Stationary Stochastic Processes, 2011
Stationary Stochastic Processes, FMS045
Table of Formulas, 2011
Chapter 2: Stationary processes
Sum of components with random phase and amplitude: If
X(t) = A0 +
n
X
Ak cos(2pfk t + fk ) ,
k=1
2
where fk Rect(0, 2p), Ak , k = 1, . . . , n, are independent, E A20 = s20 and E A2k /2 = sk ,
then
n
X
2
rX (t) = s0 +
s2k cos 2pfk t .
k=1
Let {X(t)} be a Poisson process with intensity l. Then:
E[X(t)] =
V[X(t)] =
lt
lt
l min(s, t)
rX (s, t) =
P
Estimation of expectation, mn = n1 nt=1 x(t):
V [mn ] =
nV [mn ]
1
n2
n1
X
(n |k|)rX (k)
k=n+1
rX (k)
for large n
k=
Estimation of covariance function:
? m known:
nt
X
t) = n1 (x(t) m)(x(t + t) m)
rn (
t0
t=1
? m unknown:
t) =
rn (
where
x(t) =
1
n
1
n
Pnt
x)(x(t + t) x)
t0
Pnt
x(t) )(x(t + t) x(t+) )
t0
t=1 (x(t)
t=1 (x(t)
nt
1 X
x(t),
nt
t=1
x(t+) =
nt
1 X
x(t + t)
nt
t=1
Stationary Stochastic Processes, 2011
Chapter 3: Spectral representations
Folding when sampling a continuous time process (aliasing): Let {Z(t), t = 0, d, 2d, . . .}
be Y(t) sampled with sampling interval d and sampling frequency fs = 1/d:
P
k
1
1
RZ (f) =
for 2d
< f 2d
.
k= RY f + d
P
fs
fs
R
(f
+
kf
)
for
=
<
f
.
s
k= Y
2
2
Chapter 4: Gaussian processes
The two-dimensional probability density function of a jointly Gaussian random variable,
(X, Y), with E[X] = E[Y] = 0, V[X] = V[Y] = 1 and C[X, Y] = r is
1
1
2
2
p
exp
(x 2rxy + y )
.
fX,Y (x, y) =
2(1 r2 )
2p 1 r2
Let Xk N(0,
sk ) and Yk N(0, sk ) be independent and
X(t) =
{Xk cos 2pfk t + Yk sin 2pfk t}
Ak cos(2pfk t + fk ) ,
where fk = arg(Xk iYk ) Rect(0, 2p), then Ak =
tion with probability density function
fAk (x) =
2e
k
x2 /2s2k
X2k + Yk2 has a Rayleigh distribu-
x>0 .
Campbells formulas:
X(t) =
E[X(t)] =
rX (t) =
tk t
g(t tk )
g(t) dt
g(u)g(u t) du = l
g(u)g(u + t) du
The process {X(t), t 0} is a Wiener process if
? X(0) = 0 and,
? {X(t)} has independent increments, that is X(t2 ) X(t1 ) and X(t3 ) X(t2 ), 0
t1 t2 t3 , are independent and,
? X(t) X(t + h) N(0, s h).
Stationary Stochastic Processes, 2011
Chapter 5: Linear filters - general theory
Impulse response h(u):
R
h(u)X(t u) du
Y(t) =
P
h(u)X(t u)
(continuous time)
(discrete time)
Relation between covariance functions:
R
R
u= v= h(u)h(v) rX (t + u v) du dv
rY (t) =
P P
u
v= h(u)h(v) rX (t + u v)
(continuous time)
(discrete time)
R
Derivation: X0 (t) exists (in quadratic mean) if r00X (t) exists. This is equivalent to (2pf)2 R(f)df <
. If X0 (t) exists, the following relations hold:
rX0 (t) = r00X (t) ,
RX0 (f) = (2pf)2 RX (f) ,
V X0 (t) =
(2pf)2 RX (f) df
rX,X0 (t) = r0X (t) ,
rX(j) ,X(k) (t) = (1)j rX
(j+k)
(t) .
Cross-covariance and cross-spectrum:
rX,Y (t) = C[X(t), Y(t + t)] =
RX, Y (f) = H(f)RX (f)
RX,Y (f) = AX,Y (f)eiFX,Y (f)
amplitude spectrum
AX,Y (f)
phase spectrum
FX,Y (f)
squared coherence spectrum
k2X,Y (f) = R
A2X,Y (f)
X (f)RY (f)
ei2pf t RX,Y (f) df
Stationary Stochastic Processes, 2011
Chapter 6: Linear filters - discrete time
White noise in discrete time: {et , t Z}, E [et ] = 0 and V [et ] = s2 :
s2
Re (f) =
for 1/2 f 1/2
AR(p)-process: (a0 = 1)
Xt + a1 Xt1 + a2 Xt2 + . . . + ap Xtp = et
? Yule-Walker equations for covariance function:
rX (k) + a1 rX (k 1) + . . . + ap rX (k p) =
s2
0
? Spectral density:
RX (f) = Pp
s2
k=0 ak e
i2pfk 2
MA(q)-process: (b0 = 1)
Xt = et + b1 et1 + b2 et2 + . . . + bq etq
? Covariance function:
rX (t) =
s2
jk=t bj bk
for |t| q
for |t| > q
? Spectral density:
q
2
X
RX (f) = s2
bk ei2pfk
k=0
Chapter 7: Linear filters - continuous time
Matched filter:
? with white noise:
h(u) = c s(T u)
Z
1
s(T u)2 du
SNR =
N0
for k = 0
for k = 1, 2, . . .
Stationary Stochastic Processes, 2011
? with colored noise:
Z
s(T u) = c
h(v)rN (u v) dv
Z Z
SNR = c2
h(u)h(v)rN (u v) du dv
Wiener filter:
RS (f)
RS (f) + RN (f)
R
RS (f) df
R RS (f)RN (f)
RS (f)+RN (f) df
H(f) =
SNR =
Chapter 8: Spectral estimation
Periodogram of the sequence {x(t), t = 0, 1, 2, . . . n 1},
Rx (f) =
where X (f) =
Pn1
t=0
1
|X (f)|2 ,
n
x(t)ei2pft .
E [Rx (f)]
t=
Z
kn (t)rX (t)ei2pf t ,
1/2
Kn (f u)RX (u)du,
=
1/2
P
i2pf t .
where kn (t) = 1 |nt| for n + 1 t n 1 and Kn (f) = n1
t=n+1 kn (t)e
2
RX (f) for 0 < |f| < 1/2
V [Rx (f)]
2R2X (f) for f = 0, 1/2
The distribution of the periodogram estimate is
Rx (f)
RX (f)
q2 (2) ,
2
0 < f < 1/2.
Modified periodogram
Rw (f)
2
n1
1 X
i2pft
x(t)w(t)e
,
n
t=0
Z
2
1 1/2
X
(
n
)W(f
n
)d
n
.
n 1/2
Stationary Stochastic Processes, 2011
Lag-windowing
Rlw (f)
t=
Z
1/2
=
1/2
kLn (t)rx (t)ei2pf t ,
KLn (f n)Rx (n)dn.
Averaging of spectrum
K
Rav (f) =
1 X
Rx,j (f),
K
j=1
where K different spectrum estimates,
Rav (f)
RX (f)
Rx,j (f),
q2 (2K) ,
2K
j = 1 . . . K, are used. The distribution is
0 < f < 1/2.
Trigonometric relations
cos(a + b ) = cos a cos b sin a sin b
cos(a b ) = cos a cos b + sin a sin b
sin(a + b ) = sin a cos b + cos a sin b
sin(a b ) = sin a cos b cos a sin b
Stationary Stochastic Processes, 2011
Fourier transforms
g(t)
ea|t|
(a > 0)
G(f) =
2a
a2 +(2pf)2
i2pf t g(
t) dt
a2 +t2
p e2pa|f|
a
|t|ea|t|
(2pf) )
2 ((aa2 +(2
pf)2 )2
|t|k ea|t|
k!
{(
(a2 +(2pf)2 )k+1
a + i2pf)k+1 + (a i2pf)k+1 }
p/a exp( (24paf)
eat
ea|t| cos(2pf0 t)
a
a
a2 +(2pf0 2pf)2 + a2 +(2pf0 +2pf)2
ea|t| sin(2pf0 t)
2pf0 2pf
2pf0 +2pf
a2 +(2pf0 2pf)2 + a2 +(2pf0 +2pf)2
if
if
sin(2pat)
2pt
t=0
t 6= 0
1 a|t| if |t| 1a
0
if |t| > 1a
g(t)h(t)
g(t) h(t) =
1/2
0
if |f| a
if |f| > a
a 2a
1 cos 2apf
(2pf)2
G(f) H(f) =
R
g(t)h(t t)dt G(f)H(f)
g0 (t)
i2pf G(f)
g(at)
t
a g( a )
G(af)
g(t t0 )
G(f)ei2pf t0
g(t)ei2pf0 t
G(f f0 )
a G( a )
if f = 0
if f =
6 0
G(n)H(f n)dn
Stationary Stochastic Processes, 2011
Gaussian distribution table
F(x) = F (x)
x
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.0
0.00
0.50000
0.53983
0.57926
0.61791
0.65542
0.69146
0.72575
0.75804
0.78814
0.81594
0.84134
0.86433
0.88493
0.90320
0.91924
0.93319
0.94520
0.95543
0.96407
0.97128
0.97725
0.98214
0.98610
0.98928
0.99180
0.99379
0.99534
0.99653
0.99744
0.99813
0.99865
0.99903
0.99931
0.99952
0.99966
0.99977
0.99984
0.99989
0.99993
0.99995
0.99997
0.01
0.50399
0.54380
0.58317
0.62172
0.65910
0.69497
0.72907
0.76115
0.79103
0.81859
0.84375
0.86650
0.88686
0.90490
0.92073
0.93448
0.94630
0.95637
0.96485
0.97193
0.97778
0.98257
0.98645
0.98956
0.99202
0.99396
0.99547
0.99664
0.99752
0.99819
0.99869
0.99906
0.99934
0.99953
0.99968
0.99978
0.99985
0.99990
0.99993
0.99995
0.99997
0.02
0.50798
0.54776
0.58706
0.62552
0.66276
0.69847
0.73237
0.76424
0.79389
0.82121
0.84614
0.86864
0.88877
0.90658
0.92220
0.93574
0.94738
0.95728
0.96562
0.97257
0.97831
0.98300
0.98679
0.98983
0.99224
0.99413
0.99560
0.99674
0.99760
0.99825
0.99874
0.99910
0.99936
0.99955
0.99969
0.99978
0.99985
0.99990
0.99993
0.99996
0.99997
0.03
0.51197
0.55172
0.59095
0.62930
0.66640
0.70194
0.73565
0.76730
0.79673
0.82381
0.84849
0.87076
0.89065
0.90824
0.92364
0.93699
0.94845
0.95818
0.96638
0.97320
0.97882
0.98341
0.98713
0.99010
0.99245
0.99430
0.99573
0.99683
0.99767
0.99831
0.99878
0.99913
0.99938
0.99957
0.99970
0.99979
0.99986
0.99990
0.99994
0.99996
0.99997
0.04
0.51595
0.55567
0.59483
0.63307
0.67003
0.70540
0.73891
0.77035
0.79955
0.82639
0.85083
0.87286
0.89251
0.90988
0.92507
0.93822
0.94950
0.95907
0.96712
0.97381
0.97932
0.98382
0.98745
0.99036
0.99266
0.99446
0.99585
0.99693
0.99774
0.99836
0.99882
0.99916
0.99940
0.99958
0.99971
0.99980
0.99986
0.99991
0.99994
0.99996
0.99997
0.05
0.51994
0.55962
0.59871
0.63683
0.67364
0.70884
0.74215
0.77337
0.80234
0.82894
0.85314
0.87493
0.89435
0.91149
0.92647
0.93943
0.95053
0.95994
0.96784
0.97441
0.97982
0.98422
0.98778
0.99061
0.99286
0.99461
0.99598
0.99702
0.99781
0.99841
0.99886
0.99918
0.99942
0.99960
0.99972
0.99981
0.99987
0.99991
0.99994
0.99996
0.99997
0.06
0.52392
0.56356
0.60257
0.64058
0.67724
0.71226
0.74537
0.77637
0.80511
0.83147
0.85543
0.87698
0.89617
0.91309
0.92785
0.94062
0.95154
0.96080
0.96856
0.97500
0.98030
0.98461
0.98809
0.99086
0.99305
0.99477
0.99609
0.99711
0.99788
0.99846
0.99889
0.99921
0.99944
0.99961
0.99973
0.99981
0.99987
0.99992
0.99994
0.99996
0.99998
0.07
0.52790
0.56749
0.60642
0.64431
0.68082
0.71566
0.74857
0.77935
0.80785
0.83398
0.85769
0.87900
0.89796
0.91466
0.92922
0.94179
0.95254
0.96164
0.96926
0.97558
0.98077
0.98500
0.98840
0.99111
0.99324
0.99492
0.99621
0.99720
0.99795
0.99851
0.99893
0.99924
0.99946
0.99962
0.99974
0.99982
0.99988
0.99992
0.99995
0.99996
0.99998
0.08
0.53188
0.57142
0.61026
0.64803
0.68439
0.71904
0.75175
0.78230
0.81057
0.83646
0.85993
0.88100
0.89973
0.91621
0.93056
0.94295
0.95352
0.96246
0.96995
0.97615
0.98124
0.98537
0.98870
0.99134
0.99343
0.99506
0.99632
0.99728
0.99801
0.99856
0.99896
0.99926
0.99948
0.99964
0.99975
0.99983
0.99988
0.99992
0.99995
0.99997
0.99998
0.09
0.53586
0.57535
0.61409
0.65173
0.68793
0.72240
0.75490
0.78524
0.81327
0.83891
0.86214
0.88298
0.90147
0.91774
0.93189
0.94408
0.95449
0.96327
0.97062
0.97670
0.98169
0.98574
0.98899
0.99158
0.99361
0.99520
0.99643
0.99736
0.99807
0.99861
0.99900
0.99929
0.99950
0.99965
0.99976
0.99983
0.99989
0.99992
0.99995
0.99997
0.99998
