Network selection strategies and resource management schemes in integrated heterogeneous wireless and mobile networks.

摘要

Wireless and mobile networks (WMNs) are witnessing a great success in recent years. Although there are different types of WMNs (e.g., cellular networks, WLANs, etc.) that can provide different types of services (e.g., different bandwidth and coverage), any single type of existing WMN is not able to provide all types of services such as high bandwidth with wide coverage. In order to provide more comprehensive services, a concept of integrated heterogeneous wireless and mobile network (IHWMN) is introduced by combing different types of WMNs. Additionally, with the advance of software defined radio (SDR) technologies, it is possible to integrate multiple WMN interfaces into a single mobile terminal and let the terminal be able to access multiple WMNs. It is obvious that the introduction of IHWMN as well as multi-interface terminal brings more flexible and plentiful access options for mobile users. However, it faces great challenges, such as the architecture of network integration, network selection strategies, handoff scheme, resource allocation, etc. Although IHWMN is a very promising candidate for the future WMNs, a lot of problems have to be solved before launching to the commercial market.;In this research thesis, we briefly review the existing work for IHWMN. Then, traffic and system models are proposed for IHWMNs. Based on the proposed models, we tackle the network selection problem in IHWMNs, which is required to determine which network should be accessed. A cost-function-based network selection (CFNS) strategy is proposed based on system's perspective, which also takes into account of the user's needs. Theoretical model is used to evaluate the system performance of the proposed CFNS strategy and simulations are also conducted to verify our analysis. Then, we propose preemption-based resource management schemes to support real-time and non-real-time traffic in IHWMNs. The proposed resource management schemes take advantage of heterogeneities of multiple traffic and networks as well as moving nature of users. We further design an algorithm to achieve the fairness between the real-time and non-real-time traffic. The analytical results show that the proposed resource management schemes provide the best service to non-real-time traffic while satisfying the QoS requirements of real-time traffic. In the last, we address the resource management from different perspective. A traffic-based resource management scheme (TRMS) is firstly introduced, which takes advantage of the heterogeneity of traffic and networks as well as user's satisfaction. We then use Markov decision process to model the resource management scheme of multiple traffic in IHWMNs and propose a Q-learning-based resource management scheme (QRMS) to allocate the resource. The issues addressed in this thesis provide many insights into characterizing the emerging problems in IHWMNs. The analytical and simulation methods applied in this thesis will be useful for the research and development in IHWMNs.