【作者】 董燕；

【导师】 黄载禄； 王宏远；

【作者基本信息】 华中科技大学 ， 通信与信息系统， 2007， 博士

【摘要】 作为一种国家关键的基础通信设施,以及全球移动通信的有机组成部分,卫星移动通信系统在国家安全、紧急救援、互联网、远程教学、卫星电视广播以及个人移动通信等方面得到了广泛的应用。新一代宽带卫星通信系统可以提供个人电信业务、多信道广播、互联网的远程传送,是全球无缝个人通信、互联网空中高速通道的必要手段。近年来卫星通信新技术不断发展,特别是低轨道卫星移动通信系统受到了人们的广泛关注,其研究与应用已成为各国的战略发展重点。无线资源管理是低轨卫星移动通信系统研究中的一项重要内容,这主要是由于卫星系统的资源是非常昂贵的,因此如何合理而有效地管理并利用卫星系统的资源已成为关键。本文针对目前无线资源管理中核心问题——信道分配和切换管理,开展了研究。首先,提出了一种动态信道分配策略。针对低轨卫星通信系统波束切换频繁、切换掉话率高的问题,通过给出卫星波束小区业务量均值和方差的预测公式,以及预测小区信道占用分布,结合低轨卫星的特点,提出了一种基于全局业务量预测和随机控制的动态信道分配策略PSCCA。并通过仿真验证了该策略的有效性。相对于精确切换控制方法,该策略不需要终端位置信息,降低了系统复杂度和信令负荷,对卫星移动通信系统的切换呼叫掉话率有较强的控制能力。其次,提出了一种基于效用函数的自适应信道分配策略并建立了相应的数学模型。提供多种类型的业务是宽带卫星移动通信的发展趋势,利用业务多样性以及业务带宽可变的特点,针对多种业务给出了表征带宽与用户满意度之间关系的效用函数,利用效用函数以及语音切换呼叫的高优先级,设计了一种切换呼叫接入控制策略。在此基础上,提出了一种新的、基于效用函数的自适应信道分配策略,并建立了最大化系统用户满意度的有约束非线性规划模型。该策略以少量的效用代价换取较低的切换掉话率。再次,提出了两种基于智能计算的优化算法。针对自适应信道分配策略中的非线性规划问题,根据其求解具有动态性和实时性的特点,采用智能计算方法,设计了基于遗传算法和基于粒子群优化算法的两种自适应信道分配算法进行求解。采用基于遗传算法的信道分配算法,通过仿真实验验证了自适应信道分配策略的有效性;而基于粒子群优化的信道分配算法则进一步提高了求解效率,具体措施主要是采用非线性惯性权重提高算法收敛速度,采用基于粒子约束状态信息的约束处理机制,改善求得可行解的速度。仿真实验结果表明,自适应信道策略通过动态调整带宽实现对业务量的自适应,克服了预留信道策略信道利用率低的缺点。接着,提出了一种低轨卫星-地面综合移动通信系统的体系结构。为了提高移动通信系统的容量和效率、实现全球无缝覆盖,必须实现低轨卫星移动通信系统与地面通信系统之间的有机综合。针对卫星通信系统与地面无基础设施通信系统的综合,在分析低轨卫星通信网络的业务量强度和业务源分布特点的基础上,提出了一种低轨卫星—地面无基础设施通信系统的综合体系结构,设计了基于地面中继的负荷分担机制,并给出了相应的呼叫控制协议流程。该机制在忙时启用地面中继,用于终接近距离呼叫,分担上星信道的负荷。最后,本文对主要研究成果进行了全文总结,并指出了今后进一步的研究方向。更多还原

【Abstract】 As a key national fundamental communication establishment and component of global mobile communications, satellite mobile communication systems are widely used in national security, emergent succor, internet, remote education, satellite broadcasting and personal mobile communications. The new generation wideband satellite communication system is an indispensable technique for global seamless coverage and high-speed“sky-internet”. The new technologies of satellite communication are developed quickly. Especially, the development and applications of low earth orbit satellite mobile communications are strategic consequence of countries in the world. Due to the costliness and scarceness of the satellite wireless channel, the wireless resource management is a crucial issue in the research of satellite mobile communications. In this dissertation, channel allocation and spotbeam handover management are studied, which are the hardcore of the wireless resources management.Firstly, a dynamic channel allocation scheme is proposed. In allusion order to tackle the problems of frequently spotbeam handovers and high handover call dropping probabilities, the predictive formulas of mean and variance of the spotbeam traffic are deduced. And then, the channel occupation distribution is predicted. A global traffic Prediction and Stochastic Control based Channel Allocation scheme (PSCCA) is designed according to the character of LEO satellites. The simulations are performed and the results reveal the validity of the scheme. Comparing with the termination location based handover scheme, PSCCA shows its strong control capability on handover call dropping probabilities. Because PSCCA need not the information of the termination location, it has lower system complexity and signaling overheads.Secondly, a utility function based adaptive channel allocation scheme is proposed and the corresponding mathematical model is established. Nowadays, providing multi-type services is the trend of the wideband satellite communications. With the character of variable service bandwidth, the utility function is used to express the relationship between bandwidth and user’s satisfaction and a handover call admission control scheme is designed. And then, a novel utility function based adaptive channel allocation scheme is brought forth based on the premise of the establishment of the constrained nonlinear programming model which aims at maximizing the system user’s satisfaction. The proposed scheme has the lower handover call probability at the cost of little utility.In succession, two intelligent computing algorithms are proposed to solve the constrained nonlinear programming problem in the proposed adaptive channel allocation scheme. Genetic algorithm and particle swarming optimization algorithm are designed respectively. Genetic algorithm is a general optimization algorithm by which the proposed adaptive channel allocation scheme is validated. While the utilization of the particle swarming optimization algorithm can improve the efficiency of the solving procedure further. The advantages of the designed particle swarming optimization algorithm consist in that the nonlinear inertia weight improves the convergence speed and the check of feasible of particles is introduced for constrains handling. The simulation results show that the proposed adaptive channel allocation scheme can adapt to the change of the traffic by dynamically adjusting the bandwidth of the calls and overcomes the defect of the lower channel utilization of the reservation channel scheme.Fourthly, a structure of integrated LEO satellite and terrestrial communication system is presented. It is necessary to integrate the LEO satellite communication systems and terrestrial mobile communication systems for improving the capacity and efficiency, achieving global seamless coverage. Based on the analysis of satellite network traffic and the characteristic of traffic sources distribution, an integrated structure of LEO-land non-infrastructure communication system is proposed. The load balancing scheme based on the land repeaters which is used to connect the terminations around it. And the corresponding call control protocols are designed.Finally, the main contributions of this dissertation are summarized and some suggestions and directions for the future work in this field are given.更多还原