Механизми за квалитет на сервисот и вертикален multi-homing за 5g мобилни и безжични мрежи

Abstract

Inspired by the extremely fast development of the telecommunications’ science, especially from the tremendous development in the field of wireless mobile technology, in this doctoral dissertation are presented novel scientific researches, frameworks, analysis, comparisons and mechanisms which are related to the intelligent and excellent QoS provisioning in 5G mobile and wireless networks. Considering the novel researches and papers, the nowadays telecommunications are at the point where the top 5G scientific researches are going, moreover now is the time when the first 5G prototypes and probes are launched, and the beginning of the 4G standardization process. Moreover, this is the moment when ITU is submitting the IMT-2020 specifications, together with the technical performances and requirements. In this doctoral dissertation an advanced QoS provisioning frameworks with vertical multi-homing and multi-streaming features for mobile and fixed terminals with radio network aggregation capability in next 5G mobile and wireless broadband networks is presented. The proposed 5G QoS framework is usercentric, targeted to always-on connectivity, and is leading to high performance utility networks with high level of QoS provisioning for any given multimedia service (voice, audio and video), high average throughput, seamless handovers, session continuity, low delay, minimal queue backlog, multi-RAT and mobile broadband capabilities, at the same time achieving stochastic queuing network stability. Undoubtedly, the common “heart” for all of existing and future Radio Access Technologies (RATs), which connect them all as a unifying technology, is IP. That is the main reason why the main changes and improvements we are proposing in the network layer within the future 5G mobile and fixed terminals. Also, each RAT will be responsible for handling user-mobility, while the mobile terminal will make the final choice among different existing (available) RATs for a given service or for a traffic flow, base on its smart module (here named Advanced QoS-based User-centric (RAT-) Aggregation (AQUA) module) within. Moreover, the proposed algorithms implemented within the mobile terminals and the proxy servers, having vertical multi-homing and multi-streaming capabilities are able to handle simultaneously multiple multimedia sources via multiple RAT (including any present and future RAT) interfaces and chousing the most appropriate RATs for a given multimedia service. Proposed 5G framework is using different optimization algorithms, from the optimization algorithm database. We have presented two optimization paradigms; the first is using linear programming or genetic algorithms, for solving the optimization problem, where several key QoS parameters (average system throughput and multimedia access probability) for different radio network conditions were considered. The second optimization problem is using Lyapunov drift-plus-penalty optimization technique that provides a methodology for designing algorithm to maximize the time average throughput – achieving mobile broadband capabilities and stabilizes the queuing, regarding the minimal queue delay and minimal queue backlog. The performance of our proposed 5G terminals with AQUA module is evaluated using simulations and performance analysis with multimedia traffic in different heterogeneous mobile and wireless scenarios with coexistence of multiple RATs, such as 3G, 4G as well as future 5G radio access networks. The analysis has shown significant QoS performances regarding the achievable mobile broadband throughput, minimal delays, maximal multimedia access probability ratio and other superior results regarding the key QoS parameters in heterogeneous mobile and wireless environments, achieving optimal utilization of network resources due to the optimal traffic load scheduling. Above all, the given framework for 5G terminals can be one of the feasible solutions for use in nowadays and future 5G mobile and wireless network architectures.