Bandwidth is an extremely valuable and scarce resource in multimedia networks. Therefore, efficient bandwidth management is necessary in order to provide high Quality of Service (QoS) to users. Recently, S. Kim proposed a new QoS-aware Bandwidth Allocation (QSBA) scheme for the efficient use of available bandwidth (Kim, 2010). By using the multi-objective optimization technique and Talmud allocation rule, the bandwidth is adaptively controlled to maximize network efficiency while ensuring QoS provisioning. In addition, the QSBA scheme adopts the online feedback strategy to dynamically respond to current network conditions.
Development Motivation
In recent years, the explosive growth of new services and the rapid and widespread proliferation of multimedia data have necessitated the development of an efficient network management system. The network system is expected to provide diversified traffic services and enhance network performance simultaneously. Usually heterogeneous multimedia data can be categorized into two classes according to the required Quality of Service (QoS): class I (real-time) services and class II (non-real-time) services. Different multimedia services over networks not only require different amounts of bandwidth but also have different QoS requirements (Yang, Ou, Guild, & Chen, 2009).
During network operations, the limited bandwidth has to be shared by several users. Therefore, fairness is another prominent issue for the network management. If the concept of fairness is not considered explicitly at the design stage of bandwidth allocation algorithm, different allocation requests can result in very unfair bandwidth allocations. However, fairness-oriented allocation methods may lead to a system inefficiency, which degrades total network performance quite seriously.
The QSBA scheme is developed as a new bandwidth allocation algorithm for multimedia networks. To approximate an optimal network performance, the developed algorithm has focused on the basic concept of online decision process. Based on the Modified Game Theory (MGT) (Mehmet, & Ramazan, 2001), the bandwidth is adaptively allocated to satisfy different QoS requirements. In addition, by using the Talmud allocation rule (Li, & Cui, 2009), a weight parameter is adjusted periodically to ensure the allocation fairness. Therefore, the system dynamically re-estimates the current network condition and iteratively adapts control decisions. Under dynamically changing network environments, this online strategy can find the best solution for conflicting objectives.