Adjustment of Observation Probabilities during the Lifetime of Viterbi Algorithm in Unstable Environments
Nader Rezazadeh 1 and
Omid Sojodishijani 2
1. Department of Computer, Science and Research Branch, Islamic Azad University Qazvin, Iran
2. Department of Computer and Information Technology Engineering-Qazvin Branch, Islamic Azad University, Qazvin, Iran
2. Department of Computer and Information Technology Engineering-Qazvin Branch, Islamic Azad University, Qazvin, Iran
Abstract—Typically in learning and determining the parameters in various applications of hidden Markov model, like the decoding problem solving in Viterbi algorithm, statistics such as the average are used. In this context, to
obtain the average, the stable hypotheses for the data generation process are considered. While in an unstable environment, model parameters such as the value of probability of observing event parameter generated by the state, changes directly between the successive events. For this purpose in this article an adjuster parameter of event probability has been provided in order to adjust and change the parameters after each event during the lifetime of Viterbi algorithm. Test results on the real data sets show the superior performance of the proposed method in terms of accuracy than the other methods.
Index Terms—Statistical Adjustment, Hidden Markov Model, Viterbi algorithm
Cite: Nader Rezazadeh and Omid Sojodishijani, "Adjustment of Observation Probabilities during the Lifetime of Viterbi Algorithm in Unstable Environments," International Journal of Signal Processing Systems, Vol. 2, No. 1, pp. 23-30, June 2014. doi: 10.12720/ijsps.2.1.23-30
Cite: Nader Rezazadeh and Omid Sojodishijani, "Adjustment of Observation Probabilities during the Lifetime of Viterbi Algorithm in Unstable Environments," International Journal of Signal Processing Systems, Vol. 2, No. 1, pp. 23-30, June 2014. doi: 10.12720/ijsps.2.1.23-30