Recursive Errors-In-Variables Approach for AR parameter estimation from noisy observation. Application to radar Sea Clutter Rejection

Abstract : AR modeling is used in a wide range of applications from speech processing to Rayleigh fading channel simulation. When the observations are disturbed by an additive white noise, the standard Least Squares estimation of the AR parameters is biased. Some authors of this paper recently reformulated this problem as an errors-in-variables (EIV) issue and proposed an off-line solution, which outperforms other existing methods. Nevertheless, its computation cost may be high. In this paper, we present a blind recursive EIV method that can be implemented for real-time applications. It has the advantage of converging faster than the noise-compensated LMS based solutions. In addition, unlike EKF or Sigma Point Kalman filter, it does not require a priori knowledge such as the variances of the driving process and the additive noise. The approach is first tested with synthetic data; then, its relevance is illustrated in the field of radar sea clutter rejection.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-00349756
Contributor : Eric Grivel <>
Submitted on : Sunday, January 4, 2009 - 11:21:33 AM
Last modification on : Thursday, January 11, 2018 - 6:21:07 AM

Identifiers

  • HAL Id : hal-00349756, version 1

Citation

Julien Petitjean, Roberto Diversi, Eric Grivel, Roberto Guidorzi, Patrick Roussilhe. Recursive Errors-In-Variables Approach for AR parameter estimation from noisy observation. Application to radar Sea Clutter Rejection. ICASSP 09, Apr 2009, Tai pei, Taiwan. pp. ⟨hal-00349756⟩

Share

Metrics

Record views

63