Recursive Least Squares Estimation (RLS) Algorithm

MATLAB 107K 242 views 0 downloads 1 credits
Tags:

Resource Overview

Efficient implementation for system identification using Recursive Least Squares (RLS) estimation. This algorithm is crucial for determining model order and parameter values. The program implements RLS to compute model order estimates and relevant parameter values with adaptive weight updates.

Detailed Documentation

This document discusses the Recursive Least Squares Estimation (RLS) algorithm, which plays a vital role in system identification by estimating model order and determining associated parameter values. The algorithm finds extensive applications in signal processing and communication systems, particularly in adaptive filters. Implementation typically involves programming a recursive formulation that updates parameter estimates with each new data sample, avoiding matrix inversion through the matrix inversion lemma. The program computes model order estimates and related parameters using forgetting factors to handle time-varying systems. Notably, RLS can be extended to identify nonlinear systems and track time-varying systems, making it a powerful algorithmic tool. Key implementation components include covariance matrix initialization, gain vector calculation, and parameter update equations that maintain computational efficiency.

You May Also Like