Hidden Markov Model And Its Application

Hidden Markov models whose shortened form is HMMs are statistic models. It consists of a hidden Markov process and an observation process. In 1970s, these models were applied into automatic speech recognition by Jenik and some other people. They were developed to be one of the most efficient techniques. Now, they were widely used in gene relative analysis and gene recognition, character recognition, image processing and target tracking etc. There are three problems needed to be solved by using Hidden Markov models ,which is training, decoding and recognizing. The answers for those three problems consist of the theory of hidden Markov models. Parameter estimation is the core problem of the training process.Estimation algorithm of 1-D stationary HMMs and correlation properties were discussed. In addition 2-D HMMs decoding algorithm was also discussed. Based on the 1-D discrete HMMs, MLE of one kind of HMM swhich is called Markov-modulated Poisson processes was given by changing continued-time HMMs into discrete-time HMMs. Under the weak condition, MLE algorithm of one kind of HMMs which is called Markov-modulated Poisson processes was given when its state is 2. Such processes have been proposed for modeling traffic streams in complex telecommunication networks. Based on single optimal decoding algorithm, local optimal decoding algorithm which was applied into off-line handwriting character recognition was given. The decoding algorithm of pseudo-2-D HMMs of was also discussed.