Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G
Mobile edge computing is an important element in ensuring the efficiency of the 5G network as a whole, as it enables data storage and computing at the network edge. Existing solutions do not fully address the issues of load distribution between computing nodes, and most solutions do not offer method...
Saved in:
| Date: | 2024 |
|---|---|
| Main Authors: | , , , , , |
| Format: | Article |
| Language: | English |
| Published: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2024
|
| Subjects: | |
| Online Access: | https://journal.iasa.kpi.ua/article/view/309711 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | System research and information technologies |
| Download file: |  |
Institution
System research and information technologies| _version_ | 1866302974031036416 |
|---|---|
| author | Astrakhantsev, Andrii Globa, Larysa Fedorov, Oleksandr Degtiarov, Dmytro Romanko, Yevgen Romanii, Kyrylo |
| author_facet | Astrakhantsev, Andrii Globa, Larysa Fedorov, Oleksandr Degtiarov, Dmytro Romanko, Yevgen Romanii, Kyrylo |
| author_sort | Astrakhantsev, Andrii |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2024-08-11T01:12:49Z |
| description | Mobile edge computing is an important element in ensuring the efficiency of the 5G network as a whole, as it enables data storage and computing at the network edge. Existing solutions do not fully address the issues of load distribution between computing nodes, and most solutions do not offer methods for verifying computations and controlling errors. Accordingly, this paper aims to develop an approach to the organization of mobile edge computing in a 5G mobile network that would authenticate distribution servers and computing nodes, manage the process of distributing computing nodes, have a procedure for verifying the correctness of calculations, and take into account the parameters of computing nodes during distribution. To achieve this goal, we propose to use the developed method. The method of load balancing and selection of computing nodes for edge computing via 5G allows for identifying available nodes and distributing computing blocks among them. It also provides mutual authentication of elements and includes a method of data verification and error detection for the MEC system. The provided solution allows for controlling errors during calculations and protecting the server from incorrect data. These methods are optimized according to minimum network resources and computing time criteria. These improvements increase the efficiency of mobile edge computing in a 5G network. |
| doi_str_mv | 10.20535/SRIT.2308-8893.2024.2.06 |
| first_indexed | 2025-07-17T10:28:33Z |
| format | Article |
| fulltext |
А. Аstrakhantsev, L. Globa, О. Fedorov, D. Degtiarov, Y. Romanko, K. Romanii, 2024
82 ISSN 1681–6048 System Research & Information Technologies, 2024, № 2
TIДC
ПРОБЛЕМНО І ФУНКЦІОНАЛЬНО
ОРІЄНТОВАНІ КОМП’ЮТЕРНІ СИСТЕМИ
ТА МЕРЕЖІ
UDC 004.272
DOI: 10.20535/SRIT.2308-8893.2024.2.06
AN IMPROVED APPROACH TO ORGANISING MOBILE EDGE
COMPUTING IN A 5G NETWORK
А. АSTRAKHANTSEV, L. GLOBA, О. FEDOROV, D. DEGTIAROV,
Y. ROMANKO, K. ROMANII
Abstract. Mobile edge computing is an important element in ensuring the efficiency
of the 5G network as a whole, as it enables data storage and computing at the net-
work edge. Existing solutions do not fully address the issues of load distribution be-
tween computing nodes, and most solutions do not offer methods for verifying com-
putations and controlling errors. Accordingly, this paper aims to develop an
approach to the organization of mobile edge computing in a 5G mobile network that
would authenticate distribution servers and computing nodes, manage the process of
distributing computing nodes, have a procedure for verifying the correctness of cal-
culations, and take into account the parameters of computing nodes during distribu-
tion. To achieve this goal, we propose to use the developed method. The method of
load balancing and selection of computing nodes for edge computing via 5G allows
for identifying available nodes and distributing computing blocks among them. It
also provides mutual authentication of elements and includes a method of data veri-
fication and error detection for the MEC system. The provided solution allows for
controlling errors during calculations and protecting the server from incorrect data.
These methods are optimized according to minimum network resources and comput-
ing time criteria. These improvements increase the efficiency of mobile edge com-
puting in a 5G network.
Keywords: 5G network, mobile edge computing, task allocation scheme, call flow,
load balancing, task verification.
INTRODUCTION
The new 5G cellular networks are expected to face a sharp increase in mobile
traffic and IoT user demands due to the massive growth in the number of mobile
devices and the emergence of new computing applications. Running resource-
intensive computing applications on resource-constrained mobile devices has re-
cently become a major challenge, given the stringent requirements for computing
time and the limited storage capacity of the devices.
Cloud computing allows you to store and process data on remote servers.
A large number of different applications that generate an ever-increasing amount
of data, which significantly increases network latency, uses them and places
differentiated demands on data security and manageability. Mobile Edge
Computing (MEC) technology can help prevent these problems from getting worse.
An improved approach to organising mobile edge computing in a 5G network
Системні дослідження та інформаційні технології, 2024, № 2 83
Mobile edge computing has recently emerged as a key technology to over-
come these challenges, as it enables the provision of cloud computing services
such as data storage and computing at the edge of the network. MECs have the
potential to run computationally intensive applications such as augmented and
virtual reality [1]. MEC is also an important component of the Internet of Things
(IoT), as it allows to reduce the power consumption of mobile devices.
Mobile edge computing is a data management technology that involves stor-
ing and processing data close to the source. This allows for faster response to real-
time computing needs and helps to guarantee the availability of information. In
general, MEC is a decentralized computing infrastructure in which some signal
processing, storage, management and computing applications are distributed in
the most efficient and logical way between the data source and the cloud [2]. Mo-
bile edge computing extends the concept of cloud computing by bringing the ben-
efits of the cloud closer to users in the form of the network edge, which provides
lower end-to-end latency.
The goal of the presented work is to organize mobile edge computing in
a 5G network by performing authentication of distribution servers and computing
nodes. It is also necessary to ensure the management of the process of distributing
computing units, including the procedure for checking the correctness of calculations
and taking into account the parameters of computing nodes during distribution.
In this regard, the following tasks were solved within the framework of an
improved approach to the organization of mobile edge computing in the 5G net-
work:
Development of a method for load balancing and selection of computing
nodes for MEC. The implementation of this method should not require additional
physical elements in the network.
Developing a method for data verification and error detection, as a com-
puting node may report incorrect calculation results.
Determining a method of mutual authentication for different types of
equipment in the 5G network for the process of mobile edge computing without
the use of a trusted third party.
At the same time, there are currently no existing technical solutions that
would solve all of the above problems.
ANALYSIS OF EXISTING SOLUTIONS FOR MEC IN THE 5G NETWORK
The problems that arise when organizing mobile edge computing covered in a
large number of publications. For example, [3] describes a typical MEC architec-
ture and its main elements, as well as the problems associated with the distribu-
tion of computing tasks. Paper [4] focuses on the problems of transmission delay
and computation delay with a large number of IoT devices. It also analyses the
possibility of overloading peripheral clouds due to the spatially heterogeneous
distribution of IoT tasks. To address these issues, it use game-theoretic methods
to investigate load balancing problems to minimize transmission and computation
delays in the task distribution process, given the limited bandwidth and comput-
ing resources in the edge clouds.
Work [5] solves a more complex problem of parallel offloading and load
balancing with several shared MEC servers and delay-sensitive load. A similar
А. Аstrakhantsev, L. Globa, О. Fedorov, D. Degtiarov, Y. Romanko, K. Romanii
ISSN 1681–6048 System Research & Information Technologies, 2024, № 2 84
problem is solved in [6], but it proposes a two-level model of task distribution
with delay minimization and computational cost estimation. In [5], a long-term
stochastic programming problem with an average system cost is formulated under
the conditions of stability of the battery level and delay constraints.
Another work by the same authors [7] partially solves one of the problems
studied in proposed research — it helps to create a secure reward mechanism us-
ing blockchain technology that can help to balance the load between computing
nodes.
Some works [8] propose to solve the problem of clustering and load balanc-
ing based on the charge level of computing nodes and geolocation tags.
In addition to the solutions that solve the problem of load distribution be-
tween the computing nodes of the MEC and which are presented in the above
publications, it is necessary analyzing patented solutions separately. For example,
patent [9] proposes a cloud platform with a pool of resources that connects to the
main network via a transmission network. This solution requires a special distri-
bution hub (RRH). In this article, to eliminate this drawback, it is proposed to use
a flat structure divided into zones and use the base station as an arbiter (no addi-
tional equipment is required).
Patent [10] proposes a solution based on the availability of a map that stores
information about the location, computing power and available storage of each of
the computing nodes. The disadvantage of the solution is the need for a dynami-
cally updated map with a list of nodes. This disadvantage can be overcome by
using a broadcast of the request from the MEC server. In this case, the response of
the base station to the computing nodes allows not to use the map, route table or
database — saving network resources.
The patent [11] is devoted to determining the optimal number of required
physical resource blocks during distribution, while the procedure for selecting
computing nodes is not described. In [12], the distribution of computing tasks and
resources is based on reducing the failure rate during handover, but the procedure
for allocating network resources is also not described and there is no data verifica-
tion and error control. The solution [13] offers a centralized implementation of
edge computing, where a central computer or a cloud macro base station will per-
form the main distribution tasks. This requires additional costs. In addition, this
solution uses only delay as a distribution criterion. The patent [14] also requires a
hierarchical structure and does not provide for the identification and authentica-
tion of computing nodes. In addition, this solution lacks data verification and er-
ror control.
The patent [15] describes only the process of creating a session for mobile
edge computing, does not provide solution for data verification and error control,
and does not describe procedures for load balancing and selection of computing
nodes.
Solutions [16; 17] do not provide for security measures (no identification
and authentication). In addition, in both cases, a central database is required. In
[18], it is described how a computing node should be rewarded for a completed
operation, but the data verification procedure is not described and there is no sup-
port for the 5G network, as well as no secure channel for transferring rewards.
To summarize, most solutions for selecting computing nodes and load bal-
ancing use additional physical elements, which requires additional costs. The con-
An improved approach to organising mobile edge computing in a 5G network
Системні дослідження та інформаційні технології, 2024, № 2 85
sidered technical solutions require dynamically updated maps or databases, which
requires additional network resources and increases the load on the network.
In addition, many of these publications lack authentication procedures for partici-
pants.
PROBLEM STATEMENT
Let us identify the main participants in the process of distributed mobile edge
computing (Table) and their functions according to the approach proposed in this
paper. A similar list of process participants, but with a different set of functions, is
given in [3; 19; 20].
Main participants in the process of distributed edge computing
Participant marking Participant functions and components
MEC Server
MEC Server: gather data flow from one/multiple sensors;
has 5G supported radio module;
run MEC supported application;
has identity and billing entity.
Computing Node
Computing Node: process MEC Server Application
Programming Interface (API) call;
has 5G supported radio module;
has CPU that support operability of MEC framework;
has identity and billing entity.
Base station (Cell)
Cell: assign radio resource, verify identity, sign transaction,
secure connection;
support computing unit selection;
support peer-to-peer communication;
secure and sign transaction MEC Server → Computing node;
The solution of the tasks set in this paper done by simulation and mathemati-
cal modelling for the architecture shown in Table, taking into account the short-
comings of existing solutions discussed in the previous section. In order to pre-
pare the proposed technical solution, a set of input data, a set of constraints,
dependencies between them, and a set of output values were determined. Let us
consider them in more detail.
Let the following data received at the input of the load balancing system for
distributed boundary computing:
n — a set of computing nodes available to the MEC with computing capaci-
ties ir and an initial level of trust id ;
id — initial level of trust in the computing node;
)( ixp — the probability of an error during calculations by the i-th node;
Tр — time for the distribution of computing tasks;
sRe — the amount of network resources involved in the distribution of tasks;
oT — the expected calculation time;
defT — the restriction for the expected calculation time;
V — the amount of calculations to be performed.
А. Аstrakhantsev, L. Globa, О. Fedorov, D. Degtiarov, Y. Romanko, K. Romanii
ISSN 1681–6048 System Research & Information Technologies, 2024, № 2 86
When distributing computational tasks for execution, it is necessary to en-
sure the minimum probability of calculation error )(np and minimize the network
resources used during their distribution and processing in computing nodes:
min)s(Re , subject to restrictions on the expected computing time )( defTTo .
A weighted average used to determine the probability of a calculation error:
.)(
1
)( i
n
i
xp
n
np
As a result, the proposed method should provide:
),0( nny — a set of devices that perform the calculation of distributed
computing with load balancing based on capacity ir ;
),0( ynnw — a setoff additional devices that will provide redundancy
and reliability of distributed computing;
id — change in the level of trust in the i-th node based on the results of its
work.
The expected calculation time will consist directly of the calculation time
and the time for distributing the calculation tasks defined as:
,o р
V
T T
nr
or considering the set of devices that perform the calculation and the set of addi-
tional devices:
р
ii
w
iii
y
i
o T
rwry
V
T
)()(
.
In this paper, we propose a method for organizing distributed computing that
performs the following steps:
a broadcast request from the MEC server to perform distributed computing;
a response from at least
one computing node to the base
station containing a set of parame-
ters (request ID, timestamp, etc.);
the base station checks the
available resources and provides
network parameters for the MEC
session of at least one computing
node, which will allow for further
point-to-point connection;
the MEC server can verify
the results of the calculations by
means of data validation, mirror-
ing and control code.
The essence of the method
described above showed in
Figs. 1, 2.
Fig. 1. Visualization of the principle of the pro-
posed method
An improved approach to organising mobile edge computing in a 5G network
Системні дослідження та інформаційні технології, 2024, № 2 87
The proposed approach is based on two new methods:
a method of load balancing and selection of computing nodes for edge
computing via 5G, which allows to identify available nodes and distribute com-
puting blocks among them, as well as provide mutual authentication of elements;
a method of data verification and error detection for the MEC system,
which allows to control the occurrence of errors during calculations, protect the
server from incorrect data, and prioritize and reward nodes based on the results of
the calculations performed.
Let us consider the principle of the proposed methods in more detail.
METHOD OF LOAD BALANCING AND SELECTION OF COMPUTING NODES
IN A 5G NETWORK
The principle of the load balancing method includes two stages: 1) the authentica-
tion stage and 2) the point-to-point channel establishment and the computation
and verification stage.
The first stage of authentication and channel creation (Fig. 2) involves the
following steps:
1. The MEC server broadcasts a request to perform calculations with the fol-
lowing information:
MEC identifier (temporary or permanent identifier);
type of calculation.
2. Each computing node upon receiving paging device, reply to base station
(cell) with:
,( idCFE ), idlast ET ,
where )( idC — identifier of the serving cell (base station); )( lastT — the time-
stamp of the last received slot for performing calculations; )( idE — network
identifier (temporary or permanent).
3. The base station selects computing nodes and assigns a radio channel:
selects a computing node based on the received values of E ;
assigns a radio channel based on available resources;
Fig. 2. The sequence of actions at the first stage of distributed computing — authentica-
tion and channel creation
А. Аstrakhantsev, L. Globa, О. Fedorov, D. Degtiarov, Y. Romanko, K. Romanii
ISSN 1681–6048 System Research & Information Technologies, 2024, № 2 88
notifies the MEC server and the computing node of the established infor-
mation exchange channel.
The second stage, where calculations and their verification are performed,
includes the following steps (Fig. 3):
1. The MEC server and the computing node establish a radio channel. The
radio channel is formed based on the channel configuration parameters that each
participant receives from the base station.
2. The MEC server and the computing node then perform a synchronization
procedure.
3. Based on ETSI, the computing node makes an API call (Fig. 4) and sends
a report to the base station after the computation is complete.
4. After verifying the result, the MEC server reports to the base station.
5. The reward for performing computations is calculated based on the com-
plexity of the operation, execution time, and the amount of disk space consumed.
To provide the compulsoriness of the workflow a blockchain technology is used.
The participants are rewarded proportionally their fitness/ activities.
Fig. 3. The sequence of actions at the second stage of distributed computing — per-
forming calculations and verifying results
Signing
transactiion
Unit selectiion
Valiidation
MEC server Comp node Cell
Fig. 4. The sequence of actions during an API call at the second stage
An improved approach to organising mobile edge computing in a 5G network
Системні дослідження та інформаційні технології, 2024, № 2 89
As mentioned above, during the second stage, the MEC server must check
the calculations for correctness and errors, and assign a certain level of trust to
each computing node. These procedures provided in the proposed method of data
verification and error detection for the MEC system.
DATA VALIDATION AND ERROR FOUNDING METHOD FOR MEC SYSTEM
Each task that will be processed on the MEC server contains parts that can be per-
formed independently. These parts are added to the task by software developers in
the form of an API call. The results of such external computations carry the risks
of computational errors and various types of attacks. In this paper, we propose a
combined system for verifying the results of work, which includes the analysis of
confidence levels and redundancy.
The proposed method for verifying the results of calculations and finding
errors includes the following steps:
1. The computing device of the MEC server
generates an error control code in the form of a
set of low-computing level functions.
The control code (Fig. 5) is an automatically
created task with the same complexity, format
and length of input data as the real task, with the
only difference being that the MEC server knows
the exact result, so it can be checked.
2. The MEC server distributes tasks between
the MEC computing nodes with additional redun-
dancy.
Redundancy (Fig. 6) additionally helps to
avoid mistakes in computing even on trusted de-
vices. MEC server will apply calculation results
and grant rewards only after at least 51% current
network nodes will present the same results.
3. The MEC server updates its trust level af-
ter the task is successfully completed (Fig. 7).
Each MEC server has its own “trust level”, depended on the control code ex-
ecution and the results of previously completed tasks. If the results of the check
code execution are correct, the trust level (Fig. 7) for this computing node in-
creases. Otherwise, if the device calculates the check code with errors, the level of
trust decreases until the node is completely blocked.
Fig. 5. Using a control code to
check the correctness of calcu-
lations
Fig 6. Use of additional redun-
dancy to protect calculations
from errors
Fig. 7. Using the trust level to select com-
puting nodes
А. Аstrakhantsev, L. Globa, О. Fedorov, D. Degtiarov, Y. Romanko, K. Romanii
ISSN 1681–6048 System Research & Information Technologies, 2024, № 2 90
ADVANTAGES OF THE PROPOSED APPROACH
The benefits of the proposed approach consist of two parts: the benefits to the us-
er and the benefits to the mobile operator. For the user, the proposed approach
provides:
Ease of setup: common network identifiers (e.g. IMSI) or blockchain wal-
let number can be used as the user's device identifier.
Mobility: MEC calculation can be done without bidding for mobile
phones/locations.
High level of security: The transaction is signed using blockchain tech-
nology. Mutual authentication allows verifying MEC server.
High reliability: Data verification, mirroring and check code are used to
protect against fraud and detect errors.
Network operator benefits:
Network resource economy (compare to existing solutions [9–18]): Dy-
namic map with node list (or another database / route table) does not need.
Low cost (compare to existing solutions [9–18]): no additional hardware
is required; a software upgrade can resolve this problem.
Easy UE selection and MEC load balancing: base station make a decision
based on the set of computing requirements.
Increasing Spectral Efficiency: peer-to-peer communication release high
load on cellular network.
CONCLUSION
The proposed approach for distributed edge computing in 5G allows identifying
and authenticating MEC participants, allocating additional resources for MEC
from the mobile network, including preparing point-to-point communications.
The method also assigns computing nodes and balances the load of edge computing
by modifying the messaging protocol between the base station and mobile devices.
The originality of the proposed approach is provided by two methods that
are further improvements to the methods of load balancing, selection of comput-
ing nodes for edge computing in a 5G network, data verification and error detec-
tion for the MEC system. These methods are optimized according to the criteria
of minimum network resources usage and have a time constraint.
The proposed approach allows the MEC server to verify the results of calcu-
lations and distribute data for computation according to the capacity of the com-
puting nodes.
Implementation of this approach allows the service provider to save network
resources and low cost of deployment. It also provides easy load balancing be-
tween computing nodes. This approach is more convenient for the user, as it does
not require the creation of additional identifiers and provides a high level of secu-
rity through the introduction of mutual authentication.
REFERENCES
1. F. Vhora, J. Gandhi, “A Comprehensive Survey on Mobile Edge Computing: Chal-
lenges, Tools, Applications,” 2020 Fourth International Conference on Computing
An improved approach to organising mobile edge computing in a 5G network
Системні дослідження та інформаційні технології, 2024, № 2 91
Methodologies and Communication (ICCMC), Erode, India, 2020, pp. 49–55. doi:
10.1109/ICCMC48092.2020.ICCMC-0009
2. S.S.D. Ali, H. Ping Zhao, and H. Kim, “Mobile Edge Computing: A Promising
Paradigm for Future Communication Systems,” TENCON 2018 - 2018 IEEE
Region 10 Conference, Jeju, Korea (South), 2018, pp. 1183–1187. doi:
10.1109/TENCON.2018.8650169
3. L.A. Haibeh, M.C.E. Yagoub, and A. Jarray, “A Survey on Mobile Edge Computing
Infrastructure: Design, Resource Management, and Optimization Approaches,” in IEEE
Access, vol. 10, pp. 27591–27610, 2022. doi: 10.1109/ACCESS.2022.3152787
4. D. Liu, A. Hafid, and L. Khoukhi, “Workload Balancing in Mobile Edge Computing
for Internet of Things: A Population Game Approach,” in IEEE Transactions on
Network Science and Engineering, vol. 9, no. 3, pp. 1726–1739, 1 May-June 2022.
doi: 10.1109/TNSE.2022.3150755
5. W. Zhang, G. Zhang, and S. Mao, “Joint Parallel Offloading and Load Balancing for
Cooperative-MEC Systems With Delay Constraints,” in IEEE Transactions on
Vehicular Technology, vol. 71, no. 4, pp. 4249–4263, April 2022. doi:
10.1109/TVT.2022.3143425
6. J. Ren et al., “An Efficient Two-Layer Task Offloading Scheme for MEC System
with Multiple Services Providers,” IEEE INFOCOM 2022 - IEEE Conference on
Computer Communications, London, United Kingdom, 2022, pp. 1519–1528. doi:
10.1109/INFOCOM48880.2022.9796843
7. W. Zhang, W. Fan, G. Zhang, and S. Mao, “Learning-based joint service caching
and load balancing for MEC blockchain networks,” in China Communications,
vol. 20, no. 1, pp. 125–139, Jan. 2023. doi: 10.23919/JCC.2023.01.011
8. H. Wu, L. Chen, C. Shen, W. Wen, and J. Xu, “Online Geographical Load Balancing
for Energy-Harvesting Mobile Edge Computing,” 2018 IEEE International Confer-
ence on Communications (ICC), Kansas City, MO, USA, 2018, pp. 1–6. doi:
10.1109/ICC.2018.8422299
9. “The Radio Access Network and its resource allocation method calculated based on
mixing fog,” Patent App. CN108243245A, 2020.
10. Debashish Purkayastha, Xavier De Foy, Robert G. Gazda, “Systems and methods to
create slices at a cell edge to provide computing services,” Patent App.
WO2018089417A1.
11. “Resource allocation methods based on mobile edge calculations,” Patent App.
CN109041130A, 2021.
12. “A kind of data distribution method based on MEC auxiliary in 5G networks,” Patent
App. CN108174421A, 2020.
13. “A kind of edge calculations method for 5G super-intensive networking scenes,”
Patent App. CN107333267A, 2019.
14. “A kind of 5G method of mobile communication and system based on MEC and
layering SDN,” Patent App. CN107404733A, 2020.
15. John Juha Antero Rasanen, Pekka Kuure, “Method and apparatus for implementing
mobile edge application session connectivity and mobility,” US Patent App.
US16/072,901 (US 2019/0045409 A1).
16. Mehran Moshfeghi, “System and method for discount deal referral and reward
sharing,” US Patent App. US13/484,261 (US20120316939A1).
17. Willem Jacobus van Niekerk, Marc van Niekerk, Brendon van Niekerk, Ernst Klein-
hans, “Method and system for online redistribution of data and rewards,” US Patent
App. US15/464,934 (US20170255981A1).
18. Dennis Detwiller, Robert Sparks, “Rewarding Users for Sharing Digital Content,”
US Patent App. US13/106,716 (US20120290308A1).
19. Pavel Mach, Zdenek Becvar, Mobile Edge Computing: A Survey on Architecture and
Computation Offloading. Available: https://arxiv.org/abs/1702.05309
А. Аstrakhantsev, L. Globa, О. Fedorov, D. Degtiarov, Y. Romanko, K. Romanii
ISSN 1681–6048 System Research & Information Technologies, 2024, № 2 92
20. Q.V. Pham et al., “A Survey of Multi-Access Edge Computing in 5G and Beyond:
Fundamentals, Technology Integration, and State-of-the-Art,” IEEE Access, 2020.
doi: https://doi.org/10.1109/ACCESS.2020.3001277
Received 02.02.2024
INFORMATION ON THE ARTICLE
Andrii A. Astrakhantsev, ORCID: 0000-0002-6664-3653, National Technical University
of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine, e-mail:
andrii.astrakhantsev@nure.ua
Larysa S. Globa, ORCID: 0000-0003-3231-3012, National Technical University of
Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine, e-mail: lgloba@its.kpi.ua
Oleksandr V. Fedorov, National Technical University of Ukraine “Igor Sikorsky Kyiv
Polytechnic Institute”, Ukraine
Dmytro V. Degtiarov, National Technical University of Ukraine “Igor Sikorsky Kyiv
Polytechnic Institute”, Ukraine
Yevgen O. Romanko, Interregional Academy of Personnel Management, Ukraine
Kyrylo A. Romanii, National Technical University of Ukraine “Igor Sikorsky Kyiv
Polytechnic Institute”, Ukraine
УДОСКОНАЛЕНИЙ ПІДХІД ДО ОРГАНІЗАЦІЇ МОБІЛЬНИХ
ПЕРИФЕРІЙНИХ ОБЧИСЛЕНЬ У МЕРЕЖІ 5G / А.А. Астраханцев, Л.С. Гло-
ба, О.В. Федоров, Д.В. Дегтярьов, Є.О. Романко, К.А. Романій
Анотація. Мобільні периферійні обчислення є важливим елементом забезпе-
чення ефективності мережі 5G в цілому, оскільки дозволяють зберігати дані та
виконувати обчислення на периферії мережі. В існуючих технічних рішеннях
для систем зв’язку не в повному обсязі вирішені питання розподілу наванта-
ження між обчислювальними вузлами; у більшості таких рішень не пропону-
ється метод балансування обчислювального навантаження з контролем поми-
лок у децентралізованій обчислювальній інфраструктурі з динамічно
змінюваним набором обчислювальних вузлів. Метою дослідження є розроб-
лення підходу до організації мобільних периферійних обчислень у мобільній
мережі 5G, який би виконував перевірку справжності серверів розподілу та
обчислювальних вузлів, керував процесом розподілу обчислювальних блоків,
мав процедуру перевірки коректності розрахунків та враховував параметри
обчислювальних вузлів під час розподілу. Для досягнення мети пропонується
застосувати метод балансування навантаження та вибору обчислювальних вузлів
для периферійних обчислень в мережі 5G, який дозволяє визначити наявні вузли
та здійснити розподіл обчислювальних блоків між ними, а також забезпечити
взаємну автентифікацію елементів інфраструктури з перевіркою даних та по-
шуком помилок для системи MEC, який дає змогу контролювати появу поми-
лок під час обчислень, захищати сервер від некоректних даних. Указаний ме-
тод оптимізовано за критеріями мінімуму використовуваних ресурсів мережі і
мінімального часу виконання обчислень. Такі вдосконалення дозволяють під-
вищити ефективність мобільних периферійних обчислень у 5G мережі.
Ключові слова: мережі 5G, мобільні периферійні обчислення, розподіл за-
вдань, протокол обміну, балансування навантаженням, верифікація обчислень.
|
| id | journaliasakpiua-article-309711 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2025-07-17T10:28:33Z |
| publishDate | 2024 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/01/e8c89a21b50c9a129b3c70517be03e01.pdf |
| spelling | journaliasakpiua-article-3097112024-08-11T01:12:49Z An improved approach to organising mobile edge computing in a 5G network Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G Astrakhantsev, Andrii Globa, Larysa Fedorov, Oleksandr Degtiarov, Dmytro Romanko, Yevgen Romanii, Kyrylo мережі 5G мобільні периферійні обчислення розподіл завдань протокол обміну балансування навантаженням верифікація обчислень 5G network mobile edge computing task allocation scheme call flow load balancing task verification Mobile edge computing is an important element in ensuring the efficiency of the 5G network as a whole, as it enables data storage and computing at the network edge. Existing solutions do not fully address the issues of load distribution between computing nodes, and most solutions do not offer methods for verifying computations and controlling errors. Accordingly, this paper aims to develop an approach to the organization of mobile edge computing in a 5G mobile network that would authenticate distribution servers and computing nodes, manage the process of distributing computing nodes, have a procedure for verifying the correctness of calculations, and take into account the parameters of computing nodes during distribution. To achieve this goal, we propose to use the developed method. The method of load balancing and selection of computing nodes for edge computing via 5G allows for identifying available nodes and distributing computing blocks among them. It also provides mutual authentication of elements and includes a method of data verification and error detection for the MEC system. The provided solution allows for controlling errors during calculations and protecting the server from incorrect data. These methods are optimized according to minimum network resources and computing time criteria. These improvements increase the efficiency of mobile edge computing in a 5G network. Мобільні периферійні обчислення є важливим елементом забезпечення ефективності мережі 5G в цілому, оскільки дозволяють зберігати дані та виконувати обчислення на периферії мережі. В існуючих технічних рішеннях для систем зв’язку не в повному обсязі вирішені питання розподілу навантаження між обчислювальними вузлами; у більшості таких рішень не пропонується метод балансування обчислювального навантаження з контролем помилок у децентралізованій обчислювальній інфраструктурі з динамічно змінюваним набором обчислювальних вузлів. Метою дослідження є розроблення підходу до організації мобільних периферійних обчислень у мобільній мережі 5G, який би виконував перевірку справжності серверів розподілу та обчислювальних вузлів, керував процесом розподілу обчислювальних блоків, мав процедуру перевірки коректності розрахунків та враховував параметри обчислювальних вузлів під час розподілу. Для досягнення мети пропонується застосувати метод балансування навантаження та вибору обчислювальних вузлів для периферійних обчислень в мережі 5G, який дозволяє визначити наявні вузли та здійснити розподіл обчислювальних блоків між ними, а також забезпечити взаємну автентифікацію елементів інфраструктури з перевіркою даних та пошуком помилок для системи MEC, який дає змогу контролювати появу помилок під час обчислень, захищати сервер від некоректних даних. Указаний метод оптимізовано за критеріями мінімуму використовуваних ресурсів мережі і мінімального часу виконання обчислень. Такі вдосконалення дозволяють підвищити ефективність мобільних периферійних обчислень у 5G мережі. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2024-06-28 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/309711 10.20535/SRIT.2308-8893.2024.2.06 System research and information technologies; No. 2 (2024); 82-92 Системные исследования и информационные технологии; № 2 (2024); 82-92 Системні дослідження та інформаційні технології; № 2 (2024); 82-92 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/309711/301139 |
| spellingShingle | мережі 5G мобільні периферійні обчислення розподіл завдань протокол обміну балансування навантаженням верифікація обчислень Astrakhantsev, Andrii Globa, Larysa Fedorov, Oleksandr Degtiarov, Dmytro Romanko, Yevgen Romanii, Kyrylo Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G |
| title | Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G |
| title_alt | An improved approach to organising mobile edge computing in a 5G network |
| title_full | Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G |
| title_fullStr | Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G |
| title_full_unstemmed | Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G |
| title_short | Удосконалений підхід до організації мобільних периферійних обчислень у мережі 5G |
| title_sort | удосконалений підхід до організації мобільних периферійних обчислень у мережі 5g |
| topic | мережі 5G мобільні периферійні обчислення розподіл завдань протокол обміну балансування навантаженням верифікація обчислень |
| topic_facet | мережі 5G мобільні периферійні обчислення розподіл завдань протокол обміну балансування навантаженням верифікація обчислень 5G network mobile edge computing task allocation scheme call flow load balancing task verification |
| url | https://journal.iasa.kpi.ua/article/view/309711 |
| work_keys_str_mv | AT astrakhantsevandrii animprovedapproachtoorganisingmobileedgecomputingina5gnetwork AT globalarysa animprovedapproachtoorganisingmobileedgecomputingina5gnetwork AT fedorovoleksandr animprovedapproachtoorganisingmobileedgecomputingina5gnetwork AT degtiarovdmytro animprovedapproachtoorganisingmobileedgecomputingina5gnetwork AT romankoyevgen animprovedapproachtoorganisingmobileedgecomputingina5gnetwork AT romaniikyrylo animprovedapproachtoorganisingmobileedgecomputingina5gnetwork AT astrakhantsevandrii udoskonalenijpídhíddoorganízacíímobílʹnihperiferíjnihobčislenʹumereží5g AT globalarysa udoskonalenijpídhíddoorganízacíímobílʹnihperiferíjnihobčislenʹumereží5g AT fedorovoleksandr udoskonalenijpídhíddoorganízacíímobílʹnihperiferíjnihobčislenʹumereží5g AT degtiarovdmytro udoskonalenijpídhíddoorganízacíímobílʹnihperiferíjnihobčislenʹumereží5g AT romankoyevgen udoskonalenijpídhíddoorganízacíímobílʹnihperiferíjnihobčislenʹumereží5g AT romaniikyrylo udoskonalenijpídhíddoorganízacíímobílʹnihperiferíjnihobčislenʹumereží5g AT astrakhantsevandrii improvedapproachtoorganisingmobileedgecomputingina5gnetwork AT globalarysa improvedapproachtoorganisingmobileedgecomputingina5gnetwork AT fedorovoleksandr improvedapproachtoorganisingmobileedgecomputingina5gnetwork AT degtiarovdmytro improvedapproachtoorganisingmobileedgecomputingina5gnetwork AT romankoyevgen improvedapproachtoorganisingmobileedgecomputingina5gnetwork AT romaniikyrylo improvedapproachtoorganisingmobileedgecomputingina5gnetwork |