Обеспечение качества обслуживания в условиях согласованного решения задач управления трафиком в телекоммуникационной сети

Предложена потоковая модель управления трафиком в телекоммуникационной сети, в рамках которой обеспечивается согласованное решение следующих сетевых задач: многопутевая маршрутизация; профилирование трафика; динамическое распределение пропускной способности каналов связи между потоками пользователей. В рамках полученных решений обеспечиваются требуемые значения таких ключевых показателей качества обслуживания как скорость передачи пакетов, средняя задержка и вероятность потерь пакетов.A flow model for traffic management in the telecommunications network (TCN) is proposed. This model ensures consistent solution of the following network problems: multipath routing; traffic policing; dynamic allocation of links bandwidth between threads of users. Within the obtained solutions the required values of the key indicators of quality of service such as packet rate, average delay and packet loss probability are provided. Similar nature of solutions was possible to implement using an updated version of the conservation of flow conditions on the routers in the network and TCN as a whole, thereby expanding the range of solvable tasks on the traffic profil-ing processes. This ultimately led to modification of the conditions for ensuring the quality of service that were obtained using the appa-ratus of the network tensor analyzing. The results obtained are adapted to service of SMTP / TCP, IP, FTP, TCP, HTTP flows. The efficiency of the proposed approach is demonstrated by a numerical example. In general, the use of solutions proposed in the article makes it possible to reduce failures in service on the average from 60 to 80% and to reduce the allowable value of multipath delay of packets without loss of intensity of the input stream to 15-35%. In terms of practical implementation of the developed model it is worth noting that the highest efficiency of its implementation is provided in the multiflow traffic conditions of different service classes with differenti-ated QoS-requirements for a variety of parameters: packet rate, average delay and packet loss probability