Title :
The gradient type LS algorithm for adaptive spectrum estimation
Author :
Kim, Daehoon ; Alexander, Winser E.
Author_Institution :
Dept. of Electr. & Comput. Eng., North Carolina State Univ., Raleigh, NC, USA
Abstract :
The authors develop a novel adaptive implementation of V.F. Pisarenko´s (1973) high-resolution spectrum estimation method using the least-squares (LS) criterion. This algorithm is compared to P. Thompson´s (1979) LMS algorithm and the LMS algorithm of V.U. Reddy et al. (1982). It is also shown that the proposed algorithm has no bias. A multiprocessor architecture for implementing the algorithm is presented.<>
Keywords :
computerised signal processing; convergence of numerical methods; least squares approximations; parallel algorithms; parallel architectures; spectral analysis; LS algorithm; adaptive spectrum estimation; computerised signal processing; convergence analysis; gradient type; high-resolution; least squares criterion; multiprocessor architecture; parallel processing; Autocorrelation; Eigenvalues and eigenfunctions; Frequency estimation; Lagrangian functions; Least squares approximation; Least squares methods; Random processes; Signal processing algorithms; Spectral analysis; White noise;
Conference_Titel :
Circuits and Systems, 1988., IEEE International Symposium on
Conference_Location :
Espoo, Finland
DOI :
10.1109/ISCAS.1988.15036
