Open AccessAsymptotic Analysis of the MMРР|M|1 Retrial Queue with Negative Calls under the Heavy Load Condition
Author(s) -
Ekaterina Fedorova,
Anatoly A. Nazarov,
Mais Farkhadov
Publication year - 2020
Publication title -
izvestiâ saratovskogo universiteta. novaâ seriâ. seriâ matematika. mehanika. informatika
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.255
H-Index - 4
eISSN - 2541-9005
pISSN - 1816-9791
DOI - 10.18500/1816-9791-2020-20-4-534-547
Subject(s) - poisson distribution , exponential distribution , retrial queue , queueing system , computer science , queue , orbit (dynamics) , exponential function , mathematics , real time computing , computer network , mathematical analysis , statistics , engineering , aerospace engineering
In the paper, a single-server retrial queueing system with MMPP arrivals and an exponential law of the service time is studied. Unserviced calls go to an orbit and stay there during random time distributed exponentially, they access to the server according to a random multiple access protocol. In the system, a Poisson process of negative calls arrives, which delete servicing positive calls. The method of the asymptotic analysis under the heavy load condition for the system studying is proposed. It is proved that the asymptotic characteristic function of a number of calls on the orbit has the gamma distribution with the obtained parameters. The value of the system capacity is obtained, so, the condition of the system stationary mode is found. The results of a numerical comparison of the asymptotic distribution and the distribution obtained by simulation are presented. Conclusions about the method applicability area are made.