ScholarLens Back to search
2000•Communication in Statistics- Theory and MethodsRequires access

A note on maximum entropy spectrum estimation

Dylan Shi, Junjie Song, Chi Wang

Open publisher page 0 citations

Abstract

This article studies the maximum entropy spectrum estimation. After a bnei discussion on iiow co select ciic appropriate constraints and tiie objec¬tive functions, we decide to choose the constraints containing only the first four sample moments and, consequently, to employ the second order spectral entropy as the objective function. The resulting (maximum entropy) spectral estimate is the power spectral density of an ARMA sequence. Examples for comparing our proposal with the traditional maximum entropy spectral estimate follow at the end.

About this research paper

What this paper is about

This article studies the maximum entropy spectrum estimation. After a bnei discussion on iiow co select ciic appropriate constraints and tiie objec¬tive functions, we decide to choose the constraints containing only the first four sample moments and, consequently, to employ the second order spectral entropy as the objective function. The resulting (maximum entropy) spectral estimate is the power spectral density of an ARMA sequence. Examples for comparing our proposal with the traditional maximum entropy spectral estimate follow at the end.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This article studies the maximum entropy spectrum estimation. After a bnei discussion on iiow co select ciic appropriate constraints and tiie objec¬tive functions, we decide to choose the constraints containing only the first four sample moments and, consequently, to employ the second order spectral entropy as the objective function. The resulting (maximum entropy) spectral estimate is the power spectral density of an ARMA sequence. Examples for comparing our proposal with the traditional maximum entropy spectral estimate follow at the end.

Key concepts: Maximum entropy spectral estimation, Principle of maximum entropy, Spectral density, Mathematics, Entropy (arrow of time), Spectral density estimation, Binary entropy function, Sample entropy

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
A note on maximum entropy spectrum estimation — Research Paper | ScholarLens