摘要
面向物联网高密度、大连接和差异性服务质量的移动通信网络优化具有重要的研究意义。在此条件下,移动通信网络优化是一种多参数的、针对高计算成本函数的复杂优化问题,基于计算图的移动通信网络覆盖质量评估的计算方式,为运算的并行化提供依据,进而基于计算图获得了覆盖质量指标的梯度计算方式,利用基于反向传播获取的梯度信息指导基站天线工作参数的优化,并引入动量法加速了优化问题的收敛速度。仿真结果表明,本算法适用于移动通信网络的覆盖优化计算。
The research on the mobile communication network optimization for the Internet of things large connection and differential service quality has great significance. Mobile communication network optimization is a multi-parameter complex optimization problem with high computational cost function. In order to provide the basis for parallelization of operations, the calculation method of the coverage quality assessment of the mobile communication network based on the computational graph was introduced. Based on the calculation graph, the derivative calculation method of the coverage quality index was obtained, by back propagation to guide the optimization of the antenna parameters. The momentum method was used to accelerate the convergence speed of the optimization algorithm. The simulation results show that the algorithm is suitable for the coverage optimization in mobile communication network.
引文
[1]AL-FALAHY N,ALANI O Y.Technologies for 5G networks:challenges and opportunities[J].IT Professional,2017,19(1):12-20.
[2]RATASUK R,VEJLGAARD B,MANGALVEDHE N,et al.NB-IoTsystem for M2M communication[C]//2016 IEEE Wireless Communications and Networking Conference Workshops(WCNCW).IEEE,2016.
[3]SINHA R S,WEI Y Q,HWANG S H.A survey on LPWA technology:LoRa and NB-IoT[J].ICT Express,2017,3(1):14-21.
[4]RATASUK R,VEJLGAARD B,MANGALVEDHE N,et al.NB-IoTsystem for M2M communication[C]//Wireless Communications and Networking Conference(WCNC).IEEE,2016:1-5.
[5]FAN P Z,ZHAO J,CHIH-LIN I.5G high mobility wireless communications:challenges and solutions[J].China Communications,2016,13(2):1-13.
[6]BUZZI S,CHIH-LIN I,KLEIN T E,et al.A survey of energy-efficient techniques for 5G networks and challenges ahead[J].IEEE Journal on Selected Areas in Communications,2016,34(4):697-709.
[7]BOCCARDI F,HEATH R W,LOZANO A,et al.Five disruptive technology directions for 5G[J].IEEE Communications Magazine,2013,52(2):74-80.
[8]MENGJUN Y,LEI F,WENJING L,et al.Cell outage compensation based on CoMP and optimization of tilt[J].The Journal of China Universities of Posts and Telecommunications,2015,22(5):71-79.
[9]EI-BEAINO W,EI-HAJJ A M,DAWY Z.On radio network planning for next generation 5G networks:a case study[C]//International Conference on Communications.IEEE,2015.
[10]PHAN N Q,BUI T O,JIANG H L,et al.Coverage optimization of LTE networks based on antenna tilt adjusting considering network load[J].China Communications,2017,14(5):48-58.
[11]HUANG L,ZHOU Y Q,HU J L,et al.Coverage optimization for femtocell clusters using modified particle swarm optimization[C]//IEEE International Conference on Communications.IEEE,2012:611-615.
[12]GAO M H,HUANG L F,CUI X N,et al.Intelligent coverage optimization with multi-objective genetic algorithm in cellular system[C]//Computer Science&Education(ICCSE).IEEE,2013:859-863.
[13]ABADI M,BARHAM P,CHEN J M,et al.TensorFlow:a system for large-scale machine learning[C]//OSDI.IEEE,2016:265-283.
[14]SCHAUBACH K R,DAVIS N J,RAPPAPORT T S.A ray tracing method for predicting path loss and delay spread in microcellular environments[C]//Vehicular Technology Conference.IEEE,2002.
[15]SINGH Y.Comparison of okumura,hata and cost-231 models on the basis of path loss and signal strength[J].International Journal of Computer Applications,2012,59(11):37-41.
[16]BAUER F L.Computational graphs and rounding error[J].SIAMJournal on Numerical Analysis,1974,11(1):87-96.
[17]YVES C,RUMELHART D E.Backpropagation:theory,architectures and applications[M].London:Psychology Press,2013.
[18]LéON B.Large-scale machine learning with stochastic gradient descent[J].19th International Conference on Computational Statistics,2010:177-186.
[19]KINGMA D,BA J.Adam:a method for stochastic optimization[J].Computer Science,2014.
[20]SUTSKEVER I,MARTENS J,DAHL G,et al.On the importance of initialization and momentum in deep learning[C]//International Conference on Machine Learning.IEEE,2013.