A Machine Learning Assisted Method of Coverage and Capacity Optimization (CCO) in 4G LTE Self Organizing Networks (SON)

Self-Organizing Network (SON) has been introduced more than a decade by the Next Generation Mobile Networks (NGMN) and later standardized by the 3rd Generation Partner-ship Project (3GPP). However, SON has ever never fully met the expectation from Mobile Network Operators (MNOs) since day one due to lack of suitable wireless based Machine Learning (ML) techniques which can empower SON with intelligence in the old days. The authors propose, validate, and productize a wireless ML based scheme ISO-SON to mitigate cell coverage and interference problems in 4G Long Term Evolution (LTE) networks. ISO-SON algorithm portfolio targets at maximally optimizing cell weak coverage and over-coverage collectively. To scale ISO-SON from a single pain-point optimization in a silo to joint optimization for a problem cell group, a novel clustering methodology is also developed in ISO-SON to identify the social network of the targeted problem cell through modeling neighbor cell closeness rate, pair-wised handoff frequencies between problem cell and neighbor cell candidates, and traffic density of neighbor cell candidates. Finally Sequentially Unconstrained Maximization Technique (SUMT) is applied to converge the ISO-SON loss function in a specific conditional range which is feasible for engineering implementation in tuning antenna angles, due to which the gap between 3GPP SON standards and actual operations is bridged through ISO-SON. ISO-SON is productized and now up and running on a tier-1 MNO’s 4G LTE networks. Reference Signal Received Power (RSRP), the coverage performance indicator, has been improved by 15% for the problem cells after ISO-SON is deployed.

[1]  Jorge Nocedal,et al.  Algorithm 778: L-BFGS-B: Fortran subroutines for large-scale bound-constrained optimization , 1997, TOMS.

[2]  I. Forkel,et al.  The effect of electrical and mechanical antenna down-tilting in UMTS networks , 2002 .

[3]  Yunzhou Li,et al.  System level performance of energy efficient dynamic mechanical antenna tilt angle switching in LTE-Advanced systems , 2013, 2013 IEEE International Wireless Symposium (IWS).

[4]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[5]  Rouzbeh Razavi,et al.  Utility Fair Optimization of Antenna Tilt Angles in LTE Networks , 2015, IEEE/ACM Transactions on Networking.

[6]  Rouzbeh Razavi,et al.  A Fuzzy reinforcement learning approach for self-optimization of coverage in LTE networks , 2010, Bell Labs Technical Journal.

[7]  Henning Sanneck,et al.  LTE Self-Organising Networks (SON): Network Management Automation for Operational Efficiency , 2012 .

[8]  Gerhard Fettweis,et al.  Concurrent Load-Aware Adjustment of User Association and Antenna Tilts in Self-Organizing Radio Networks , 2013, IEEE Transactions on Vehicular Technology.

[9]  Lianfen Huang,et al.  Intelligent coverage optimization with multi-objective genetic algorithm in cellular system , 2013, 2013 8th International Conference on Computer Science & Education.

[10]  Eitan Altman,et al.  Self-organization in wireless networks: A flow-level perspective , 2012, 2012 Proceedings IEEE INFOCOM.

[11]  Rudolf Mathar,et al.  Autonomous Self-Optimization of Coverage and Capacity in LTE Cellular Networks , 2013, IEEE Transactions on Vehicular Technology.

[12]  M. Hata,et al.  Empirical formula for propagation loss in land mobile radio services , 1980, IEEE Transactions on Vehicular Technology.

[13]  Fredrik Athley,et al.  Impact of Electrical and Mechanical Antenna Tilt on LTE Downlink System Performance , 2010, 2010 IEEE 71st Vehicular Technology Conference.

[14]  Gerhard Fettweis,et al.  Online Antenna Tilt-Based Capacity and Coverage Optimization , 2014, IEEE Wireless Communications Letters.

[15]  Honggang Zhang,et al.  Spatial modeling of the traffic density in cellular networks , 2014, IEEE Wireless Communications.

[16]  Xiaohu You,et al.  Joint optimization on load balancing and network load in 3GPP LTE multi-cell networks , 2011, 2011 International Conference on Wireless Communications and Signal Processing (WCSP).

[17]  John A. Nelder,et al.  A Simplex Method for Function Minimization , 1965, Comput. J..

[18]  Andreas Mitschele-Thiel,et al.  Cooperative Fuzzy Q-Learning for self-organized coverage and capacity optimization , 2012, 2012 IEEE 23rd International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC).

[19]  Satoshi Konishi,et al.  Load Balancing Techniques Based on Antenna Tilt and Handover Timing Control , 2013, 2013 IEEE 78th Vehicular Technology Conference (VTC Fall).

[20]  J. Doye,et al.  Global Optimization by Basin-Hopping and the Lowest Energy Structures of Lennard-Jones Clusters Containing up to 110 Atoms , 1997, cond-mat/9803344.

[21]  Anja Klein,et al.  On the potential of traffic driven tilt optimization in LTE-A networks , 2013, 2013 IEEE 24th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).

[22]  Tong Zhang,et al.  Accelerating Stochastic Gradient Descent using Predictive Variance Reduction , 2013, NIPS.

[23]  Vera Stavroulaki,et al.  An opportunistic approach for coverage and capacity optimization in Self-Organizing Networks , 2013, 2013 Future Network & Mobile Summit.

[24]  Mehdi Amirijoo,et al.  Effectiveness of cell outage compensation in LTE networks , 2011, 2011 IEEE Consumer Communications and Networking Conference (CCNC).

[25]  Yiqing Zhou,et al.  Coverage optimization for femtocell clusters using modified particle swarm optimization , 2012, 2012 IEEE International Conference on Communications (ICC).

[26]  Dritan Kaleshi,et al.  International Symposium on Wireless Communication Systems , 2012 .

[27]  Kaamran Raahemifar,et al.  Antenna placement optimization for cellular networks , 2013, 2013 26th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE).

[28]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[29]  Gerhard Fettweis,et al.  Improving coverage and load conditions through joint adaptation of antenna tilts and cell selection rules in mobile networks , 2012, 2012 International Symposium on Wireless Communication Systems (ISWCS).

[30]  Olav N. Østerbø,et al.  Benefits of Self-Organizing Networks (SON) for Mobile Operators , 2012, J. Comput. Networks Commun..

[31]  Gerhard Fettweis,et al.  Joint Downlink and Uplink Tilt-Based Self-Organization of Coverage and Capacity Under Sparse System Knowledge , 2016, IEEE Transactions on Vehicular Technology.

[32]  Sebastian Ruder,et al.  An overview of gradient descent optimization algorithms , 2016, Vestnik komp'iuternykh i informatsionnykh tekhnologii.