This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Chapter 12 Spectrum Estimation 12.1 Ergodicity & CS K data a RP a statistical parametersdp 5 a parametersdp function of ( ) t X h expectationd p 5 expectationd p 5 sample space a average (ensemble average)a a For a specific t E ( ) t X is an RVa { } ( ) ( ) t E t η = X t a observed n samples ( , ) i t ξ X of ( ) t X h mean of ( ) t X by 1 ˆ( ) ( , ) i i t t n η ξ = X a point estimation a consistent estimate of ( ) t η n E E variance a a n dp 5 estimatea L dp 5 sample of ( ) t X { } ( ) E t X h { } ( ) E t X depends on t dp 5 E ( ) t X h regular stationary processa time average a { } ( ) E t X as length a Ergodicity a Meanergodic process Given a real stationary process ( ) t X dp 5 { } ( ) ( ) t E t η = X h form time average 1 ( ) 2 T T T t dt T = η X { } { } 1 ( ) 2 T T T E E t dt T η = = η X h ∴ T η h unbiased estimator of η variance of T η 2 T σ as T E E T η η in the MS sensea 1 ( ) t X h meanergodica i.e.a ( ) t X is meanergodic if T η h ensemble average η as T E ∴ meanregodic a T σ T σ as T E Ex. 12.1 Let c be an RV with mean c η and ( ) t = X c h { } { } ( ) c E t E η η = = = X c . For a specific sample ( ) t c = X 1 (constant)a a ( ) ( ) T ξ ξ = η c (i.e.D1 5 determined by ξ )a ( ) T ξ η η h ∴ ( ) t X h meanergodic Ex. 12.2 Given two meanergodic processes 1 ( ) t X h 2 ( ) t X h mean a 1 η 2 η form 1 2 ( ) ( ) ( ) t t t = + X X cX h c h RV independent of 2 ( ) t X h = c h 1a probabilities a 0.5 a { } { } { } { } { } 1 2 1 2 1 2 ( ) ( ) ( ) ( ) 0.5 E t E t E t E E t η η η = + = + = + X X cX c X h for a specific samplea if ( ) ξ = c h 1 ( ) ( ) t t = X X h 1 T η η as T E E E ( ) 1 ξ = c h 1 2 ( ) ( ) ( ) t t t = + X X X h 1 2 T η η = + η h ( ) t X h mean ergodic 2 T σ of T η for an RP ( ) t X D1 form 1 ( ) ( ) 2 t T t T t d T α α + = w X h moving average of ( ) t X h (0) T = η w h h ( ) t w h ( ) t X h LTI system with impulse response a 2 1   ( ) 2 0 o.w. t T h t T E E E = ∴ 2 2 1   ( ) ( )(1 ) 2 2 T T C C d T T α τ τ α α = ww a ( ) t w autocovariance a ( ) C τ ( ) t X autocovariance and 1   (1 ) ( ) ( ) 2 2 t h t h t T T = 2 ( (0)) (0) T Var C σ = = ww w Q ( ) ( ) C C α α = ∴ 2 2 2 1   ( )(1 ) 2 2 T T T C d T T α σ α α = 2 1 ( )(1 ) 2 T C d T T α α α = (124) a ( ) t X is meanergodic iff 2 1 ( )(1 ) 2 T T C d T T α α α V (125) T σ Tchebycheff’s inequality a confidence interval for the estimate T η of η η 10 T T σ η É 5 0.99a ∴ T E E T σ η = T η η ← 5 Ex. 123 If   ( ) c C qe τ τ = 2 2 2 1 (1 ) (1 ) 2 2 cT T c T q q e e d T T cT cT τ τ σ τ = = ∴ 2 T σ as T E a ( ) t X is meanergodic If 1 T c ?...
View
Full Document
 Fall '08
 SinHorngChen

Click to edit the document details