On the performance of variable forgetting factor recursive least-squares algorithms

Camelia Elisei-Iliescu; Constantin Paleologu; Răzvan Tamaş

doi:10.1117/12.2242960

14 December 2016 On the performance of variable forgetting factor recursive least-squares algorithms

Camelia Elisei-Iliescu, Constantin Paleologu, Răzvan Tamaş

Proceedings Volume 10010, Advanced Topics in Optoelectronics, Microelectronics, and Nanotechnologies VIII; 1001023 (2016) https://doi.org/10.1117/12.2242960
Event: Advanced Topics in Optoelectronics, Microelectronics, and Nanotechnologies 2016, 2016, Constanta, Romania

Abstract

The recursive least-squares (RLS) is a very popular adaptive algorithm, which is widely used in many system identification problems. The parameter that crucially influences the performance of the RLS algorithm is the forgetting factor. The value of this parameter leads to a compromise between tracking, misadjustment, and stability. In this paper, we present some insights on the performance of variable forgetting factor RLS (VFF-RLS) algorithms, in the context of system identification. Besides the classical RLS algorithm, we mainly focus on two recently proposed VFF-RLS algorithms. The novelty of the experimental setup is that we use real-world signals provided by Romanian Air Traffic Services Administration, i.e., voice and noise signals corresponding to real communication channels. In this context, the Air Traffic Control (ATC) communication represents a challenging task, usually involving non-stationary environments and stability issues.

Citation Download Citation

Camelia Elisei-Iliescu, Constantin Paleologu, and Răzvan Tamaş "On the performance of variable forgetting factor recursive least-squares algorithms", Proc. SPIE 10010, Advanced Topics in Optoelectronics, Microelectronics, and Nanotechnologies VIII, 1001023 (14 December 2016); https://doi.org/10.1117/12.2242960

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available