Research On Key Technologies For QoS Provisioning In Heterogeneous Wireless Networks

In the near future, multiple network is coexisted. Research of key technologies in Heterogeneous Wireless Networks(HWNs) is needed for future communication network. For the rapid development of wireless networks and the increasing demand of mobile call admission, our works have important applications for provisioning quality of service.This kind of approaches stand at the view of the user and have become a hot point in research community recently. Although many research achievements have been made, there are still a lot of challenges, such as the complex resource management optimization model and the inefficient model solution algorithm. And it is difficult to implement in real circumstance of heterogeneous wireless networks.In this thesis, based on the characteristic of heterogeneous wireless networks and the demand of practical system, we study joint call admission control scheme, the resource management in heterogeneous wireless networks, multi-network resource management and intelligent network selection. According to the above mentioned facts, we propose new theory and new method in thesis. The new method can get better performance. In experiments, we also confirm its effectiveness.Our main works and contributions are as follows:1)Based on the architecture of heterogeneous wireless networks, we first analyze the system reward of call admission in multi-traffic. A joint call admission control(JCAC) model based on semi-Markov decision process for new calls and handoff calls is proposed. First, the two-type network of heterogeneous wireless networks is introduced. Secondly, we propose an optimal JCAC strategy for new calls and handoff calls, to analyze the network resource utilization and the quality of service(QoS) parameters. The optimal policy has a multi-dimensional threshold structure to improve the algorithm of value iteration. The performance of the proposed model is evaluated by the simulation result, it can provide high quality of service and minimize the blocking/dropping probabilities, which also shows that the efficiency and effectiveness can be more applicable for reality.2)In order to solve the issue that existing model can not combine breakoff and integrity very well, a partially observable markov-modulated poisson process(PO-MMPP) traffic model is proposed. This model captures the burstiness of the overflow traffic under the imperfect observability of the network states. The optimal objective is minimum network cost. It is shown that the optimal control policy can balance the network cost and traffic QoS requirement. By derivation and confirmation, we induce the relations in call arrival rate and transition rates. Simulation results indicate that the proposed model provided better performance compared with complete sharing algorithm, in terms of minimizing the network cost, new call blocking and handoff call dropping probabilities. The network cost is reduced by 30%. The optimal control policy can be adaptive with network state and QoS requirement.3) Based on the characteristic of multiple network cooperation, we study multi-domain optimal resource management in heterogeneous wireless networks. We analyze the situation of optimal resource management in heterogeneous wireless networks. We propose an optimal model to allocate the whole network resource. The model can improve both the entire rewards of the system and the quality of service(QoS) by allocating the heterogeneous wireless network resources between network domains. We propose a new Semi-Markov Decision Process(SMDP) based model for the optimal management of resource to achieve the optimal decision-making strategies between multiple network domains in heterogeneous wireless network. Theoretical analysis and experiment results indicate that the optimal strategies can increase the overall long-term rewards of the system, but also reduce the call blocking probability of service in order to increase the quality of service(QoS) of heterogeneous wireless networks, compared with the complete sharing algorithm, the new call blocking probabilities have fallen by at least 50%.4)Because lacking of unified, mature decision systems in network selection strategy for heterogeneous wirless networks, a decision model of the integrated network and end user is proposed. First, we analyze the network state, the QoS of different traffics and the state of mobile terminal in order to form the system decision scheme. Then based on the pre-selection the new model analyze the QoS requirement of different traffics, the different traffic prameters are utilized by parallel fuzzy logic controllers to select the best network as the target of handoff. Simulation results indicate that the proposed scheme provide better performance compared with the existing algorithm.