IEEE Systems, Man and Cybernetics Magazine - January 2023 - 28

Table 1. A summary of relevant works.
Research
Works
[7]
[8]
Content Description
The study surveyed some of the challenges and possible solutions for ultralow-latency
and high-reliability services in millimeter-wave cellular systems in terms of the media
access control, congestion control, and core network architecture.
The study focused on an amalgam of diverse applications and services with
heterogeneous performance requirements and discussed the challenges that
next-generation emergency services need to overcome to fulfill the requirements for
rich-content and real-time applications.
[9]
[10]
[11], [12]
The study proposed an ultrareliable and low-latency communication approach in the
time division duplexing systems by encoding a grant signal in the form of a sparse vector.
The study introduced a generalized path-permutation code scheme to increase reliability
and achieve higher throughput, which aims to increase the reliability of opportunistic
links in cognitive radio networks and low-latency vehicular networks.
The first study [11] believed that an ultralow end-to-end delay can be met by considering
new technologies such as network function virtualization. Through the idea of virtualizing
wireless access resources, the second study [12] discussed the challenges for realizing
dynamic and intelligent resource management to achieve the goal of ultralow
end-to-end delay.
[13]
The study investigated the joint edge caching and computation management and
developed an online and distributed approach to jointly manage service caching,
request scheduling, and resource allocation, which aims to satisfy the ultralow delay
requirements of Internet of Vehicles services in a cost-efficient way.
[14]
The study combined mobile edge computing with end-user resources to the corresponding
solution, including buffer-nadir-based multicast mechanism, multicast-aware
transcoding offloading algorithm, and crowd-assisted delivery algorithm,
which aims to provide the virtual reality live-cast services with ultrahigh data
rates and ultralow delay requirements.
[15]
The study proposed a network architecture consisting of a resource cognitive engine and
a data engine, which aims to achieve ultralow end-to-end delay. The resource cognitive
engine focuses on a global view of computing, caching, and communication resources
in wireless networks, while the data cognitive engine offers personalized and
smart services toward specific domains.
[16]
The study combined cognitive computing along with virtualization and smart spectrum
resource management to address the spectrum resource challenge and explored how
to precisely predict the cells' traffic demand so as to optimize the carrier resource allocation.
Its purpose is to meet the requirements of high transmission data rates, ultrareliable
low-latency communications, and high connection density.
[17]
Emerging applications with increasingly high requirements in terms of latency, reliability,
peak data rate, and service continuity require the 6G core network to have cognitive
capabilities to achieve the integrated scheduling of communication resources and
computing resources, so this study proposed cognitive service architecture for a 6G core
network, including cognitive services, knowledge graphs, and cognitive scheduling
as the key concepts.
[18]
Considering the requirements for response delay and reasoning accuracy of different
users in industrial IoT scenarios with the heterogeneity of computing resources and the
randomness of communication environments, this study proposed a collaborative cloudedge
service cognition framework, which aims to provide dynamic and flexible computing
services for users with different service requirements.
[19]
To satisfy the requirements of the high reliability and low latency of communications in
5G applications, this study proposed an attention-based long short-term memory (LSTM)
algorithm to predict the service traffic flow with different requested data types and
designed a cognitive caching strategy based on LSTM and collaborative filtering.
[20]
To satisfy multitask parallel decision demand in complex dynamic environments, this
study proposed an artificial intelligence-driven cognitive networking collaborative
decision-making scheme, including terminal cognitive layer, edge cognitive layer,
and cloud cognitive layer.
28 IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE January 2023
Yes
Yes
Yes
No
No
No
No
Is There Any Intelligent
Cognition Service?
No
No
No
Yes
Yes
Yes
IEEE Systems, Man and Cybernetics Magazine - January 2023

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - January 2023
