Title :
Power spectrum estimation with randomly spaced correlation samples
Author :
Ning, Taikang ; Nikias, Chirysostomos L.
Author_Institution :
Dept. of Eng. & Comput. Sci., Trinity Coll., Hartford, CT, USA
fDate :
6/1/1990 12:00:00 AM
Abstract :
Three algorithms are developed to reproduce uniformly spaced correlation measurements for power spectrum estimation. Some general relations between the randomly spaced correlation samples and the power spectrum are derived. The details of the three algorithms that can be used to reproduce uniformly spaced correlation measurements are included. Simulation examples using sinusoidal signals are given to examine the adverse effects due to the correlation randomness as well as to demonstrate improvements achieved by the suggested algorithms
Keywords :
correlation methods; estimation theory; spectral analysis; correlation randomness; ensemble average algorithm; extended region algorithm; minimum variance algorithm; power spectrum estimation; randomly spaced correlation samples; Algorithm design and analysis; Displacement measurement; Frequency estimation; Power measurement; Random variables; Sampling methods; Signal design; Signal processing algorithms; Signal sampling; Spectral analysis;
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
