IEEE Computational Intelligence Magazine - February 2018 - 60

applied to mobile network optimization. The simplest way to
apply classification systems for performance prediction is to discretize the numerical values used into categories that are meaningful to the application at hand. For example, user data rates
could be discretized based on the bandwidth requirements of
commonly used applications (e.g., video codecs). Often predicting performance in terms of such larger categories is more
robust compared to regression for point estimates [45]. However,
we are not aware of much especially empirical research of the
effectiveness of such methods in applications discussed here.
VII. Conclusions

In this paper, we discussed the application of machine learning
techniques for performance prediction problems in wireless networks. We studied the performance of existing machine learning algorithms for these problems and proposed a simple
categorization of main problem types between spatial, temporal
and multidimensional prediction tasks. Using an extensive realworld drive test data set, we showed that classical machine
learning methods such as Gaussian process regression, exponential smoothing of time series, and random forests can yield very
good prediction results for drive test data. We also discussed the
key challenges for future work, especially with the focus of
practical deployment of machine learning techniques for performance prediction in mobile wireless networks. We are currently working toward integrating prototype implementations
of the proposed mechanism as a part of our radio environment
map work [46].
References

[1] B. Bojovi´c, E. Meshkova, N. Baldo, J. Riihijärvi, and M. Petrova, "Machine learningbased dynamic frequency and bandwidth allocation in self-organized LTE dense small
cell deployments," EURASIP J. Wireless Commun. Netw., vol. 2016, no. 183, pp. 16, Dec.
2016.
[2] R. Razavi, S. Klein, and H. Claussen, "Self-optimization of capacity and coverage
in LTE networks using a fuzzy reinforcement learning approach," in Proc. IEEE Int.
Symp. Personal Indoor Mobile Radio Communications, Istanbul, Turkey, Sept. 2010, pp.
1865-1870.
[3] S. Deb and P. Monogioudis, "Learning-based uplink interference management in 4G
LTE cellular systems," IEEE/ACM Trans. Netw., vol. 23, no. 2, pp. 398-411, Apr. 2015.
[4] A. Attar, V. Krishnamurthy, and O. N. Gharehshiran, "Interference management using cognitive base-stations for UMTS LTE," IEEE Commun. Mag., vol. 49, no. 8, pp.
152-159, Aug. 2011.
[5] O. Sallent, J. Pérez-Romero, R. Ferrús, and R. Agustí, "Learning-based coexistence
for LTE operation in unlicensed bands," in Proc. IEEE Int. Conf. Communications Workshops, London, U.K., June 2015, pp. 2307-2313.
[6] P. Garcia-Teodoro, J. Diaz-Verdejo, G. Maciá-Fernández, and E. Vázquez, "Anomalybased network intrusion detection: Techniques, systems and challenges," Comput. Secur.,
vol. 28, no. 1-2, pp. 18-28, Feb.-Mar. 2009.
[7] S. Shamshirband, N. B. Anuar, M. L. M. Kiah, and A. Patel, "An appraisal and design
of a multi-agent system based cooperative wireless intrusion detection computational intelligence technique," Eng. Appl. Artif. Intell., vol. 26, no. 9, pp. 2105-2127, Oct. 2013.
[8] M. A. Alsheikh, S. Lin, D. Niyato, and H.-P. Tan, "Machine learning in wireless sensor networks: Algorithms, strategies, and applications," IEEE Commun. Surv. Tutorial, vol.
16, no. 4, pp. 1996-2018, Apr. 2014.
[9] S. Rajasegarar, C. Leckie, and M. Palaniswami, "Anomaly detection in wireless sensor
networks," IEEE Wireless Commun., vol. 15, no. 4, pp. 34-40, Aug. 2008.
[10] D. Janakiram, V. Reddy, and A. P. Kumar, "Outlier detection in wireless sensor networks using Bayesian belief networks," in Proc. Communication System Software Middleware,
New Delhi, India, Jan. 2006, pp. 1-6.
[11] J. Mitola, "Cognitive radio: An integrated agent architecture for software defined
radio," Ph.D. dissertation, KTH Royal Inst. Technol., 2000.
[12] S.-H. Liao, "Expert system methodologies and applications: A decade review from
1995 to 2004," Expert Syst. Appl., vol. 28, no. 1, pp. 93-103, Jan. 2005.
[13] G.-B. Huang, Q.-Y. Zhu, and C.-K. Siew, "Extreme learning machine: Theory and
applications," Neurocomputing, vol. 70, no. 1-3, pp. 489-501, Dec. 2006.
[14] C. Clancy, J. Hecker, E. Stuntebeck, and T. O'Shea, "Applications of machine learning
to cognitive radio networks," IEEE Wireless Commun., vol. 14, no. 4, pp. 47-52, Aug. 2007.

60

ieee CompUtAtionAl intelligenCe mAgAzine | FeBRUARY 2018

[15] P. Mähönen, M. Petrova, J. Riihijärvi, and M. Wellens, "Cognitive wireless networks: Your network just became a teenager," in Proc. IEEE INFOCOM, Barcelona,
Spain, Apr. 2006.
[16] R. W. Thomas, D. H. Friend, L. A. Dasilva, and A. B. Mackenzie, "Cognitive networks: Adaptation and learning to achieve end-to-end performance objectives," IEEE
Commun. Mag., vol. 44, no. 12, pp. 51-57, Dec. 2006.
[17] M. Zorzi, A. Zanella, A. Testolin, M. D. F. De Grazia, and M. Zorzi, "Cognitionbased networks: A new perspective on network optimization using learning and distributed intelligence," IEEE Access, vol. 3, pp. 1512-1530, Aug. 2015.
[18] J. van de Beek, T. Cai, S. Grimoud, B. Sayrac, P. Mähönen, J. Nasreddine, and J. Riihijärvi, "How a layered REM architecture brings cognition to today's mobile networks,"
IEEE Wireless Commun., vol. 19, no. 4, pp. 17-24, Aug. 2012.
[19] J. Perez-Romero, A. Zalonis, L. Boukhatem, A. Kliks, K. Koutlia, N. Dimitriou,
and R. Kurda, "On the use of radio environment maps for interference management in
heterogeneous networks," IEEE Commun. Mag., vol. 53, no. 8, pp. 184-191, Aug. 2015.
[20] J. Friedman, T. Hastie, and R. Tibshirani, The Elements of Statistical Learning. Springer
Series in Statistics, vol. 1. Springer: Berlin, 2001.
[21] C. E. Rasmussen and C. K. I. Williams, Gaussian Processes for Machine Learning. Cambridge, MA: MIT Press, 2006.
[22] N. Cressie and C. K. Wikle, Statistics for Spatio-Temporal Data. New York: John Wiley,
2015.
[23] A. Achtzehn, J. Riihijärvi, and P. Mähönen, "Improving accuracy for TVWS geolocation databases: Results from measurement-driven estimation approaches," in Proc. IEEE
DySPAN, McLean, VA, Apr. 2014, pp. 392-403.
[24] H. Braham, S. B. Jemaa, G. Fort, E. Moulines, and B. Sayrac, "Fixed rank kriging
for cellular coverage analysis," IEEE Trans. Veh. Technol., vol. 66, no. 5, pp. 4212-4222,
May 2017.
[25] T. J. Hastie and R. J. Tibshirani, Generalized Additive Models. Boca Raton, FL: CRC
Press, 1990.
[26] S. J. Taylor and B. Letham. (2017). Forecasting at scale. Facebook Research Tech.
Rep. [Online]. Available: https://facebookincubator.github.io/prophet/static/prophet_
paper_ 20170113.pdf
[27] C. C. Holt, "Forecasting seasonals and trends by exponentially weighted moving
averages," DTIC, Tech. Rep. AD0127522, Apr. 1957.
[28] P. R. Winters, "Forecasting sales by exponentially weighted moving averages," Manage. Sci., vol. 6, no. 3, pp. 324-342, Apr. 1960.
[29] S. Makridakis, S. C. Wheelwright, and R. J. Hyndman, Forecasting Methods and Applications. New York: Wiley, 2008.
[30] R. J. Hyndman, A. B. Koehler, R. D. Snyder, and S. Grose, "A state space framework
for automatic forecasting using exponential smoothing methods," Int. J. Forecasting, vol.
18, no. 3, pp. 439-454, July-Sept. 2002.
[31] R. J. Hyndman and Y. Khandakar, "Automatic time series for forecasting: The forecast package for R," J. Statist. Softw., vol. 27, no. 3, pp. 1-22, July 2008.
[32] L. Breiman, "Random forests," Mach. Learn., vol. 45, no. 1, pp. 5-32, Oct. 2001.
[33] A. Liaw and M. Wiener, "Classification and regression by RandomForest," R. News,
vol. 2, no. 3, pp. 18-22, Dec. 2002.
[34] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553,
pp. 436-444, May 2015.
[35] C. Phillips, D. Sicker, and D. Grunwald, "Bounding the error of path loss models,"
in Proc. IEEE DySPAN, Aachen, Germany, May 2011, pp. 71-82.
[36] C. Phillips, D. Sicker, and D. Grunwald, "A survey of wireless path loss prediction and coverage mapping methods," IEEE Commun. Surv. Tutorials, vol. 15, no. 1, pp.
255-270, Mar. 2013.
[37] A. Molisch, L. Greenstein, and M. Shafi, "Propagation issues for cognitive radio,"
Proc. IEEE, vol. 97, no. 5, pp. 787-804, May 2009.
[38] J. Nasreddine, J. Riihijärvi, and P. Mähönen, "Transmit power control for secondary
use in environments with correlated shadowing," in Proc. IEEE Int. Conf. Communications,
Kyoto, Japan, June 2011.
[39] R. Fraile, J. Monserrat, N. Cardona, and J. Nasreddine, "Impact of slow fading modelling on TD-CDMA system-level simulations," in Proc. Int. Symp. Wireless Communication Systems, Valencia, Spain, Sept. 2006.
[40] M. Sooriyabandara, T. Farnham, P. Mähönen, M. Petrova, J. Riihijärvi, and Z.
Wang, "Generic interface architecture supporting cognitive resource management in
future wireless networks," IEEE Commun. Mag., vol. 49, no. 9, pp. 103-113, Sept. 2011.
[41] S. Rosen, S.-J. Lee, J. Lee, P. Congdon, Z. M. Mao, and K. Burden, "MCNet:
Crowdsourcing wireless performance measurements through the eyes of mobile devices,"
IEEE Commun. Mag., vol. 52, no. 10, pp. 86-91, Oct. 2014.
[42] W. A. Hapsari, A. Umesh, M. Iwamura, M. Tomala, B. Gyula, and B. Sebire, "Minimization of drive tests solution in 3GPP," IEEE Commun. Mag., vol. 50, no. 6, pp. 28-36,
June 2012.
[43] T. Condie, P. Mineiro, N. Polyzotis, and M. Weimer, "Machine learning on big
data," in Proc. IEEE Int. Conf. Data Engineering, Brisbane, Australia, Apr. 2013, pp.
1242-1244.
[44] S. Shalev-Shwartz, "Online learning and online convex optimization," Found. Trends
Mach. Learn., vol. 4, no. 2, pp. 107-194, Feb. 2012.
[45] S. Fidler, D. Skocaj, and A. Leonardis, "Combining reconstructive and discriminative subspace methods for robust classification and regression by subsampling," IEEE
Trans. Pattern Anal. Mach. Intell., vol. 28, no. 3, pp. 337-350, Mar. 2006.
[46] A. Achtzehn, J. Riihijärvi, I. A. Barriía Castillo, M. Petrova, and P. Mähönen,
"CrowdREM: Harnessing the power of the mobile crowd for f lexible wireless network
monitoring," in Proc. ACM HotMobile Conf., Santa Fe, NM, Feb. 2015, pp. 63-68.
https://facebookincubator.github.io/prophet/static/prophet_paper_ 20170113.pdf https://facebookincubator.github.io/prophet/static/prophet_paper_ 20170113.pdf
Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - February 2018
