The Community for Technology Leaders
Green Image
Issue No. 05 - May (2012 vol. 11)
ISSN: 1536-1233
pp: 807-820
Eng Hwee Ong , University of Newcastle, Australia, Newcastle
Jamil Y. Khan , University of Newcastle, Australia, Newcastle
Kaushik Mahata , University of Newcastle, Australia, Newcastle
Recently, the IEEE 1900.4 standard specified a policy-based radio resource management (RRM) framework in which the decision making process is distributed between network-terminal entities. The standard facilitates the optimization of radio resource usage to improve the overall composite capacity and quality of service (QoS) of heterogeneous wireless access networks within a composite wireless network (CWN). Hence, the study of different RRM techniques to maintain either a load- or QoS-balanced system through dynamic load distribution across a CWN is pivotal. In this paper, we present and evaluate three primary RRM techniques from different aspects, spanning across predictive versus reactive to model-based versus measurement-based approaches. The first technique is a measurement-based predictive approach, known as predictive load balancing (PLB), commonly employed in the network-distributed RRM framework. The second technique is a model-based predictive approach, known as predictive QoS balancing (PQB), typically implemented in the network-centralized RRM framework. The third technique is a measurement-based reactive approach, known as reactive QoS balancing (RQB), anchored in the IEEE 1900.4 network-terminal distributed RRM framework. Comprehensive performance analysis between these three techniques shows that the IEEE 1900.4-based RQB algorithm yields the best improvement in QoS fairness and aggregate end-user throughput while preserving an attractive baseline QoS property.
IEEE 1900.4, radio resource management, reactive, predictive, load distribution, WLANs.

E. H. Ong, J. Y. Khan and K. Mahata, "Radio Resource Management of Composite Wireless Networks: Predictive and Reactive Approaches," in IEEE Transactions on Mobile Computing, vol. 11, no. , pp. 807-820, 2011.
81 ms
(Ver 3.3 (11022016))