Hindawi

Wireless Communications and Mobile Computing

Volume 2022, Article ID 4397610, 14 pages
https://doi.org/10.1155/2022/4397610

Research Article
Sixth-Generation (6G) Mobile Cloud Security and Privacy
Risks for AI System Using High-Performance
Computing Implementation

Srinivasa Rao Gundu,1 Panem Charanarur,2 Kashinath K. Chandelkar,2

Debabrata Samanta ,3 Ramesh Chandra Poonia ,3 and Partha Chakraborty 4

1
Government Degree College Sithaphalmandi, Hyderabad, Telangana, India
2
School of Cyber Security and Digital Forensic, National Forensic Sciences University, Goa Campus, Goa, India
3
Department of Computer Science, CHRIST (Deemed to Be) University, Bengaluru, Karnataka 560029, India
4
Department of Computer Science and Engineering, Comilla University, Cumilla-3506, Bangladesh

Correspondence should be addressed to Partha Chakraborty; [email protected]

Received 22 February 2022; Accepted 18 April 2022; Published 5 May 2022

Academic Editor: Mohammed Hammoudeh

Copyright © 2022 Srinivasa Rao Gundu et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

The exchange of information from one person to another is called communication. Telecommunication makes it possible with
electronic devices and their tools. The scientist Alexander Graham Bell has invented the basic telephone in 1876 in the USA.
Telephones now have the new format in the form of mobile phones, which are the primary media for communicating and
transmitting data. We are using 5th-generation mobile network standards. Still, there are some requirements for the users that
are believed to be solved in the 6th-generation mobile network standards. By 2030, all of the people would be using 6G. The
computing model in the cloud is not dependent on either the location or any specific device that would provide the service. It
is an on-demand computational service-oriented mechanism. Combining these two technologies as mobile cloud computing
provides customized options with more flexible implementations. Artificial intelligence is being used in devices in many fields.
AI can be used in mobile network services (MNS) to provide more reliable and customized services to the users, such as
network operation monitoring, network operation management, fraud detection, and reduction in mobile transactions and
security to the cyber devices. Combining cloud with AI in mobile network services in the 6th generation would improve
human beings’ lives, such as zero road accidents, advanced level special health care, and zero crime rates in society. However,
the most vital needs for sixth-generation standards are the capability to manage large volumes of records and excessive-
statistics-fee connectivity in step with gadgets. The sixth-generation mobile network is under development. This generation has
many exciting features. Security is the central issue that needs to be sorted out using appropriate forensic mechanisms. There
is a need to approach high-performance computing for improved services to the end-user. Considering three-dimensional
research methodologies (technical dimension, organizational dimension, and applications hosted on the cloud) in a high-
performance computing environment leads to two different cases such as real-time stream processing and remote desktop
connection and performance test. By ‘narrowing the targeted worldwide audience with a wide range of experiential
opportunities,’ this paper is aimed at delivering dynamic and varied resource allocation for reliable and justified on-demand
services.

1. Introduction other one listens, and vice versa. However, if these conver-
sants are located at two different geographical locations
Exchange of information from one person to another is and conversation is needed, it is done using some telecom-
called communication. It is done in many ways. The first munication approach. Telecommunication is done using
form of communication is verbal. It means one speaks, the some electronic tools that would easily express the required
2 Wireless Communications and Mobile Computing

Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6

Figure 1: Pictorial representation of telephones.

information or message quickly. The electronic devices age. Speedwise, the 1st generation has experienced 2 kbps.
transmit the message to a longer distance within a short The speed 200 kbps in 3G and in 5G is aimed at being as fast
period. ‘Tele’ means ‘distance’and ‘phone’means ‘communi- as 35.46 Gbps [8]. Figure 2 shows analog signal, digital sig-
cation’ [1]. Therefore, telephone means distant communica- nal, and analog-to-digital signal conversion.
tion. It means communicating with a person who is in a long In the earlier days, people used to have lesser transac-
distance [2]. Alexander Graham Bell is behind all these sci- tions on mobiles. They were realistic, and their works have
entific investigations today about mobile communications. existed in the real world. But today’s world is turned much
He invented this device in 1876 in the USA. Today, many more virtual rather than practical [9]. Perhaps much of the
people might not have seen, but till the early 1990s, rotary shopping is done on mobiles today. The characteristics of
dial telephones has dominated almost all areas in public the goods are provided with all minute details and photos
places, government offices, and individual private houses. on the commercial sites they are being purchased with digi-
Figure 1 shows a pictorial representation of telephones. tal currency. As a result, a considerable amount of data
Later, push-button phones have dominated the market. would be generated, and such data need to be used. Hence,
These are a few examples of landline phones. Later, mobile there is a need for cloud computing. These cloud services
phones came into existence. Mobile phones are portable could store a large amount of data related to the customers
phones; they do not have any wires connected to any tele- and the sellers online [10, 11]. Processing these data and
phonic links. They run on a wireless network process. These the applications would increase the speed enormously. The
mobile phones will have a simcard which will provide a data is acquired from much more remote locations using a
unique identity to the subscriber. From anywhere on this simple internet connection with some simple secured proto-
globe, no other person would have this number. These sim- cols. The cloud space is enough to maintain many volumes
cards are removable and insertable into the mobile [3]. All of information for many years. A cloud is formed from the
today’s mobile phones would have CPUs. But these CPUs collaboration of many data centers to provide a reliable ser-
will run on lesser electrical or battery power with less mem- vice to customers. This enormous amount of data stored on
ory and more sophisticated work. Today’s modern phones the servers is not possible; thereby, 3rd party assistance is
will have many features like radio, music, navigational tools, used in the form of the cloud to support the existing services
and video games. Mobile phones are the primary media for to the customers [12]. When a user enters some details for
the communication and transmission of data. Cloud com- his commercial or financial services in his/her mobile needs
puting is a device-independent and location-independent, to connect to a web application, thereby which it should
on-demand computational service-oriented mechanism. connect to a server that is at a remote location. Inserting
Combining these two technologies as mobile cloud comput- noninformation bits into data is known as bit stuffing. Keep
ing provides customized options with more flexible imple- in mind that overhead bits and packed bits are two very dif-
mentations. The recent trends in this field are ‘narrowing ferent things. Bits that do not really contain data but must be
the targeted global audience with a wide range of experien- sent are known as overhead. It is possible to use bit stuffing
tial opportunities’ [4, 5]. Recent times suggest that signifi- to synchronize several channels before multiplexing, to
cant businesses are being carried out using mobile cloud match the rates of two separate channels, and to perform
computing. Most companies do not own a cloud and rely length-limited coding when necessary. Run-length-
on cloud service providers. Lower investments and usage- restricted programming was done in order to restrict the
based billing are the major attraction for major businesses number of consecutive bits of the same value in the data that
migrating to the cloud [6]. Still, there are unsolved problems is to be conveyed. After the maximum allowable number of
in this field, such as the need to improve the GPS with low consecutive bits, a bit of the opposite value is added. Bit suf-
electricity consumption, screens with higher resolution, focation does not guarantee that the data delivered is intact
and three-dimensional cameras to mobiles. In the 1st- when it reaches its destination. For the most part, this is just
generation mobiles, the network signals used are analog by a means to guarantee that the transmission begins and stops
nature. The 2nd-generation mobiles are digital networks. in the right locations. The delimiting flag sequence in a data
These networks use approximately 12-15 kbps. In this gener- connection frame typically consists of six or more consecu-
ation, people have enjoyed text messaging. 3rd generation tive 1 s. A single bit is tucked away in the message to distin-
has increased new frequency bands thereby data transfer guish it from the flag in the event of a similar sequence. At
rates have increased. 4th generation has given away to the the conclusion of every string of five consecutive 1 s, an addi-
internet accessibility, HD TV, games, and cloud-based ser- tional 0 bit is added for good measure. The stuffed 0 s are
vices [7, 8]. 5th generation is under development. It is aimed removed by the receiver after each five-one sequence. When
at reducing the latency and increase the efficiency of cover- the message has been unstuffed, it is forwarded up to the
Wireless Communications and Mobile Computing 3

Signal
Signal

Time Digital signal Time

Analog signal

0100111101

In ADC Out

Analog to digital signal conversion

Figure 2: Analog signal, digital signal, and analog-to-digital signal conversion.

higher tiers. As a result of the coding rate instability caused resources is important in MEC systems because their effects
by bit stuffing, it is not always a reliable method of data are tied together. According to the majority of current
transmission. All the server services need to be achieved research, network slicing and MEC optimization simply
using some web browser that works like a mediator. Gener- takes into account mobile devices’ resource slicing, energy
ally, Web application service is provided with a blend of scheduling, and power allocation without factoring the oper-
added service-oriented architecture. Thereby, it can be ator’s income. We cannot have a different network for each
known as a sophisticated internet-based application. application situation since it is not viable. Using network
slicing to overcome this problem has been suggested as an
1.1. Aim of the Study. The aim of study is by ‘narrowing the option. In network slicing, numerous conceptually distinct
targeted worldwide audience with a wide range of experien- networks are operated on top of a shared physical infrastruc-
tial opportunities,’ the goal was to deliver dynamic and var- ture, which is the primary characteristic. Network resources
ied resource allocation for reliable and justified on-demand may be dynamically and flexibly distributed to logical net-
services. In fields including manufacturing automation, work slices in response to specific on-demand service needs
health care technology, and transportation, the fast growth via the use of network slicing. There has been a shift in net-
of the mobile information business has spawned a variety work edge service needs due to the fast growth of Internet of
of mobile applications. Mobile devices may not be able to Things (IoT) and cyberphysical systems. Existing works, on
handle the processing demands of these apps, which often the other hand, have not been able to keep up with the ever-
need large calculations and severe latency requirements. changing demands of these applications. Therefore, how to
Mobile edge computing (MEC) is a potential technology enable various apps in a shared physical infrastructure is still
for supporting these kinds of applications since it installs an unanswered question. If you are looking for a way to sup-
edge servers near mobile devices to facilitate computation port edge services with special needs, mobile edge comput-
offloading. MEC has the ability to significantly increase the ing (MEC) is a potential solution. With MEC, latency and
computational power of mobile devices. There are draw- energy usage are lowered since it is closer to the edge of
backs to offloading computations from mobile devices to the network than traditional cloud computing systems. In
edge servers, such as increased transmission delay and both academics and business, the combination of network
energy usage. This may lead to nonnegligible compute time slicing with MEC has piqued a lot of interest. Large-scale
due to the limited CPU power available on the edge servers. energy harvesting fog computing networks have been devel-
A cooperative allocation of processing and communication oped with dynamic network slicing design to enhance
4 Wireless Communications and Mobile Computing

resource efficiency and balance workloads across fog nodes. High bandwidth and low latency are essential for edge appli-
Fog radio access networks were the subject of more study cations. Distributed data centers, or distributed clouds, are
into hierarchical radio resource allocation (F-RAN). Net- created by service providers to accomplish this goal. To
work slicing resource allocation in MEC systems was inves- put it another way, there is no such thing as “the cloud”; it
tigated by most studies, but they did not consider the is a collection of resources that can be located in any loca-
computing resources. However, the dynamic demand of ser- tion, including the customer’s location. With the MEC plat-
vices has not been addressed in most recent publications on form, you have the option of employing either a server or
the use of network slicing to MEC systems. Cloud forensics CPE for edge computing. It is possible to employ a
and high-performance computing would offer improve- software-defined access layer as an extension of a cloud.
ment. However, the authors of this paper suggest consider- Open source hardware and software, including SDN and
ing the fog computing-based dynamic network slicing NFV, are being used in the majority of edge computing pro-
design would enhance the computational potentialities. jects. There are a number of popular MEC use cases that
may be listed, including data analytics, location tracking ser-
1.2. Motivation for the Study. Since the world has become vices, IoT, and augmented reality. Hosting stuff locally, such
very gregarious for handling impeccable knowledge to savvy as movies, on a server in your area, driving patterns, road
and scrupulous educators, in this era of scientific inventions. conditions, and other vehicle movements may all be moni-
Today, it is aimed at gravitating to innovative ideas in the tored by a connected automobile as an example of an Inter-
field of science and technologies with human touch. It is net of Things (IoT) application. On-time delivery of
the time to tease off and endorse to bring out precocious predictive and prescriptive information is critical. Sensor
advancements in the field of science and technologies. data must be gathered, processed, and analyzed at the edge
Today, all are committed to the creation, dissemination, in order to deliver low latency insights to the driver. This
and acquisition of knowledge through our research and means that MEC may be used for a broad range of applica-
developments. As unprecedented methodologies phase in tions requiring immediate reaction time, such as autono-
to create a pathfinder way, new innovative and gravitate mous cars, virtual reality, robots, and other immersive
ideas are needed. It is needed to focus on thriving technolog- media. Indeed, new technical ideas might bring about a par-
ical trends for the prosperity and opportunities for the adigm change from 4G to 5G. An ongoing effort is needed
human endeavor in recent times and in the coming years, from both academic and industrial sectors to successfully
too. use MEC in 5G networks. MEC technology and its prospec-
tive uses and applications are first discussed in this study.
2. Review of the Literature This is followed by a summary of the most recent research
findings on the integration of MEC with 5G and other
According to Laxman Shankar [13], the Bigonet cloud-based emerging technologies. In addition, we provide an overview
mobile framework was established to meet the requirements of edge computing testbeds, experiments, and open source
of the subsequent standard level in cloud apps. Such follow- efforts. This section sums up findings from current research
ing level apps would require parallel processing apart from and discusses problems and future perspectives for MEC
several connected systems needing to be handled in the dis- studies. According to Chihani et al. [15], contextual infor-
tributed environment. They show how the network-based mation is used to characterize the conditions of entities
activity analysis in the operating system and robust multi- (end-users and systems) and their responses. Apps require
point to multipoint apps in the distributed switching, highly such data to make adjustments in their processing approach
accessible, scalable infrastructure, and user-friendly inter- as a response to make change in the context of the end-users.
faces make developing and managing numerous parallel On the other hand, like the machine-to-machine-based
and concurrent processes easier [14]. While network slicing communication approach, the cohesiveness of many systems
allows on-demand services, many of their apps need multi- makes keeping the established contextual information
access edge computing (MEC) structural design to be extremely challenging. The lack of a scalable and simply cus-
deployed in the 5th-generation distributed system. Origi- tomizable solution for efficient context management is a
nally, edge computing is a key force behind the 5th- show-stopper in a big and dispersed system. This research
generation and 6th-generation mobile standard distributed presents a scalable cloud-based context management para-
systems; its role in network slicing remains unknown. The digm for dealing with contextual data in large distributed
5th-generation distributed system will use multiaccess edge contexts. This framework was developed and proven to
computing as its structural architecture. As a result of multi- allow programs to subscribe to context changes and declare
access edge computing (MEC), traffic and service processing how context data should be handled using an XML-based
is moved away from the cloud and closer to the end user. programming language. Since there is a need of enormous
The network edge processes, analyzes, and saves data instead data upload and download in all the time, there is also a need
of transmitting it all to the cloud. High-bandwidth applica- of transmission of data continuously. It is always needed to
tions benefit from reduced latency and real-time perfor- use extensible markup language- (XML-) based program-
mance since data is collected and processed closer to the ming language. It is now required for a file format that can
client. In addition, MEC provides a network edge IT service store, transport, and reconstruct arbitrary data. Since then,
environment and cloud computing capabilities. Distributed it establishes a set of guidelines for encoding documents in
data centers at the edge are often used to implement MEC. a human- and machine-readable manner. Serialization is
Wireless Communications and Mobile Computing 5

the primary goal of XML. XML is a markup language that high-speed TCP/IP. Empirical studies imply that the TCP/IP
labels, categorizes, and arranges information structurally. protocol can process data at up to 85 Mbps under the UNIX
The data structure is represented by XML tags, which also operating system when a good data connection layer interface
contain information. The data included within the tags is is provided, needing a memory bandwidth of 172 Mbyte/s.
encoded according to the XML standard. A separate XML According to Shivabhai and Babu [20], high-
schema (XSD) specifies the metadata required for interpret- performance computing (HPC) is utilized to address vast
ing and validating XML. The canonical schema is another and difficult computer issues by combining scientific
name for this. A well-formed XML document follows basic research and industry innovation. The ultimate objective is
XML rules, while a valid XML document follows its schema. to connect a high-performance computing cluster to a
Characters from the unicode repertoire are used exclusively web-based interface that hides the complexities of high-
in XML documents. Except for a tiny number of control performance computing. The massive resource broker
characters that are explicitly banned, every unicode charac- (MRB) bridges the gap by providing a web-based task sub-
ter can appear in an XML document’s content. The encoding mission, administration, and monitoring interface for high-
of the unicode characters that make up the document can be performance computing. Naive users submit jobs, monitor
identified using XML. Therefore, it is advised to use XML- clusters, and produce reports using the MRB portal. Saving
based programming language for all the needs of 6G. time, delivering more productive output, reducing mistakes,
According to Segal et al. [16], heterogeneous computing is and improving consistency are all priorities for large
a potential approach for high-performance and energy- resource brokers. The relevance of MRB, its implementation,
efficient computing. Till now, the high-performance hetero- and significant elements such as job submission, monitoring,
geneous computing industry was dominated by discrete analysis, benefits, and workflow are all discussed in this arti-
GPUs, but new options based on APUs and FPGAs have cle. An approach for systematically finding and isolating
emerged. These innovative concepts have the potential to floating point implementation errors in high-performance
increase energy efficiency significantly. Heterogeneous com- multiple CPU computing systems has been devised, accord-
puting based on FPGA has a lot of promise since it allows ing to Ghoshal [21]. A validation suite has been created and
one-of-a-kind hardware for data-centric parallel applica- put through its paces. The results reveal that the implemen-
tions to be designed. Most significant impediment to FPGA tation was flawed. Proposed and prototyped guidelines for
acceptance as high-performance computing systems is their proper implementation are presented.
programming difficulties. According to El-Araby et al. [17],
parallel computers with integrated FPGA chips are high- 2.1. Artificial Intelligence Usage in Mobile Networks. AI is an
performance reconfigurable computers. The Cray XT5 h acronym for artificial intelligence. It is an antonym for natural
and Cray XD1, the SRC-7 and SRC-6, and the SGI Altix/ intelligence. Natural intelligence is an exhibiting nature of nat-
RASC are examples of such systems. In classic high- ural living beings, and artificial intelligence is done by the
performance computers, the single-program multiple data devices. These devices are generally called agents [22]. Devices
method is used to execute parallel programs on HPRCs are aimed at achieving their targets more precisely than natu-
(HPCs). FPGAs have previously been used as coprocessors ral beings. These agents can have the capacity to store in mem-
in similar systems. Overall, system resources are typically ory, they can learn, and they can make a decision and express
underutilized because reconfigurable processors are appropriately. All these activities simply follow human intelli-
deployed unevenly compared to standard processors. gence thereby possible to deploy in many other fields where
Because of the asymmetry, the SPMD programming human involvement is not possible. AI can be used in mobile
approach cannot be utilized on these devices. In this paper, network services (MNS) such as to provide more reliable and
we describe a resource virtualization technique. Underuti- customized services to the users. Some of them can be (i) net-
lized processors will be able to share reconfigurable proces- work operation monitoring, (ii) network operation manage-
sors thanks to this method. According to Ba et al. [18], the ment, (iii) fraud detection and reduction in mobile
message forwarding interface has already become a common transactions, (iv) security to the cyber devices, (v) customer
communication library for distributed memory computing services, (vi) marketing management, (vii) digital assistance,
systems. Since new MPI standard versions are published, and (viii) customer relationship management.
several MPI implementations have been made publicly avail-
able. In various implementations, different approaches are 2.2. The Sixth-Generation Mobile Network. The fifth-
employed. Communication performance is the key in mes- generation network standard would provide new functional-
sage passing-based systems, so choosing an appropriate ities; along with this, it would also provide improved service
MPI implementation is crucial. According to Kim et al. quality in contrast with the fourth generation network stan-
[19], the TCP/IP protocol suite is the most widely utilized dard. The fifth-generation network standard would encom-
networking computer. They looked on the role of TCP/per- pass numerous new additional strategies, the latest
formance IPs in data sharing under the UNIX operating sys- frequency bands, such as the mmWave and the optical spec-
tem once a connection was established in this study. By tra, superior spectrum utilization and control, and the com-
measuring accurate data for various areas of the protocol bination of licensed and unlicensed bands. Nevertheless, the
implementation, we identify the key bottlenecks and define fastest boom of the data center-based centric and automated
the maximum performance limitations. They took memory systems can also exceed the competencies of 5G Wi-Fi struc-
bandwidth requirements into account when processing tures [23].
6 Wireless Communications and Mobile Computing

A few devices, together with virtual reality (VR) devices, (5) Energy capability needs to be increased. Present-day
would go to head past 5G because they would require not less batteries are not competent enough for 6G
than 10 Gbps facts charge. The key drivers of sixth generation
(6) Security has to be increased. There, it should not
might be the convergence of past capabilities, including com-
leave a chance for hackers and crackers
munity densification, excessive throughput and reliability at a
high level, lesser energy consumption, and higher data for the (7) The spectrum share must be focused. There, it
connectivity. The sixth-generation machine could additionally should no race but cooperation between the
maintain the traits of the previous generations, which included operators
new offerings with new technology.
As per reliable sources like Cisco, these features include (8) Governing consortium: until now, no formal entity
driving and maintenance of different types of vehicles, exists that would govern the technology in the com-
assigning various tasks and achieving the targets more pre- ing days
cisely by robots, running and maintaining other drones in (9) It is needed to be established
commercial and noncommercial areas, maintenance and
safeguarding of home appliances, and supporting them in
IOT, supporting many intelligent devices in the fields of 3. Service Requirements for 6G
constructions, care, and industries [23]. The feature also
includes upcoming technologies like augmented reality, The 6G wireless system will have the following key factors:
extended reality, and virtual reality. Speed of internet access
would increase in geometrical proportion. The influence of (1) Need of security improvement in 6G: 6G is expected
this technology would bring out some exciting offers to soci- to contain sophisticated technology that will provide
ety such as (1) zero road accidents, (2) advanced level special considerably greater privacy and security to the user’s
health care, and (3) zero crime rates in community. data, such as channel coding and estimation in the
physical layer, as well as multiple access in the MAC
2.3. Sixth Generation and Its Challenges. By 2030, all of the layer, among other things. It is widely assumed that
people would be using 6G. The upcoming 6G needs more 6G will employ quantum communication
sophisticated services by using 1 TB per second [24]. This
(2) Mobile broadbands need to increased in 6G: 6G will
means one would have the devices which would receive its
link devices with extremely low data rates, such as
signals 8000 GB per second. This prediction is based on a
biosensors and IoTs, as well as devices with large
study at Sydney University. It would have decentralized net-
data rates, such as HD video streaming in smart cit-
works. Not based on one single operator, perhaps a collec-
ies. As a result, 6G is now supported by mobile
tion of operators would cohesively provide the services to
broadbands. They ought to be elevated.
the user. Science fiction like communicating with others in
space could be easily possible with this 6G. China has (3) Lesser latency communications that are ultrareliable:
already started the 6G development project. Very recently, 6G services should rely on ultrareliable low-latency
China has launched 5G. There is going to be a tough chal- communications (URLLC) services, which have a
lenge for the implementation in 6G. This new wireless com- latency of less than 1 millisecond and a 99.999 per-
munication will require ultrareliable low-latency cent reliability, according to the Electronics and
communication networks. Not only this, the upcoming Telecommunications Research Institute (ETRI). This
devices should possess the speed of terabit/second speeds URLLC focuses on communications and gives strin-
[25]. This requires making much more advancement in the gent performance guarantees.
field of electronics.
(4) Machine type machine communication in 6G:
machine type communication, which includes both
2.4. Major Demanding Advancements. Some more aspects
mission essential and huge connectivity characteris-
need to be advanced.
tics, is expected to be a crucial cornerstone of 6G
(1) Computational power has to be increased. The pres- development, driven by a desire to supply vertical-
ent computational power is not sufficient even for specific wireless network solutions [26].
5G. Accommodating present-day computational (5) More energy capability need to be increased in 6G:
power for 6G would be unimaginable due to their ability to operate in higher-frequency
(2) The reliability has to be increased. Mission critical bands than previous generations, 6G gadgets require
tasks in 6G needs a high level of reliability and significantly more energy than previous generations.
consistency As a result, energy consumption and efficiency are
key concerns that need to be addressed right away
(3) The network coverage needs to be widespread.
Antennae numbers and density have to rise more (6) Lesser network access overcrowding in 6G: the key
concern is user service delivery. After establishing
(4) The network speed needs to be much faster. It connectivity, speed, capacity, and latency are used
requires THz of speed to assess a network’s efficiency
Wireless Communications and Mobile Computing 7

Cloud

Figure 3: Pictorial representation of sample 6G network usage.

(7) Communication integrated with artificial intelligence: 100 GHz, (f) area traffic capacity 1000 Mbps/m2,
artificial intelligence, automated systems, and 6G and (g) connection density 107 devices/km2
mobile communications can all be said to be intercon-
nected. The fundamental technology for automated (ii) System latency: (a) end-to-end latency 0.1 ms and
systems is artificial intelligence. The fundamental driv- (b) delay jitter 10-3
ing factor behind automation is a variety of machine (iii) System management: (a) energy efficiency 1 TB/J,
learning algorithms and deep learning principles. (b) reliability 10-9 packet error rate, and (c) mobility
Real-time learning is a principle that allows an auto- 1000 km/h.
mated system to perform well. When discussing 6G
communication technologies, automated systems are Figure 3 shows the pictorial representation of sample 6G
important. To get the most out of the potential of 6G network usage.
communications, many systems must be automated
when linked throughout the world. 3.2. Key Factor Requirements in Sixth Generation. Essential
(8) Lesser backhaul in 6G: mobile backhauling is a phys- requirements of the sixth-generation mobile communication
ical channel that connects radio controllers and base standards could be as given below:
stations. It is often implemented using optical fibers
or microwave radio connections. Backhaul systems (1) High-performance networking: compared with fifth-
today often rely on cost-effective packet-switched generation communications, sixth-generation com-
technologies (e.g., Wi-Fi and WiMAX). munications would help networking and connect
most people. Presently, in highly populated areas,
this task may not be easy. Even in the case of less
3.1. Key Performance Indicators in 6G. These below are key populated regions deep below the water surface, the
performance indicators in upcoming 6G. They are as follows: communication signals are also not possible to con-
nect. Sixth-generation communications will use
(i) System capacity: (a) peak data rate in 1000 Gbps, (b) novel conversation networks to support different
experienced data rate 1 Gbps, (c) peak spectral effi- data types such as audio and video. This would reach
ciency 60 b/s/Hz, (d) experienced spectral efficiency a new kind of experience in touch using virtual net-
3 b/s/Hz, (e) maximum channel bandwidth working technology and involvement everywhere
8 Wireless Communications and Mobile Computing

(2) Higher energy efficiency: in sixth-generation mobile tics based on game theory, contract ideas, and
network standards, higher energy capability necessi- many more are widely employed to achieve a high level
ties for Wi-Fi gadgets with charging limits would exist. of network operation. However, those optimization
Apart from this, battery for the mobile has become lost theories will not yield the best results in large-scale
for a lesser time. Hence, lengthy battery existence and time-varying variables and multiobjective scenarios
usage would be the most considerable points for
(ii) Application intelligence: applications connected to
research in this standard communications. Consider
fifth-generation networks are becoming increasingly
the case of unmanned aerial vehicles (UAVs) and elec-
intelligent. Intelligent applications are one of the
tric vehicles (EVs) in the Wi-Fi era. Recently, a new
applicational needs for sixth-generation networks.
technology called symbiotic radio (SR) was released
FL enabled Wi-Fi communication technology, allow-
to overcome power issues for wireless devices [27].
ing devices to connect to sixth-generation networks
The authors coined the word “unmanned aerial vehi-
and execute a wide range of intelligent applications
cles” with reference to power issue to the upcoming
6th-generation mobile technoloical standard. (iii) Furthermore, as a human-centric network, the sixth-
Unmanned aerial vehicles recently started using sym- generation community’s excessive intelligence will
biotic radio (SR) technology; it is particularly with ref- provide intelligent services satisfaction-oriented and
erence to wireless devices to overcome the power issue. individualized. FL, for example, would provide clients
Smart electricity control is every other promising with individualized healthcare and referral services
strategy for dynamically optimizing the stability of
electricity needs for supply. AI-based solutions for 3.3. Key Factor Requirements in Sixth Generation. In terms of
black communication technologies would be critical evaluation, respondents in business units are less inclined to
for optimizing power utilization and power usage transfer power to appropriate IT for picking public clouds
scheduling for all Wi-Fi devices in an ever-changing (41%), determining/advising on which apps pass to the cloud
technological environment and more complex optimi- (45%), and selecting private clouds (40%). Overall, cloud chal-
zation goals. For the optimization, available and lenges are declining: expertise, security, and spend tie for first
updated machine learning-based technologies, such place lack of assets/understanding, the top cloud assignment
as deep reinforcement learning (DRL), could be in recent years. Security concerns have also decreased to
applied. This would optimize the computing task 25%, down from 29% last year. The most frequently stated
devolvement in determining the Wi-Fi gadgets and venture among seasoned cloud users (24 percentages). Users
an improved working and suspended time scheduling are focusing on costs resulting from significant waste in cloud
solution, lowering energy consumption. It integrates spending. The latest trends in this field are server computing
passive backscatter devices with a lively transmission and multivendor approach. Here are some predictions for
device. Ambient backscatter communication is a clas- cloud computing for a near future as follows:
sic example of SR, which allows network gadgets to
use ambient RF indicators to transfer data without (1) Hybrid infrastructure: hybrid infrastructures are
requiring vigorous RF transmission, allowing for now restricted to public and private cloud infrastruc-
battery-free communication. Smart electricity control tures. Hybrid infrastructures will broaden its reach
is every other promising strategy for dynamically opti- to meet the agencies’ demands for efficiency, safety,
mizing the stability of electricity needs for supply control, and cost-effectiveness. The services would
(3) A high-level security and privacy: the available be improved in terms of performance while also pro-
research particularly specialized in network through- viding dependability and scalability
put, reliability, and delay in 4G and 5G communica- (2) No. of apps would increase on cloud: cloud comput-
tions. However, wireless communicational exchange ing is the way of the future for businesses, and com-
security and privacy issues have been disregarded panies have begun to prepare their programs to be
to some extent in recent years cloud-compatible. Generally, roughly 70% of busi-
(4) A high-level intelligence: The sixth generation’s nesses consider cloud to be a distinguishing factor,
high-level intelligence would be useful in providing and 65% of businesses spend about 10% of their bud-
users with high-quality, tailored, and intelligent ser- get on cloud services
vices. As shown below, the sixth-generation high-
intelligence standard would contain (i) operational 4. Research Methodology
intelligence, (ii) application intelligence, and (iii) ser-
vice intelligence There are three dimensions to the research methodology:

(1) Dimensional technical: it is necessary to collect

information that is flexible, static, and verifiable. In
(i) Traditional network operations entail a slew of a remote computing context, such data should be
resource optimization and multigoal overall perfor- processed using accessible criminal-related logical
mance optimization difficulties. Optimization tac- methods
Wireless Communications and Mobile Computing 9

Specialized compute nodes (Fat, GPU etc)

Fast interconnet

Login node
Attached storage

Data transfer node

Compute nodes

Figure 4: High-performance computing-based scalable ‘cloud forensics-as-a-service’ readiness framework.

(2) Dimension of organization: it is primarily consid- licated across many servers to improve distributed
ered in the context of distributed computing. They parallelism, fault tolerance, and disaster recovery. In order
are cloud customers and cloud service providers, to get started with a typical virtual machine, the administra-
respectively. The programs are hosted in the cloud tor must write down or describe the sources of configuration
information. Inefficient network configurations almost
(3) As a result, the services of one cloud service provider always lead to overloading or underutilization when addi-
are linked to those of other cloud service providers. tional VMs join a network. EC2 is an excellent example of
The inquiry process is complicated by the reliance a web service that delivers scalable computing capacity in a
on multiple parties. As a result, it becomes quite cloud, such as Amazon’s EC2. EC2 allows customers to con-
difficult struct virtual machines and manage user accounts through-
out their usage. In order to create a virtual cluster, one or
4.1. Testing Environment Plan. The following are the exper- more physical clusters must host a virtual machine (VM).
imental plans for testing important components of the cloud The borders of the virtual clusters are shown as different
forensic tools operating mechanism. To verify the viability of lines. In a virtual cluster, virtual machines (VMs) are
the technique’s execution, a few tests must be carried out at dynamically provisioned. Nodes in the virtual cluster might
various stages. Some tests are also designed to set bench- be both real and virtual computers. On a single physical
marks for the underlying components. The systems that node, many VMs each running a separate operating system
must be tested before cloud forensics may be implemented may be set up. To operate in a virtual machine, you need a
are listed in the table below. guest operating system, which is typically different from
There should be an HPC cluster with a large number of the operating system of your host computer. Consolidating
nodes. This cluster will need to be hosted on a high- numerous functions on a single server is the goal of virtual
performance computing machine. On the virtualization machines. This will considerably increase the server’s usage
cluster nodes, virtualization software is installed, allowing and the flexibility of the application. Parallelism, fault toler-
one node to function as several nodes. Amazon’s Elastic ance, and disaster recovery may be achieved by colonizing
Compute Cloud (EC2) offers cloud computing with elastic- (replicating) VMs over numerous servers. Virtual cluster
ity. VMs from one or more physical clusters are put on dis- deployment, monitoring, and administration across huge
persed servers to form virtual clusters. The goal of clusters, as well as resource scheduling, load balancing,
employing virtual machines is to integrate multiple func- server consolidation, and fault tolerance, are all part of this.
tions on a single server. Virtual machines (VMs) can be rep- The ability to efficiently store a large number of virtual
10 Wireless Communications and Mobile Computing

Table 1: Systems needs of cloud forensic tool implementation.

Name RAM CPU Cores Threads Disk space

HPC cluster 24 GiB 2 Intel Xeon [email protected] GHz 12 24 100 Mib/s
Virtualization 128 GiB [email protected] Hz 16 32 1 GiB/s
Cluster work station SGiB Intel Core i7- [email protected] GHz 4 8 100 Mib/s

Table 2: Data transfer calculations theoretically. cessing requirements. There are two sorts of parallel
Size of the evidence 100 Mbit/s 10 Gbit/s hardware and software designs.
100 GiB 02 : 56 : 00 00 : 01 : 43
4.2. Testing. This test is designed to determine how many
500 GiB 14 : 45 : 22 00 : 08 : 30 nodes each forensic cloud user has. Begin by evenly distrib-
1 TiB 30 : 16 : 32 00 : 16 : 43 uting all nodes over all conditions, starting with one.
5 TiB 153 : 25 : 30 01 : 26 : 43 Increase the number of instances one by one until the pro-
cessing speed is too slow to bear. Keep track of the number
Table 3: Evidences needed to use in testing processing. of nodes required to make a scenario operate. The HPC clus-
ter and the workstation performed similarly. It is unable to
Ubnistl.gen3.E01 Nps-209-domexusers.E01 give it with the same or equal storage device that the virtua-
Compressed 855 Mib 4.05 GiB lization cluster provided due to technological restrictions. If
Uncompressed 1.98 GiB 42 GiB
a quicker storage was used, processing performance would
improve. Figure 4 shows high-performance computing-
based scalable ‘cloud forensics-as-a-service’ readiness frame-
work. Table 1 shows the system needs of cloud forensic tool
machine images is critical in a virtual cluster system. The implementation.
nodes in the various virtual clusters are shown in the diagram Simultaneous imaging and upload and encryption are
by the various colors. The most crucial thing to consider when observed. Scenario 1 includes setting up the first processing
dealing with a big number of virtual machine images is how to activities that need be conducted on the evidence, as well
store them in the system. Operating systems and user-level as potential imaging and encryption of the evidence for
programming libraries are among the most popular installs Internet transmission. The data processing should begin at
for most users and applications. Preinstalled templates for the same time as the imaging procedure. To speed up pro-
these software programs are available. Users may create their cessing, the client should allow for simultaneous data trans-
own software stacks with the help of these templates. The tem- mission and picture creation. The client should be run twice:
plate VM may be used to create new OS instances. These once to just upload the test data to the server, and again to
instances allow the installation of user-specific components, submit the test data while concurrently creating an image
such as programming libraries and apps. of it. This will be used to test if the client can perform both
This would allow virtualization-based cluster systems to tasks at the same time while maintaining a 120 MB/s upload
have more nodes, each with 24 gigabytes of RAM. It is nec- pace. Each node had 24 Gigabytes of RAM and 12 threads.
essary to set up task processing from the beginning. These
responsibilities will be carried out on the evidences, as well 4.3. Cloud Security Mechanism in HPC Environment. Once
as the possible imaging and encryption of the proof for the image has finished uploading, tools that would need
internet broadcast. This level’s testing must be completed the entire image would be run. Because the devices will func-
entirely on a workstation. Data can be processed quickly, tion in a cluster, the results will be stored in the working
and complex computations can be completed quickly, directory. The virtual computer’s connection will be safe-
thanks to high-performance computing. This high- guarded. Only the virtual computer holding the evidence
performance computation can be done on any computer submitted by the cloud forensics expert will be accessible.
with a 3 GHz CPU. This computer is capable of performing The virtual machine displays the results of the tools as they
over 3 billion operations per second. A supercomputer, on are complete on the cluster. They require access to tools that
the other hand, has similar powers. are unable or unwilling to operate on a collection during this
High-performance computers can now do miracles. It virtual machine. They cannot access evidence that belongs to
advances significantly in a wide range of sectors, including another cloud forensics specialist since virtual machines are
science and technology. It benefits technology such as the separated from one another. The information will be
Internet of Things, artificial intelligence, and 3D imaging, encrypted using the analysis virtual machine for the cloud
to name a few. The high-performance computing operating forensics professional.
system includes computation, networking, and data storage.
A cluster is a group of computers connected together to per- 5. Results Obtained
form high-speed processing. Workstation clusters are now a
viable substitute for supercomputers. In general, these work- To protect the data’s secrecy and integrity throughout trans-
station clusters were preferred for all high-performance pro- mission, encryption must be applied to the evidence being
Wireless Communications and Mobile Computing 11

Table 4: One node for one cluster outputs.

Parameter-1 Parameter-2 Parameter-3 Parameter-4 Parameter-5 Parameter-6

Factors HPC Virtualization Work station HPC Virtualization Work station
Sec 114.2 43.7 102.1 892.2 559.2 889.9
Mebibyte per second 19.22 46.1 19.06 47.3 71.9 47.2
Mebibyte 2106 2106 2106 42949 42949 42949
Bound CPU CPU CPU CPU -None- CPU
Threads 12 12 8 12 12 8
RAM 24v 24 8 24 24 8

Table 5: Outputs of workstation clusters.

Ubnist1.gen3.E01 Nps-2009-domexuser.E01
Factors Node∼1 Node∼2 Node∼3 Node∼4 Node∼1 Node∼2 Node∼3 Node∼4
Sec 34.8 43.7 25.8 4.9 459.1 22.8 198.4 44.2
Mebibyte per second 13.9 12.8 19.2 108.4 21.7 51.3 51.7 222.3
Mebibyte 520 520 520 546 10737 10737 10737 10737
Bound C.P.U. C.P.U. C.P.U. -none- C.P.U. C.P.U. C.P.U. Input and output

transferred between the client and server. The advanced expensive VPN products. Using the latest technology, firms
encryption system (AES) is a well-known symmetric key are able to do the following: for both View VDI and RDS
encryption standard with a proven track record. For data secu- published desktops, the PCM facilitates communication
rity during network transmission, this is deemed to be suffi- between the view connection server (VCS) and end points.
cient. Table 2 shows data transfer calculations theoretically. Remote access to the PCM may be achieved by imple-
In real-time stream processing: to speed up the analysis menting the PCM on the corporate network or in a DMZ.
process, data must be processed as soon as it becomes avail- PCoIP Security Gateway (PSG) must be configured if the
able to the server. There are two methods for processing dig- PCM is placed in a firewalled environment (DMZ). The
ital evidence. The two possibilities are bulk data analysis and PSG may be added to the PCM as an optional component.
a file-based technique. Data chunks are treated as a whole in VMware View Security Gateway may be replaced with the
bulk data analysis, regardless of their structure. The data ova package without the need for a VPN connection, and
must be chunked and supplied to each tool, with suitable customers may access their remote workstations from the
nodes assigned. To handle data at the file level, several pro- Internet. The nodes are the clients in this scenario. The files
grams use file-based approaches. By extracting files from a can then be handed to appropriate tools for file-centric pro-
stream in real time, the forensic cloud cluster can perform cessing after they have been made available. Table 4 shows
file-level analysis. Table 3 shows evidences needed to use in one node for one cluster outputs. More functionalities, such
testing processing. as USB redirection and multimonitor support, are available
In remote desktop connection, performance test, the with PCoIP. The usefulness of the remote desktop protocol
cloud forensic investigator connects to a distant virtual will be determined in this test. We connect to the available
machine using VDI’s built-in remote desktop protocol. cloud forensic tools to find out. Table 5 shows outputs of
There are two protocol alternatives for VMware Horizon workstation clusters. Table 6 shows outputs of high perfor-
View VDI: PCoIP and Microsoft’s RDP. As an alternative mance clusters. Table 7 shows outputs based on virtualiza-
to Microsoft RDP, the PCoIP protocol for Microsoft RDS tion cluster.
is available. For terminal services, the PCoIP protocol is
now available for use with PCoIP devices, including low- 5.1. Advantages of Present Study. We studied reports and
maintenance, ultrasecure zero clients, for increased perfor- employed forensic tools after making my judgments to verify
mance across any network. Customers of Teradici Arch are if the virtual machine was still useful. The virtual machine
able to do the following: by employing the PCoIP protocol should react quickly and without any noticeable lag. Before
to broker and manage both VMware View and RDS session deciding on the optimum processing option for digital data
desktops with VMware View Manager, you can provide a in forensic cloud tools, it is critical to establish a benchmark
rich, interactive experience across any network. The APEX for how rapidly a forensics workstation can handle digital
server offload card can protect and ensure a consistent user evidence. When parallelization is employed, both the virtua-
experience regardless of task or activity level while comply- lization and HPC clusters must test the speeds that each tool
ing with stringent government and security mandates and can achieve.
being virus-resistant; it can reduce TCO with low- A variety of forensic cloud workflow components must
maintenance PCoIP zero clients and eliminate the need for be studied to identify the particular forensic cloud system.
12 Wireless Communications and Mobile Computing

Table 6: Outputs of high-performance clusters.

Ubnist1.gen3.E01 Nps-2009-domexuser.E01
Factors Node∼1 Node∼2 Node∼3 Node∼4 Node∼1 Node∼2 Node∼3 Node∼4
Seconds 44.7 63.7 29.2 9.8 479.5 259.3 167.2 54.3
Mebibyte per second 12.4 7.9 14.7.8 48.7 21.8 39.4 64.9 199.8
Mebibyte 520 520 520 546 10737 10737 10737 10737
Bound CPU CPU CPU -None- CPU CPU CPU Input and output

Table 7: Outputs based on virtualization cluster.

Ubnist1.gen3.E01 Nps-2009-domexuser.E01
Factors Node-1 Node-2 Node-3 Node-4 Node-1 Node-2 Node-3 Node-4
Sec 19.0 25.8 13.5 5.9 239.2 169.2 164.3 46.8
Mebibyte per second 24.6 18.8 34.8 89.4 43.9 61.6 65.3 224.8
Mebibyte 519 520 520 546 10737 10737 10737 10737
Bound C.P.U. C.P.U. C.P.U. -none- C.P.U. C.P.U. C.P.U. Input and output

These tests show discover just how much capacity and pro- upload sites by one. Keep track of how long it takes for each
cessing power a forensic cloud installation can handle. file to upload. Continue to add one facility at a time until any
user’s upload time becomes prohibitive. Therefore, there is a
need to search an alternative way to conduct more compre-
6. Limitations of Present Study
hensive comparison.
The advantages of this model are as follows:

(i) The total analyzing time of large volumes of data is

7. Conclusions
reduced by utilizing the capabilities of a high- There is a need of such a congregation of cohesive redressal
performance computing environment and modify- prognostication which would surely make a solid impression
ing current tools to work inside this context with a comprehensive knowledge on many issues and will
(ii) If a smaller department does not have access to increase the knowledge of sharing for a cohesive success
commercial software, it can use it remotely due to global technological trends. The world is being
embarked on to a greatest reset. This would be an ever-
(iii) Allows for teamwork. Because the data is stored in changing innovation and technological advancement. After
the cloud, anyone with the right authority from a detailed study, it is observed that the method of resource
the case owner can access it and perform extra management techniques are differed one from the other
analysis and one strategy may be useful for real-time interactive
application that may not be suitable for some other applica-
6.1. Recommendations. The goal of this test is to see if data tion area. Therefore, there is a need to find a solution using
can be uploaded to a forensic cloud. It uses several data set high-performance computing which would be suitable for
sizes and uploads and downloads each data set separately mobile cloud computing also. The sixth-generation mobile
from different forensic cloud providers. Keep track of how network is under development. By 2030, all of the people
long it takes for each file to upload. Calculate how long a sin- would be using 6G. Combination of mobile networks with
gle user would take to perform the task. This test is used to cloud computing environment provides customized options
see if a forensic cloud environment can handle a large num- with more flexible implementations. AI can be used in
ber of uploads. Upload and download a 500 GB or larger mobile network services (MNS) such as to provide more reli-
data collection from two forensic cloud facilities at the same able and customized services to the users. However, the most
time. Keep track of how long it takes for each file to upload. vital needs for sixth-generation standards is the capability of
Increase the number of upload sites by one and reupload the managing large volumes of records and very excessive-
files. Looking at the limitation of this present study, the goal statistics-fee connectivity in step with gadgets. Sixth genera-
of the nodes per job test is to conclude the optimal node tion has many exciting features. Security is the major issue
numbers for every task. This test is aimed at examining if that needs to be sorted out using appropriate forensic mech-
data can be uploaded to a forensic cloud environment. It is anisms. There is a need to approach high-performance com-
using various data set sizes, we upload and download each puting for the improved services to the end user. This
data set and upload and download a data collection with a approach would provide dynamic and versatile resource
size of 500 GB or more from two forensic cloud facilities allocation for reliable and warranted on-demand services
simultaneously. Keep track of how long each file takes to by narrowing the targeted global audience with a wide range
upload. Reupload the files after increasing the number of of experiential opportunities.
Wireless Communications and Mobile Computing 13

8. Scope for the Future Work ence on Mobile Cloud Computing, Services, and Engineering,
pp. 235-236, San Francisco, CA, USA, 2015.
The combination of 6th-generation mobile network stan- [7] S. Abolfazli, Z. Sanaei, M. Alizadeh, A. Gani, and F. Xia, “An
dards with cloud computing along with artificial intelligence, experimental analysis on cloud-based mobile augmentation
cloud forensics, and high-performance computing would in mobile cloud computing,” IEEE Transactions on Consumer
offer improvement in the user experiences; however, for Electronics, vol. 60, no. 1, pp. 146–154, 2014.
managing the huge volumes of records and very excessive- [8] P. Bondada, D. Samanta, S. A. Chaudhry, Y. B. Zikria, and
statistics-fee connectivity in step with gadgets, there is a need F. Ishmanov, “Efficient neighbour feedback based trusted
to check with machine learning which would automate the multi authenticated node routing model for secure data trans-
analytical model building. mission,” Sustainability, vol. 13, no. 23, p. 13296, 2021.
[9] J. Y. Zhang, P. Wu, J. Zhu, H. Hu, and F. Bonomi, “Privacy-
Data Availability preserved mobile sensing through hybrid cloud trust frame-
work,” in 2013 IEEE Sixth International Conference on Cloud
The evaluation data that support the findings of this study Computing, pp. 952-953, Santa Clara, CA, USA, 2013.
are available on request from the corresponding author. [10] S. S. Qureshi, T. Ahmad, and K. Rafique, “Mobile cloud com-
puting as future for mobile applications - implementation
Disclosure methods and challenging issues,” in 2011 IEEE International
Conference on Cloud Computing and Intelligence Systems,
The funder had no role in the design of the study; in the col- pp. 467–471, Beijing, China, 2011.
lection, analyses, or interpretation of data, in the writing of [11] H. S. Alqahtani and G. Kouadri-Mostefaou, “Multi-clouds
the manuscript, or in the decision to publish the results. mobile computing for the secure storage of data,” in 2014
IEEE/ACM 7th International Conference on Utility and Cloud
Computing, pp. 495-496, London, UK, 2014.
Conflicts of Interest
[12] M. M. Fuad and D. Deb, “Cloud-enabled hybrid architecture
The authors declare no conflict of interest. for in-class interactive learning using mobile device,” in 2017
5th IEEE International Conference on Mobile Cloud Comput-
ing, Services, and Engineering (MobileCloud), pp. 149–152,
Acknowledgments San Francisco, CA, USA, 2017.
We wish to thank the late Mr. Panem Nadipi Chennaih for [13] L. Shankar, “Bigonet mobile cloud framework,” in 2014 Sixth
his continued support for the development of this research International Conference on Communication Systems and Net-
paper, and it is dedicated to him. works (COMSNETS), pp. 1–3, Bangalore, India, 2014.
[14] S. D. A. Shah, M. A. Gregory, and S. Li, “Cloud-native network
slicing using software defined networking based multi-access
References edge computing: a survey,” IEEE Access, vol. 9, pp. 10903–
10924, 2021.
[1] V. L. Shirokov, “The basics of methodology for estimation and
choosing parameters of multi-service telecommunication sys- [15] B. Chihani, E. Bertin, and N. Crespi, “Enhancing M2M com-
tems,” 2nd International Conference on Dependability of Com- munication with cloud-based context management,” in 2012
puter Systems (DepCoS-RELCOMEX 07), vol. 10, pp. 191–197, Sixth International Conference on Next Generation Mobile
2007. Applications, Services and Technologies, pp. 36–41, Paris,
[2] K. Trapezon, T. Gumen, and A. Trapezon, “Basics of using France, 2012.
connected me system in the transmission of information sig- [16] O. Segal, N. Nasiri, M. Margala, and W. Vanderbauwhede,
nals,” in 2018 14th International Conference on Advanced “High level programming of FPGAs for HPC and data centric
Trends in Radioelecrtronics, Telecommunications and Com- applications,” in 2014 IEEE High Performance Extreme Com-
puter Engineering (TCSET), pp. 229–232, Lviv-Slavske, puting Conference (HPEC), pp. 1–3, Waltham, MA, USA,
Ukraine, 2018. 2014.
[3] T. Minde, S. Bruhn, E. Ekudden, and H. Hermansson, [17] E. El-Araby, I. Gonzalez, and T. El-Ghazawi, “Virtualizing and
“Requirements on speech coders imposed by speech service sharing reconfigurable resources in highperformance reconfi-
solutions in cellular systems,” in 1997 IEEE Workshop on gurable computing systems,” in 2008 Second International
Speech Coding for Telecommunications Proceedings. Back to Workshop on High-Performance Reconfigurable Computing
Basics: Attacking Fundamental Problems in Speech Coding, Technology and Applications, pp. 1–8, Austin, TX, USA, 2008.
pp. 89-90, Pocono Manor, PA, USA, 1997. [18] Z. Ba, H. Zhou, H. Zhang, and Z. Yang, “Performance evalua-
[4] M. Bahrami, “Cloud computing for emerging mobile cloud tion of some MPI implementations on workstation clusters,”
apps,” in 2015 3rd IEEE International Conference on Mobile in Proceedings Fourth International Conference/Exhibition on
Cloud Computing, Services, and Engineering, pp. 4-5, San High Performance Computing in the Asia- Pacific Region,
Francisco, CA, USA, 2015. pp. 392–394, Mississippi State, MS, USA, 2000.
[5] R. Gurunath, A. H. Alahmadi, D. Samanta, M. Z. Khan, and [19] K. Kim, H. Sung, and H. Lee, “Performance analysis of the
A. Alahmadi, “A novel approach for linguistic steganography TCP/IP protocol under UNIX operating systems for high
evaluation based on artificial neural networks,” IEEE Access, performance computing and communications,” in Proceed-
vol. 9, pp. 120869–120879, 2021. ings High Performance Computing on the Information
[6] S. Li and J. Gao, “Moving from mobile databases to mobile Superhighway. HPC Asia 97, pp. 499–504, Seoul, Korea
cloud data services,” in 2015 3rd IEEE International Confer- (South), 1997.
14 Wireless Communications and Mobile Computing

[20] P. V. Shivabhai and M. R. Babu, “An open source web-based

massive resource broker (MRB) for high performance com-
puting (HPC),” in 2016 International Conference on Research
Advances in Integrated Navigation Systems (RAINS), pp. 1–3,
Bangalore, India, 2016.
[21] S. Ghoshal, “A floating-point validation suite for high-
performance shared and distributed memory computing sys-
tems,” in Proceedings Fourth International Conference on
High-Performance Computing, pp. 88–93, Bangalore, India,
1997.
[22] S.-J. Han and S.-Y. Oh, “Evolutionary algorithm based neural
network controller with selective sensor usage for autonomous
mobile robot navigation in IJCNN’01,” in International Joint
Conference on Neural Networks. Proceedings (Cat.
No.01CH37222), Washington, DC, USA, 2001.
[23] S. V. Anand and S. P. S. Kumar, “A modular data link layer
(M-DALL) for NEXT GEN mobile terminals enabling wireless
aware applications: a platform independent software design,”
in 2010 Global Mobile Congress, Shanghai, China, 2010.
[24] J. Teng, X. Xuan, and Y. Bai, “A fast Q algorithm based on EPC
generation-2 RFID protocol,” in 6th International Conference
on Wireless Communications Networking and Mobile Comput-
ing (WiCOM), pp. 1–4, Chengdu, China, 2010.
[25] S. R. Gundu, C. A. Panem, A. Thimmapuram, and R. S. Gad,
“Emerging computational challenges in cloud computing and
RTEAH algorithm based solution,” Journal of Ambient Intelli-
gence and Humanized Computing, 2021.
[26] S. R. Gundu, C. A. Panem, and A. Thimmapuram, “Real-time
cloud-based load balance algorithms and an analysis,” SN
Computer Science, vol. 1, no. 4, p. 187, 2020.
[27] S. R. Gundu, C. A. Panem, and A. Timmapuram, “Robotic
technology-based cloud computing for improved services,”
SN Computer Science, vol. 1, no. 4, p. 190, 2020.