Brno University of Technology

Electrical Engineering and Telecommunication

Project Name :The importance of Edge Computing and the way

of improving

by: Jamal Baraznji

master student in telecommunication and networking

Supervisor :[Link] Horváth Tomáš

DEC 2023

Brno university of Technology E 1

lectrical Engineering and Telecom
munication
Contents:

1- Introductions:
2- Edge computing description
functionality
3- Edge Computing vs. Cloud
Computing
4-Edge computing in 5G.
6-How to improve Ede computing.
7- Content catching.
9-Resources.

2
Brno University of Technology Electrical Engineering and Telecommunication
1-Introduction

With the growing on technology and increasing demanding for high internet needs

which required high processing and responses on time , edge computing introduced

to speed up the processes and to provide a computing closer to user side , this

solution have many advantageous and its very cost effective in all sectors of the life

from telecommunication to healthcare , farming ,smart cities , etc , This project will

clarify different topics and aspects of the edge computing , the advantageous and

challenged and the last update of this solution. I have used the last update

information and articles as a reference for preparing this project.

Brno university of Technology Electrical Engineering and Telecommunication
3
2– What is Edge computing?

Edge computing is a distributed open platform integrating core capability of network ,computing ,storage

and application, provides edge intelligence service at network edge close to devices or data sources. In

other words , edge computing directly analyses the collected data from the terminals at or near the

networks where the data is exist and didn’t send them to the cloud center.[1]

In a traditional computing model, data is sent to a centralized data center or a cloud provider for storage

and processing. However, this can result in latency and bandwidth constraints.

With edge computing, devices and sensors located at the edge of the network can process and analyze data

in real time. Because data doesn’t have to be sent to a central location for processing, organizations can

process it locally, which can improve cost savings, response times and application performance.[2]
Brno university of Technology Electrical Engineering and Telecommunication 4
2-1 Architecture of Edge Computing

Terminal : The terminals are various types of IoT devices like sensors ,
smartphones , cameras and in this layer IoT is provide only sensing
capabilities without computing.

Edge computing node: Edge computing nodes enable the response to basic
services by installing computing and storage capabilities ar the network
edge.

Network node: A network node uploads useful data through edge computing
nodes to cloud computing for processing and analysis.

Cloud computing node: The data reported by the edge computing layer
permanently stored on cloud computing nodes , the task which are beyond
the capabilities of edge computing nodes are processing by cloud computing
nodes because can dynamically adjust the applied configurations and
algorithm at edge comouting layes base on network resources availability . [5]
5
Brno university of Technology Electrical Engineering and Telecommunication
 2-2 why we need Edge computing?
With the growth of technology especially IoT , more common Objects are introduced and interconnected and also
industry terminal massively increased , this accelerated the digital transformation in the industry and in the networks as
well.

Nevertheless , as the IoT is very complex system and the scenarios change according to the industrial needs , and
according to the statistics analyzing there will be more than 100 Billion devices connected to the internet and the data
volume will reach to 300ZB by 2025 , so all data needs to be transferred to the biggest platforms and transmitted to the
clouding system for analysis , and as these data are huge so the could is face many challenges like insufficient
bandwidth , high power consumption and also hanging out due to CPU utilization.

And due to aforementioned reasons the Edge introduced to overcome those challenges.

Edge computing a distributed open platform integrating core capabilities of network ,computing , storage and application
, provide edge services at the network edge close to devices or data sources , in brief , it will analyze the data near the
device and no need to send them to data center for processing.
Brno university of Technology Electrical Engineering and Telecommunication
6
 2-3 Advantageous of Edge
Computing
Allocating computing at the edge allows the organizations or the network to manage and improve their service
easily below are some most advantageous:

1- Reduce Latency :

2-Improved data Security: the most important data keep in the local and doesn't have to be send to the cloud
server where it can be susceptible to cyberattack

3-Cost-effective: Edge computing reduces the storage and bandwidth costs associated with transmitting data to
the cloud , making it more cost effective option for organization and industries.

4-Scalability : the organization can design and scale their resources dynamically based on their needs which
make them to easily handle any requirements.

5-Reliability: the reliability of application can be improved with Edge computing because the applications run
independently of the cloud . 7
Brno university of Technology Electrical Engineering and Telecommunication
Challenges of Edge computing

Hint: Effect Options gives you even more options for Morph.

Brno university of Technology Electrical Engineering and Telecommunication 8

2-4 The importance of Edge computing
Edge computing is important because it creates new and improved ways for industrial and enterprise-level
businesses to maximize operational efficiency, improve performance and safety, automate all core business
processes, and ensure “always on” availability. It is a leading method to achieve the digital transformation of
how you do business.[3]

Brno university of Technology Electrical Engineering and Telecommunication 9

3- Edge Computing vs. Cloud Computing

The explosive growth of data brought about by IoT magnifies these disadvantages of cloud computing:

Cloud computing cannot efficiently process increasingly massive volumes and sources of data. With the
integration between the Internet and various industries, especially after the popularization of IoT, the demand
for computing power is increasing exponentially, which is beyond the current processing capability of the
traditional cloud computing architecture.

Cloud computing cannot implement real-time data [Link] the traditional cloud computing architecture,
terminals collect IoT data and then transmit the data to the cloud computing cluster in a faraway DC. The
computing cluster then sends back the result. This inevitably results in a long response time. However, some
emerging application scenarios, such as unmanned driving and smart mining, have demanding requirements
on the response time. As such, cloud computing is unsuitable for them.

Edge computing can solve these problems to some extent. As shown in flowing figure, data generated by IoT
terminals does not need to be transmitted to remote cloud DCs for processing. Instead, data is analyzed and
processed at the network Brno
edge, which ofisTechnology
university more efficient and secure
Electrical thanand
Engineering cloud computing. [4]
Telecommunication 10
Edge Computing vs. Cloud Computing

Brno university of Technology E 11

lectrical Engineering and Telecom
munication
3-1 some basic different between cloud computing and Edge computing

Item Edge computing Cloud computing

Data Storage Only useful information All collected information

transfer to the remote end stored
Deployment cost Low high

Security and privacy Relative high Relatively low

Latency Low High

Computing method Distributed computing Centralized computing

Processing Close to terminals Cloud DC

12
Brno university of Technology Electrical Engineering and Telecommunication
 3-2 CDN vs. Edge Computing.
Even
closer Edge
to thecomputing
end usersand CDN
, they have
are aim todifferent
basically deliver content and services
in their architecture
and uses cases.
In traditional
servers networks
for hosting and ,data
the service providers
distribution were
fortheir
their depend
end users on
, socentralized
theback
userfor
even
a located
long distance in the far place to the server request should
to the central server for processing ,consequence it faced be
latency and slowness.
With
popular thewebsites
growing ,ofservers
internet when
faces the users
hanging andfrom globe tried
unexpected to access
traffic leadinga to
downtime.
Based
Their on that is
function CDN introduced
to brought in late 90sto and overcome those challenges.
location quickly and no needthe to content
check with the
theuser
mainincentral
wherever the
server.
However
and the CDN
not have function
feature is only fast
for computing responding
.but Edge storagecome
computing on the tocloud
demand
as response
processing. [7] to the internet of things IoT , especially for real-time

[Link]

Brno university of Technology Electrical Engineering and Telecommunication 13

CDN vs. Edge Computing.
Aspect CDN Edge Computing
Speed up web content delivery by Process data closer to the source to
Primary Objective
reducing latency. reduce latency.

Network of geographically distributed Local processing on devices or

Architecture
servers. nearby edge servers.
Web content delivery, video Real-time processing for IoT,
Use Cases streaming, web application healthcare devices, autonomous
acceleration. vehicles, etc.
Data Handling Caches and delivers content. Processes and may store data.
Reduced load times, DDoS Reduced load times, DDoS
Benefits protection, global content protection, and global content
distribution. distribution.
Scales by adding more server nodes Scales by adding processing
Scalability
to the network. capabilities at the edge.
Provides security through traffic Can enhance security by processing
Security dispersion, DDoS protection, and sensitive data locally, reducing
content encryption. exposure.
It may involve costs for edge devices, Wearable health devices, smart
Examples
local servers, and data transfer. factories, autonomous vehicles.
Latency Reduced latency in content delivery. Reduced latency in data processing.
Typically integrates with cloud It can enhance security by processing
Integration with Cloud services for content sourcing and sensitive data locally, reducing
updates. exposure.
Real-time data processing, reduced
Often based on bandwidth usage and
3-3 MEC and Fog computing

Fog computing deploys local fog nodes which are local hardware devices, such as switches and routers,
to provide local computation. According to the OpenFog Consortium, fog computing is ``a system-level
horizontal architecture that distributes resources and services of computing, storage, control, and
networking anywhere along the continuum from cloud to [Link] computing shares similar benets to
other edge computing variants (e.g., MEC) to provide low latency and real-time analytics; however, it
has low storage capacity.

MEC provides storage and computational capacities at the edge of the network, such as the radio
access networks (RANs) and BSs, to improve context awareness and reduce latency. The MEC servers,
which are usually co-located with multiple hosts (e.g., BSs), use a virtualized interface to access
storage and computation facilities. A MEC orchestrator overlooks the MEC hosts by gathering and
providing real-time information regarding the services offered by each host, the available resources
(e.g., network capacity and load), the network topology (e.g., UEs connected to the servers including
their location and networking information), as well as managing MEC applications.[10]
17
Brno university of Technology Electrical Engineering and Telecommunication
3-4 Fog vs MEC

18
Brno university of Technology Electrical Engineering and Telecommunication
Use cases of Edge computing

Farming: Manufacturing
Network Optimization

Brno university of Technology Electrical Engineering and Telecommunication

19
Use cases of Edge computing [7]

1- Autonomous Vehicles : Edge computing can enable self-driving vehicles to process data from
sensors and cameras in real time, allowing them to make rapid decisions and respond to changes in
their environment.

2-Video streaming: organization can use edge computing for contet delivery networks (CDNs) to
improve streaming performance and reduce latency by catching and processing data at the edge of
the network.

3-Healthcare : ege computing have a big role in the health sector which can monitor patient in real-
time and analysis a related data and involve in the surgery .

4-IoT Devices :Edge computing can be used to process and analysis data locally , reducing latency and
improving response time, this is very useful for those applications which require a real-time data
processing and response , such as smart cities and network monitoring OSS.
Brno university of Technology Electrical Engineering and Telecommunication 20
 Use cases of Edge computing[7]
5-Traffic Management Systems

Another area of an edge computing example is traffic management systems. If we look at the most
sophisticated application, the New York City traffic management system is where we see the use case
of edge computing devices. It can perform functions such as adjusting the timing of traffic signals and
opening and closing extra traffic lanes on an ad-hoc basis. Here if we notice, through edge computing,
traffic is managed intelligently and on a real-time basis to avoid any sort of congestion.

6- Agriculture and Farming

Agriculture and farming is also an area where we can see a significant use case of edge computing.
Using the concept of precision agriculture technology, farmers can leverage edge computing here and
monitor various parameters such as the soil’s temperature, humidity, and level of nutrients in the soil.
This can further yield in taking decisions on the watering quantity and level of fertilizers to be applied
for farming.
Brno university of Technology Electrical Engineering and Telecommunication 21
Improving Edge computing

Try it yourself with these two simple “planets”:

Brno university of Technology Electrical Engineering and Telecommunication 22

Edge computing evolution

Brno university of Technology Electrical Engineering and Telecommunication 23

 Roles of Edge Computing in 5G:
Edge computing has Six major roles in 5G which Briefly described below.:
[10]

1- Local storage: Edge computing offloads a massive amount of data from

UEs to edge clouds . The edge server has a local storage for a significant
amount of data but its much lower than the cloud storage capacity.

2-Local computation: : Edge computing offloads computation and process

from UEs to edge cloud. The edge cloud is an intelligent computing system
that provide s local computation and data processing capabilities close to UEs
in an independent way.

3-Local data analysis : Edge computing processes and perform a real-time

important data analysis on a massive amount of raw data gathered from
different applications in close proximity to generate valuable information ,
this will reduce latency required to send to and wait response from the cloud.

5G-multi-access edge
Brno university of Technology Electrical Engineering and Telecommunication
computing architecture24
.
Roles of Edge Computing in 5G.

4-Local decision making :Edge computing helps entities to make real-time decisions and corresponding
actions in an automated manner based on well-processed data , consequently this will make the decision
will be easier as reduces involvement from more components and data exchange ultimately this leads to
improve system availability.

5-local operation : Edge computing enables remote control and monitoring-particularly critical devices
including those are far away or in remote area.

6-Local security enhancement : Edge computing serves as an additional layer between the cloud and
connected devices in order to improve network security ,because Edge can serve as secured distributed
platforms that provide security credentials management , can do malware detection , software patches
distributions and detect countermeasure attacks.

The advantage is that , due to close procimity of edge computing malicious can be detected and isolated.

Brno university of Technology Electrical Engineering and 25

Telecommunication
Important enabler of Edge Computing in 5G :[11]

1-Software-defined network (SDN) defines a network architecture that separates a network into control and

data planes to provide flexible and agile networks,which helps to simplify network management and deploy

new services. In general, the control plane handles policy on cloud, while the data plane forwards traffic

according to decisions made by the control plane. Network functions (e.g., routing) that require real-time

response can be handled by edge servers.

2- Network function virtualization (NFV) performs network functions (also known as virtual functions) in virtual

machines on servers, which can handle a massive amount of data to provide flexible, automated, and scalable networks

Network

demands can be processed either at the cloud or at the edge, which prevent all data and information from being sent to

the cloud.
Brno university of Technology Electrical Engineering and Telecommunication 26
 Important enabler of Edge Computing in 5G :[11]

3- Massive MIMO deploys multiple antenna elements to increase an antenna array at transmitter and receiver.

This is in accordance to the Shannon theorem in which the signal-to-noise ratio increases without the need to

increase the transmission power, leading to increased network capacity and energy . Using massive MIMO,

multiple UEs can offload tasks to an edge server simultaneously to reduce latency and energy consumption.

4- Dynamic access to radio access technologies provide access to conventional access technologies, such as

Wi-Fi, and new radio access technology (RAT) in 5G, such as NR ,5G NR is a new standard that provides

connection to a diverse range of devices for achieving low latency and scalable networks, which allow future

extension to existing networks.

Brno university of Technology Electrical Engineering and Telecommunication

27
 Important enabler of Edge Computing in 5G :[11]

5- throughput, energy efficiency, and spectrum utilization [50], [51]. UEs can offload tasks and computations to
edge servers in order to empower UEs

with computational capabilities. This feature of edge computing ensure successful D2D communication.

Brno university of Technology Electrical Engineering and Telecommunication 28

Taxonomy of Edge computing for 5G [11]

Brno university of Technology Electrical Engineering and Telecommunication 29

4-MEC ,IoT ,Fog

4-1 : MEC and IoT.

4-2 : Main characteristic.

4-3 :Emerging MEC application in IoT.

4-4 :Fog
Brno university of Technology Electrical Engineering and 30
Telecommunication
4-1 MEC and IoT.

IoT is complex and not consistence interconnection of smart devices , it’s a great platform to develop
novel mobile applications , However the limit resources of devices become a barriers for realize such
developments especially when required high computational and low-latency. MEC [mobile Edge
computing] is a promising approach to address such challenges.

Mobile edge computing (MEC) brings the possibility to enable applications that require low latency with
high bandwidth utilization. Even so, ensuring low latency as required can be quite a challenge, especially
in an “Internet of Everything” .[15].

At MEC world congress in 2016 , the European Telecommunication Standard Institute(ETSI) officially
changed its MEC name from Mobile Edge Computing to Multi-Access Edge computing. Considering the
enlargement of connected devices that are not mobile devices and they decide to focus more on
integrating MEC resources to non-cellular networks, after Expansion MEC can deployed at various places
in RAN , Base station ,cell towers , WiFi access point. [16]
31
Brno university of Technology Electrical Engineering and Telecommunication
4-2 Characteristic of MEC

Connected smart IoT devices often require information processing. The computing capacity of nearby
MEC servers helps them realize sensing, interaction, and control efficiently [5]. The main
characteristics of MEC are as follows :

Proximity: Being deployed in proximity to the smart devices, MEC is suitable for analyzing key
features in IoT data generated by these [Link] Latency: Shifting data offloading from core
network, MEC speeds up application response, improves user experience, and reduces

potential network congestion.

High Bandwidth: The data transmission between MEC servers and smart devices fully exploits the
bandwidth of access networks, and gains high speed.
Location and Network Context Awareness: MEC servers use received information from smart devices within
the local access network to determine the specific location of these devices. Based on the local real-time
information, applications running on the devices can estimate states of the network.
Brno university of Technology Electrical Engineering and Telecommunication 32
4-2 Characteristic of MEC
Flexible Deployment: MEC is readily applicable for hosting novel mission-critical applications which are
deployed by both cellular operators and authorized third-party developers

Real-Time Response: IoT emergency applications require real-time response [8]. As the latency of the
offloading process is not only affected by the states of various resources, but also by the strategies of the
other devices, designing computation

offloading schemes that can guarantee low latency requirements is not trivial. Next, we introduce six
distinct potential IoT.

Location and Network Context Awareness: MEC servers use received information from smart devices
within the local access network to determine the specific location of these dev ices. Based on the local real-
time information, applications running on the devices can estimate states of the network

Brno university of Technology Electrical Engineering and

Telecommunication
33
4-3 Emerging MEC application in IoT

Brno university of Technology Electrical Engineering and

34
Telecommunication
 Emerging MEC application in IoT

Brno university of Technology Electrical Engineering and

Figure ( 4) Emerging MEC applications in IoT. 35
Telecommunication
4- Caching system for mobile Edge computing

Caching contents at the edge of mobile networks is an efficient mechanism that can alleviate the backhaul links

load and reduce the transmission delay. For this purpose , choosing an adequate caching strategy becomes an

important issue. Recently the tremendous growth of mobile Edge computing (MEC) empowers the edge

network nodes with more computation capabilities and storage capacities, Caching is considered a critical

bottleneck for the development of MEC systems, since edge caches are located at the edge of the networks and

physically closer to the end user. With limited storage, edge caches use different strategies of placement and

replacement, which are in general designed to allow spreading only popular contents around the network. In

addition,MEC paradigm presents an exclusive opportunity to implement edge caching and to design caching

placement andreplacement strategies. Therefore, caching popular contents at MECS is considered as a cost

effective solution due to the benefits of avoiding network congestion and alleviating the backhaul links.. [17]

36
Brno university of Technology Electrical Engineering and Telecommunication
Caching system for mobile Edge computing

In general, edge caching means that popular contents can be cached in edge
nodes such as macro base stations, small base stations or even user equipment.,
caching a content which is close to the end users can effectively reduce the
redundant data traffic and greatly improve the QoE of users Concerning what to
cache, the range of contents is widely increasing including videos, audio files
and Internet-of- Things data, which makes the number of reachable contents
over the internet extremely huge. Not all of the available contents should be
cached, regarding to the fact of limited storage space of edge nodes . [17]

37
Brno university of Technology Electrical Engineering and Telecommunication
Categorizing Caching policy base on their characteristics

Caching policies can be divided into categories depending on their characteristics:

• Depending on cache coordination:

Coordinated: caches must exchange information to be efficient, that is, to have a good estimation of where to place
the

content and to avoid caching the same content in too many caches. Coordinating cache is represented by a
distributed

session feature that allows multiple instances of a session to broadcast content changes among each other so that
each

cache is updated.

Uncoordinated: in uncoordinated scheme, each cache is working in an individual way. [17]

Brno university of Technology Electrical Engineering and
Telecommunication 38
Categorizing Caching policy base on their characteristics
Depending on cache size:

Homogeneous: all caches in base stations have the same size.

Heterogeneous: each cache has a different size.

• Depending on the cooperation between caches

Cooperative: caches cooperate with each other by establishing a cache state that allows other caches to
know the different states .

Non-cooperative: caching decisions are made independently with no need to advertise the information of
cache state.

• Depending on where the content is cached:

On path: caching only the contents caught along the downloading path.

Off path: caching the content caught outside the downloading path. [17]
39
Brno university of Technology Electrical Engineering and Telecommunication
 Categorizing Caching policy base on their characteristics
In general, conventional caching policies are divided into two main phases: placement phase and replacement
phase:

• The placement phase which is the process that decides whether we should cache and how and when we can

cache the contents.

• The replacement phase which is the process that decides which data to drop if there is no free storage space.
[17]

Brno university of Technology Electrical Engineering and

40
Telecommunication
Content Caching in 5G Networks with Mobile Edge Computing

Along with modern wireless networks being content-cenetric , the demand for rich multimedia service has
been growing at a tremendous pace, which brings significant challenges to mobile networks in term of the
need for massive content [Link] caching has emerged as a promising approach to alleviate the heavy
burden on data transmission through caching and forwarding contents at the edg of networks.

In recent year wireless has been witnessing explosive growth of smart devices and mobile users. Along with
the development of mobile networks, the advancements in wireless technologies as well as the Internet of
Things (IoT) bring us a wide variety of powerful mobile applications and multimedia services. These

attractive applications and services heavily rely on high-speed data rates and low-latency transmission. which
pose significant challenges to mobile networks.

Brno university of Technology Electrical Engineering and Telecommunication 41

Cooperative Caching Networks

In below figure it shows the architecture of the proposed caching network., there are two types of

resources. One is caching resources and the other one is MEC resources. They serve end users under

the management of the controllers. Caching resources are composed of various types of content-

storage-enabled entities. The first one is the caching servers located in the core network. Although

these servers are far away from the end users, the servers still play a vital role in the content serving

process. As edge nodes always have limited caching capabilities, they cannot cache all the contents

themselves at the same time. Furthermore, the popularity of the contents is time-varying. Thus, the

contents cached on the

edge nodes should be updated adaptively.

Brno university of Technology Electrical Engineering and 42
Telecommunication
Cooperative Caching Architecture

Brno university of Technology Electrical Engineering and

43
Telecommunication
Benefits for Content-Centric 5G Networks

Benefits for Content-Centric 5G Networks The proposed cooperative edge caching

architecture leverages MEC resources to alleviate the content caching burden on edge
nodes with strained storage capability. With centralized control of both computing and
the caching process, a flexible and efficient caching mechanism can be achieved. The
benefits of this architecture for content-centric 5G networks are summarized as follows.

Heterogeneous Capability Integration: Caching and computing capabilities, which

are heterogeneous but ubiquitous at the edge nodes, are integrated in the content
caching process. The MEC capability of the nodes acts as an enabler to improve caching
capability through content information processing. In this regard, MEC can also
contribute toward enhancing caching.
Brno university of Technology Electrical Engineering and Telecommunication 44
Benefits for Caching Content-Centric 5G Networks
Efficient Resource Utilization: The storage resources of heterogeneous integration of a large number of
things are fully utilized for content caching. In particular, the resources of moving smart vehicles as well as of
handheld devices are exploited to cache and spread popular contents to a wide area. Hence, resource
utilization can be improved. Flexible Cache Scheduling: This architecture abstracts the caching and
computing capabilities of the network, and manages content caching in

a centralized manner. Various types of resources from heterogeneous edge nodes can be adaptively arranged
in independent or cooperation modes according to the characteristics of the content, the requirements of end
users, and the features of communication scenarios.

Flexible Cache Scheduling: This architecture abstracts the caching and computing capabilities of the
network, and manages content caching in a centralized manner. Various types of resources from
heterogeneous edge nodes can be adaptively arranged in independent or cooperation modes according to
the characteristics of the content, the requirements of end users, and the features of communication
scenarios. Brno university of Technology Electrical Engineering and Telecommunication 45
Benefits for Content-Centric 5G Networks

Brno university of Technology Electrical Engineering and Telecommunication 46

 Resources and References
1 : [Link]

2- [Link]

3-[Link]

4-[Link]

5-[Link]

7-[Link] Streaming / By Krishna Rao Vijayanagar / August 28,

2023

• 8 –IEEE Access :Edge Computing Technology Enablers: A Systematic Lecture Study by SALMANE DOUCH 1,3, (Graduate Student Member,
IEEE),

MOHAMED RIDUAN ABID2,3, (Member, IEEE), KHALID ZINE-DINE 4, DRISS BOUZIDI1, AND DRISS BENHADDOU 5, (Senior Member, IEEE)

47
Received 23 May 2022,Brno
accepted 12 June
university 2022, date of publication
of Technology 16 June 2022,
Electrical Engineering date
and of current version 5 July 2022. Digital Object Identifier
Telecommunication
10.1109/ACCESS.2022.3183634
Resources and References
10 -IEEE Access Edge Computing in 5G: A Review by NAJMUL HASSAN 1, KOK-LIM ALVIN YAU 1, (Senior Member, IEEE),

1Department of Computing and Information Systems, Sunway University, Bandar Sunway 47500, Malaysia

2Graduate School of Informatics and Engineering, The University of Electro-Communications, Tokyo 182-8585, Japan

Corresponding author: Kok-Lim Alvin Yau (koklimy@[Link])

Received August 2, 2019, accepted August 26, 2019, date of publication August 30, 2019, date of current version September 18,
2019.

Digital Object Identifier 10.1109/ACCESS.2019.2938534

16-IEEE Access :Edge Computing Technology Enablers: A Systematic Lecture Study by SALMANE DOUCH 1,3, (Graduate Student
Member, IEEE),

MOHAMED RIDUAN ABID2,3, (Member, IEEE), KHALID ZINE-DINE 4, DRISS BOUZIDI1, AND DRISS BENHADDOU 5, (Senior Member,
IEEE)

Received 23 May 2022, accepted 12 June 2022, date of publication 16 June 2022, date of current version 5 July 2022. Digital Object
Brno university of Technology Electrical Engineering and
Identifier 10.1109/ACCESS.2022.3183634 48
Telecommunication
13[Link]

7 Edge Computing Uses You Should Know Discover seven different examples of edge computing and how each one lands in
the edge computing landscape. Dec 13th, 2022 11:41am by Anirudh Men

15- Mobile Edge Computing and Networking for Green and Low-Latency Internet of Things [Ke Zhang, Supeng Leng, Yejun
He, Sabita Maharjan, and Yan Zhang]

Mobile Edge Computing and Networking for Green and Low-Latency Internet of Things , IEEE Communications Magazine •
May 2018.

by Ke Zhang, Supeng Leng, Yejun He, Sabita Maharjan, and Yan Zhang

17- Modified reinforcement learning based- caching system for mobile edge computing Sarra Mehamel, Samia Bouzefrane, Soumya
Banerjee, Mehammed Daoui, Valentina Bala HAL Id: hal-03112767 [Link] Submitted on 17 Jan 2021

Brno university of Technology Electrical Engineering and

49
Telecommunication