This thesis studies intelligent on-demand radio resource provisioning involving sleep mode operation in ultra Small Cell Networks (SCNs). Sleep modes are low power states of base stations. The purpose of the research is to investigate how appropriate traffic information can be adopted in sleep mode operation schemes for SCNs with different architectures.ududA novel protocol-friendly sleep mode operation algorithm based on Adaptive Traffic Perception is proposed for distributed SCN architectures. It is proved robust to different SCN layouts with the reduction in the average power consumption of base stations being more than 35% while maintaining the Quality of Service. ududThe Traffic-aware Cell Management scheme adopting Direction of Arrival information is particularly designed to eliminate the necessity of computation for sleeping base stations. This scheme is shown to significantly reduce the side effects associated with the sleep mode operation, including system overheads and the increasing user transmission power.ududFor SCNs using centralised architectures, such as Cloud Radio Access Networks, Hotspot-oriented Green Frameworks are proposed for different information availabilities, which achieve almost 80% reduction in power consumption of Remote Radio Heads at low traffic levels. A clustering technique is utilised for the optimisation of the placement of active Remote Radio Heads, lowering the average user transmission power. The amount of reduction depends on the completeness of the information and can exceed 70% compared with the state-of-the-art. ududA type II Matern Hard-core Point Process is used for modelling SCNs. The derivation and approximation of its distance distributions are also proposed. The distance distributions are used for the probabilistic theoretical analysis of some metrics of the sleep mode operation.
展开▼
机译：本文研究了超小型小区网络（SCN）中涉及睡眠模式操作的智能按需无线电资源配置。睡眠模式是基站的低功耗状态。本研究的目的是研究在具有不同体系结构的SCN的睡眠模式操作方案中如何采用适当的流量信息。 ud ud提出了一种基于自适应流量感知的新型协议友好型睡眠模式操作算法，用于分布式SCN体系结构。它被证明对不同的SCN布局具有鲁棒性，基站的平均功耗降低了35％以上，同时保持了服务质量。 ud ud采用到达方向信息的流量感知小区管理方案是专门为消除休眠基站计算而设计的。该方案显示可以显着减少与睡眠模式操作相关的副作用，包括系统开销和增加的用户传输功率。 ud ud对于使用集中式架构的SCN（例如云无线电接入网），提出了面向热点的绿色框架针对不同的信息可用性，在低通信量水平下，远程无线电头的功耗几乎降低了80％。利用群集技术来优化活动远程无线电头的位置，从而降低平均用户传输功率。减少的数量取决于信息的完整性，与最新技术相比，减少的数量可能超过70％。 ud udII型Matern Hard-core Point Process用于SCN建模。还提出了其距离分布的推导和近似。距离分布用于睡眠模式操作的某些度量的概率理论分析。
展开▼
