Estimation quality of a weighted least-square parameter estimation method based on binay observations

Abstract : In this paper, we investigate the quality of a weighted leastsquare (WLS) parameter estimation method based on binary observations when only a finite number of samples are available. An upper bound of the number of samples that are necessary for identifying system with a given accuracy is theoretically derived. The accuracy is defined in the sense of correlation coefficient between the system parameters and our estimated system parameters. Furthermore, we compare theoretical results with simulations in order to study the validity of the results practically.
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal-supelec.archives-ouvertes.fr/hal-00516976
Contributor : Karine El Rassi <>
Submitted on : Monday, September 13, 2010 - 12:00:44 PM
Last modification on : Monday, February 25, 2019 - 4:34:17 PM
Long-term archiving on : Tuesday, December 14, 2010 - 2:42:11 AM

File

Eusipco2010_final_camera_ready...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00516976, version 1

Collections

Citation

Jérôme Juillard, Kian Jafaridinani, Eric Colinet. Estimation quality of a weighted least-square parameter estimation method based on binay observations. European Signal Processing Conference (EUSIPCO'10), Aug 2010, Aalborg, Denmark. pp.1043-1047. ⟨hal-00516976⟩

Share

Metrics

Record views

309

Files downloads

130