euclidean norm.
con't Spectral MinNorm f!gn=1 = ^k the locations of the n highest peaks in the “pseudospectrum” 2 1 = a (!) 1 ^ g " # Root MinNorm f!gn=1 = ^k the angular positions of the n roots of the polynomial aT (z. Moses. Stoica and R. Prentice Hall.MinNorm Method. Lecture notes to accompany Introduction to Spectral Analysis by P.1) 1 ^ g " # that are closest to the unit circle. 1997 Slide L6–9 .
(S S ).1 = =(1 .S =(1 .1 gm " # " # MinNorm Method: Determining g ^ Then: 1 2 R(G) ) S 1 = 0 ^ ^ ^ ^ g g ) S ^= . 1997 Slide L6–10 . for N 1. k k2) g Lecture notes to accompany Introduction to Spectral Analysis by P. at least.1 exists iff = k k2 6= 1 (This holds.1 g + S S . k k2) = (S S ) ) (S S ). Moses.Let S ^= S g1 . Prentice Hall.) ^^ I=S S= Multiplying the above equation by gives: (1 .S(S S ). k k2) ) ^ = . Stoica and R. g " # MinNorm solution: As: ^ = .
Moses.1DC {z } So has the same eigenvalues as D . where e.1]S 2 6 4  3 7 5 Recall S = AC with jC j 6= 0. Stoica and R.1S1S2 Lecture notes to accompany Introduction to Spectral Analysis by P. D= 0 e. Prentice Hall.1]A Then A2 Also.i!n S1 = Im... let = A1D.ESPRIT Method Let A1 = Im. determined as is uniquely = (S1S1).1 0]S S2 = 0 Im.i!1 0 . Then S2 = A2C = A1DC = S1 C .1 0]A A2 = 0 Im. 1997 Slide L6–11 .
arg(^k ) ^ k ESPRIT Advantages: computationally simple no extraneous frequency estimates (unlike in MUSIC or Min–Norm) accurate frequency estimates Lecture notes to accompany Introduction to Spectral Analysis by P.ESPRIT Implementation From the eigendecomposition of R. Stoica and R. ^ ^ ^ ^ The frequency estimates are found by: where f^kgn=1 are the eigenvalues of k ^ = (S1S1). then S1 and S2 .1S1S2 ^^ ^^ f!kgn=1 = . ﬁnd S . Moses. 1997 Slide L6–12 . Prentice Hall.
Stoica and R. 1997 . Moses.Summary of Frequency Estimation Methods Method Periodogram Nonlinear LS YuleWalker Pisarenko MUSIC MinNorm ESPRIT Computational Accuracy / Risk for False Burden Resolution Freq Estimates small mediumhigh 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 mediumresolution applications Use ESPRIT for highresolution applications Slide L6–13 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.
Prentice Hall. 1997 Slide L7–1 . Stoica and R. Moses.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. (!) by a 2. 1997 Slide L7–2 . !+ ] N >1 1 2 is more general than 1. Lecture notes to accompany Introduction to Spectral Analysis by P. but 2 requires to ensure that the number of estimated values = = ) is < N . Nonparametric Approach: Implicitly smooth ! !=. ( =2 2 =1 N > 1 leads to the variability / resolution compromise associated with all nonparametric methods. by assuming that ! is nearly constant over the bands f ( )g ( ) !. Moses. Parametric Approach: Parameterize ﬁnitedimensional model. Prentice Hall.
c ( )d = (') (!) (a) consistent power calculation (b) Ideal passband ﬁlter with bandwidth (c) ( ) constant on 2 ! . as well as (b) and (c). 1997 . !c @Q Q ! . Stoica and R. Moses.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. Z 6 ~ (! ) @ ! (b) Z !c ^FB (!) ' 1 2 (a) . Prentice Hall.with varying ! Power Calculation Bandpass Filter c and xed bandwidth  Power in the band  Division by lter bandwidth ^(!c )  6 y y (! ) . 2 ! + 2 ] Note that assumptions (a) and (b). are conﬂicting. jH ( )j2 ( )d = ' 21 !+ !.
Filter Bank Interpretation of the Periodogram 1 N y(t)e. 1 ~ hk = N otherwise 0 1 center frequency of H 3dB bandwidth of H (!) = ! ~ (!) ' 1=N Slide L7–4 Lecture notes to accompany Introduction to Spectral Analysis by P.i!t 2 ~ ^p(~) = ! 4 N t=1 2 N 1 ~ = N y(t)ei!(N .t) t=1 X X = N where ( X 1 k=0 hk y(N . Stoica and R. k) 2 ! . 1 H (!) = hk e ! N ei(~. 1997 . Moses.!) . Prentice Hall.i!k = 1 eiN (~. 1 k=0 X 1 ei!k k = 0 : : : N .!) .
con't jH (!)j as a function of (~ .Filter Bank Interpretation of the Periodogram. and: narrow ﬁlter passband. Stoica and R. for N = 50. Prentice Hall. ! 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. ^( ) Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 Slide L7–5 . Moses.
Moses. Use several bandpass ﬁlters on the whole sample. and bandpass ﬁlter each subsample. Prentice Hall. Both approaches compromise bias for variance.Possible Improvements to the Filter Bank Approach 1. 2. and in fact are quite related to each other: splitting the data sample can be interpreted as a special form of windowing or ﬁltering. Stoica and R. ~ provides several samples for power calculation. more data points for the power calculation stage. This “multiwindow approach” is similar to the Daniell method. 1997 Slide L7–6 . Split the available sample. Each ﬁlter covers a small band centered on ! . This approach leads to Bartlett and Welch methods. Lecture notes to accompany Introduction to Spectral Analysis by P.
1997 Slide L7–7 . y(t .i!k = h a(!) Lecture notes to accompany Introduction to Spectral Analysis by P. yF (t) = X m k=0  hk y(t .i! : : : e. Stoica and R.Capon Method Idea: Datadependent bandpass ﬁlter design. 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. Moses.im! ]T hk e. Prentice Hall. k) 2 6 4  . = h0 h1 : : : hm] .
1a(!) which should be the power of y (t) in a passband centered on ! . Stoica and R.m t=m+1 X Lecture notes to accompany Introduction to Spectral Analysis by P. 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. Prentice Hall.1 a (!)R a(!) with N 1 ^ y(t)~ (t) ~ y R = N . 1997 Slide L7–8 .1a=a R. Moses.Capon Method. n The Bandwidth 1 ' m+1 = (ﬁlter 1 length) Conclusion Estimate PSD as: +1 ^(!) = m^.
bias decreases and variance increases.Capon Properties m is the user parameter that controls the compromise between bias and variance: – as m increases. Prentice Hall. . Capon uses one bandpass ﬁlter only. Stoica and R. but it splits the N data point sample into (N m) subsequences of length m with maximum overlap. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. 1997 Slide L7–9 .
1997 Slide L7–10 . Prentice Hall.i!k ^ k=.m m + 1 1 m m ^(t .i!(t. j )] ^ r = m+1 X X X Then we have ^ a (!)Ra(!) ^BT (!) = m+1 m+1 ^C (!) = ^ a (!)R. s)e. Stoica and R. Moses.1a(!) Lecture notes to accompany Introduction to Spectral Analysis by P. jkj ^BT (!) = r(k)e.Relation between Capon and BlackmanTukey Methods Consider ^BT (!) with Bartlett window: m m + 1 .s) = m+1 r t=0 s=0 ^ a (!)Ra(!) R = ^(i .
^( ) ^( ) Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.Relation between Capon and AR Methods Let 2 ^k ^k (!) = ^ 2 jAk(!)j be the kth order AR PSD estimate of y (t). C ! generally has worse resolution and bias properties than the AR method. 1997 Slide L7–11 . Due to the loworder AR terms in the average. Moses. Stoica and R. C ! generally has less statistical variability than the AR PSD estimator. AR Then ^C (!) = 1 1 m 1= ^AR(!) m + 1 k=0 k X Consequences: Due to the average over k.
1997 Slide L8–1 . Moses. Stoica and R.Spatial Methods — Part 1 Lecture 8 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall.
R . . sonar. @. @. Sensor 1 @. 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. Stoica and R. seismometers Applications: Radar. @. @.. Sensor 2 Sensor m Problem: Detect and locate n radiating sources by using an array of m passive sensors.@ . . . mechanical Receiving sensors: Hydrophones. @. Emitted energy: Acoustic.. v Source n @. Prentice Hall. Moses. communications. electromagnetic. seismology..The Spatial Spectral Estimation Problem Source 1 vSource 2 B B B B N B cccc ccc c c c c c v @ . 1997 Slide L8–2 . antennas.@ . .
Prentice Hall. (We do not treat the detection problem) The sensors are linear dynamic elements with known transfer characteristics and known locations (That is.Simplifying Assumptions Farﬁeld sources in the same plane as the array of sensors Nondispersive wave propagation Hence: The waves are planar and the only location parameter is direction of arrival (DOA) (or angle of arrival.) Lecture notes to accompany Introduction to Spectral Analysis by P. The number of sources n is known. the array is calibrated. Stoica and R. AOA). 1997 Slide L8–3 . Moses.
() f kg with hk(t) This is a timedelay estimation problem in the unknown input case. Stoica and R. etc. Moses. thermal noise in sensor electronics.. k) + ek (t) ( = convolution operator). 1997 Slide L8–4 . Lecture notes to accompany Introduction to Spectral Analysis by P. at the “ﬁrst” sensor) the delay between the reference point and the kth sensor the impulse response (weighting function) of sensor k “noise” at the kth sensor (e. Then the output of sensor k is yk (t) = hk (t) x(t .g. Prentice Hall.Array Model — Single Emitter Case x(t) = k= the signal waveform as measured at a reference point (e.. background noise.) hk (t) = ek (t) = Note: t 2 R (continuoustime signals).g. Basic Problem: Estimate the time delays known but x t unknown.
Stoica and R. Moses. k ) ' (t) cos !ct + '(t) . k ) ' jHk (!c)j (t)cos !ct + '(t) . Prentice Hall. !c k + argfHk (!c)g] where Hk function (!) = Ffhk(t)g is the kth sensor's transfer Hence. Then: x(t) = (t) cos !ct + '(t)] where (t) '(t) vary “slowly enough” so that (t .Narrowband Assumption Assume: The emitted signals are narrowband with known carrier frequency !c . !c k + arg Hk (!c)] + ek(t) Slide L8–5 Lecture notes to accompany Introduction to Spectral Analysis by P. 1997 . k ) ' '(t) Time delay is now ' to a phase shift !c k : x(t . the kth sensor output is yk (t) = jHk (!c)j (t) cos !ct + '(t) . k) ' (t) '(t . !c k ] hk (t) x(t .
i!ct ~ = yk(t) cos(!ct) . 1997 Slide L8–6 or . by lowpass ﬁltering and sampling the signal yk (t)=2 = yk(t)e.i !ct+ (t)] 2 Demodulate z (t) (translate to baseband): 2z(t)e.Complex Signal Representation The noisefree output has the form: z(t) = (t) cos !ct + (t)] = = (t) ei !ct+ (t)] + e.i!ct to obtain (t)ei (t) lowpass highpass Hence. Prentice Hall.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). Lecture notes to accompany Introduction to Spectral Analysis by P. 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.i 2!ct+ (t)] g n o  {z }  {z } Lowpass ﬁlter 2z(t)e. Moses.!ct = (t)f ei (t) + e. Stoica and R.
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
Moses. 1997 Slide L8–10 . Stoica and R. 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 . 1) c Lecture notes to accompany Introduction to Spectral Analysis by P.
Stoica and R.1)!s ]T By direct analogy with the vector a(! ) made from uniform samples of a sinusoidal time series.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(m. Prentice Hall. Moses.i!s : : : e. !s = spatial frequency The function !s 7! a( ) is onetoone for j!sj $ dj sin j 1 d =2 As =2 d =2 is a spatial Shannon sampling theorem. 1997 . Slide L8–11 d = spatial sampling period Lecture notes to accompany Introduction to Spectral Analysis by P.
Stoica and R. Moses. 1997 Slide L9–1 .Spatial Methods — Part 2 Lecture 9 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 timeseries 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. 1997 Slide L9–2 . Stoica and R.
Prentice Hall. Stoica and R. m + 1)]T If u m.i! : : : e. Moses.1] y(t) = u(t) : : : u(t .Analogy between Temporal and Spatial Filtering Temporal FIR Filter: yF (t) = hk u(t .i(m. 1997 Slide L9–3 . k) = h y(t) k=0 h = ho : : : hm. Lecture notes to accompany Introduction to Spectral Analysis by P.1)! ]T We can select h to enhance or attenuate signals with different frequencies ! .1 X (t) = ei!t then yF (t) = h a(!)] u(t)  {z } ﬁlter transfer function a(!) = 1 e.
i!c 2 : : : e. Stoica and R. Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. 1997 Slide L9–4 .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. Spatial FIR Filter output: yF (t) = X m k=1 hk yk (t) = h y(t) Narrowband Wavefront: The array's (noisefree) response to a narrowband ( sinusoidal) wavefront with complex envelope s t is: () y(t) = a( )s(t) a( ) = 1 e. Prentice Hall.Analogy between Temporal and Spatial Filtering Spatial Filter: fyk(t)gm=1 = k the “spatial samples” obtained with a sensor array.
1997 Slide L9–5 .i! C @ h a(!)]u(t) R @ C . C C B C B C B C B C B A @ .  e {z } ? a(!) 1 ? 0 qq ( 1) 1 (Temporal sampling) (a) Temporal ﬁlter narrowband source with DOA= Z z Z Z Z ~ aa 1reference !! h 0 0 1 B aa 2.@ 0 B 1 C h @@ C B B e.1 ?  1 .B ..i m.1 q .. Cu(t) C B . Stoica and R.Analogy between Temporal and Spatial Filtering u(t) = ei!t z 1 s q. s qq h  ? 1 ?  m. ! hm.i! Be 2 B !! B B B B . Moses.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. B . ? 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 . B B @ .
h a 0 . Moses. where 0 j ( )j = ( ) = 25 0 30 −30 Th et a (d eg ) 60 −60 90 0 2 4 6 Magnitude 8 10 −90 Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. Here. con't Example: The response magnitude h a of a spatial ﬁlter (or beamformer) for a 10element ULA. 1997 Slide L9–6 . Prentice Hall.Spatial Filtering.
1997 Slide L9–7 . Lecture notes to accompany Introduction to Spectral Analysis by P. by sweeping the ﬁlter through the DOA range of interest (“goniometer”). Prentice Hall. Stoica and R. Moses. hence ﬁltering out interferences located outside the ﬁlter's beam (but possibly having the same temporal characteristics as the signal). To locate an emitter in the ﬁeld of view.Spatial Filtering Uses Spatial Filters can be used To pass the signal of interest only.
n o n o The (dominant) peaks of h DOAs of the sources. ( )Rh( ) give the Slide L9–8 Lecture notes to accompany Introduction to Spectral Analysis by P. Prentice Hall. Moses. 1997 . Stoica and R. 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.Nonparametric Spatial Methods A Filter Bank Approach to DOA estimation. and evaluate the powers of the ﬁltered signals: E jyF (t)j2 = E jh ( )y(t)j2 = h ( )Rh( ) with R = E fy (t)y (t)g.
Stoica and R. Moses. () 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. y t can be considered as impinging on the array from all DOAs.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. 1997 Slide L9–9 .
Stoica and R. pj > = wavelength array length array beamwidth Inconsistency problem: Beamforming DOA estimates are consistent if n inconsistent if n > . 1 = 1. Resolution Threshold: inf j k .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. 1997 . but Slide L9–10 Lecture notes to accompany Introduction to Spectral Analysis by P. Moses. ( )^ ( ) This is the direct spatial analog of the BlackmanTukey periodogram.
Stoica and R.1a( ) E jyF (t)j2 = 1=a ( )R. Moses.1a( ): Performance: Slightly superior to Beamforming.1a( ) n o Implementation: f^kg = the locations of the n largest peaks of ^ 1=a ( )R. They do not make assumptions on the covariance properties of the data (and hence do not depend on them). 1997 Slide L9–11 .1a( )=a ( )R. 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. Prentice Hall. Both Beamforming and Capon are nonparametric approaches.
Moses. Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. MINNORM and ESPRIT methods of frequency estimation can be used. YW. 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. almost without modiﬁcation. Prentice Hall. 1997 Slide L9–12 . for DOA estimation. Then: R = E fy(t)y (t)g = APA + 2I Thus: The NLS.
1A y(t) t = 1 : : : N s X Then 1 N k I . Lecture notes to accompany Introduction to Spectral Analysis by P. this is precisely the form of the NLS method of frequency estimation. f ^k g = arg max trf A(A A).1A ]y(t)k2 f ( ^) = N s t=1 1 N y (t) I . Moses. 1997 Slide L9–13 =1 . A(A A).Nonlinear Least Squares Method 1 N ky(t) . Prentice Hall.1A ]Rg X f kg For N . A(A A).1A ]y(t) = N t=1 ^ = trf I .1A ]Rg ^ Thus. Stoica and R. 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). A(A A).
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. Prentice Hall. 1997 Slide L9–14 . Stoica and R. Moses.
n ~ ~ Properties: A V2 = 0 V1 2 R(A) Then: rank {z} {z} Lecture notes to accompany Introduction to Spectral Analysis by P. Stoica and R. and the SVD of .) = n. is . 1997 Slide L9–15 . Moses.= y (M L) Also assume: M > n L > n ( ) m = M + L > 2n) rank ~ (A) = rank(A) = n " # " # (.n 0 V2 g L n M .YuleWalker 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 ~ . = U1 U2 ] 0 n n 0 V1 g n . Prentice Hall.
1997 Slide L9–16 . Moses. Stoica and R.YWMUSIC DOA Estimator f^kg = where the n largest peaks of 1=~ ( )V2V2 ~( ) a ^^a ~( ).=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. Prentice Hall. is the “array transfer vector” for y(t) a ~ at DOA ^ V2 is deﬁned similarly to V2. (L 1). using 1 N y(t)~ (t) ^ .
that is. Prentice Hall. Spectral forms of MUSIC and MinNorm can be used for arbitrary arrays Root forms can be used only with ULAs MUSIC and MinNorm break down if the source signals are coherent. 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. Moses.MUSIC and MinNorm Methods Both MUSIC and MinNorm methods for frequency estimation apply with only minor modiﬁcations to the DOA estimation problem. Stoica and R. 1997 Slide L9–17 .
Prentice Hall... Lecture notes to accompany Introduction to Spectral Analysis by P. 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. Moses. 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.i!c ( 1) 0 . D= 0 e.
1 A1 = Im. Moses. 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. Stoica and R. so m m and .1 0]A = .1 A2 = 0 Im.1]A Lecture notes to accompany Introduction to Spectral Analysis by P. the two subarrays are often the ﬁrst m and last m array elements. Prentice Hall.1 . 1997 Slide L9–19 .ESPRIT Method.