IEEE Computational Intelligence Magazine - February 2020 - 33

Abstract-With the exponential expansion of the number of
Internet of Things (IoT) devices, many state-of-the-art communication technologies are being developed to use the lowerpower but extensively deployed devices. Due to the limits of
pure channel characteristics, most protocols cannot allow an
IoT network to be simultaneously large-scale and energyefficient, especially in hybrid architectures. However, different
from the original intention to pursue faster and broader connectivity, the daily operation of IoT devices only requires stable
and low-cost links. Thus, our design goal is to develop a comprehensive solution for intelligent green IoT networking to satisfy the modern requirements through a data-driven mechanism,
so that the IoT networks use computational intelligence to realize self-regulation of composition, size minimization, and
throughput optimization. To the best of our knowledge, this
study is the first to use the green protocols of LoRa and ZigBee
to establish an ad hoc network and solve the problem of energy
efficiency. First, we propose a unique initialization mechanism
that automatically schedules node clustering and throughput
optimization. Then, each device executes a procedure to
manage its own energy consumption to optimize switching in
and out of sleep mode, which relies on AI-controlled service
usage habit prediction to learn the future usage trend. Finally,
our new theory is corroborated through real-world deployment
and numerical comparisons. We believe that our new type of
network organization and control system could improve the
performance of all green-oriented IoT services and even
change human lifestyle habits.

I. Introduction

he Internet of Things (IoT) helps make the world
that people live in smarter than ever. The world residents living in are incessantly, automatically and collaboratively provided with customized urban services.
The IoT contains smart devices (e.g., wearable devices and
mobile devices), communication technologies (e.g., wireless ad
hoc networks (WANETs) and next-generation cellular networks (NGCNs)), computation infrastructure (e.g., cloud
computing and edge computing) and other core technologies
(identification, sensing and personalization) [1]. In addition to
facilitating these common smart services in daily life, IoT
devices have appeared in various environments. Such devices
create new industry forms, such as smart healthcare, smart
grids [2], and the overall concept of smart cities, and influence
home appliances [3], food supply chains and industrial automation [4]. In contrast to the pure IoT networks' need for
faster speeds and larger bandwidth, these daily services also
need stable, low-cost, and energy-efficient links [5]. The green
IoT initiatives are motivated by the full deployment and everincreasing demand for smart services when the power and
bandwidth rather than resources are considered as a limitation
[6]. Therefore, new energy-efficient procedures (whether
involving hardware or software) should be adopted during the
design of IoT services to eliminate waste of resources or
reduce the impact of the green procedure itself.

To make the IoT 'green,' more state-of-the-art technologies
are being considered and newly developed to fulfill the demand
of these energy-hungry devices [7]. Regarding wireless communications, numerous IoT nodes are intelligently assembled to
transfer the data collected during monitoring [8]. A base station
(BS) provides various applications to facilitate communications,
and it intelligently adjusts power, controls the protocols, schedules activities and allocates resources. These tasks are easily performed by technologies in the D2D domain. However, most of
these tasks consume more energy than a simple IoT device can
afford. According to a survey of all the wireless communication
technologies [9], medium-distance connectivity methods (e.g.,
IEEE 802.15.4) are good choices for meeting the demand for
low-rate and low-power communications [10] usually involved
in smart parking and tracking. Furthermore, long-distance
machine-type communications (e.g., LoRa and eMTC) are
areas of new and ongoing research. They represent effective
approaches applied in meters and public facilities.
Nevertheless, if we use medium-distance communication
technologies (such as ZigBee) to form a WANET, the actual
delay over multiple hops will significantly increase, and
throughput will decline [11]; many redundant packets will have
to be generated to achieve a higher delivery rate, which will
lead to additional energy consumption. On the other hand, the
pure low-power wide-area (LPWA) protocols could avoid the
above multiple relaying problems, but the unlicensed narrow
band significantly limits the performance of multinode networks [12]. In addition to the throughput limitation of the
node-to-BS model, the IoT nodes have to perform long-term
channel detection and endure single-channel interference,
which necessitates more sophisticated power control mechanisms [13]. The green initiatives are still in their infancy, and
many theoretical and practical issues need to be thoroughly discussed; such analyses will boost the development of large-scale
and hybrid IoT networks.
To the best of our knowledge, there is no design for a combined network of LoRa and ZigBee that addresses the energy
efficiency issues in IoT devices. We take advantage of suitable
distance and channel characteristics of both communications
technologies to form a QoS-aware green IoT network. The
design goal of this article is to develop a comprehensive design
of a new type of green IoT network. The study contains node
clustering, throughput optimization, energy self-management,
and a field test. This network reduces the energy consumption
of the IoT network while at the same time guaranteeing QoS.
We believe that our new type of IoT node organization and
control process could improve the performance of green-oriented IoT services and even change people's service use habits.
The remainder of this paper is organized as follows. Section II discusses the motivation of our design and introduces
the organization of LoRa and ZigBee in a moderate-density
IoT network. Afterwards, the implementation of multiplehop data transmission is discussed. Section III gives a brief
overview of the solution regarding the problem of organizing a large number of nodes, and a center node selection

FEBRUARY 2020 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE

IEEE Computational Intelligence Magazine - February 2020

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - February 2020