Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Power Spectrum
Estimation PRESENTED BY:-
Duha Hassan
Power Spectrum of signal 2
ESTIMATION ?
To estimate the spectral characteristics of signal characterized as
random processes.
To estimation of spectra in frequency domain when signals are
random in nature.
Power Spectral Estimation method is to obtain an approximate
estimation of the power spectral density of a given real random
process .
Introduction to Estimation 5
Estimation
parametric or model-based methods teqniques are based on the use of models for
the data
Let assume that the data is the output for a linear time invariant system with
frequency response in response to a wait noise input sequins
The assumption that the input has a flat spectrum implies that the power spectrum
of the model output is shaped entirely by the frequency response of the model
The input–output relation of a generalised discrete linear time-invariant 9
model is given by
where x(m) is the model output, e(m) is the input, and the ak and bk are the
parameters of the model. is known as an auto-regressivemoving-average (ARMA)
model.
The system function H(z) of the discrete linear time-invariant model of Equation is
given by
where 1/A(z) and B(z) are the autoregressive and moving-average parts of H(z)
respectively.
10
The power spectrum of the signal x(m) is given as the product of the power
spectrum of the input signal and the squared magnitude frequency response of
the model:
where H(f) is the frequency response of the model and PEE(f) is the input
power spectrum.
model based approach
Once the model is selected, the next step is to estimate the model
parameters from the given data.
The final step is the estimate the power spectrum by incorporating
the estimated parameters into the parametric form for the spectrum.
Example: An ARMA(p , q) model with ap(k) and bq(k) estimated ,
the spectrum estimate would be
Power spectrum estimation (cont.) 14
u(m): is a zero-mean,
unit-variance random signal,
G: a gain term
is the square root of the variance of e(m):
24
One of the most important consideration is the choice of the number 25
of terms in the AR model, this is known as its order p.
If a model with too low an order,
We obtain a highly smoothed spectrum.
If a model with too high an order,
There is risk of introducing spurious low-level peaks in the
spectrum.
The power spectrum of an autoregressive
process is given by
the power spectrum of a moving-average process may be obtained directly from the Fourier
transform of the autocorrelation function as
29
Determine the mean and the autocorrelation of the sequence jc (n) generated by the
MA(2) process described by the difference equation
x(n) = w(n) - 2 w(n - 1) + w(n — 2) 30
where w(n) is a white noise process with variance a*.
Autoreg-ressive Moving Average 31
(ARMA)
The ARMA, or pole–zero, The relationship between the ARMA parameters and the
autocorrelation sequence can be obtained by multiplying both sides of Equation by x(m–j)
and taking the expectation
It consists of two parts, an autoregressive (AR) part and a moving average (MA) part.