IEEE Consumer Electronics Magazine - January/February 2023 - 44

An SDN-Based Framework in UAV Networks
due to the high dynamics of UAVs. Extensive simulation results show that the traffic load
is distributed evenly on multiple links connecting different systems with less battery
consumption.
& IN RECENT YEARS, unmanned aerial vehicles
(UAVs) attracted significant interest and research
in consumer and military domains.1 They can provide
disaster warnings on time, speed up recovery
and rescue operations, and carry medical supplies
in case of adisasterorthe absence of public communication.
Public safety, police, and transportation
management are public uses. Similarly, in the
military, these elements are deployed to perform
operations, such as surveillance, battlefield inspection,
andmapping of inaccessible areas. Compared
to a single UAV system, multiple UAV systems are
more reliable due to redundancy and survivability
to efficiently completeatask. ThesemultipleUAVs
form a UAV network (UAVNet),2 which introduces
issues, such as uneven load on wireless links, battery
limitations, frequent changes in topology, and
high dynamics. To overcome these issues, a centralized
and programmable solution is required.
Software-defined networking (SDN)3 is a good candidate
that separates control logic from data plane
elements and provides high-level programming
abstractions. Data plane elements become simple
forwarding nodes, whereas decision-making is
shifted to a centralized control plane. Data plane
updates control plane by using well-defined application
programmable interfaces.4 In response to
this information, a global view is generated, and forwarding
rules are pushed by the control plane. In
this article, considering the abovementioned problems,
a dynamic SDN framework in UAV communication
is proposed for path selection based on the
UAV elements' link load and battery level. Moreover,
a flight control mechanism is proposed. The
main contributions of this article are as follows.
A novel load balancing algorithm is proposed
for path selection based on link load and battery
level of UAV elements.
It uses OpenFlow statistics to dynamically
adjust the paths in case of congestion, link
failure, and/or availability of a better path.
The battery consumption of UAV elements is
reduced by shifting the load on elements
with high battery levels and validating the
44
effectiveness of the proposed approach with
an example.
The proposed solution takes decisions based
on the link load and battery level. It also
takes heuristic decisions, but the objective is
to select the optimal path.
It handles the high dynamics of UAV flights
and uses the programmability of SDN to
avoid collisions.
The rest of this article is organized as follows.
The " Related Work " section describes the
related work. Then, the " Background and Problem
Statement " section presents background
and problem statement, followed by the " System
Overview " section. After that, the performance
evaluation and analysis are described in the
" Performance Evaluation and Analysis " section.
Finally, the conclusion and future work are discussed
in the " Conclusion " section.
RELATED WORK
SDN has been extensively deployed in wired
networks; however, in recent years, it has also
been applied to wireless scenarios. For example, it
has diverse applications (e.g., path selection, channel
allocation, etc.) in MANET, VANET, and UAVNets.
In UAVNets, there are frequent changes in
topology due to battery drainage and intermittent
links. Moreover, the programmability of UAVNet
can reduce the threat of collisions, performance
improvement, path optimization for data routing,
and change packet transmission range due to
energy constraints. Gupta et al.5 exploited a survey
on various issues in UAVNets and discusses different
SDN capabilities to solve these issues. For
example, mobility support, flexible strategies for
routing and switching, dealing with unreliable wireless
links, network greening, and interference
reduction. Rehman et al.6 addressed the controller
placement problem in SDN-based UAVNets. The
communication overhead and end-to-end delay
of the control packets exchanged betweenUAV elements
and the controller is considered. Results
IEEE Consumer Electronics Magazine
IEEE Consumer Electronics Magazine - January/February 2023

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2023
