A simplified optimization for resource management in cognitive radio network-based internet-of-things over 5G networks

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 13, No. 6, December 2023, pp. 6513~6524
ISSN: 2088-8708, DOI: 10.11591/ijece.v13i6.pp6513-6524  6513
Journal homepage: http://ijece.iaescore.com
A simplified optimization for resource management in cognitive
radio network-based internet-of-things over 5G networks
Vani Bengaluru Purushothama, Sundaraguru Ramakrishnan
Department of Electronics and Communication Engineering, Faculty of Engineering, Sir Mokshagundam Visvesvaraya Institute of
Technology, Bangalore, India
Article Info ABSTRACT
Article history:
Received Mar 3, 2023
Revised May 24, 2023
Accepted Jun 4, 2023
With increasing evolution of applications and services in internet-of-things
(IoT), there is an increasing concern of offering superior quality of service to
its ever-increasing user base. This demand can be fulfilled by harnessing the
potential of cognitive radio network (CRN) where better accessibility of
services and resources can be achieved. However, existing review of literature
shows that there are still open-end issues in this regard and hence, the
proposed system offers a solution to address this problem. This paper presents
a model which is capable of performing an optimization of resources when
CRN is integrated in IoT using five generation (5G) network. The
implementation uses analytical modeling to frame up the process of topology
construction for IoT and optimizing the resources by introducing a simplified
data transmission mechanism in IoT environment. The study outcome shows
proposed system to excel better performance with respect to throughput and
response time in comparison to existing schemes.
Keywords:
Cognitive radio network
Five generation communication
Internet of things
Optimized resource allocation
Resource management
This is an open access article under the CC BY-SA license.
Corresponding Author:
Vani Bengaluru Purushothama
Department of Electronics and Communication Engineering, Faculty of Engineering, Sir Mokshagundam
Visvesvaraya Institute of Technology
Bangalore, Karnataka, India
Email: vaniganesh2003@gmail.com
1. INTRODUCTION
With an increasing adoption of internet-of-things (IoT), the dependency towards fifth generation (5G)
communication is increasing targeting towards facilitating computational resources by mobile edge computing
(MEC) [1]. The idea is to enhance quality-of-service (QoS) and reduce latency along with an efficient traffic
control system over backhaul networks [2]. At present, MEC is known to perform aggregation of idle resources
e.g., processing and storage from the mobile user over similar applications. In IoT environment, the availability
of cloud services, calculation of statistical processing, and other environmental monitoring is carried out by
MEC [3]. In this perspective, the MEC resources can be harnessed for increasing the capabilities of certain
mobile users using resources from other mobile users in IoT. However, there is a limitation associated with
this process of MEC owing to their restricted capacity of resources to facilitate service delivery to the mobile
user. Therefore, it is essential to develop an effective scheme of resource management while working over 5G
in IoT in order to control these MEC resources. This will also result in reduction of unwanted resource
allocation by user node. This problem can be sorted by resource management mechanism in IoT by using
aggregation and allocation strategies. The MEC is actually constructed by the resource allocation steps which
generates resource pool from one to another mobile device. The process of aggregation involves identifying
with declaring the idle resources and increases the resource pool mainly [4], [5]. The process of resource
allocation in IoT is all about provisioning services on the basis of available resources in MEC in order to cater

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 13, No. 6, December 2023: 6513-6524
6514
up the demands of the user’s request [6]. The performance of the services in an IoT significantly depends upon
the selection of allocation of services with higher specifications. However, the problem arrives in an IoT
environment in presence of dynamic users that result in potential resource variations with an increasing chance
of intermittent connectivity from MEC. Presence of this problems give rise to various other adverse issues with
respect to attended and allocated services in IoT. Irrespective of various work towards resource management
in IoT and 5G, not much optimal solution is explored [7]–[9]. Majority of the existing studies have been carried
out towards addressing specific problems of resource allocation and not much on aggregation. This results in
significance less availability of criteria leading to unavailability of services owing to resource insufficiency.
This could also result in blocking of services while attempting exploration of devices at the time of allocation
of resources. It may also degrade the QoS performance owing to service relocation. IoT is also witnessed to
improvise on operationalities while using cognitive radio network (CRN); however, one of the potential
challenges in this regard is the degradation of quality of channel in case of more number of secondary users
[10]. Apart from this, there is also a need to improve upon the optimization performance of the radio power so
that better QoS performance can be delivered.
This section discusses about the various approaches being carried out towards resource allocation
considering the case study of deployment in IoT. At present, there has been an investigation towards resource
management concerning the narrowband IoT. The discussion of Migabo et al. [11] discussed the utilization of
narrowband IoT that targets towards cost and energy efficiency in IoT. According to the study, there are still
open-end issues in energy efficiencies due to the reasons viz: i) fluctuation of spatial and temporal factors
associated with harvested energy, ii) latency issues, and iii) inclusion of iterative operation for offering reliable
coverage. Such issues are addressed to some extent with usage of cognitive radio network (CRN) as discussed
in work of Ejaz and Ibnkahla [12] where an optimization using cross layer is used for introducing a
reconfigurable approach that is capable of allocation resources in dynamic order with an explicit quality of
service. The study contributes toward formulation of a policy where a game logic is used for modelling the
reconfiguration for supporting better allocation of resources. Study towards narrow band IoT was also carried
out by Malik et al. [13] with a target of controlling overhead by formulating a data rate that is feasible to be
achieved. The study also incorporates interference aware scheme where iterative scheme is formulated towards
obtaining optimal solution for allocation of radio resources. Resource allocation has been studied with respect
to Khan et al. [14] where mobility environment has been considered. The author has mechanized a policy where
a software defined network (SDN) to leverage 5G technologies has been used for constructing framework for
flow control in vehicular network. Apart from this, it is also seen that energy harvesting through radio
frequencies also contributes towards resource management when IoT is integrated with cognitive networks.
Study towards this direction is carried out by Alzahrani and Ejaz [15] with a core goal towards increasing the
data transmission performance with better quality of service. The presented technique utilizes greedy approach
and integer linear programming. The paper also increases the energy efficiency for all the IoT nodes. Adoption
of machine learning scheme has also been witnessed towards improving resource management for nodes
deployed in ultra-dense network of 5G. The study of Yu et al. [16] conjointly uses artificial intelligence along
with blockchain for constructing this model. This technique utilizes less training time in order to ensure that
there is reduced delay of cumulative offloading while optimizing the resources. Similar adoption of
reinforcement learning is also found to be used in the work of Khumalo et al. [17] where fog computing is for
testifying the architecture of radio access network. The study addresses the challenges associated with such
architecture. The presented technique contributes towards allocation of the autonomous resources using
Q-learning where the outcome is witness to exhibit reduced delay. Nassar and Yilmaz [18] have also
implemented Reinforcement learning approach over fog computing where different types of Markov decision
models are used for allocation of resources. Usage of artificial intelligence is also seen in work of Lu et al. [19]
where a gated recurrent unit is used, the technique makes use of recurrent neural network and long short-term
memory (LSTM) in order to carry out predictive operation for better form of resource management.
Apart from this, the study also uses federate learning in order to secure the privacy of the data.
Li et al. [20] have discussed a model where game theory has been used for offering a unique power allocation
approach along with reduction of computational complexity. A three-dimensional matching problem has been
used for modelling the channel allocation while iterative Hungarian method is used to solve this problem over
the virtual devices. Apart from this, bio-inspired algorithms are also witnessed towards improving resource
allocation strategies in 5G network. The work carried out by Wu et al. [21] has used characteristics of slices for
allocation of resources. The study also makes use of the evolutionary approach towards relationship and
preference of personalized services using cellular automation for resource management. Lieria et al. [22] have
also utilized bio-inspired approach for optimizing resource management. A unique study carried out by
Sarrigiannis et al. [23] has discussed the connectivity of resource allocation with network function virtualization
in 5G network. The study has presented an embedding mechanism on the basis of latency by allocating specific
network virtualization factor followed by scaling and migration of it. The work carried out by Liu and

Int J Elec & Comp Eng ISSN: 2088-8708 
A simplified optimization for resource management in cognitive … (Vani Bengaluru Purushothama)
6515
Zhang [24] have addressed the problems associated with allocation and power of IoT node in order to maximize
the data transmission in 5G. The study implements Lagrange dual optimization scheme in order to solve issues
associated with power reduction and data transmission increment. Abozariba et al. [25] have used a multiple
access technology in order to perform resource management in IoT. The study has used non-orthogonal multiple
access for this purpose to reduce the number of sub-carriers and thereby control better resource management.
Song et al. [26] have developed a resource allocation scheme using virtualized network with cluster formation.
Liu et al. [27] have developed a scheme that enhances the resource utilization over fiber wireless. The model
also implements Q-Learning method for predicting traffic behavior and its dynamics using an embedding of
virtual network in IoT. Miuccio et al. [28] have implemented a sparse code multiple access technology for
allocation of resources in dynamic order. Kherraf et al. [29] have presented an assignment of workload as well
as provisioning of resources. The study has addressed the problems associated with multi-access edge computing
towards the computational burden. The study outcome shows better delay control. Study towards similar form
of problem is also solved using resource provisioning scheme presented by Ugwuanyi et al. [30]. This study has
presented deadlock avoidance scheme towards resource provisioning in order to offer maximized reliability. The
study has also targeted towards controlling of communication overhead.
All the above-mentioned techniques significantly assist in accomplishing targeted resource utilization
and still there are certain issues associated with existing literatures. The identified research problems are as
follows: i) Less emphasis towards CRN: The dynamicity of the traffic management involved in IoT are better
managed by CRN with respect to its cognitive ability and re-configurability. The facilitation of dynamic
spectrum access is enabled by CRN along proper utilization/reconfiguration of it. However, there are also
challenges in CRN which is related to fulfilling the demands for structured topology for handling smart city in
an IoT. At present, there is no report of any standard model towards optimizing the data transmission
performance over CRN based IoT use case scenario; ii) Non-inclusion of randomness in modelling: At present,
existing studies in IoT has either considered mobile nodes or considered fully static IoT nodes. There is no much
consideration of dynamic topology in smart city use case. Non-inclusion of this consideration will lead to
impractical modelling in data transmission of IoT. An efficient modelling should have an inclusion of random
number of users that are nearby to various forms of cells. Apart from this, there is also a need of considering
interference in this process in order to ensure practical mapping of environment in IoT; iii) Lack of structured
communication model: Existing studies on 5G in IoT mainly emphasized on using conventional access
technologies in order to perform uplink and downlink transmission. The communication mainly involves either
one or bidirectional transmission via macro and small cells in IoT environment where CRN is used. However,
this will contradict for running distributed application in the user equipment while using 5G communication.
There is a need of structured communication model which could perform a proper grouping and clustering of
the IoT devices as well as various network devices (e.g., switches, and gateway) for better management of
resources; iv) Adoption of sophisticated methodologies: Almost all the existing research work has used heavily
iterative as well as complex algorithm formation in order to achieve better resource management. However, it
should be noted that normal IoT devices (or smart appliances) are characterized by low processing capability as
well as restricted resource availability. Hence, execution of complex and heavily iterative operation will result
to drainage of excessive resources; and v) Low emphasis on optimization: Various machine learning and bio-
inspired approaches were used for optimizing outcomes but not towards cumulative resource management in
large scale. There is a need of cost-effective optimization approach which can balance computation and
communication demands in IoT
Hence, there are various dedicated research attempts towards resource management scheme in IoT with
claimed benefits. However, there are also explored issues reviewed from these literatures. The next section briefs
about issues identified from existing approaches. Therefore, the proposed system presents a novel framework
that introduces a unique process of implementing an optimization process with a target of better resource
management when IoT is integrated with CRN. The proposed system implements an analytical research
methodology in order to construct a model which is capable of performing optimization of the resources in IoT.
The core idea of the study is to ensure development of cost-effective optimization process towards computational
modelling harnessing CRN. According to the architecture exhibited in Figure 1, the first part of implementation
is about IoT deployment where a use-case of smart city has been designed. The second stage of implementation
is about developing a dual mode of connectivity i.e., inter and intra connectivity among the IoT nodes and base
station within the deployed area of smart city. The deployed area will mainly consist of primary network of
macro cell and small cell. The proposed study also considers availability of home access point user, core cell
users, and random users in its deployment area of CRN. The architecture further has a block of operation for
framing up the IoT ward considering mainly smart home, deployment area of smart home, and cumulative area
of smart city. The proposed study will use intra connectivity for communication in deployment area within a
smart phone while it will use inter-connectivity mode for cumulative area of smart city. The architecture of
proposed system is shown in Figure 1.

 ISSN: 2088-8708
6516
Control
System
IoT Deployment
Intra-Connectivity
Inter-Connectivity
Primary Network
(Macro-Cell)
CRN (Small Cell)
CRN deployment
Smart
Home
Deployment
Area of smart
home
Cumulative
area of
smart city
IoT Ward formation
Algorithm for Topology
Construction of Smart City
Algorithm for Optimizing
Resources in IoT
Mode of Connectivity
Figure 1. Proposed architecture
Further, the architecture offers a control of communication system by deploying two simplified
algorithms. The algorithms are responsible for topology construction followed by optimization of resources.
An extensive test environment is developed where various access control techniques are used over small cell
base station for assessing its throughput performance. Signal quality over presence of interference is further
assessed along with processing time of algorithmic operation in order to prove the optimization performance.
The next section elaborates about the system design involved in proposed system.
2. METHOD
This section discusses about the complete system design of proposed study which aims for optimizing
the process of resource management in IoT environment. The study assumes the deployment of 5G network as
a media of communication where the case study of smart city is considered as an assessment environment. The
complete implementation is carried out via two sequential algorithms where the first algorithm is associated
with communication set up in IoT environment while the second algorithm is associated with the optimization
of the resource management in IoT. In order to optimize management of resources within the devices involved
in IoT environment, the proposed system hypothesizes that a better placement of the nodes with an efficient
topology in IoT will significantly structure the process of resource-efficient data dissemination. The first part
of this implement is associated with intra-connectivity among the IoT nodes. Basically, intra-connectivity
communication refers to communication within a specific domain or location in a smart city. In a typical
household in smart city, many devices or appliances need to communicate with each other in order to control
it ubiquitously. In order to build such intra-device communication a setup of local IoT provides a cost-effective
solution. In the local IoT setup, devices with unique-identity number get collaborated through a controller
device with capacity of gateway as well as access point. The Figure 1 illustrates a scenario of such local IoT.
In order to construct a deployment scenario of all the IoT nodes with dedicated connection among
them, the proposed system makes use of graph theory. The diagrammatic representation of the above scenario
exhibited in Figure 2 is implemented using graph theory represented by 𝐺(𝑣, 𝑒), where the vertices v is further
classified into v1 representing a set of devices and v2 representing the base station. The variable e will represent
the connectivity link between the all the vertices as in the Figure 3.
It should be noted that above exhibited graph G is representation of one local IoT with specific number
of vertices. In case of smart city, there are possibilities of various such graphs which could represent houses or
a defined location with multiple IoT devices. The typical smart city possesses many locations (e.g., smart
home) with smart appliances running (vertices) which are connected by certain access points. A pictorial
representation of this is shown in Figure 4, where it can be seen that each smaller regions (local IoT) is arranged
in structure of an array.
The next operation of the proposed study is to develop an inter-connectivity functionality. The local
IoT requires connectivity to the global level to achieve many of the application goals; therefore, a global IoT
system is formed. In simpler representation, global IoT is an interconnected system of available local IoT
system to offer extensive coverage using 5G. The global IoT exploits the benefit of the CRN capacity of the

6517
respective base stations in both local IoT and global IoT to realize a connection between the base station of the
small-cell and the large cell using CRN and 5G communication standard. The justification behind the usage of
CRN in proposed system is to exploit the following capabilities viz. i) CRN offers the capabilities where the
available frequencies, which is basically a spectrum that is under-utilized, is identified by sensing operation of
an IoT devices. The decision towards selection a specific band, either occupied or idle, is carried out by CRN
at given location and time and ii) CRN also facilitates dynamic programming capability of the radio for the
purpose of enabling the communication system with another set of frequencies and access technologies
specified by the underlying hardware. The typical processing blocks of the node with cognitive technology is
as in Figure 5.
E-Lock
Wearable
RFID
Control System
[Control Algorithm, Communication Module]
E-Wash
Basin
E-
wardro
be
E-
Televisi
on
E-
Refrige
tor
Control System
[Control Algorithm, Communication Module]
E-Kitchen
(e-Appliance)
E-BedRoom
(e-Appliance)
v1
v2
v3
v4
v5
BS
e1
e2
e3
e4
e5
Figure 2. A context of local IoT in the smart home Figure 3. Graph modeling of local IoT as 𝐺(𝑣, 𝑒)
Data Aquisition
Processing Unit
Communication Unit
Node with Conginitve
Technology : CR-Node
Figure 4. Array of smart homes in a smart city Figure 5. Processing units/blocks of
the node with CR technology
The capacity of the CRN to share the spectrum resources comes from a spectrum sensing technology,
where the user equipment (UE) in the small cell is popularly known as secondary or the cognitive user (SU/CU)
and the UE of bigger cell is known primary user (PU). The Figure 6 shows a typical communication scenario
of the 5G based CR enabled IoT system.
Primary Network-Macro Cell CRN-Small Cell
Figure 6. 5G based cognitive radio enabled IoT

 ISSN: 2088-8708
6518
The design of the algorithm is carried out in sequential approach where primary emphasis is given
towards topology construction of smart city in IoT while the secondary approach is meant for optimizing the
resource management while performing data dissemination process. The briefings of the inclusive algorithms
are as in algorithm 1.
Algorithm 1. Topology construction of smart city
Input: λ1, da, ca
Output: (px, py)
Start
1. init λ1, da, ca
2. For i =1: h: ca
3. For j=1: h: da
4. ca=g1(i, j, λ1, da, ca)
5. [px, py]=g2(π1, π2)
6. End
7. End
End
The algorithm takes multiple input i.e., spatial distance λ between two local IoT (smart home),
deployment area of smart home da, and cumulative area of smart city ca. The algorithm gives an outcome of
effective coordinates of the IoT ward (px, py). The initial operation of this algorithm is to consider the spatial
distance λ along with deployment area da of local IoT. The spatial distance λ1 is summed up with deployment
area da in order to obtain the area of smart city (Line-1). It should be noted that proposed system considers
the distance between all individual IoT homes h to be uniform as exhibited in Line-2/3. An explicit function
g1(x) is developed that can carry out an update operation with regular interval of time. The updating is carried
out for the matrix considering all the input arguments λ1, da, and ca (Line-4). The outcome of this operation
is updated version of the topology with respect to the cumulative deployment area ca (Line-4). A unique
function g2(x) is deployed considering dual form of input arguments π1 and π2, where the former argument
represents a structured memory with retention of information related to deployment area da maintained in the
form of an array. It also retains information associated with (i+λ1*da). The latter argument (π2) represents a
structured memory with deployment area da maintained in the form of an array along with information
associated with (j+λ1*da). This part of the matrix-operation results in generation of a better position
information associated with the IoT ward i.e. (pxm, pym). The advantage of using these algorithmic steps is
that the model contributes towards better disposition of finding the better placement of the local IoT ward for
a given smart city. This strategy of implementation will further contribute towards superior form of scope as
well as extensive flexibility towards IoT node placement. This will also provide the user to consider the CRN
in order to carry out data transmission. The primary idea of this algorithm thereby performs structured
management of resources in presence of IoT nodes which is better position to perform cost effective data
transmission in IoT environment. The next step of the algorithmic operation is to carry out an optimization of
the resource management in dynamic form over 5G network. The briefing of the algorithmic steps is shown
in algorithm 2.
Algorithm 2. Optimizing resources in IoT
Input: ca, rg
Output: dprop
Start
1. init rg
2. field=g3(ca, rg)
3. Mm=field(ψ);
4. Pward→g4(Mm)
5. Ahome→ ηward* θ
6. deploy → Mhome(nrand)
7. If node>Fr
8. node→ha
9. Assign δ →node
10. Check γ
11. If δ =unused
12. assign δ →node
13. End
14. Estimate Dm=Euc(N(Px, Py), δs)
15. initiate dprop nodei→nodej
End
This part of the algorithmic operation is basically a continuation of the prior algorithm where currently
a novel strategy of optimization is carried out. In order to process this algorithm, the input arguments are

6519
cumulative deployment area ca associated with use case of smart city and range of transmission rg. The
resultant of successful data propagation dprop is formed after processing this algorithm. The computation of the
positional information of the IoT ward is initiated by this algorithm considering the spatial distance between
them along with their respective range of transmission r (Line-2). This process is carried out in order to evaluate
the coverage region of the smart city. The study considers ψ number of masks obtained in Line-3 which is
further continued by computing the positional information of the IoT ward Pward by applying function of
concatenation g4(x) as seen in Line-4. The next step of the operation is to deploy an access point in the local
IoT Ahome which is carried out on the basis of number of involved local IoT ward ηward and density θ (Line-5).
A specific matrix Mhome is formulated in order to retain all the positional information of the access point of
local IoT considering random IoT nodes nrand (Line-6). The resource management is carried out in further steps
of operation of this algorithm. A link of communication for all the user is set up by this algorithm on the basis
of received power. This is further continued by computing the distance on the basis of both access point within
local IoT and core cell. The access point of local IoT that is located in higher proximity is explored followed
by assessing a logical condition if the user’s position falls within the transmission range of the access point of
the local IoT. If the user is positively found to reside the transmission range than the access point of local IoT
is used by the user (Line-8). This operation is further continued by setting up assignment channel and
communication link (Line-9).
On the other hand, if there is no IoT device within this range of transmission of access point of local
IoT (Line-7), then core cell is communicated by that IoT node which is further followed by provisioning of
channels in an arbitrary fashion. The received signal strength γ is computed by the algorithm (Line-10); further
an assessment is carried out for unutilized communication channel δ. Therefore, if the condition finds the
channel to be unutilized (Line-11) the user is assigned to this channel in next step (Line-12). The minimal
spatial distance between the IoT node with allocated arbitrary channel as well as the other node is computed in
next step (Line-14). Further, the algorithm assesses the proximity of the target node with the core cell followed
by enabling the communication with the core cells. This operation is further continued by extraction of the
information of the deployed region over the smart city. The signal to interference noise ratio is computed
concerning all the nodes followed by obtaining the user channel as well as it also looks for another user who
is also utilizing the similar channel of communication. Further, all the positional information of the source IoT
nodes are obtained which results in interference along with interference power associated with all the IoT
nodes. The algorithm obtains the Euclidean distance for all the IoT nodes from main source node that ultimately
yields to confirmation of the communication path for transmitting the data packet. This operation finally ends
up with the data transmission between the transmitting and destination node (Line-15).
On the basis of the complete operation, it can be said that proposed algorithm offers supportability of
both centralized as well as decentralized management of resources. According to this algorithmic operation,
the network centrality of the process is carried out by the core cell while the access point of local IoT controls
the local network. The core cell in this model is established with the communication with the access point of
local IoT using a backhaul connection. The determination of the relative position of the access point of the
local IoT is carried out by the core cell. This process is further followed up by the synchronization of the
significant radio signals by access point of local IoT for obtaining expected core signal. By assigning a
particular capacity of the access point of local IoT that is directly proportional to the arbitrary number of users.
A coefficient of the resource distribution is considered in proposed model that is computed by ratio of
maximum capacity of the arbitrary user to the capacity of the access point of local IoT. A conditional logic is
formulated which attempts to find the equivalency of infinity value associated with coefficient of resource
distribution. In such case, all the arbitrary users will be provisioned with the resources from the access point
of local IoT. On the other hand, if the coefficient of resource distribution is found to be zero then home access
point will represent a closed access while if the value of coefficient of resource distribution is found to be 1
then the home access point will represent open access. The value of coefficient of resource distribution is
resides between 0 and 1 when it is a case of hybrid access. The random users are permitted to use limited
capacity allocated for all the type of user. The switching between the station can be done for any arbitrary user
in order to obtain enhance service from core base station. A user is permitted to assess other user in case it is
found to be active within the transmission range Tx as per the list of identity of owner of home access point.
However, the identity of the subscriber will be different while the right to reserve the capacity of home access
point is done by home access point itself while that of random users too.
Upon identifying the cumulative number of users in smart city deployed, the home access point
chooses number of random users and estimates throughout of owner as well as arbitrary users. The availability
of service for the arbitrary users is computed by core base station. In number of arbitrary users are minimized
by home access point to one every time when it finds the reserved throughput for the arbitrary user in home
access point is minimal compared to core base station. Hence, an effective resource management is carried out
for the considered deployed scenario of smart city in IoT using 5G communication system. The next section
discusses about outcome. A closer look at the proposed methodology and system design shows that proposed

 ISSN: 2088-8708
6520
system is highly progressive and extremely less iterative which will offer more benefits towards faster data
transmission along with optimizing resources. This will also justify the cost effectiveness of the model. The
next section discusses out outcomes achieves.
3. RESULT
This section discusses about the outcomes obtained after implementing the proposed system
discussed in prior section. The core idea of the proposed model is to offer resource optimization in an IoT
environment. A specific testbed is constructed for this purpose with the simulation parameters exhibited in
Table 1.
Table 1. Simulation parameters
Parameters Values
1 Number of base stations 1
2 No. of access point in local IoT 20 to 4,500
3 No. of active user 600
4 Interval of distributed time 500 s
3.1. Individual outcomes
The study introduces three different layers of communication introduce various kinds of noises as
well interferences and the suitable medium access control techniques (MACT) used in the small cell base
station or local IoT base station alleviates the challenges of such interferences. The effectiveness of the local
IoT base station largely depends on the technique used for establishing connectivity between the IoT nodes as
secondary user (SU) or cognitive user (CU) with a base station (BS) of a particular cell. The system model
introduces three types of MACT namely: i) MACT-O, ii) MACT-C, and iii) MACT-H.
In MACT-O approach, it permits all the registered nodes to get communicated within the
communication range of local IoT base station. In MACT-C approach, the local IoT base station does not
permits the communication to the registered nodes from a specific group of the local IoT node even if a better
signal quality is available. In MACT-H approach, the communication is granted for both registered and
nonregistered node with a constrained resource of the local IoT base station. The Figure 7 illustrates the
observations made for throughput by varying the number of local IoT base station respective to the methods of
MACT-O, MACT-C, and MACT-H.
Table 2 highlights the SINR outcomes obtained from individual analysis of proposed model. The
outcome shows that at the value of number of local IoT base station is 500 Figure 7, both MACT-O and
MACT-H gives same throughput whereas the MACT-C provides much higher throughput. The pattern of
throughput for MACT-O is nonlinear incremental, whereas MACT-C provides much higher throughput as
compare to MACT-O and MACT-H.
Figure 7. Evaluation of throughput for all IoT nodes based on resource allocation methods

6521
Apart from this, Figure 8 highlights the throughput evaluation with respect to various node densities.
It can be seen that higher throughput is achieved in proposed local IoT cell whereas increase of node density
affects the throughput slowly. This outcome evidently shows that proposed system is not dependent on increment
of node density to increase the throughput. Hence, a better optimization outcome can be exhibited here.
However, the pattern of increment is highly nonlinear as compare to both MACT-O and MACT-H.
The MACT-H provides linearly consistent increment in throughput with respect to increasing number of L-IoT
base station, therefore MACT-H is a consistent medium access control technology to achieve optimal resource
allocation in the context of 5G based CRN enabled-IoT ecosystem. This evaluation is further validated with
respect to comparative analysis with existing system.
Table 2. Comprehensive SINR performance
Noise (dB) SINR
0.1 0.4 0.9
-20 -0.1027 -0.098878 -0.09878
-15.6 -0.1027 -0.1027 -0.09488
-11.2 -0.09878 -0.09098 -0.09488
-6.8 -0.07929 -0.1027 -0.09488
-2.4 -0.04421 -0.07929 -0.09878
2 -0.07274 -0.01302 -0.09878
6.4 0.252 0.1351 -0.08708
10.8 0.4119 0.2598 -0.02082
15.2 0.5678 0.4275 0.1039
19.6 0.7081 0.5561 0.1936
24 0.8017 0.6574 0.3144
28.4 0.8446 0.7315 0.3651
32.8 0.8952 0.7705 0.4197
37.2 0.9264 0.7861 0.4625
41.6 0.9498 0.8134 0.4664
46 0.9615 0.8601 0.4742
50.4 0.9771 0.903 0.5171
54.8 0.9849 0.9576 0.5678
59.2 0.9693 0.9693 0.6419
63.6 0.9888 0.981 0.8718
68 0.9771 0.981 0.9888
72.4 0.9771 0.9849 0.9888
76.8 0.9771 0.981 0.9888
81.2 0.9732 0.981 0.9927
85.6 0.9849 0.981 0.9888
90 0.9849 0.9771 0.9849
Figure 8. Evaluation of throughput for all IoT nodes based on resource allocation methods
3.2. Comparative analysis
The proposed system is scripted in MATLAB where the idea of implementation is to carry out
assessment the outcome associated with resource management optimization. The assessment consists of two

 ISSN: 2088-8708
6522
scenarios of channel assignment that is confirmed to be highly suitably working with IoT when integrated with
CRN i.e. i) spectrum-based sharing and ii) spectrum-based allocation [31]. In order to evaluate the effectiveness
of proposed study, a comparative analysis is carried out considering the recent approach implemented by
i) Mansukhani and Ray [32] who has used spectrum-based sharing (SS) and ii) Tang et al. [33] who has used
spectrum allocation approach (SA). The implementation of existing system considers similar simulation
parameters in order to arrive at unbiased outcome of performance parameters with respect to throughput and
processing time. The analysis is hypothetically carried out emphasizing over the above-mentioned channel
assignment schemes from literatures.
Figure 9 highlights the outcome of throughput trend with respect to increasing number of home access
point. A better form of resource management in CRN can be ensured if the data delivery services are found to
be enhanced considering the larger dimension of IoT environment. The outcome shows that proposed system
offers significantly better throughput in contrast to existing mechanism of channel assignment. The prime
justification behind this is: The conventional mechanism of spectrum sharing scheme has used a complex
mechanism in order to compute the amount of the spectrum resources in random fashion to be shared among the
users. In this process, the system over utilizes the available channel capacity which degrades the throughput
performance over long run with increase of home access point. Similarly, the mechanism of conventional
spectrum allocation considers sharing of spectrum by primary users to cognitive users in order to enhance the
throughput performance. However, with increasing usage of the home access point, there is only a slight
improvement in obtaining optimal channel resulting in better mechanism compared to conventional spectral
sharing method by reduced in contrast to the proposed system. This is because the proposed system ensures the
controlling of the core cell controller to determine the relative position associated with home access point. This
results in significant improvement of throughout with increasing home access points and thereby the effect of
optimization is noted in this outcome. Apart from this, the proposed system offers a simplified binary logic for
representing home access point to either open access point or closed access point as well as it also supports
hybrid access mechanism. This operation of optimization carried out in proposed system also leads to faster
algorithm processing time in contrast to existing system as shown in Figure 10.
Figure 9. Comparative analysis of throughput Figure 10. Comparative analysis of processing time
A closer look into Figure 10 highlights that performance of both conventional channel assignment
scheme is nearly same as compared to that of proposed system. This reason behind this is that proposed system
performs resource allocation in dynamic approach which offers faster decision making for switching the channel
based on the current demand of the user (for both home user, hybrid users, and random user). Apart from this,
the proposed system assumes the block of resources from standard 5G concept where same channel capacity is
used for operation of home access point which are randomly deployed. Similar frequencies are also used by the
home access point users. This will mean that proposed system is capable of offering better data delivery services
over 5G standards with an effective power usage and without using any form of iterative and sophisticated
method. This causes the proposed system to be in cost effective in its operation towards resource management
in CRN deployed over IoT.

6523
4. CONCLUSION
A closer look into the existing literature towards 5G implementation in IoT shows higher inclination
towards achieving increases rate of data transmission and lower inclination towards computational cost involved
in controlling resource consumption. Hence, the proposed study addresses this problem by presenting a
simplified model where CRN is used over IoT for optimizing resources. The contribution of the proposed model
are as follows: i) the proposed scheme deploys CRN over a use case of smart city management in IoT targeting
towards achieving faster rate of data transmission along with resource conservation, ii) the proposed scheme
facilitates equal emphasis towards random user as well as home user by identifying the idle resources available
in network resulting in throughput improvement, and iii) the proposed system has a better control of the users
connected with access point of local IoT by mitigating interference to offer higher quality signal transmission.
REFERENCES
[1] D. Wu, X. Huang, X. Xie, X. Nie, L. Bao, and Z. Qin, “LEDGE: Leveraging edge computing for resilient access management of
mobile IoT,” IEEE Transactions on Mobile Computing, vol. 20, no. 3, pp. 1110–1125, Mar. 2021,
doi: 10.1109/TMC.2019.2954872.
[2] T.-D. Nguyen, E.-N. Huh, and M. Jo, “Decentralized and revised content-centric networking-based service deployment and
discovery platform in mobile edge computing for IoT devices,” IEEE Internet of Things Journal, vol. 6, no. 3, pp. 4162–4175,
Jun. 2019, doi: 10.1109/JIOT.2018.2875489.
[3] X. Tao, K. Ota, M. Dong, H. Qi, and K. Li, “Performance guaranteed computation offloading for mobile-edge cloud computing,”
IEEE Wireless Communications Letters, vol. 6, no. 6, pp. 774–777, Dec. 2017, doi: 10.1109/LWC.2017.2740927.
[4] M. Marjanovic, A. Antonic, and I. P. Zarko, “Edge computing architecture for mobile crowdsensing,” IEEE Access, vol. 6,
pp. 10662–10674, 2018, doi: 10.1109/ACCESS.2018.2799707.
[5] P. Bellavista, D. Belli, S. Chessa, and L. Foschini, “A social-driven edge computing architecture for mobile crowd sensing
management,” IEEE Communications Magazine, vol. 57, no. 4, pp. 68–73, Apr. 2019, doi: 10.1109/MCOM.2019.1800637.
[6] Y. Fu et al., “Energy-efficient offloading and resource allocation for mobile edge computing enabled mission-critical internet-of-
things systems,” EURASIP Journal on Wireless Communications and Networking, vol. 2021, no. 1, Dec. 2021, doi: 10.1186/s13638-
021-01905-7.
[7] P. Padmanabhan, K. S. Kumar, and A. John, “Effective resource virtualization for dynamic IoT devices in road network,” Indonesian
Journal of Electrical Engineering and Computer Science (IJEECS), vol. 30, no. 3, pp. 1777–1785, Jun. 2023,
doi: 10.11591/ijeecs.v30.i3.pp1777-1785.
[8] M. Elkholy and M. A. Marzok, “Light weight serverless computing at fog nodes for internet of things systems,” Indonesian Journal
of Electrical Engineering and Computer Science (IJEECS), vol. 26, no. 1, pp. 394–403, Apr. 2022, doi:
10.11591/ijeecs.v26.i1.pp394-403.
[9] F. Anwar, M. H. Masud, B. U. Islam, R. F. Olanrewaju, and S. A. Latif, “Game theory for resource allocation in heterogeneous
wireless networks - a review,” Indonesian Journal of Electrical Engineering and Computer Science (IJEECS), vol. 12, no. 2,
pp. 843–851, Nov. 2018, doi: 10.11591/ijeecs.v12.i2.pp843-851.
[10] M. A. Kalam Azad, A. Majumder, J. Krishna Das, and M. I. Islam, “Improving signal detection accuracy at FC of a CRN using
machine learning and fuzzy rules,” Indonesian Journal of Electrical Engineering and Computer Science (IJEECS), vol. 21, no. 2,
pp. 1140–1150, Feb. 2021, doi: 10.11591/ijeecs.v21.i2.pp1140-1150.
[11] E. M. Migabo, K. D. Djouani, and A. M. Kurien, “The narrowband internet of things (NB-IoT) resources management performance
state of art, challenges, and opportunities,” IEEE Access, vol. 8, pp. 97658–97675, 2020, doi: 10.1109/ACCESS.2020.2995938.
[12] W. Ejaz and M. Ibnkahla, “Multiband spectrum sensing and resource allocation for IoT in cognitive 5G networks,” IEEE Internet
of Things Journal, vol. 5, no. 1, pp. 150–163, Feb. 2018, doi: 10.1109/JIOT.2017.2775959.
[13] H. Malik, H. Pervaiz, M. M. Alam, Y. Le Moullec, A. Kuusik, and M. Ali Imran, “Radio resource management scheme in NB-IoT
systems,” IEEE Access, vol. 6, pp. 15051–15064, 2018, doi: 10.1109/ACCESS.2018.2812299.
[14] S. Khan et al., “5G Vehicular network resource management for improving radio access through machine learning,” IEEE Access,
vol. 8, pp. 6792–6800, 2020, doi: 10.1109/ACCESS.2020.2964697.
[15] B. Alzahrani and W. Ejaz, “Resource management for cognitive IoT systems with RF energy harvesting in smart cities,” IEEE
Access, vol. 6, pp. 62717–62727, 2018, doi: 10.1109/ACCESS.2018.2874134.
[16] S. Yu, X. Chen, Z. Zhou, X. Gong, and D. Wu, “When deep reinforcement learning meets federated learning: Intelligent
multitimescale resource management for multiaccess edge computing in 5G ultradense network,” IEEE Internet of Things Journal,
vol. 8, no. 4, pp. 2238–2251, Feb. 2021, doi: 10.1109/JIOT.2020.3026589.
[17] N. N. Khumalo, O. O. Oyerinde, and L. Mfupe, “Reinforcement learning-based resource management model for Fog radio access
network architectures in 5G,” IEEE Access, vol. 9, pp. 12706–12716, 2021, doi: 10.1109/ACCESS.2021.3051695.
[18] A. Nassar and Y. Yilmaz, “Reinforcement learning for adaptive resource allocation in fog RAN for IoT with heterogeneous latency
requirements,” IEEE Access, vol. 7, pp. 128014–128025, 2019, doi: 10.1109/ACCESS.2019.2939735.
[19] Y. Lu, L. Liu, J. Panneerselvam, B. Yuan, J. Gu, and N. Antonopoulos, “A GRU-based prediction framework for intelligent resource
management at cloud data centres in the age of 5G,” IEEE Transactions on Cognitive Communications and Networking, vol. 6,
no. 2, pp. 486–498, Jun. 2020, doi: 10.1109/TCCN.2019.2954388.
[20] S. Li, Q. Ni, Y. Sun, G. Min, and S. Al-Rubaye, “Energy-efficient resource allocation for industrial cyber-physical IoT systems in
5G era,” IEEE Transactions on Industrial Informatics, vol. 14, no. 6, pp. 2618–2628, Jun. 2018, doi: 10.1109/TII.2018.2799177.
[21] D. Wu, Z. Zhang, S. Wu, J. Yang, and R. Wang, “Biologically inspired resource allocation for network slices in 5G-enabled Internet
of Things,” IEEE Internet of Things Journal, vol. 6, no. 6, pp. 9266–9279, Dec. 2019, doi: 10.1109/JIOT.2018.2888543.
[22] D. D. Lieira, M. S. Quessada, A. L. Cristiani, and R. I. Meneguette, “Algorithm for 5G resource management optimization in edge
computing,” IEEE Latin America Transactions, vol. 19, no. 10, pp. 1772–1780, Oct. 2021, doi: 10.1109/TLA.2021.9477278.
[23] I. Sarrigiannis, K. Ramantas, E. Kartsakli, P.-V. Mekikis, A. Antonopoulos, and C. Verikoukis, “Online VNF lifecycle management
in an MEC-enabled 5G IoT architecture,” IEEE Internet of Things Journal, vol. 7, no. 5, pp. 4183–4194, May 2020,
doi: 10.1109/JIOT.2019.2944695.
[24] X. Liu and X. Zhang, “Rate and energy efficiency improvements for 5G-based IoT with simultaneous transfer,” IEEE Internet of
Things Journal, vol. 6, no. 4, pp. 5971–5980, Aug. 2019, doi: 10.1109/JIOT.2018.2863267.

 ISSN: 2088-8708
6524
[25] R. Abozariba, M. K. Naeem, M. Patwary, M. Seyedebrahimi, P. Bull, and A. Aneiba, “NOMA-based resource allocation and
mobility enhancement framework for IoT in next generation cellular networks,” IEEE Access, vol. 7, pp. 29158–29172, 2019,
doi: 10.1109/ACCESS.2019.2896008.
[26] S. Song, C. Lee, H. Cho, G. Lim, and J.-M. Chung, “Clustered virtualized network functions resource allocation based on context-
aware grouping in 5G edge networks,” IEEE Transactions on Mobile Computing, vol. 19, no. 5, pp. 1072–1083, May 2020,
doi: 10.1109/TMC.2019.2907593.
[27] Y. Liu, Y. Yang, P. Han, Z. Shao, and C. Li, “Virtual network embedding in fiber-wireless access networks for resource-efficient
IoT service provisioning,” IEEE Access, vol. 7, pp. 65506–65517, 2019, doi: 10.1109/ACCESS.2019.2915333.
[28] L. Miuccio, D. Panno, and S. Riolo, “Joint control of random access and dynamic uplink resource dimensioning for massive MTC
in 5G NR based on SCMA,” IEEE Internet of Things Journal, vol. 7, no. 6, pp. 5042–5063, Jun. 2020,
doi: 10.1109/JIOT.2020.2974402.
[29] N. Kherraf, H. A. Alameddine, S. Sharafeddine, C. M. Assi, and A. Ghrayeb, “Optimized provisioning of edge computing resources
with heterogeneous workload in IoT networks,” IEEE Transactions on Network and Service Management, vol. 16, no. 2,
pp. 459–474, Jun. 2019, doi: 10.1109/TNSM.2019.2894955.
[30] E. E. Ugwuanyi, S. Ghosh, M. Iqbal, and T. Dagiuklas, “Reliable resource provisioning using bankers’ deadlock avoidance
algorithm in MEC for industrial IoT,” IEEE Access, vol. 6, pp. 43327–43335, 2018, doi: 10.1109/ACCESS.2018.2857726.
[31] D. Tarek, A. Benslimane, M. Darwish, and A. M. Kotb, “Survey on spectrum sharing/allocation for cognitive radio networks internet
of things,” Egyptian Informatics Journal, vol. 21, no. 4, pp. 231–239, Dec. 2020, doi: 10.1016/j.eij.2020.02.003.
[32] J. Mansukhani and P. Ray, “Censored spectrum sharing strategy for MIMO systems in cognitive radio networks,” IEEE
Transactions on Wireless Communications, vol. 18, no. 12, pp. 5500–5510, Dec. 2019, doi: 10.1109/TWC.2019.2936839.
[33] F. Tang, L. Chen, X. Li, L. T. Yang, and L. Fu, “Intelligent spectrum assignment based on dynamical cooperation for 5G-satellite
integrated networks,” IEEE Transactions on Cognitive Communications and Networking, vol. 6, no. 2, pp. 523–533, Jun. 2020,
doi: 10.1109/TCCN.2020.2988480.
BIOGRAPHIES OF AUTHORS
Vani Bengaluru Purushothama completed B.E. in Electronics and Communications
in 2005 at Alpha College of Engineering VTU and M.Tech in Digital Electronics (2011) at
SSIT, Tumkur, VTU and Completed Her Ph.D. in Wireless Communication and Computer
networks under VTU in 2023. Overall Teaching experience is 15 years. Area of research includes
wireless communication, networking and IoT. She has presented papers in national/international
journal and conferences and Guided UG projects. She is a member of different professional bodies
like IE, ISTE and IEEE. She can be contacted at email: vaniganesh2003@gmail.com.
Sundaraguru Ramakrishnan received AMIE in Electronics and Communications
in 1993, M.E in Communication Systems in 1997 at Thiagarajar College of Engineering
Madurai and Ph.D. in wireless communication from Anna University Chennai. He is currently
working as professor and Head in the department of ECE, Sir MVIT. He has around 3 years of
industrial experience as a service/assistant engineer and around 24 years in teaching. His area
of research includes wireless communication, image processing, networking, VLSI and
embedded systems. To his credit he has published a paper in reputed international journals and
IEEE Conferences, about 7 papers in International Journal and 15 paper in National and
International conferences. He has coordinated number of Faculty development programmed.
Guided more than 75 UG and 30 PG projects. He is a member of different professional bodies
like Institute of Engineers, Member of IETE and ISTE. He can be contacted at email:
sugursg@gmail.com.

A simplified optimization for resource management in cognitive radio network-based internet-of-things over 5G networks

More Related Content

Similar to A simplified optimization for resource management in cognitive radio network-based internet-of-things over 5G networks (20)

More from IJECEIAES (20)

Recently uploaded (20)

A simplified optimization for resource management in cognitive radio network-based internet-of-things over 5G networks