Fundamentals of Statistical Signal Processing, Volume III

Chapter 15. Case Studies - Spectral Estimation Problem

15.1. Introduction

In this chapter we discuss the use of spectral analysis to obtain information about the data of interest. Since spectral analysis is a preliminary data analysis tool, we do not investigate in detail an actual algorithm but instead use our analysis to formulate a noise power spectral density (PSD) model. Other necessary information about the noise such as its probability density function (PDF) is not studied. We do describe, however, the use of a matched filter to improve the data quality and why it ultimately proved unsuccessful for the heart rate monitor problem. In contrast to the previous two chapters in which we followed the flow chart of Figure 2.1 to design an actual ...