Title :
Blind identifiability of a quadratic stochastic system
Author :
Bondon, Pascal ; Krob, Michel
Author_Institution :
Lab. des Signaux et Syst., Ecole Superieure d´´Electr., Gif-sur-Yvette, France
fDate :
1/1/1995 12:00:00 AM
Abstract :
Quadratic systems appear in various fields of signal processing, in particular in detection and estimation. In this paper, we establish some conditions to determine the order and identify the coefficients of any discrete and finite extent quadratic system driven by a sequence of independent and identically distributed random variables. The input sequence is assumed unobservable and the conditions are based on some properties of the output cumulant functions up to the third order. Necessary and sufficient conditions ensuring the determination of the order are given. A sufficient condition of uniqueness to recover the parameters is presented. Lastly, some examples are given
Keywords :
higher order statistics; parameter estimation; signal detection; signal processing; stochastic systems; blind identifiability; coefficients; discrete quadratic system; finite extent quadratic system; independent identically distributed random variables; input sequence; necessary conditions; order determination; output cumulant functions; parameters recovery; quadratic stochastic system; signal detection; signal estimation; signal processing; sufficient conditions; Bonding; Gaussian noise; Higher order statistics; Linear systems; Nonlinear systems; Predictive models; Random variables; Signal processing; Stochastic systems; Sufficient conditions;
Journal_Title :
Information Theory, IEEE Transactions on
