We develop an efficient algorithm that calculates the maximal state sequences in an imprecise hidden Markov model by means of coherent lower previsions. Initial results show that this algorithm is able to robustify the inferences made by a classical precise model. Keywords— Imprecise hidden Markov model, coherent lower prevision, epistemic irrelevance, maximal state sequence, Viterbi algorithm.

The classical Viterbi algorithm (ML sequence estimation) can be computed using a forward-backward structure, similar to that of the classical hidden Markov model forward-backward algorithm (MAP state estimation). This similarity is exploited to develop a hybrid algorithm which provides a mathematical connection between ML sequence estimation and MAP state estimation.

This paper, based on the Viterbi algorithm, selects the most likely combination of different wording from a variety of scenarios. In this regard, the Bi-gram and Unigram tags of each word, based on the letters forming the words, as well as the bigram and unigram labels After the breakdown into the composition or moment of transition from the decomposition to the combination obtained from th...

Abstract—The free distance of (n, k, l) convolutional codes has some connection with the memory length, which depends on not only l but also on k. To efficiently obtain a large memory length, we have constructed a new class of (2k, k, l) convolutional codes by (2k, k) block codes and (2, 1, l) convolutional codes, and its encoder and generation function are also given in this paper. With the he...

The long term tracking of sparse local features in an image is important for many applications including camera calibration for stereo applications, camera or global motion estimation and people surveillance. The majority of existing tracking frameworks are based on some kind of prediction/correction idea e.g. KLT and Particle Filters. However, given a careful selection of interest points throu...

The principle of Per-Survivor Processing (PSP) provides a general framework for the approximation of likelihood-based data detection (search) algorithms whenever the presence of unknown quantities prevents the precise use of the classical Viterbi Algorithm (VA). The PSP concept is based on the realization that decision-directed estimation of the relevant unknown parameters should be embedded in...

Combined linear-Viterbi equalizer (CLVE) is a class of receivers employing a linear pre-filter before the Viterbi algorithm (VA) in order to shape the original channel impulse response to some shorter desired impulse response which allows a lower complexity VA. In this paper, we present a new approach for optimizing CLVEs by making use of the Genetic Algorithm to minimize the effective mean squ...

In this report, I have studied a simplification, suggested by Viterbi [1], to the BCJR algorithm for the bitwise MAP decoder for convolutional and turbo codes. The simplifications involve reduction in both computational complexity and memory requirements, using, first, a good approximation to the metric update rules for BCJR algorithm and, second, a scheduling algorithm to reduce memory require...