Research On The Technology Of Radio Resource Management In Heterogeneous Wireless Networks

The next-generation wireless network will be the heterogeneous wireless networks (HWNs). Itaims to effectively exploit radio network resource, and provide various services for subscribers.Therefore, it is necessary that the HWNs resource can be rationally scheduled through thecomprehensive and unified platform for radio resource management (RRM). One of the importantresearch hotspots in RRM is about network selection algorithm, and the reasonable networkselection algorithm can provide subscribers with satisfied quality of service (QoS). The thesisdeeply discusses and studies both RRM and network selection scheme according to mobility model,performance analysis optimization, and decision method and etc.Terminal mobility in heterogeneous networks is a key factor for wireless resource allocation.The thesis builds a terminal angle mobility model, and analyzes the characteristic of the directionangle for terminal mobility in this model. According to the characteristic of the direction angle forterminal mobility, the theoretical closed-formula is derived for vertical handoff rate in the conditionthat users’ moving direction is unchanging in the fixed time interval T. And the theoretical ultimatevalue of handoff rate is taken as a special case when the fixed time interval T tends to zero. It can beclearly demonstrated that the relationship between handoff rate and the users’ distributed density,moving speed, cell radius from the theoretical ultimate value of handoff rate. Then it furtherdeduces the closed-formula of vertical handoff rate in heterogeneous networks, and reveals thevalidity of theoretical analysis by simulation. The simulation results show that the theoreticalultimate value of handoff rate can be used to approximate to the practical handoff rate under certaincondition, thus it can effectively reduce the complexity of theoretical calculation for handoff rate.The thesis proposes a method of performance optimization based on mobility analysis andMarkov model in Heterogeneous Wireless Networks (HWNs) consisting of diverse radio accessnetworks (RANs). The method takes users’ mobility as an important factor affecting systemperformance, and users’ mobility is embodied by handoff rate. Then the system performance ofHWNs based on Markov process is analyzed, and the expression of total throughput is obtained forthe performance analysis of HWNs. When users perform network selection in the overlappingregion, it will involve the reasonable allocation for radio resource. Hence the proportion of servicearrival rate accessing to one of the networks in the overlapping region is regarded as an optimizedfactor for performance analysis. The throughput of the HWNs’ system is maximized by choosing appropriate factor. There are three types of HWNs scenarios analyzed, namely one overlappingregion within two RANs, single overlapping region within multiple RANs and multiple overlappingregions within multiple RANs. Finally, three corresponding RRM algorithms are given according togolden section method, subgradient approach and quadratic programming. The simulation resultsshow that the optimization algorithm can effectively enhance the throughput of HWNs, andimprove the resources’ utilization of wireless networks.This thesis puts forward a network selection algorithm based on multiple attribute decisionmaking (MADM) and group decision making (GDM). Firstly, the entropy weighting (EW) methodis adopted to make objective decision, and the analytic hierarchy process (AHP) is used to makesubjective decision, then the objective and subjective decision results are synthesized to be a newattribute weights’ vector by using GDM. Considering that the results of GDM should be reasonableand convincible, the criterion of consistency is adopted for judging the compatibility of groupjudgments. More specifically, the algorithm takes into account not only objective attributes ofnetworks but also the preference of subscribers and traffic class. Hence it guarantees that thesubscribers can not select the networks with poor performance depending on their preference. Thesimulation results show that the proposed algorithm can effectively reduce the handoff number andprovide subscribers with satisfactory quality of service (QoS).Based on load balance and service characteristics, the thesis proposes a common radio resourcemanagement (CRRM) algorithm for HWNs consisting of CDMA2000and IEEE802.11. Theproposed algorithm selects appropriate access and handoff networks for new and handoff callsaccording to users’ mobility characteristic, location information, service type and network load. Theperformance of the proposed algorithm is analyzed based on a three-dimensional Markov chainmodel. Based on the HWNs model and handoff rate analysis, the model’s parameters aredetermined. In order to resolve the steady-state distribution of the three-dimensional Markov model,successive over-relaxation (SOR) iteration method is adopted to work out steady state equations.According to the steady state probabilities of the Markov model, the closed formulas of HWNsperformance metrics are given to evaluate the proposed CRRM algorithm, e.g. new call blockingprobability, handoff blocking probability, throughput, and service cost. For the purpose ofillustrating advantages of the proposed algorithm, two other CRRM algorithms are employed forcomparison. The simulation results show that the proposed algorithm can effectively reduce callblocking probability, and improve HWNs throughput.