Femtocell网络的异构性研究

发布时间：2018-06-08 13:57

本文选题：毫微微蜂窝 + 随机几何　；参考：《电子科技大学》2014年硕士论文

【摘要】：作为下一代移动电信网络的关键技术,femtocell可以有效分流宏蜂窝流量,并为室内用户提供高质量的通信环境。Femtocell网络覆盖在宏蜂窝区域内,使得蜂窝网络结构异构化。相比其它无线接入技术,femtocell的覆盖半径十分有限,一般不会超过30米。这些覆盖有限的接入点,通常由非专业用户直接购买并安装,所以与传统蜂窝的位置特征十分不同。对于传统蜂窝网络,基站的位置和部署是由运营商精心规划并测试通过,而femtocell网络是无规划的网络。Femtocell位置的随机性使得性能分析变得棘手。本文首先研究了femtocell网络的基站规划和位置建模等问题。其中涉及了自动femtocell蜂窝部署概念,以及在实际中部署室内和室外femtocell的策略。本文同时探讨了3种建模femtocell位置的方法:基于仿真的蒙特卡罗模型,基于确定性能的单小区线性模型,以及基于统计的随机几何建模方法。此外,本文分别为每种位置建模方法构造了下行容量模型来阐明其在femtocell位置建模中的主要思想和特征。通过联合使用蒙特卡罗方法和随机几何方法,本文继而研究了femtocell-宏蜂窝混合网络的覆盖容量和孤立概率。覆盖容量对于任何蜂窝网络都是一项重要的指标,然而由于femtocell网络位置的随机特征,其覆盖容量特征将与众不同;而孤立概率对于femtocell网络尤其重要,因为femtocell部署在宏蜂窝下,而孤立的femtocell必然导致用户频繁的切换和糟糕的用户体验。本文假设femtocell采用开放式接入,主要用于宏蜂窝网络的分流。本文使用2种方法来建模femtocell网络的位置特征。第一种方法使用齐次泊松点过程模型,第二种方法使用了Matérn硬核模型。齐次泊松点过程对应于femtocell的位置在给定区域内是均匀分布(即femtocell间没有最小距离约束);Matérn硬核模型对应于femtocell间存在着最小距离约束。通过综合考虑femtocell网络的覆盖率、孤立概率以及实际约束,覆盖半径为20米的femtocell网络性能较优。对于覆盖半径为20米的femtocell网络,如果最小距离约束为18米,那么在500×500平方米的区域内,当平均femtocell数目为200个时,其覆盖率从无约束情况下的63.4%增加到75.6%,而孤立概率却没有增加。本文同时考虑了不同最小距离约束对覆盖半径为20米的femtocell网络的影响并且发现了有趣的现象:femtocell网络的孤立概率与无约束情况下的差异并不随最小距离约束的变化成正相关关系。
[Abstract]:As the key technology of the next generation mobile telecommunication network, femtocell can effectively distribute macro cellular traffic, and provide indoor users with a high quality communication environment. Femtocell network is covered in the macro cell area, which makes the cellular network structure isomerization. Compared with other wireless access technologies, femtocell has a very limited coverage radius, generally less than 30 meters. These access points with limited coverage are usually purchased and installed directly by non-professional users, so they are very different from traditional cellular location features. For traditional cellular networks, the location and deployment of base stations are carefully planned and tested by the operator, while the randomness of the location of the unplanned network, .Femtocell, makes performance analysis difficult. In this paper, the base station planning and location modeling of femtocell network are studied. The concept of automatic femtocell cellular deployment and the strategy of deploying indoor and outdoor femtocell in practice are discussed. This paper also discusses three methods of modeling femtocell location: Monte Carlo model based on simulation, single cell linear model based on deterministic performance, and stochastic geometric modeling method based on statistics. In addition, a downlink capacity model is constructed for each location modeling method to illustrate its main ideas and characteristics in femtocell location modeling. By combining Monte Carlo method and stochastic geometry method, we study the coverage capacity and isolation probability of femtocell-macro cellular hybrid network. Coverage capacity is an important index for any cellular network. However, because of the random feature of the location of femtocell network, the coverage capacity feature will be different, and the isolation probability is especially important for femtocell network, because femtocell is deployed under macro cell. Isolated femtocell will inevitably lead to frequent switching and poor user experience. This paper assumes that femtocell uses open access, which is mainly used for the distribution of macro cellular networks. In this paper, two methods are used to model the location characteristics of femtocell networks. The first method uses homogeneous Poisson point process model, and the second method uses Mat 茅 rn hard kernel model. The position of homogeneous Poisson point process corresponding to femtocell is uniformly distributed in a given region (that is, there is no minimum distance constraint between femtocell and Mat 茅 rn hard kernel model corresponding to the minimum distance constraint between femtocell. By considering the coverage, isolation probability and practical constraints of femtocell network, the performance of femtocell network with coverage radius of 20 meters is better. For a femtocell network with a coverage radius of 20 meters, if the minimum distance constraint is 18 meters, then in the area of 500 脳 500 square meters, when the average number of femtocell is 200, the coverage rate increases from 63.4% to 75.6%, but the probability of isolation does not increase. In this paper, we also consider the influence of different minimum distance constraints on the femtocell networks with coverage radius of 20 m and find that the isolation probability of the: femtocell network is not positively correlated with the variation of the minimum distance constraints.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5

【相似文献】