System Identification via Polynomial Transformation Method

TL;DR

This paper introduces a polynomial approximation-based method for system identification that accurately extracts system poles from impulse response data, outperforming existing methods under various sampling and noise conditions.

Contribution

The paper presents a novel minimum-variance polynomial approximation technique for system pole extraction, demonstrating its robustness and superiority over existing methods in uniform sampling scenarios.

Findings

Outperforms existing methods in pole extraction accuracy

Handles general sampling and noise conditions

Shows statistical superiority in uniform sampling cases

Abstract

We propose a method based on minimum-variance polynomial approximation to extract system poles from a data set of samples of the impulse response of a linear system. The method is capable of handling the problem under general conditions of sampling and noise characteristics. The superiority of the proposed method is demonstrated by statistical comparison of its performance with the performances of two exiting methods in the special case of uniform sampling.