Informatics and Machine Learning

6Analysis of Sequential Data Using HMMs

Generalized Hidden Markov model (HMM) methods are described for both signal feature extraction and structure identification. The generalized HMMs described also enable a new form of carrier‐based communication, where the carrier is stationary but not periodic (further details in Chapter 12). HMM‐with‐binned‐duration, and meta‐HMM generalizations, shown in Chapter 7, enable practical stochastic carrier wave encoding/decoding, where the generalized HMM methods have generalized Viterbi algorithms with all of the inherent benefits of an efficient dynamic programming implementation, as well as Martingale convergence properties when used for filtering and robust feature extraction.

Numerous prior book, journal, and patent publications by the author are drawn upon extensively throughout the text [1–68]. Almost all of the journal publications are open access. These publications can typically be found online at either the author’s personal website (www.meta‐logos.com) or with one of the following online publishers: www.m‐hikari.com or bmcbioinformatics.biomedcentral.com.

6.1 Hidden Markov Models (HMMs)

6.1.1 Background and Role in Stochastic Sequential Analysis (SSA)

HMMs have been used in speech recognition since the 1970s [128], and in bioinformatics since the 1990s [134], and have an extensive, and growing, breadth of applications in other areas (especially as more computational resources become available). Other areas of HMM application ...