An SDN-approach for QoE management of multimedia services using resource allocation

Abstract

Future networks will be accompanied by new heterogeneous requirements in terms of end-users Quality of Experience (QoE) due to the increasing number of application scenarios being deployed. Network softwarization technologies such as Software Defined Networks (SDNs) and Network Function Virtualization (NFV) promise to provide these capabilities. In this paper, a novel QoE-driven resource allocation mechanism is proposed to dynamically assign tasks to virtual network nodes in order to achieve an optimized end-to-end quality. The aim is to find the best combination of network node functions that can provide an optimized level of QoE to the end users though node cooperation. The service in question is divided in tasks and the neighbor nodes negotiate the assignment of these considering the final quality. In the paper we specifically focus on the video streaming service. We also show that the agility provided by SDN/NFV is a key factor for enhancing video quality, resource allocation and QoE management in future networks. Preliminary results based on the Mininet network emulator and the OpenDaylight controller have shown that our approach can significantly improve the quality of a transmitted video by selecting the best path with normalized QoS values.

Adapting reinforcement learning for multimedia transmission on SDN

“…It maximized QoE considering link capacities, delay, network topology, and service utility functions. Finally, Grigoriou et al 19 presented an SDN-based resource management mechanism for multimedia services that optimizes the QoE of end users. Therefore, video quality may be degraded while audio quality is maintained high enough when the resource availability of the network decreases.…”

“…Results showed an improvement in terms of QoS of the proposed framework compared with the best-effort approach. Finally, Grigoriou et al 19 presented an SDN-based resource management mechanism for multimedia services that optimizes the QoE of end users. To do so, the service was divided into different tasks assigned to the neighbor nodes taking into consideration the final quality.…”

Optimized Energy Aware Resource Allocation Algorithm Using Software Defined Network Technology

“…QoS and QoE are two important factors when considering end user satisfaction and when rating DC capacity. QoS is determined by diminishing packet loss, delay and high latency levels [8], [23]. When subjected to any of those factors, the service offered to the end user subject to deterioration, due to the distorted or damaged nature of information received [23].…”

“…The exponential increase in Internet traffic calls for the construction of more DCs to ensure that massive storage capacities and fast processing speeds are guaranteed. However, building more DCs will increase the operational costs due to increased energy consumption [6]- [8]. Extensive research has been conducted to ensure energy efficient operation of DCs [7], [9]- [11], taking into consideration various, energy-intensive DC systems, such as the cooling installation, for instance [5], [6].…”

“…However, switching off some devices might cause degradation in Quality of Service (QoS) and Quality of Experience (QoE), especially during the peak load periods. Thus, such algorithms must consider maintaining acceptable QoS and the QoE levels [8], [10], [11].…”