计算机应用   2017, Vol. 37 Issue (1): 97-102  DOI: 10.11772/j.issn.1001-9081.2017.01.0097 0

### 引用本文

LIAN Xiaocan, ZHANG Pengyuan, TAN Guoping, LI Yueheng. Antenna down-tilt angle self-optimization method based on particle swarm in long term evolution network[J]. JOURNAL OF COMPUTER APPLICATIONS, 2017, 37(1): 97-102. DOI: 10.11772/j.issn.1001-9081.2017.01.0097.

### 文章历史

Antenna down-tilt angle self-optimization method based on particle swarm in long term evolution network
LIAN Xiaocan, ZHANG Pengyuan, TAN Guoping, LI Yueheng
Communication and Information Systems Institute, Hohai University, Nanjing Jiangsu 211100, China
Abstract: To solve the coverage and capacity optimization problem of Self-Organizing Network (SON) in the 3rd Generation Partnership Project (3GPP), an active antenna down-tilt angle optimization method based on Particle Swarm Optimization (PSO) algorithm was proposed. First, the number of User Equipments (UE) served by evolved Node B (eNB) was determined, and the Reference Signal Received Power (RSRP) and position measured from the UE were obtained. Second, the Spectral Efficiency (SE) was regarded as the fitness function which defined by optimization goals. Then, down-tilt angle optimization was regarded as multidimensional optimization problem, and antenna down-tilt angle was regarded as the set of particles to resolve the optimal value by the PSO algorithm. Finally, the capacity and coverage self-optimization of Long Term Evolution (LTE) networks was achieved by adjusting down-tilt angle independently. By simulations, different objective functions made different optimization results. When the average spectrum efficiency was set as the optimization goal, the spectral efficiency of traditional golden section algorithm increased by 12.9% than primary settings, the spectral efficiency of PSO was increased by 22.5%. When the weighted average spectral efficiency was set as the optimization goal, the spectral efficiency of golden section algorithm was not significantly improved but that of PSO was increased by 19.3% for edge users. The experimental results show that the proposed method improves cell throughput and system performance.
Key words: Long Term Evolution (LTE)    down-tilt angle optimization    Self-Organizing Network (SON)    coverage and capacity optimization    Particle Swarm Optimization (PSO)
0 引言

1 系统建模

 ${{\theta }_{m,u}}=\arctan \left( \left( {{h}_{m}}-{{h}_{u}} \right)/{{d}_{m,u}} \right)$
 图 1 系统模型 Figure 1 System model

 ${{S}_{m,u}}({{\theta }_{m}})={{G}_{m,u}}({{\theta }_{m}}){{\rho }_{m,u}}{{P}_{m}}$

 ${{I}_{u}}\left( \Theta \right)=\sum\limits_{m\in M\backslash \overset{\_}{\mathop{m}}\,}{{{S}_{m,u}}({{\theta }_{m}})}$

 ${{\gamma }_{u}}\left( \Theta \right)=\frac{{{S}_{m,u}}({{\theta }_{m}})}{{{I}_{u}}\left( \Theta \right)+{{n}_{0}}}$

 ${{r}_{u}}\left( \Theta \right)={{w}_{u}}\lg (1+{{\gamma }_{u}}(\Theta ))$

1) 当优化目标是系统的总体吞吐量时，有：

 $f({{\theta }_{m}})=\sum\limits_{u\in {{U}_{m}}}{{{r}_{u}}({{\theta }_{m}})}=\sum\limits_{u\in {{U}_{m}}}{{{w}_{u}}\lg (1+{{\gamma }_{u}}({{\theta }_{m}}))}$

2) 当优化目标是系统的平均频谱效率时，有：

 \begin{align} & f({{\theta }_{m}})=\frac{1}{N}\sum\limits_{u\in {{U}_{m}}}{{{r}_{u}}({{\theta }_{m}})/{{w}_{u}}}= \\ & \frac{1}{N}\sum\limits_{u\in {{U}_{m}}}{\lg (1+\gamma ({{\theta }_{m}}))} \\ \end{align}

3) 当优化目标是系统的加权吞吐量时，有：

 $f\left( \Theta \right)=\sum\limits_{u\in U}{{{\alpha }_{u}}{{r}_{u}}(\Theta )}$

4) 优化目标是系统的加权平均频谱效率：

 \begin{align} & f({{\theta }_{m}})=\frac{1}{N}\sum\limits_{u\in {{U}_{m}}}{{{\alpha }_{u}}{{r}_{u}}({{\theta }_{m}})/{{w}_{u}}} \\ & =\frac{1}{N}\sum\limits_{u\in {{U}_{m}}}{{{\alpha }_{u}}\lg (1+{{\gamma }_{u}}({{\theta }_{m}}))} \\ \end{align}

 $\theta _{m}^{*}=\underset{_{{{\theta }_{m}}}}{\mathop{\arg }}\,\max f({{\theta }_{m}})$

2 粒子群算法

 \begin{align} & v_{id}^{t+1}=\omega v_{id}^{t}+{{c}_{1}}{{r}_{1}}(p_{id}^{t}-x_{id}^{t}) \\ & +{{c}_{2}}{{r}_{2}}(p_{gd}^{t}-x_{id}^{t}) \\ \end{align} (1)
 \begin{align} & X_{id}^{t+1}=X_{id}^{t}+v_{id}^{t+1} \\ & \omega ={{\omega }_{start}}-\left( {{\omega }_{start}}-{{\omega }_{end}} \right)/{{t}_{\max }}\times t; \\ \end{align} (2)

 图 2 基站天线调整流程 Figure 2 Antenna adjustment process of eNB

1) 确定基站m中正在传输数据的用户集合Um，设有Nm=Um个用户；

2) 根据Um中所有用户上报的邻小区测量报告的情况，确定基站m的邻小区集合，记为Mm

2) 用户测量位置信息，并上报给基站；

3) 基站根据用户上报的位置信息，估算用户与基站间的垂直角度θm,u

2) OAM侧使用粒子群优化，寻找基站的最优下倾角。求解得到最优下倾角集合。

1) 将基站的下倾角调整为优化后的角度。

2) 等待预定的时间段之后，或满足预定义的事件触发条件，返回步骤1，再次进入调整循环。

3 仿真与性能分析

 图 3 系统框架 Figure 3 System frame

3.1 设定优化目标为用户平均频谱效率

 图 4 用户频谱效率累积分布 Figure 4 Cumulative distribution of spectral efficiency
 图 5 用户信干噪比累积分布 Figure 5 Cumulative distribution of signal to interference plus noise ratio

 图 6 平均频谱效率对比 Figure 6 Comparison of average spectral efficiency

3.2 设定优化目标为加权平均频谱效率

 图 7 加权条件下频谱效率累积分布 Figure 7 Cumulative distribution of weighted spectral efficiency
 图 8 加权条件下信干噪比累积分布 Figure 8 Cumulative distribution of weighted signal to interference plus noise ratio

 图 9 加权条件下平均频谱效率对比 Figure 9 Comparison of weighted average spectral efficiency

 图 10 PSO收敛性 Figure 10 Astringency of PSO

4 结语

 [1] CELENTANO J M. Carrier capital expenditures[J]. IEEE Communications Magazine, 2008, 46 (7) : 82-88. doi: 10.1109/MCOM.2008.4557047 [2] 夏永康, 梁晓雯. LTE网络天线下倾角自优化算法[J]. 无线电工程, 2014, 44 (3) : 11-14. ( XIA Y K, LIANG X W. An algorithm of antenna tilt self-optimization in LTE networks[J]. Radio Engineering of China, 2014, 44 (3) : 11-14. ) [3] ECKHARDT H, KLEIN S, GRUBER M. Vertical antenna tilt optimization for LTE base stations[C]//VTC 2011:Proceedings of the 73rd IEEE Vehicular Technology Conference. Piscataway, NJ:IEEE, 2011:1-5. [4] GANDHI A D. Significant gains in coverage and downlink capacity from optimal antenna downtilt for closely-spaced cells in wireless networks[C]//WOCC 2014:Proceedings of the 23rd Wireless and Optical Communication Conference. Piscataway, NJ:IEEE, 2014:1-6. [5] KASEM F, HASKOU A, DAWY Z. On antenna parameters self optimization in LTE cellular networks[C]//ICCIT 2013:Proceedings of the Third International Conference on Communications and Information Technology. Piscataway, NJ:IEEE, 2013:44-48. [6] LUKETIC I, SIMUNIC D, BLAJIC T. Optimization of coverage and capacity of self-organizing network in LTE[C]//MIPRO 2011:Proceedings of the 34th International Convention. Piscataway, NJ:IEEE, 2011:612-617. [7] RAZAVI R, KLEIN S, CLAUSSEN H. Self-optimization of capacity and coverage in LTE networks using a fuzzy reinforcement learning approach[C]//PIMRC 2010:Proceedings of IEEE 21st International Symposium on Personal Indoor and Mobile Radio Communications. Piscataway, NJ:IEEE, 2010:1865-1870. [8] MITSCHELE-THIEL A. Cooperative fuzzy Q-learning for self-organized coverage and capacity optimization[C]//PIMRC 2012:Proceedings of IEEE 23rd International Symposium on Personal Indoor and Mobile Radio Communications. Piscataway, NJ:IEEE, 2012:1406-1411. [9] LI J, ZENG J, SU X, et al. Self-optimization of coverage and capacity in LTE networks based on central control and decentralized fuzzy Q-learning[J]. International Journal of Distributed Sensor Networks, 2012, 2012 : 1018-1020. [10] ISLAM M N U, MITSCHELE-THIEL A. Reinforcement learning strategies for self-organized coverage and capacity optimization[C]//WCNC 2012:Proceedings of the 2012 IEEE Wireless Communications and Networking Conference. Piscataway, NJ:IEEE, 2012:2818-2823. [11] PAN R J, JIANG H L, PEI S Y. Modified particle swarm optimization-based antenna tilt angle adjusting scheme for LTE coverage optimization[J]. Journal of Southeast University (English Edition), 2015 (4) : 443-449. [12] 3GPP. TR25.996 V12.0.0, 3rd generation partnership project; technical specification group radio access network; spatial channel model for Multiple Input Multiple Output (MIMO) simulations[EB/OL].[2016-04-25] . http://www.qtc.jp/3GPP/Specs/25996-900.pdf. [13] HAR D, WATSON A M, CHADNEY A G. Comment on diffraction loss of rooftop-to-street in COST 231-Walfisch-Ikegami model[J]. IEEE Transactions on Vehicular Technology, 1999, 48 (5) : 1451-1452. doi: 10.1109/25.790519