{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T15:52:19Z","timestamp":1783007539158,"version":"3.54.5"},"reference-count":41,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2018,11,5]],"date-time":"2018-11-05T00:00:00Z","timestamp":1541376000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>With the dramatic increase of connected devices, the Internet of things (IoT) paradigm has become an important solution in supporting dense scenarios such as smart cities. The concept of heterogeneous networks (HetNets) has emerged as a viable solution to improving the capacity of cellular networks in such scenarios. However, achieving optimal load balancing is not trivial due to the complexity and dynamics in HetNets. For this reason, we propose a load balancing scheme based on machine learning techniques that uses both unsupervised and supervised methods, as well as a Markov Decision Process (MDP). As a use case, we apply our scheme to enhance the capabilities of an urban IoT network operating under the LoRaWAN standard. The simulation results show that the packet delivery ratio (PDR) is increased when our scheme is utilized in an unbalanced network and, consequently, the energy cost of data delivery is reduced. Furthermore, we demonstrate that better outcomes are attained when some techniques are combined, achieving a PDR improvement of up to about 50% and reducing the energy cost by nearly 20% in a multicell scenario with 5000 devices requesting downlink traffic.<\/jats:p>","DOI":"10.3390\/s18113779","type":"journal-article","created":{"date-parts":[[2018,11,5]],"date-time":"2018-11-05T10:43:45Z","timestamp":1541414625000},"page":"3779","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":44,"title":["Machine Learning Aided Scheme for Load Balancing in Dense IoT Networks"],"prefix":"10.3390","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2666-947X","authenticated-orcid":false,"given":"Cesar A.","family":"Gomez","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Abdallah","family":"Shami","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4890-0748","authenticated-orcid":false,"given":"Xianbin","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2018,11,5]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1109\/MCOM.2012.6211486","article-title":"Heterogeneous cellular networks: From theory to practice","volume":"50","author":"Ghosh","year":"2012","journal-title":"IEEE Commun. Mag."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"136","DOI":"10.1109\/MCOM.2013.6476878","article-title":"Seven ways that HetNets are a cellular paradigm shift","volume":"51","author":"Andrews","year":"2013","journal-title":"IEEE Commun. Mag."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1109\/MWC.2014.6812287","article-title":"An overview of load balancing in hetnets: Old myths and open problems","volume":"21","author":"Andrews","year":"2014","journal-title":"IEEE Wirel. Commun."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1018","DOI":"10.1109\/COMST.2016.2516538","article-title":"User Association in 5G Networks: A Survey and an Outlook","volume":"18","author":"Liu","year":"2016","journal-title":"IEEE Commun. Surv. Tutor."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"e3192","DOI":"10.1002\/ett.3192","article-title":"A distributed learning\u2013based user association for heterogeneous networks","volume":"28","author":"Arani","year":"2017","journal-title":"Trans. Emerg. Telecommun. Technol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"2522","DOI":"10.1109\/COMST.2016.2571730","article-title":"Ultra-Dense Networks: A Survey","volume":"18","author":"Kamel","year":"2016","journal-title":"IEEE Commun. Surv. Tutor."},{"key":"ref_7","unstructured":"De Silva, J.C., Rodrigues, J.J.P.C., Alberti, A.M., Solic, P., and Aquino, A.L.L. (2017, January 12\u201314). LoRaWAN\u2014A low power WAN protocol for Internet of Things: A review and opportunities. Proceedings of the 2017 2nd International Multidisciplinary Conference on Computer and Energy Science (SpliTech), Split, Croatia."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"855","DOI":"10.1109\/COMST.2017.2652320","article-title":"Low Power Wide Area Networks: An Overview","volume":"19","author":"Raza","year":"2017","journal-title":"IEEE Commun. Surv. Tutor."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1109\/MWC.2016.7721743","article-title":"Long-range communications in unlicensed bands: The rising stars in the IoT and smart city scenarios","volume":"23","author":"Centenaro","year":"2016","journal-title":"IEEE Wirel. Commun."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"34","DOI":"10.1109\/MCOM.2017.1600613","article-title":"Understanding the Limits of LoRaWAN","volume":"55","author":"Adelantado","year":"2017","journal-title":"IEEE Commun. Mag."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1186\/1687-1499-2013-61","article-title":"Cell range expansion using distributed Q-learning in heterogeneous networks","volume":"2013","author":"Kudo","year":"2013","journal-title":"EURASIP J. Wirel. Commun. Netw."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2706","DOI":"10.1109\/TWC.2013.040413.120676","article-title":"User Association for Load Balancing in Heterogeneous Cellular Networks","volume":"12","author":"Ye","year":"2013","journal-title":"IEEE Trans. Wirel. Commun."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1334","DOI":"10.1109\/TVT.2016.2560900","article-title":"Gibbs-Sampling-Based CRE Bias Optimization Algorithm for Ultradense Networks","volume":"66","author":"Jiang","year":"2017","journal-title":"IEEE Trans. Veh. Technol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2417","DOI":"10.1109\/TWC.2018.2794983","article-title":"Poisson Cluster Process Based Analysis of HetNets with Correlated User and Base Station Locations","volume":"17","author":"Afshang","year":"2018","journal-title":"IEEE Trans. Wirel. Commun."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"633","DOI":"10.1109\/TVT.2017.2748570","article-title":"Load-Balancing Scheme with Small-Cell Cooperation for Clustered Heterogeneous Cellular Networks","volume":"67","author":"Park","year":"2018","journal-title":"IEEE Trans. Veh. Technol."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1999","DOI":"10.1109\/LCOMM.2016.2571287","article-title":"Joint Effect of User Activity Sensing and Biased Cell Association in Energy Efficient HetNets","volume":"20","author":"Fan","year":"2016","journal-title":"IEEE Commun. Lett."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"16469","DOI":"10.1109\/ACCESS.2017.2722007","article-title":"Optimization of Resource Allocation and User Association for Energy Efficiency in Future Wireless Networks","volume":"5","author":"Yang","year":"2017","journal-title":"IEEE Access"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"5241","DOI":"10.1109\/TVT.2016.2614696","article-title":"Cell Association with Load Balancing in Nonuniform Heterogeneous Cellular Networks: Coverage Probability and Rate Analysis","volume":"66","author":"Muhammad","year":"2017","journal-title":"IEEE Trans. Veh. Technol."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Sun, Y., Xia, W., Zhang, S., Wu, Y., Wang, T., and Fang, Y. (2018). Energy Efficient Pico Cell Range Expansion and Density Joint Optimization for Heterogeneous Networks with eICIC. Sensors, 18.","DOI":"10.3390\/s18030762"},{"key":"ref_20","unstructured":"LoRa Alliance (2017). Technical Committee LoRaWANTM 1.1 Specification, LoRa Alliance."},{"key":"ref_21","unstructured":"LoRa Alliance (2015). Technical Marketing Group LoRaWAN, What Is It? A Technical Review of LoRa and LoRaWAN, LoRa Alliance."},{"key":"ref_22","unstructured":"(2018, July 16). Building a Global Internet of Things Network Together. Available online: https:\/\/www.thethingsnetwork.org\/."},{"key":"ref_23","unstructured":"(2018, August 07). LoRaWAN Gateways. Available online: https:\/\/tektelic.com\/iot\/lorawan-gateways\/."},{"key":"ref_24","unstructured":"(2018, July 17). TTN Mapper. Available online: https:\/\/ttnmapper.org\/."},{"key":"ref_25","unstructured":"Bishop, C. (2006). Pattern Recognition and Machine Learning, Springer."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Marsland, S. (2015). Machine Learning: An Algorithmic Perspective, CRC Press. [2nd ed.].","DOI":"10.1201\/b17476"},{"key":"ref_27","unstructured":"(2018, July 25). Network Architecture. Available online: https:\/\/www.thethingsnetwork.org\/docs\/network\/architecture.html."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Pop, A., Raza, U., Kulkarni, P., and Sooriyabandara, M. (2017, January 4\u20138). Does Bidirectional Traffic Do More Harm Than Good in LoRaWAN Based LPWA Networks?. Proceedings of the 2017 IEEE Global Communications Conference, Singapore.","DOI":"10.1109\/GLOCOM.2017.8254509"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Rogers, S., and Girolami, M. (2012). A First Course in Machine Learning, CRC Press.","DOI":"10.1201\/9781466506299"},{"key":"ref_30","unstructured":"Murphy, K.P. (2012). Machine Learning: A Probabilistic Perspective, MIT Press."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1016\/j.ipm.2009.03.002","article-title":"A systematic analysis of performance measures for classification tasks","volume":"45","author":"Sokolova","year":"2009","journal-title":"Inf. Process. Manag."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Van Otterlo, M., and Wiering, M. (2012). Reinforcement Learning and Markov Decision Processes. Reinforcement Learning, Springer.","DOI":"10.1007\/978-3-642-27645-3_1"},{"key":"ref_33","unstructured":"Russell, S., and Norvig, P. (2016). Artificial Intelligence: A Modern Approach, Pearson."},{"key":"ref_34","unstructured":"Sheskin, T.J. (2011). Markov Chains and Decision Processes for Engineers and Managers, CRC Press."},{"key":"ref_35","unstructured":"Tanenbaum, A.S., and Wetherall, D.J. (2012). Computer Networks, Pearson Higher."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1830","DOI":"10.1109\/JIOT.2018.2815150","article-title":"Improving Reliability and Scalability of LoRaWANs through Lightweight Scheduling","volume":"5","author":"Reynders","year":"2018","journal-title":"IEEE Internet Things J."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Bor, M.C., Roedig, U., Voigt, T., and Alonso, J.M. (2016, January 13\u201317). Do LoRa Low-Power Wide-Area Networks Scale?. Proceedings of the 19th ACM International Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems, Valletta, Malta.","DOI":"10.1145\/2988287.2989163"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Casals, L., Mir, B., Vidal, R., and Gomez, C. (2017). Modeling the Energy Performance of LoRaWAN. Sensors, 17.","DOI":"10.3390\/s17102364"},{"key":"ref_39","unstructured":"(2018, July 27). Python Data Analysis Library\u2014Pandas. Available online: https:\/\/pandas.pydata.org\/index.html."},{"key":"ref_40","first-page":"2825","article-title":"Scikit-learn: Machine Learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"916","DOI":"10.1111\/ecog.00888","article-title":"MDPtoolbox: A multi-platform toolbox to solve stochastic dynamic programming problems","volume":"37","author":"Chapron","year":"2014","journal-title":"Ecography"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/18\/11\/3779\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T15:28:03Z","timestamp":1760196483000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/18\/11\/3779"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,11,5]]},"references-count":41,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2018,11]]}},"alternative-id":["s18113779"],"URL":"https:\/\/doi.org\/10.3390\/s18113779","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,11,5]]}}}