RUDN University mathematicians have proposed a model for calculating the optimal parameters of a 5G network for virtual reality applications. It is simpler than analogues and at the same time has the same efficiency. The results are published in Future Internet.

5G technologies are designed to support applications that need high speeds. To do this, various techniques are used, including multicasting. This is a routing scheme in which a station sends data not to one user (unicast) and not to all at once (broadcasting), but to a dedicated group of users. However, the use of multicasting can negatively affect network performance, since the data transfer rate in a group of users is always limited by the speed of the subscriber device with the weakest signal. RUDN University mathematicians proposed a model based on the example of virtual reality applications, in which both multicast and unicast transmission schemes are combined. It allows one to estimate how many base stations are needed for a given quality of services provided. The algorithm makes it possible to obtain estimates no worse than alternative options using simulation modelling, but at the same time it is simpler, since it is based on an analytical approach.

“Most of the studies performed so far for multi-layer multicast/unicast services concentrated on optimizing an already provisioned deployment for given traffic conditions. However, for network operators, the question of estimating the required density base stations for a given stochastic traffic load in a given area is equally important," said Daria Ostrikova, PhD, Deputy Director of Applied Mathematics & Communications Technology Institute of the RUDN University.

Mathematicians have proposed a model with several base stations. Each of them has three antennas covering 120°. Users of VR devices move within the service stations randomly according to the Poisson point process model, sometimes blocking each other's line of sight with the base station. The network in the proposed model operates on two levels. At the basic level, multicasting is used, and unicast delivery is used to improve the quality of communication. RUDN mathematicians calculated the parameters of such a system and proposed an algorithm for grouping users for multicasting and calculating the required density of base stations.

User groups for multicasting are created based on the maximum allowable beam angle at which the signal retains at least half the power at the required distance from the station. All users who fall into the sector specified in this way become part of the same group. Sectors that no user has entered are considered inactive. The calculation of the optimal number of base stations is based on an iterative principle — at each new step, the density of base stations increases until the probability of signal loss reaches the required minimum. In a numerical experiment, the new algorithm turned out to be no less effective than other schemes, but at the same time much simpler.

“Our numerical results show that the proposed grouping scheme, in which the number of groups and their configurations can be explicitly calculated, makes it possible to evaluate the performance of millimeter communication systems no worse than when using simulation algorithms. Our work provides a simple yet efficient algorithm for estimating the density of base stations needed to support a given density of VR devices and certain service quality parameters,” said Daria Ostrikova, PhD, Deputy Director of Applied Mathematics & Communications Technology Institute of the RUDN University.

Journal Link: Future Internet 2021, 13(7), 185