DOI: 10.1002/ett.

3757

EDITORIAL

Machine Learning/AI for IoT, M2M, and Computer

Communication

Recent advances in machine learning (ML) have been endorsed by other computing areas, such as robotics and speech
recognition. However, the application of ML techniques on networking is recently getting momentum and requires com-
prehensive investigation. Currently, most networking algorithms and policies work on predefined sets of rules, which are
not efficient nor adaptive to support the explosive traffic demand due to the growth of smart devices and new services
such as Internet of Things (IoTs) and machine-to-machine (M2M). On the contrary, ML techniques can learn from data
to solve application-specific and generic network issues.
Artificial intelligence (AI), specifically ML, has been one of the driving forces in developing 5G systems. However, a
true intelligent and fully automated network system will only be realized in future systems such as beyond 6G. Therefore,
it is crucial to investigate the potential applications of AI/ML algorithms in computer communication and networks.
An intelligent network demands smart decision making everywhere from edge to core networks and in between. While
software-defined networking (SDN) can utilize ML techniques to enhance routing, hosts can use ML for analyzing big
data such that cloud computing is enhanced. Access networks and smart devices can apply ML to improve transmission
ranges and energy efficiency. In addition, new network services networks such as IoTs, M2Ms, vehicle-to-vehicles (V2Vs),
as well as networks such as cognitive radio networks (CRNs) and mobile ad hoc networks (MANETs) can use advanced
ML algorithms to improve communication. This special issue (SI) on “Machine learning/AI for IoT, M2M, and computer
communication” is an attempt to solicit new proposals on the issues discussed above.
In this SI, the first two papers investigate the use of ML in beamforming for directional communication. The first paper,
“Supervised learning and graph signal processing strategies for beam tracking in highly directional mobile communi-
cations” by Ortega et al,1 applies supervised learning to identify historically similar samples and predict channel state.
Then K-nearest neighbor (K-NN) is used to reduce beam search spaces such that feedback channel usage is improved.
The second paper, “On the machine learning-based smart beamforming for wireless virtualization with large-scale MIMO
system” by Sapavath et al,2 takes advantage of ML to lease portions of large-scale MIMO to mobile virtual operators in a
bid to increase spectrum utilizations.
The next four papers discuss the use of ML for infrastractureless self-organizing ad hoc networks. The third paper,
“Improving positioning accuracy based on self-organizing map and inter-vehicular communication” by Ma et al,3 applies
an ML technique named constrained self-organizing map (C-SOM) to estimate a location with lower error in a V2V net-
work. Each vehicle collects location though GPS and ranging sensors, then this information is shared among neighbor
vehicles to be used for accurate estimation of location. Khuntia and Hazra4 took leverage of deep q-learning to propose a
resource sharing scheme to solve channel and power allocation problems in article 4, “An efficient Deep reinforcement
learning with extended Kalman filter for device-to-device communication under-laying cellular network.” Another use
of q-learning with reinforcement learning to enhance connectivity in D2D is stated in the fifth paper, “Reinforcement
learning algorithm for 5G indoor device-to-device communications” by Sreedevi and Rama Rao.5 The sixth paper, “Effi-
cient artificial fish swarm based clustering approach on mobility aware energy-efficient for MANET” by Gupta et al,6 uses
artificial fish swarm optimization to increase the network lifetime of the network.
The following four papers deal with the application of ML in SDN and network virtualization. Benayas et al7 presented
autonomous SDN fault management platform that uses Bayesian diagnosis in the seventh paper, “A semantic data lake
framework for autonomous fault management in SDN environments.” Article 8, “Adapting reinforcement learning for
multimedia transmission on SDN,” by Rego et al,8 enhances decisions of OpenFlow and guarantees QoS by applying
ML on multimedia transmissions of SDN. The ninth paper, by Zangiabady et al,9 “Self-adaptive online virtual network
migration in network virtualization environments,” leverages reinforcement learning to minimize the cost associated with
migrating virtual network resource from one to another. The 10th paper, “A rate-maximizing spectrum sharing algorithm
for cognitive radio networks with generic resource constraints” by Halloush et al,10 proposed a spectrum sharing scheme
for CRNs that is compatible with SDN and reduces computational complexity by applying distributed computing.

Trans Emerging Tel Tech. 2019;30:e3757. wileyonlinelibrary.com/journal/ett © 2019 John Wiley & Sons, Ltd. 1 of 3
https://doi.org/10.1002/ett.3757
2 of 3 EDITORIAL

Here we discuss two papers that use ML algorithms in cloud computing issues. Article 11, by Al-Jarrah et al,11
“Integrated network and hosts energy management for cloud data centers,” proposes a network-aware virtual machine
reallocation algorithm to conserve energy of servers and switches. Shenbaga Moorthy and Pabitha,12 in the 12th paper,
“Optimal provisioning and scheduling of analytics as a service in cloud computing,” apply particle swarm optimization
in a bid to optimize execution time, cost, and other resources during analytic request.
Data-driven networking issues together with ML techniques are investigated by the next two papers. Kong et al13 sug-
gested the use of long short-term ML algorithm on big data to predict traffic flow in the 13th paper, “Big data-driven
machine learning-enabled traffic flow prediction.” Article 14, “Fuzzy logic-based efficient interest forwarding (FLEIF)
in named data networking” by Qureshi et al,14 proposes a neuro-fuzzy logic technique to reduce congestion caused by
sending multiple interest packet in Named Data Networking (NDN).
The applications of ML on IoT are demonstrated in the following two papers. The 15th paper, “Performance analysis of
machine learning and deep learning classification methods for indoor localization in Internet of Things environment” by
Turgut et al,15 compares the indoor localization accuracy of ML with deep learning classification algorithms where the
dataset is collected from IoT devices. The 16th paper, “Design and implementation of music teaching assistant platform
based on Internet of Things” by Li,16 uses immune algorithm to improve energy consumption of router nodes in IoT of
music teaching assistant platform.
The rest of the papers discuss other networking aspects of ML. Article 17, “The evaluation and application of node influ-
ence in dynamic networks based on evolving communities” by Huang et al,17 applies learning-based ML to rank local
experts. The 18th paper, “Local experts finding using user comments in location-based social networks” by Cao et al,18
proposed an unsupervised ML algorithm to detect overlapping community in self-organizing networks. Choudhury et al19
proposed convolutional neural network (CNN)–based learning to differentiate hard negative backgrounds from true
pedestrian on the 19th paper, “Scale aware deep pedestrian detection.”
First and foremost, we would like to thank the Transactions on Emerging Telecommunications Technologies journal and
its current editor-in-chief Luigi Alfredo Grieco for his unrelenting giving us the chance to organize this SI and relentless
assistance. We would also like to extend our gratitude the authors, reviewers, and editors for their contribution to this SI.
The list and order of papers in this SI are presented in the reference list.

Zelalem Jembre Yalew1

Syed Hassan Ahmed2
Young-June Choi3
Jaime Lloret4
1 Department of Electronic Engineering, Keimyung University, Daegu, South Korea

2 Department of Computer Science, University of Central Florida, Orlando, Florida

3 Department of Information and Computer Engineering, Ajou University, Suwon, South Korea
4 Department of Communications, Polytechnic University of Valencia, València, Spain
Correspondence
Zelalem Jembre Yalew, Keimyung University, Daegu 42601, South Korea.
Email: zizutg@kmu.ac.kr

REFERENCES
1. Ortega YR, Guerreiro IM, Hui D, Cavalcante CC, Cavalcanti FRP. Supervised learning and graph signal processing strategies for beam
tracking in highly directional mobile communications. Trans Emerging Tel Tech. 2019;e3687. https://doi.org/10.1002/ett.3687
2. Sapavath NN, Safavat S, Rawat DB. On the machine learning-based smart beamforming for wireless virtualization with large-scale MIMO
system. Trans Emerging Tel Tech. 2019;e3713. https://doi.org/10.1002/ett.3713
3. Ma S, Lee HJ. Improving positioning accuracy based on self-organizing map and inter-vehicular communication. Trans Emerging Tel Tech.
2019;e3733. https://doi.org/10.1002/ett.3733
4. Khuntia P, Hazra R. An efficient Deep reinforcement learning with extended Kalman filter for device-to-device communication under-
laying cellular network. Trans Emerging Tel Tech. 2019;e3671. https://doi.org/10.1002/ett.3671
5. Sreedevi AG, Rama Rao T. Reinforcement learning algorithm for 5G indoor device-to-device communications. Trans Emerging Tel Tech.
2019;e3670. https://doi.org/10.1002/ett.3670
EDITORIAL 3 of 3

6. Gupta D, Khanna A, Lakshmanaprabu SK, Shankar K, Furtado V, Rodrigues JJPC. Efficient artificial fish swarm based clustering approach
on mobility aware energy-efficient for MANET. Trans Emerging Tel Tech. 2019;e3524. https://doi.org/10.1002/ett.3524
7. Benayas F, Carrera Á, García-Amado M, Iglesias CA. A semantic data lake framework for autonomous fault management in SDN
environments. Trans Emerging Tel Tech. 2019;e3629. https://doi.org/10.1002/ett.3629
8. Rego A, Sendra S, Garcia L, Lloret J. Adapting reinforcement learning for multimedia transmission on SDN. Trans Emerging Tel Tech.
2019;e3643. https://doi.org/10.1002/ett.3643
9. Zangiabady M, Garcia-Robledo A, Gorricho J-L, Serrat-Fernandez J, Rubio-Loyola J. Self-adaptive online virtual network migration in
network virtualization environments. Trans Emerging Tel Tech. 2019;e3692. https://doi.org/10.1002/ett.3692
10. Halloush R, Halloush M, Almalkawi I, Musa A, Bany Salameh H. A rate-maximizing spectrum sharing algorithm for cognitive radio
networks with generic resource constraints. Trans Emerging Tel Tech. 2019;e3602. https://doi.org/10.1002/ett.3602
11. Al-Jarrah O, Al-Zoubi Z, Jararweh Y. Integrated network and hosts energy management for cloud data centers. Trans Emerging Tel Tech.
2019;e3641. https://doi.org/10.1002/ett.3641
12. Shenbaga Moorthy R, Pabitha P. Optimal provisioning and scheduling of analytics as a service in cloud computing. Trans Emerging Tel
Tech. 2019;e3609. https://doi.org/10.1002/ett.3609
13. Kong F, Li J, Jiang B, Zhang T, Song H. Big data-driven machine learning-enabled traffic flow prediction. Trans Emerging Tel Tech.
2019;e3482. https://doi.org/10.1002/ett.3482
14. Qureshi Z, Shah PA, Hassan N, et al. Fuzzy logic-based efficient interest forwarding (FLEIF) in named data networking. Trans Emerging
Tel Tech. 2019;e3577. https://doi.org/10.1002/ett.3577
15. Turgut Z, Ustebay S, Aydin MA, Aydin ZG, Sertbaş A. Performance analysis of machine learning and deep learning classification methods
for indoor localization in Internet of Things environment. Trans Emerging Tel Tech. 2019;e3705. https://doi.org/10.1002/ett.3705
16. Li W. Design and implementation of music teaching assistant platform based on Internet of Things. Trans Emerging Tel Tech. 2019;e3606.
https://doi.org/10.1002/ett.3606
17. Huang Y, Xi H, Shah SC, Ye F. The evaluation and application of node influence in dynamic networks based on evolving communities.
Trans Emerging Tel Tech. 2019;e3556. https://doi.org/10.1002/ett.3556
18. Cao J, Yang Y, Cao B. Local experts finding using user comments in location-based social networks. Trans Emerging Tel Tech. 2019;e3600.
https://doi.org/10.1002/ett.3600
19. Choudhury SK, Padhy RP, Sangaiah AK, Sa PK, Muhammad K, Bakshi S. Scale aware deep pedestrian detection. Trans Emerging Tel Tech.
2019;e3522. https://doi.org/10.1002/ett.3522