2024 15th International Conference on Information and Knowledge Technology (IKT)

A New Routing Protocol in Internet of Vehicles

Inspired of Spread Model of the Covid-19 Virus
Esmaeil Zeinali Reza Ebrahimi Atani
Taha Yasin Rezapour
Department of Computer and Department of Computer Engineering
Department of Computer and
Information Technology Engineering University of Guilan
Information Technology Engineering
Qazvin Branch, Islamic Azad Rasht, Iran
Qazvin Branch, Islamic Azad
University rebrahimi@guilan.ac.ir
University
Qazvin, Iran
Qazvin, Iran
zeinali@qiau.ac.ir
trezapour@pmo.ir
Mohammad Mehdi Gilanian Sadeghi
Department of Computer and
Information Technology Engineering
Qazvin Branch, Islamic Azad
University
Qazvin, Iran
msadeghi@ieee.org

Abstract— In recent years, the Internet of Vehicles(IoV) has interaction is needed among the heterogeneous platforms and
emerged as a critical subset of the Internet of Things(IoT), services in this domain to support common components and
playing a key role in enhancing the safety and efficiency of ensure coordinated operations [1-5].
intelligent transportation systems. Unlike other IoT applications
that typically operate in static and controlled environments, IoV With the surge in Internet of Things (IoT) connected
operates in complex, dynamic, and highly variable vehicles, the core concept of VANETs has evolved into a new
environments, necessitating unique solutions for routing paradigm known as the Internet of Vehicles (IoV)[6]. IoV, as
management. This paper proposes a novel routing approach for a branch of the Internet of Things (IoT), creates a dynamic
IoV networks that, for the first time, leverages the coronavirus intelligent network connecting vehicles, users, and various
propagation model to improve Quality of Service (QoS) smart devices through internet connectivity. The integration
parameters, particularly end-to-end delay. Inspired by the of advanced information and communication technologies in
rapid propagation characteristics of the coronavirus, this IoV has proven beneficial in addressing various traffic and
method establishes a fast information dissemination mechanism driving challenges, resulting in enhanced driving convenience
within the network, leading to an exponential decrease in data and passenger safety. The objectives of IoV include
transfer delay. Through this model, routing information is improving road safety, route planning, entertainment,
rapidly disseminated among vehicles with minimal hops, navigation, smart traffic management and etc. Vehicle-to-
resulting in significant improvements in IoV.
vehicle (V2V) and vehicle-to-infrastructure (V2I)
Keywords—IoV, IoT, Routing, COVID-19, ITS. communications form the backbone of IoV, enabling vehicle-
to-vehicle communication, intra-vehicle communication, and
I. INTRODUCTION mobile vehicular internet [7-10].
Over recent years, Intelligent Transportation Systems Although the number of IoV-connected vehicles continues
(ITS) applications have gained significant momentum in to grow, there are various challenges and opportunities for IoV
many aspects. Advances in electronic technology and the technology that remain unclear. The report in [11] suggests
integration of telecommunications, ad-hoc networks, that the global number of (passenger and commercial)
computing, and communication systems have led to dramatic vehicles is anticipated to reach 2 billion by 2035; in other
improvements in real-time communication and decision- words, the scale of intelligent transportation networks will
making processes. The growing prevalence of physical increase significantly. Therefore, ensuring quality of service
devices connected to the internet, in conjunction with the in these networks is essential to prevent accidents and
advancement of numerous essential technologies including unfortunate events in urban and suburban (road)
wireless sensor networks, RFID networks, Vehicular Ad Hoc environments, as well as in large industrial environments with
Networks (VANETs), cloud computing, and fog computing, extensive traffic areas such as large shipping ports where the
together with the development of a new class of cost-effective safety of operators and workers is of paramount importance.
and compact wireless devices utilizing diverse Additionally, in non-safety applications, it is crucial to ensure
communication protocols, has led to the emergence of the IoT that the required services are provided to users effectively.
paradigm. IoT is the network of physical objects that Consequently, delivering data and routing it from source to
encompass embedded electronics, software, sensors, and destination with the least delay and the highest packet delivery
connectivity, and the ability to sense and record data, process ratio is a critical issue.
information obtained, and react based on the data. In order to Because of the distributed nature of nodes and their radio
provide high-quality services to users, a comprehensive constraints, multi-hop communication becomes part of IoV

979-8-3315-2225-4/24/$31.00 ©2024 IEEE

for enabling data exchange between individual nodes. An then, the network model and the proposed method for data
efficient and reliable multi-hop communication significantly routing with a focus on selecting the next node inspired by the
depends on the final routing strategy between a source- virus transmission model will be presented in section IV, and
destination node pair, which usually requires low latency and finally, the proposed protocol will be evaluated in terms of
high delivery rate. However, the inherent features of IoV, like quality of service parameters and section V focuses on the
frequent topology changes and a high mobility of nodes, pose performance analysis of the proposed scheme. Finally,
many challenges in the design of such protocols. On the other Section VI discusses the conclusions drawn from this
hand, the design of routing protocols for complex urban IoV research.
environments is more difficult due to intersections, buildings,
and other barriers that result in a higher signal attenuation and II. RELATED WORKS
packet loss [12]. Therefore, optimally selecting the next hop
node from among the available and selectable neighboring In this section, we delve into several existing approaches
nodes is a crucial step in providing a routing approach. designed to meet the quality of service requirements such as
delay and packet delivery ratio in IoV.
Nature-inspired optimization algorithms have
demonstrated a remarkable ability to solve complex problems. The AODV protocol, introduced in RFC 3561[14], is a
These algorithms are generally iterative in nature, meaning dynamic and reactive routing protocol designed for mobile ad
that they follow a repeated cycle to obtain the best hoc networks (MANETs). In these networks, nodes move
solution[13]. Therefore, in this work, it is decided to use an around and communicate wirelessly. AODV's primary goal is
innovative approach inspired by a natural phenomenon, to establish and maintain optimal communication paths
namely the coronavirus propagation model, for the first time between nodes in ad hoc networks, minimizing resource
to present a new approach for data routing in IoV networks, consumption and latency. AODV employs an "on-demand"
which are networks with mobile nodes. The main objective of routing approach, meaning routes are created dynamically
this work is to present a new routing approach in the Internet only when needed. This reduces control traffic and conserves
of Vehicles, in a way that improves the end-to-end delay network bandwidth. AODV uses sequence numbers to ensure
parameter and provides a quality service so that this approach route freshness and prevent routing loops. This protocol can
can be used in the management and operation of intelligent effectively adapt to network topology changes, such as node
transportation in urban and industrial environments with mobility, and quickly adjust to new conditions. AODV uses
heavy traffic such as busy ports and improve the quality of several types of control messages for routing: route request,
service in the network; The network structure in this work, as route reply, route error, and route reply acknowledgment.
shown in Figure 1, consists of an urban/industrial environment AODV offers a dynamic and efficient routing solution for
of vehicles or the same mobile nodes, roadside infrastructure mobile ad hoc networks. The use of sequence numbers and
(RSUs), base station (BS), intersections, and road sections specific control messages ensures route freshness and
between intersections, and vehicle-to-vehicle (V2V) prevents routing loops. These features make AODV a suitable
communications as well as vehicle-to-infrastructure (V2I) choice for wireless ad hoc networks with constantly changing
communications form wireless communication between these topologies.
components, all of which are possible between network nodes Chen et al. [9], propose a Traffic-aware and Link
in a two-way manner. Preference Routing Protocol (TLRP) for urban IoV,
considering how to avoid sending packets to sparsely
populated road segments and avoid significant data
congestion in vehicular networks. The authors introduce a
new routing metric, which they call Link Transmission
Quality (LTQ), and which chooses a route based on the
relative positions of links in the routing process, while
integrating the transmission costs and packet delivery rates
into the routing decisions. Therefore, in order to adapt to the
dynamic nature of IoV, a road weight evaluation scheme is
proposed for evaluating every road segment based on real-
time traffic and link quality information provided by LTQ and
selects the route with the lowest cumulative weight as the
routing candidate. The study further proposes two methods of
data transmission based on the locations of source and
destination nodes, which are particularly suitable for packet
forwarding in large modern metropolitan areas.
Wang et al. [15], introduce in their work the concepts of a
vehicle mobility probability matrix, a vehicle location
Fig. 1. IoV network model in an urban/industrial environment with V2V dependency matrix, and latent factors to analyze the impact of
and V2I communications.
future vehicle location attributes. They combine these
The structure of this paper is as follows: In section II, the attributes with potential vehicle dependency attributes and
existing IoV based routing protocols will be reviewed. Then, propose a routing algorithm based on vehicle location
evaluation, namely RAVP. This algorithm can precisely
considering the goal of the paper, a brief review of the
predict the future trajectory of a vehicle and then compute the
coronavirus propagation model and the model of virus
distance of the vehicle on the basis of a predicted trajectory.
transmission from an infected person to a susceptible person
The results obtained in this study indicate that the vehicle
through a super-spread event will be presented in section III;
trajectory is subject to a Gaussian distribution. node and
location information are used to construct a probability matrix individual who has been in contact with an infected person, as
of the nodes' movement. The effectiveness of the algorithm is official statistics of infected people show. Therefore, the
ensured by addressing the objective function with posterior coronavirus will be transmitted to the adjacent person in the
estimation, gradient descent, and regularization, and using a shortest time and with the highest probability. This is the main
predefined numerical threshold to adjust the prediction level; motivation for addressing this issue in this research. In short,
the route delivery and maintenance mechanisms are also if you consider individuals as moving nodes in a human
realized in their work. network, you will see a very close similarity to IoV networks
with moving vehicles that communicate with each other when
Kabbaj et al.[16] suggest a dynamic source-destination they are within range. Also, data can be considered analogous
path-aware opportunistic data dissemination scheme for an to the virus itself, which is transferred from a node to its
industrial environment. They use a single-path solution for neighboring nodes. Candidate nodes to receive data from the
establishing opportunistic connections for the transmission of current node are also similar to healthy individuals who,
single-hop data. This paper proposes a data forwarding depending on the distance from the infected person (the
scheme for multi-sources and -destinations in an industrial current node holding the data packets), can be exposed to
environment with intermediate nodes buffering data packets infection and themselves become the next transmitter.
temporarily in order to enable their successive transmission to
Consequently, by using a mathematical model of coronavirus
the next node with repeated retransmissions. This paper transmission from an infected person to a target person,
modifies the AODV routing protocol architecture. Type of considering the high speed and high probability of
hop count, message, reserved, destination and source address transmission of this virus, an optimal model can be used to
fields are carried in messages. In addition to the above fields, select the next node for data transmission in IoV. Given that
additional fields have been added including Id, road list, in the virus transmission model, different individuals can exist
record list, the distance traveled by the message since its in a specific environment, for example, multiple paths can be
creation, the timestamp of when the message was generated drawn from the source to the destination, and the path with the
by the source, and the timestamp of when the message was least time required for transmission can be selected as the
received by the destination. This study has carried out the optimal path.
communication time and communication energy modeling
and proposed a framework of data forwarding, including To analyze the spread of COVID-19, quite a number of
initialization, buffer management, next-hop selection, mathematical models have been applied. The
shortest-path-based protocol, and data fusion. This work SIR(Susceptible-Infected-Removed) model is one of the
considers parameters such as packet delivery ratio, end-to-end simple epidemiological models showing the dynamics of an
delay, hop count, and network congestion in the evaluation. infectious epidemic, given a large population that is initially
susceptible, infected, and recovered. Assuming the virus
Wang et al.[17] propose a protocol based on Nakagami-m usually infects only once, an infected person either dies or
outage probability, named VCNOP. In this work, routing with recovers. As the disease propagates, individuals who are
RSU-assisted path selection is applied, and outage probability susceptible are increasingly likely to contract the infection,
is considered to measure the impact of physical problems on whereas those who are infected generally transition to the
vehicular communications. Outage probability is defined as an removed category (through either death or recovery). It is
event when the link capacity is unable to satisfy the user's posited that the overall population remains a constant figure
demanded rate. In this paper, dynamic guidance message N, and ultimately, the entire infected population transitions to
mechanism design is carried out with the consideration of the removed class. This model separates the homogeneous and
vehicle speed and density. The mechanism can predict vehicle specified population into three classes: (1) Susceptible (S):
collisions, so the routing decisions can be more accurate. In These are individuals who are not infected with the virus but
this study, RSU-assisted path selection mechanism is adopted might become infected due to transmission from an infected
for multi-hop communications between source and individual; (2) Infected (I): These individuals have been
destination vehicles. This mechanism can be employed in previously infected with the disease; and (3) Removed (R):
intelligent transportation systems to coordinate with Individuals infected with this virus either recover or die,
infrastructures such as RSUs to assist road-to-infrastructure
which are removed in this model. The population of these
communications. VCNOP takes into consideration outage classes at time t is denoted by S(t), I(t), and R(t), respectively,
probability, vehicle position, relative speed, and vehicle and for all t, it is assumed that S(t) + I(t) + R(t) = N; This
direction. model is described by the following equations, where β is the
contact rate and μ is the removal rate[19]:
III. COVID-19 EPIDEMIC MODEL

=−
In 2019, global media reported the outbreak of a new virus
⎧
⎪
in Wuhan, China, which rapidly spread throughout the city

= −
and caused a high mortality rate among those infected. Within
⎨
⎪
a short time and at an alarming rate, this virus spread to (1)

⎩ =
countries worldwide, resulting in a global pandemic. Every
day, we witnessed new cases and deaths from this virus in
every corner of the globe. As of September 21, 2023, the Research [16] examines the probability of coronavirus
WHO reported 770,778,396 confirmed cases of COVID-19 transmission through coughing by an ordinary individual and
globally, including 6,958,499 deaths[18]. a super-spreader. In summary, this study models the spread
rate of a cough cloud with distance using an exponential decay
A key point that has drawn the attention of health experts equation and shows that the cough cloud changes over time;
and even ordinary people in this pandemic is the high speed changes in concentration within the cough cloud are also
of transmission of this virus from an infected person to considered as a Gaussian distribution, and the concentration
another, as well as the high probability of infection in a healthy
width is approximately 1.2 times the speed width; at each • S(t) = Number of nodes that have not received data
distance, the amount of pollutant flow is calculated by (susceptible) at time t.
combining the local values of speed and concentration in the
cloud cross-section; only a fraction of the pollutants expelled We can also make the following assumptions:
by a person is inhaled by another. This amount, "quantum of • Nodes can be in one of two states: infected(I) or
infection", is also calculated considering a hemispherical susceptible(S).
breathing zone. The probability of infection is calculated using
the Wells-Riley equation, which relates the probability to the • A susceptible node becomes infected through contact
inhaled quanta through the Poisson distribution; parameter with an infected node.
values such as cough volume, inhalation volume, and duration
• The rate of contact/transmission depends on factors
are used for calculations. In this model, the probability of virus
such as the density of vehicles on the roads.
transmission through coughing from an infected person to a
susceptible person is formulated. First, the amount of viral • We define the transmission rate as β.
particles entering a person's body (denoted by µ) is calculated
Then, the basic dynamics of the model will be:
Rate of new infections per time step = βI(t)S(t) (3)
as µ=CR/CS; Where CR is the amount of inhaled contamination
by the neighbor and CS is the maximum amount of particles at
the source (the contaminating person). Then, the Wells-Riley
equation is used to calculate the probability of transmission as which represents the probability of transmission at each
follows: contact; S(t) decreases while I(t) increases with the occurrence

= 1−
of transmission and N(t) = S(t) + I(t) remains constant. We can
(2) write the differential equations as follows:

= (*) (*)
Where is P the probability of transmission, e is the

) (4)
exponential number, and µ is the amount of particles entering
=− (*) (*)
the person's body.

IV. THE PROPOSED MODEL Solving these equations over time gives us I(t) and S(t),
allowing us to model and predict data dissemination through
In general, models used to calculate the spread of a virus
the IoV network based on the transmission rate β.

t=β/data size ; To select the next node, inspired by the model

or information in a population typically consider these factors:
Consequently, we can formulate the data transmission time as
(1) the number of individuals infected with the virus at any
given time (such as the number of nodes in an IoV network
presented in research [20] for calculating the probability of
that possess data); (2) the probability of transmission from one
coronavirus transmission based on coughing to an individual
individual to another, which depends on factors such as the
within the range of a cough cloud, the probability of data
level of contact (such as nodes in an IoV network that are one
transmission from the source node to the next node within the
hop away from the source node and are candidates for
communication range of the current node and at a specific
receiving data); (3) the transmission time from one individual
distance from it is calculated. The node with the highest
to another (such as the single-hop delay in IoV networks); and
probability is selected as the next node, and this process is
(4) the number of individuals who have not yet received the
repeated for subsequent nodes. Additionally, after finding the
virus (such as nodes in IoV networks that have not yet
next nodes in the grid where the source node is located, in the
received the data).
next step, based on the grid in which the destination node is
If we divide the network into a grid-based model, located, the next grid and subsequent nodes are selected in
considering each region as a grid, vehicles can be categorized order to be delivered to the destination. The probability of data
into several states. The source vehicle is considered as the transmission from the current node to the next node is
vehicle that has data to send. Based on the objectives of this calculated as follows:

,,.=1− /
research, in the first step, we need to find the next node, which
is the next vehicle to which the data packet should be (5)
transferred. For this purpose, vehicles that are within the which is the amount of data received by node j from node
communication range of the current vehicle can be considered i, and depends on factors such as the distance between the two
as candidate nodes for the next node. So far, if we want to take vehicles and their relative speed. It is worth noting that the
inspiration from the coronavirus transmission model, we can procedure should be defined as follows: to select the next
model the current vehicle as an infected individual (I) and the node, control packets are first sent within the communication
next vehicle as a susceptible individual (S). With this range of the source node. Nodes within the communication
modeling, we can model the current node and the next node at range that receive these packets calculate this probability and
this stage. inform the source node. The source node then selects the
option with the highest calculated probability as the choice for
Here, a basic model can be presented that can be used to routing packets. This process is repeated within the grid until
calculate data transmission between nodes in IoV based on an intra-grid path is formed. Therefore, the steps of the process
epidemiological models of disease transmission. The can be described as follows:
necessary definitions can be presented as follows:
• First, the probability of virus propagation from the
• N(t) = Total number of nodes/vehicles in the network source node to each neighboring node is calculated.
at time t. This probability is calculated based on factors such as
• I(t) = Number of nodes that have received data distance, relative speed of vehicles, and the amount of
(infected) at time t. data sent.
 • For each subsequent node, the node with the highest 2. Store all possible paths, including nodes and RSUs,
probability of propagation is selected. as a graph.
• This process is repeated until the destination node is 3. For each path, calculate the sum of the transition
reached, thus finding the complete path. probabilities between nodes as the score of that path.
• To minimize delay, nodes with shorter transmission
times are prioritized. 4. Select the path with the highest score as the best
path.
• Environmental information is collected by sensors
located in the vehicles and used to optimize the route. The proposed model can be seen in Figure 2.
The algorithm for selecting the next node can be presented
in the following steps: V. PERFORMANCE EVALUATION OF THE PROPOSED
MODEL
1. Designate the source node as node i.

Create a list of all neighbors of i as 012 , 13 , … , 15 6.

In IoV networks, the goal is for vehicles to quickly access
and utilize route-related information. Assuming that infected
2. vehicles (I(t)) can infect other vehicles (S(t)) at a rate of β,
meaning they can provide route information to them, the
3. For each neighboring node, calculate P(i→j), which equation governing the increase in the number of infected
represents the probability of virus (data) vehicles at time t is as follows:

= (*) (*)
( )
transmission from node i to node j.
(6)
4. Select the node with the highest P(i→j) as the next In which β is equivalent to the rate of spreading route
node. information (similar to the rate of virus spread), S(t) is the
number of susceptible vehicles (nodes that are susceptible to
receiving data), and I(t) is the number of infected vehicles
(nodes that possess the data and can deliver it to the next
node); over time, the number of vehicles that have access to
route information increases, and consequently, the number of
susceptible vehicles decreases. In a simple model, end-to-end
delay in IoV networks typically depends on the number of
hops (or intermediate nodes). As the number of infected
vehicles increases, the number of hops required to reach the
destination decreases. Now, let's assume that the end-to-end
delay D is inversely proportional to the number of infected
vehicles I(t):
8(*) ∝
2
( )
(7)
Given that I(t) increases exponentially over time due to the
rate β:
(*) = (0) (8)
Therefore, the delay D(t) decreases as follows:
8(*) ∝ = −
2 2
(;)< = (;)
(9)
This equation demonstrates that the end-to-end delay
decreases exponentially over time, resulting from the rapid
dissemination of route information throughout the network
through the application of the coronavirus model. Based on
this analysis, it can be concluded that employing the
coronavirus propagation model for designing routing
protocols in IoV networks leads to a significant reduction in
end-to-end delay. This is due to the rapid dissemination of
route information to neighboring vehicles, which can
Fig. 2. The Proposed Routing protocol based on Covid19 Model.
exponentially reduce delay and increase packet delivery rates.
This approach can also yield significant improvements in
The node selection algorithm should be updated terms of security, as route information is quickly disseminated
periodically to account for environmental changes. The next throughout the network, leaving insufficient time for potential
step is to find the optimal path among all existing grids. attacks.
Therefore, we have the following algorithm for selecting the To compare the end-to-end delay of the proposed method
best path: based on the coronavirus propagation model with AODV
routing, let's assume TAODV is the end-to-end delay in the
1. Determine the source and destination nodes. AODV protocol, and Proposed is the end-to-end delay in the
proposed method. If, for an urban/industrial environment, n
represents the number of intermediate nodes (vehicles) destination, the packet delivery ratio can be modeled as
between the source and destination, d is the average time taken follows:
8\cdef = ∏5,T2 ,
by each node to send a message in AODV, and β is the
information propagation rate in the proposed method, then in (14)
AODV, the end-to-end delay can be calculated as follows: In the proposed method, the number of vehicles that have

T?@AB ≈ D × F (due to N ). In contrast, in the AODV protocol, PDR

received route information increases exponentially with time
(10)
In the proposed method, considering the virus propagation depends on the success rate of individual links, which may
model: decrease with the number of intermediate nodes and link
TGHIJIKLM ∝
2 N
quality. In AODV, as the number of nodes increases, the
(;)
(11) probability of PDR decreasing also increases because the
Now, for comparison, in AODV, if the network is very likelihood of packet loss at each node becomes higher.
congested (considering the nature of an urban network where However, in the proposed method, as time increases, due to
traffic and consequently the number of vehicles and nodes are the rapid dissemination of route information, the number of
high), the delay increases linearly with the number of nodes; vehicles with route information increases and PDR quickly
whereas, in the proposed method, the delay decreases approaches 1.
exponentially and reaches a minimum over time as The proposed method, due to the rapid and exponential
information is rapidly disseminated among nodes. dissemination of route information, has a very high probability
Next, we compare our proposed method with the TLRP of maximizing the packet delivery ratio, especially in
protocol. As previously detailed, this protocol is designed for conditions where the network is congested and dynamic.
routing in urban Internet of Vehicles (IoV) and utilizes two Based on the mathematical models, it can be concluded that
primary metrics for path selection: (1) Link Transmission the proposed method outperforms AODV in terms of packet
Quality (LTQ), which encompasses packet transmission cost delivery ratio in dynamic and congested network conditions,
and packet delivery ratio, and (2) road weight evaluation, especially when the number of nodes or topology changes
calculated for each segment of road by link information and increase.
real-time traffic. Then, this protocol selects paths with the However, the TLRP protocol uses traffic information and
lowest total weight as candidate paths. The mathematical link quality for routing. In this method, PDR is influenced by
model for delay in this method can be expressed as follows:
OPQ R = ∑5,T2 F, + OQPV
two main factors: the first is Link Transmission Quality

∑5,T2 F,
(12) (LTQ), which determines the percentage of packets that are

from source to destination, and OQPV represents the additional

where represents the sum of delays along the path successfully transmitted from one node to another; and the
second is the traffic condition, which includes delays in routes
delay caused by link quality evaluation and decision-making and the capacity of roads for transmitting packets. If we
based on it. For comparison, assume in an urban/industrial assume Pi is the probability of successful transmission of a
network, the number of nodes and traffic conditions are the packet from node i to node i+1 and n is the number of
same for both methods. In the TLRP method, the delay intermediate nodes between the source and destination, the
packet delivery ratio can be modeled as follows:
( OPQ R = D. F + OQPV ), while in the proposed method, the 8\PQ R = ∏5,T2 ,
increases linearly with the number of intermediate nodes

N
(15)

decreases exponentially, with time, TGHIJIKLM decreases

delay decreases exponentially with time. Since This equation shows that the overall PDR is the product of
the success probabilities for each node in the path. The higher
rapidly. In contrast, in the TLRP method, the delay increases the link quality, the closer Pi is to 1, and the higher the PDR
linearly with the number of nodes (n). Therefore: will be. Over time, due to the exponential increase of I(t) in
TGHIJIKLM < OPQ
the proposed method, PDR quickly approaches 1 (meaning all
R vehicles receive route information). In contrast, in the TLRP
If: method, PDR depends on the product of the success
* > lnZD. F + OQPV [
2 probabilities of individual links, which may decrease over
N time due to network changes or decreased link quality. In
This inequality shows that for appropriate values of TLRP, PDR is directly dependent on link quality and may
propagation rate β and time t, the proposed method has lower decrease with an increase in the number of nodes due to the
delay compared to the TLRP method. multiplication of probabilities of intermediate links. However,
in the proposed method, PDR only depends on the information
The proposed method is based on the coronavirus dissemination rate (β) and the total number of vehicles (N),
propagation model, aiming to achieve faster access to optimal and an increase in the number of nodes will have a lesser
routes for vehicles and improve Packet Delivery Ratio (PDR) impact on it, as the rapid dissemination of information leads
through rapid dissemination of route information. The packet to a rapid increase in the number of vehicles with route
delivery ratio in the proposed method depends on the number information.
of vehicles that have received the optimal route at time t.
Therefore, the packet delivery ratio is defined as follows: Table 1 presents a comparison of the proposed scheme

8\R]^_^`< = =
( ) (;)< b=
with AODV and TLRP regarding some factors like delay,
a a
(13) packet delivery ratio (PDR), computational complexity,
This equation shows that the PDR in the proposed method congestion management, and its suitability in complex
increases rapidly and exponentially over time. In the AODV environments. It can be observed that the delay of AODV
method, if we assume Pi is the probability of successful increases linearly with an increase in the number of nodes
transmission of a packet from node i to node i+1 and n is the while TLRP reduces the delay using higher quality links. In
number of intermediate nodes between the source and contrast, the proposed scheme has exponentially decreased
delay with an increase in the number of compromised nodes. REFERENCE
Regarding PDR, AODV is not good for dense networks, and [1] P. K. Dhillon, and S. Kalra, “A lightweight biometrics based remote
its PDR degrades as the number of nodes increases. The PDR user authentication scheme for IoT services,” Journal of Information
of TLRP depends on link quality and, thus, depends on Security and Applications, vol. 34, pp. 255-270, 2017/06/01/, 2017.
computed path weights; in the proposed method, PDR [2] M. Tao, J. Zuo, Z. Liu, A. Castiglione, and F. Palmieri, “Multi-layer
presents a high increasing slope with the growth of data and cloud architectural model and ontology-based security service
number of nodes. Among the three routing methods, AODV framework for IoT-based smart homes,” Future Generation Computer
Systems, vol. 78, pp. 1040-1051, 2018/01/01/, 2018.
has the least computational complexity, and embedded
[3] V. K. N. Jeba, “A Survey on Routing Protocols for Internet of Things,”
computers aboard vehicles can support the computational International Journal of Advanced Research in Science, Engineering
requirements imposed by the other two protocols. In terms of and Technology vol. 3, no. 5, 2016.
congestion management, AODV is only for simple and stable [4] S. Sicari, A. Rizzardi, D. Miorandi, C. Cappiello, and A. Coen-Porisini,
networks, while TLRP uses real-time traffic information to “A secure and quality-aware prototypical architecture for the Internet
reduce congestion. The proposed scheme also reduces of Things,” Information Systems, vol. 58, pp. 43-55, 2016/06/01/,
congestion using distributed data dissemination. As 2016.
mentioned, AODV is not suitable for dense urban/industrial [5] T. Y. Rezapour, R. E. Atani, and M. S. Abolghasemi, "Secure
positioning for shipping containers in ports and terminals using WSN."
environments; TLRP applies in high-traffic urban cities, and pp. 10-14.
the proposed scheme is highly suitable for dynamic and [6] J. Contreras-Castillo, S. Zeadally, and J. A. Guerrero-Ibañez, “Internet
complex urban/industrial networks. of Vehicles: Architecture, Protocols, and Security,” IEEE Internet of
Things Journal, vol. 5, no. 5, pp. 3701-3709, 2018.
TABLE I. PERFORMANCE COMPARISON RESULTS OF THE PROPOSED [7] F. Yang, S. Wang, J. Li, Z. Liu, and Q. Sun, “An overview of Internet
PROTOCOL WITH AODV AND TLRP PROTOCOLS. of Vehicles,” China Communications, vol. 11, no. 10, pp. 1-15, 2014.
[8] J. H. Park, Q. Yang, and S. J. Yoo, “RHRA-DRL: RSU-Assisted
Features AODV TLRP Proposed Protocol Hybrid Road-Aware Routing Using Distributed Reinforcement
Learning in Internet of Vehicles,” IEEE Access, vol. 12, pp. 25385-
Based on LTQ (Link 25396, 2024.
Based on rapid data
Routing Tables Transmission
Routing propagation inspired [9] M. Mazouzi, K. Mershad, O. Cheikhrouhou, and M. Hamdi, “Agent-
and Quality),
Metrics by the coronavirus Based Reactive Geographic Routing Protocol for Internet of Vehicles,”
RREQ/RREP Road Weight
spread model IEEE Access, vol. 11, pp. 79954-79973, 2023.
Messages Evaluation
[10] T. Sharma, M. R. Kumar, S. Kaushal, D. Chaudhary, and K. Saleem,
Linear delay Reduce delay Exponential delay “Privacy Aware Post Quantum Secure Ant Colony Optimization Ad
depending on by using reduction with Hoc On-Demand Distance Vector Routing in Intent Based Internet of
Delay
the number of better quality increasing number of Vehicles for 5G Smart Cities,” IEEE Access, vol. 11, pp. 110391-
nodes links infected nodes 110399, 2023.
Decreasing with Dependent on [11] J. Voelcker, 1.2 Billion Vehicles On World's Roads Now, 2 Billion By
Exponentially 2035: Report, 2014.
increasing link quality
PDR increasing with rapid
number of nodes and road [12] C. Chen, L. Liu, T. Qiu, J. Jiang, Q. Pei, and H. Song, “Routing With
data expansion
or weak links weight Traffic Awareness and Link Preference in Internet of Vehicles,” IEEE
Transactions on Intelligent Transportation Systems, vol. 23, no. 1, pp.
Medium 200-214, 2022.
Low (due to the (calculation Medium (calculation
Computational [13] I. U. Khan, I. M. Qureshi, M. A. Aziz, T. A. Cheema, and S. B. H.
use of simple of road of data transfer
Complexity
tables) weight and probability)
Shah, “Smart IoT Control-Based Nature Inspired Energy Efficient
link quality)
Routing Protocol for Flying Ad Hoc Network (FANET),” IEEE
Access, vol. 8, pp. 56371-56378, 2020.
Reduce [14] I. E. T. F. (IETF), "Ad hoc On-Demand Distance Vector (AODV)
Data congestion Reduce congestion Routing," RFC Editor, 2003.
No congestion
Congestion using real- with distributed data [15] L. L. Wang, J. S. Gui, X. H. Deng, F. Zeng, and Z. F. Kuang, “Routing
management
Management time traffic dissemination Algorithm Based on Vehicle Position Analysis for Internet of
information
Vehicles,” IEEE Internet of Things Journal, vol. 7, no. 12, pp. 11701-
11712, 2020.
Suitable for
Application in Limited to Ideal for dynamic [16] S. Kabbaj, A. U. Rahman, A. W. Malik, A. I. Baba, and S. D. Ravana,
high-traffic
Complex simple and and complex “Time-bound single-path opportunistic forwarding in disconnected
urban
Environments stable networks networks
environments industrial environments,” Vehicular Communications, vol. 27, pp.
100302, 2021/01/01/, 2021.
[17] T. Wang, J. Zhang, Y. Zhang, Y. Cao, L. Fu, A. Hussain, and G. Chen,
VI. CONCLISION “Design and performance evaluation of V2X communication protocol
based on Nakagami-m outage probability,” Journal of Ambient
The main objective of this work is to present a novel Intelligence and Humanized Computing, vol. 12, no. 10, pp. 9405-
solution for addressing routing challenges in the Internet of 9421, 2021/10/01, 2021.
Vehicles (IoV) in smart transportation and vehicular ad-hoc [18] "WHO Coronavirus (COVID-19) Dashboard,"
networks. This paper emphasizes the importance of https://covid19.who.int/.
employing IoV in optimizing smart transportation [19] N. Wang, Y. Fu, H. Zhang, and H. Shi, “An evaluation of mathematical
management in urban/industrial environments. After models for the outbreak of COVID-19,” Precision Clinical Medicine,
reviewing existing data routing methods, we propose a new vol. 3, no. 2, pp. 85-93, 2020.
model inspired by the natural phenomenon of coronavirus [20] A. Agrawal, and R. Bhardwaj, “Probability of COVID-19 infection by
propagation. By utilizing the mathematical model of virus cough of a normal person and a super-spreader,” Physics of Fluids, vol.
33, no. 3, 2021.
propagation and the transmission model via super-spreading,
we aim to develop a model that routes data with low end-to-
end delay and high Packet delivery ratio.