ADSP UNIT 2

A stationary random process is one in which:

• Mean and variance change with time

• Autocorrelation depends only on the time difference

• Power spectrum is not defined

• None of the above

The autocorrelation function R_x(τ) of a wide-sense stationary (WSS) process is always:

• Odd and real

• Even and real

• Complex and odd

• None of the above

The power spectral density (PSD) of a random process is the:

• Laplace transform of the autocorrelation

• Fourier transform of the autocorrelation

• Z-transform of the autocorrelation

• Cosine transform of the autocorrelation

The AR(1) process is defined as:

• x[n]=a1 x[n−1]+w[n]

• x[n]=b0 w[n]+w[n−1]

• x[n]=a1 x[n−1]+b1 w[n]

• None of the above

The MA(q) process can be generated by:

• Passing white noise through an all-pole filter

• Passing white noise through an all-zero filter

• Adding a deterministic signal to white noise

• Differencing a stationary process

The ARMA(p,q) model combines:

• Autoregressive process only

• Moving average process only

• Both AR(p) and MA(q) processes

• Neither

The autocorrelation function of a pure MA process is:

• Infinite in length

• Finite in length

• Always zero

• None of the above

For an AR process, the autocorrelation satisfies a:

• Nonlinear relation

• Yule-Walker equation

• Fourier equation

• Differential equation