| Review: |
Hidden Markov models (HMMs) are discrete-state, discrete-time, stochastic dynamical systems. At a symposium on hidden Markov models in 2001, someone asked for a simple reference book to explain the ideas and algorithms for applying HMMs. This book is the result of that request. It is an introduction for undergraduates working in probability, linear algebra and differential equations in engineering, mathematics and science. It introduces the basic ideas of HMMs and the algorithms for using them and explains the derivations of the algorithms with sufficient supporting theory to enable readers to develop their own variants. There are chapters on: basic algorithms; variants and generalisations; continuous states and observations and Kalman filtering; performance bounds and a toy problem; and identifying obstructive sleep apnoea by ECGs. |