Informatics and Applications2020, Volume 14, Issue 1, pp 17-23
NUMERICAL SCHEMES OF MARKOV JUMP PROCESS FILTERING GIVEN DISCRETIZED OBSERVATIONS II: ADDITIVE NOISE CASE
AbstractThe note is a sequel of investigations initialized in the article Borisov, A. 2019. Numerical schemes of Markov jump process filtering given discretized observations I: Accuracy characteristics. Inform. Appl. 13(4):68-75.
The basis is the accuracy characteristics of the approximated solution of the filtering problem for the state of homogeneous Markov jump processes given the continuous indirect noisy observations. The paper presents a number of the algorithms of their numerical realization together with the comparative analysis. The class of observation systems under investigation is bounded by ones with additive observation noises. This presumes that the observation noise intensity is a nonrandom constant. To construct the approximation, the authors use the left and midpoint rectangle rule of the accuracy order 2 and 3, respectively, and the Gaussian quadrature of the order 5. Finally, the presented numerical schemes have the accuracy of the order 1 /2, 1, and 2.
[+] References (9)
[+] About this article
TitleNUMERICAL SCHEMES OF MARKOV JUMP PROCESS FILTERING GIVEN DISCRETIZED OBSERVATIONS II: ADDITIVE NOISE CASE
JournalInformatics and Applications
2020, Volume 14, Issue 1, pp 17-23
PublisherInstitute of Informatics Problems, Russian Academy of Sciences
Key wordsMarkov jump process; optimal filtering; additive and multiplicative observation noises; stochastic differential equation; analytical and numerical approximation
AuthorsA. V. Borisov
Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation