Title :
Representation and estimation for harmonizable type processes
Author_Institution :
Dept. of Math., California Univ., Riverside, CA, USA
Abstract :
The harmonizable type class contains Karhunen processes and Cramer classes, and if it has some (complex) measure or bimeasure /spl mu/, then the process is called (strongly) harmonizable or weakly harmonizable respectively. Finally if /spl mu/ concentrates on the diagonal so that it becomes a positive bounded measure /spl mu/~, the process is (weakly or Khintchine) stationary, and /spl mu/~is called the spectral distribution function, and in the general case, /spl mu/ is termed a bispectral distribution. The harmonizable-type processes are considered in this paper.
Keywords :
integral equations; parameter estimation; probability; signal representation; spectral analysis; stochastic processes; Cramer classes; Karhunen processes; Khintchine stationary processes; bispectral distribution; estimation; integrals; representation; spectral distribution function; strongly harmonizable processes; weakly harmonizable processes; Convergence; Fourier transforms; Mathematics; Stochastic processes;
Conference_Titel :
Signals, Systems and Computers, 2002. Conference Record of the Thirty-Sixth Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-7803-7576-9
DOI :
10.1109/ACSSC.2002.1197040