Multipath routing protocol based on backward routing with optimal fuzzy logic in medical telemetry systems for physiological data measurements


ACTA IMEKO 
ISSN: 2221-870X 
March 2022, Volume 11, Number 1, 1 -8 

 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 1 

Multipath routing protocol based on backward routing with 
optimal fuzzy logic in medical telemetry systems for 
physiological data measurements 

Suryadevara Kranthi1, Pallavi Deshpande2, Shadab Pasha Khan3, Amedapu Srinivas4 

1 Department of IT, V R Siddhartha Engineering College, Vijayawada-520007, Andhra Pradesh, India  
2 Department of Electronics and Telecommunication, Bharati Vidyapeeth (Deemed to be University) College of Engineering, Pune-411043, 
  Maharastra, India  
3 Department of IT, Oriental Institute of Science & Technology, Bhopal-462021, Madya Pradesh, India 
4 Department of CSE, Sreenidhi Institute of Science and Technology, Hyderabad-501301, Telangana, India 

 

 

Section: RESEARCH PAPER 

Keywords: Authentication; quality of service; mobile ad hoc network; multipath routing; optimal fuzzy logic 

Citation: Suryadevara Kranthi, Pallavi Deshpande, Shadab Pasha Khan, Amedapu Srinivas, Multipath routing protocol based on backward routing with 
optimal fuzzy logic in medical telemetry systems for physiological data measurements, Acta IMEKO, vol. 11, no. 1, article 23, March 2022, identifier: IMEKO-
ACTA-11 (2022)-01-23 

Section Editor: Md Zia Ur Rahman, Koneru Lakshmaiah Education Foundation, Guntur, India 

Received November 20, 2021; In final form February 22, 2022; Published March 2022 

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, 
distribution, and reproduction in any medium, provided the original author and source are credited. 

Corresponding author: Suryadevara Kranthi, e-mail: kranthisri41@gmail.com  

 

1. INTRODUCTION 

A mobile ad hoc network (MANET) has several sensor nodes 
[1] that are connected without central infrastructure via wireless 
media. To execute a fundamental routing activity like transferring 
data packets from one source to a location, routing techniques 
such as dynamic basis [2] and ad hoc remote vector [3] were 
devised. The MANET properties should be taken into account 
in routing methods. The fundamental requirements of a 

MANET are data dependability, multipath selection [4] and 
security [5] that improve network presentation. To attain this 
purpose many of the investigations have been developed. One 
of the important MANET functionalities is on-demand routing 
[6]. In an ad hoc setting, the routing arrangements [7] must be 
confident, resilient and flexible. Dynamic topology and node 
connection failure restricts the routing function. The mobility of 
the node makes it harder to route because it causes the nodes' 
connection failure. This frequent link failure leads to expensive 
routing and topology management, degrades reliability of data 

ABSTRACT 
Mobile ad hoc network (MANET) performance is critically affected by the mobility and resource constraints of nodes. Node mobility will 
impact on connecting stability, and node resource limitations will result in congestion, so the development of a routing protocol that 
promotes quality of service (QoS) in MANET is quite difficult. In particular, frequent interrupting connection may degrade QoS 
performance in the high-speed node drive scenario, so it is required to build the MANET protocol routing which can be adapted for 
changes in the networking architecture to support QoS. Moreover, MANET's multipath routing is the most necessary for secure 
transmission that can be achieved if self-centered nodes in the MANET network are ignored. Secure routing guarantees the reliability of 
data, confidentiality, authorization and authentication and non-denial. A novel safe multi-way method for dependable data transfer, 
which depends on the quality of service, is offered in this study. The Ad Hoc on Demand Backward Routing protocol with Optimal Fuzzy 
Logic (OFL) is also designed to provide multi-path routing. The hybridization of the bat optimization approach provides the optimum 
route by generating rules in OFL, and hence chooses an optimal rule. The efficiency of the planned technique is evaluated for factors 
such as final delay, the packet delivery ratio and so on. When the node is 150, the proposed hybrid bat algorithm achieved 0.61 of 
average packet latency, where the existing technique Particle Swarm Optimization achieved nearly 1.2 to 1.3 packet latency. In this 
communication network, measurement of packet delivery ratio is an important task for establishing the communication network. The 
results show that our suggested work has more energy efficiency and network lifetime than existing ones. 

mailto:kranthisri41@gmail.com


 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 2 

transfer and reduces network efficiency. The failure of the 
connection in MANET therefore becomes a serious issue. 
Furthermore, such a link failure often results in a trajectory of 
tragedy. As a result, data transmission reliability is diminished; 
the lower the packet transmission ratio the greater the delay from 
end to end. The measurement of data packets in a MANET 
increases the overhead control message pricey and reduces 
routing function performance. This is in turn more useful 
method in physiological data measurement and transmission via 
cognitive radio-based networks.  

Therefore, the selection of an alternative way or the creation 
of several paths is vital to ensure that the data transfer is reliable 
when the link is unsuccessful. And there is a lot of interest in 
loop-free paths to determine an ideal path between several routes 
between the source and destination of a network. In order to 
improve the reliability of the data transmission, multipath routing 
is constructed in a MANET [8], providing for a load balance 
between the nodes. Using several split pathways, the data has 
been sent parallel and the delivery rate is greatly improved. The 
problem of scalability, confidentiality, integrity and the life of the 
network is addressed in the multi-routing systems [9]. The 
dependability of data communication in a MANET is ensured by 
multi-way [10] routing between source and destination. Existing 
multipath routing systems in a MANET cause flooding, empty 
neighbourhoods, flat management, widespread data, high energy 
consumption, meddling and load balance concerns. The efficient 
multi-way routing strategy is therefore presented to tackle one or 
more of these problems. In the dynamic environment changes 
and frequent path failures [11]-[12] also the existing multi-path 
routers are not good. They also produce an overhead route in the 
network. The overhead routing takes up a significant part of the 
bandwidth of the network and the power of the mobile node 
quickly exhausts. Therefore, in order to limit the mobile node's 
participation in a route discovery phase to maintain dependability 
of data communication, with minimum overall costs, a reliable 
multi-path protocol is essential [13]. 

Multi-routing protocols are continuously improved for one 
decade by updating the quality of service (QoS) support demand. 
As a matter of fact, the multi path routing protocol may liken the 
available node resources in various paths, such as the available 
bandwidth, battery level, etc., and choose the optimum path for 
data to satisfy QoS. Today, in the MANET path selection 
parameters not only the available path node resources but path 
stability is included. Researchers are looking for different 
parameters to measure path stability effectively. However, the 
routing protocol for adaptive link state, which is changing quickly 
to enable QoS, has not yet been fully researched in the high-
speed movements of resource limited nodes scenario. Finding 
adequate criteria for road stability would also be harder. The 
major goal of this study is the development of a safe, nearby 
position verification protocol for the QoS performance of high-
speed MANET, by measuring the packet delivery ratio. A 
backward routing technique is devised in this routing protocol to 
determine the confident nodes. In this work, the optimum Fuzzy 
[14]-[15] system is used to improve the QoS routing procedure. 
In fuzzy logic system, the optimal rule generation is identified by 
using the hybrid bat algorithm (HBAT) technique and compared 
its efficiency with other optimization techniques.  

The paper is structured as follows: literary review is presented 
in Section 2. Section 3 delivers an explanation of the methods 
proposed. Section 4 provides validation of the methods 
proposed with current methodologies and measuring the packet 

delivery ratio. Finally, Section 5 offerings the conclusion of this 
study and its future effort. 

2. LITERATURE REVIEW 

This section mainly examines several routing methods in 
order to deal with changes in network topology and analyses their 
quality of service. Gomes et al. [16] introduces the industrial 
Wireless Sensor Network (WSN) Link Quality Estimator, and 
the dedicated WSN Node assesses connection quality depending 
on the signal intensity received. The estimate of connecting 
stability by distance alone, however, is not adequate for scenarios 
with high speeds such nodes that move with an average speed of 
20m/s. Since node speed can be reflected in QoS performance 
through path life. Long lifetime paths can give superior QoS due 
to the long-term stability of the broadcast path. To improve the 
QoS, Cross-layer multicast routing (CLMR) protocol based on a 
tree is proposed in [17]. They take advantage of data on battery 
life, physical layer bandwidth, routing layer stability and 
application layer overhead to upgrade the tree and then calculate 
the cost function based on the attributes of each layer. To 
develop robust routing pathways, CLMR selects low-cost nodes. 
A Topological change Adaptive Ad-hoc multi-path distance 
vector protocol is developed in [18], which aims to minimize the 
data traffic by utilizing QoS. The protocol's limits are that it does 
not function well in dynamic arrangements requiring stability and 
node density both on routes. In general, this protocol improves 
a little when other protocols do far better in many circumstances. 
The notion of finding the efficient route with least energy 
consumption and shortest distance was proposed for the Fitness 
Function Adaptive Ad-hoc On-demand Multipath Distance 
Vector protocol (FF-AOMDV) [19]. This protocol employed 
AOMDV, and the broadcast will be performed over the alternate 
route with the shortest path in the routing table in the event of 
any breakdown or break in connection. The FF-AOMDV model 
took into account few QoS characteristics, which means that its 
presentation is not as high as AOMDV, while network 
improvement is very limited. The authors of [20] devised a 
routing technique which improves the network's quality with 
GA. This work took into account several circumstances 
including mobility speed and failure of the nodes. Compared to 
other protocols, the network performance was enhanced. 
However, the energy consumption issue was not taken into 
account. In the multi-track transmission control protocol in [21], 
the energy-efficient mobbing control technique was projected. 
The results show that this algorithm works better than Multipath 
Transmission Control Protocol because of its reduced energy use 
and improved throughput. The suggested approach did not 
however take into account the random loss regularly occurring 
in wireless connections in loss networks. Therefore, loss of 
packets is supposed to decrease the congestion window and 
degrade data performance as a result of congestion lost. 

A routing procedure that takes into account the energy 
efficiency and certain QoS routing factors was proposed by the 
authors in [22]. In order to update nodes to QoD state, the 
method relies on the transmission of topological information to 
the whole network. This influx of information would raise the 
traffic overhead, especially with big networks. A protocol known 
as the Dynamic Energy Ad-hoc On-demand Distance Vector 
(AODV) protocol was suggested in [23]. The major goal is to 
reduce the time required to transmit packets, reduce the 
consumption of energy as well as to maximize the life of the 
network. It determines the shortest distance path and selects the 



 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 3 

intermediate nodes with high residual power and network 
authority. In case of any inattentive link breaking or low-energy 
nodes during packet delivery, this procedure gives external 
energy to nodes. This enhances the reliability of the route and 
the lifespan. This external energy supply can also be considered 
to minimize the cost of the network. 

Authors of [24] proposed to do away with congestion and 
clustering in WSN a new protocol called Congestion and 
Clustering Routing. Congestion control and model-based 
bandwidth estimates were presented in sender side to TCP-
traffic via containment-free Multiple-Access Time Division 
based MANETs [25]. Both protocols offer acceptable routing 
options based on preventing congestion but do not take account 
of the random loss that would unduly narrow the congestion 
window. Masood Ahmad et al. [26] introduced a clustering 
technique based on honeybee and genetic algorithm. The 
overhead of topology maintenance is reduced by this structure. 
Dynamic solutions with improved quality are provided by this 
combined algorithm, therefore, stable and balanced clusters are 
formed. However, multipath routing is done by this technique 
with high energy consumption. The energy issues and workload 
of CH are focused by Amutha et al. [27] by developing a Cluster 
Manager-Based CH selection scheme. The CH execute the 
packet transmission between sensor nodes in the network. Low 
bandwidth, reliable throughput and low energy are provided by 
this technique. However, there is a major problem developed in 
this study, which is use of extra mechanism for selection process 
that leads automatically low bandwidth. 

2.1. Problem Identification 

MANET's volatility and limited assets make it very difficult to 
communicate information via such networks. 

– Because of inconsistent wireless channel, integrated 
device shortages, channel quarrels, portability, node 
control and affirmation device, assurance of QoS to 
MANET applications is also highly complex. 

– The MANET mobile nodes do not provide adequate 
power and risk repetitive node failures, resulting in a 
variety of network topologies, network allocations, 
packet loss problems and the lowest signal quality. 

– The current key organisation strategy of addressing an 
awkward node successfully does not deprive users of 
creating useful identities or the stealing of the 
individuality of individuals who do not participate in 
network activities. 

– Moreover, MANETs generally require fresh limitations 
on problems related with QoS in wire-based networks. 
This is the projected dynamic behaviour of the 
respective networks and the controlled assets. 

3. PROPOSED METHODOLOGY 

In this study, a new optimum multi-track routing protocol is 
provided for MANET to improve the QoS, where network 
failures are reduced and will not use the information of 
individuals, who are not participate in the network activities. This 
is the four phases of the suggested model: The trust ideal for 
routing, the multipath routing, the optimal rule and Data 
Transmission. The phases of the proposed model are shown in 
Figure 1. 

The secure adjacent position trust verification protocol 
(SAPTV) for analysis of trust value process is considered here. 
In our trust model each node maintains a value of trust for each 
of its neighbours. The Ad-hoc On-demand Distance Vector - 

Backward Routing (AODV-BR) is employed for the 
development of confidential packets in the multi-way outing 
process. If source S wishes to connect with destination D, the 
route delivery is initiated by the source at that time. To improve 
the QoS routing procedure, the optimum Fuzzy system is used. 
The rules are made in a fuzzy source- and target-based system. 
An inspired optimization model is proposed to optimize these 
rules.  

3.1. Trust Model for Multipath Routing 

The source node begins the network-wide flood with the 
distribution of the route query packet, the most important 
waiting packets for the route response. The simulation process, 
the SAPTV protocol, is now intended. The data clustering 
updates the entry value. As demonstrated in (1), the trust 
measure is then taken into account 

𝑇𝑣1~𝑇𝑣2 = 1 − (1 − 𝑇𝑣2)
𝑇1  , (1) 

where T 1 is the Trust Value suggestion and T 2 is the Trust 
Value Direct. The values are between 0 and 1. The innovative 
technique specifies an explicit trust relationship as the trust 
suggestion among two nodes in the similar collection, but the 
trust references are regarded by deliberation as a confidence 
association among nodes in the diverging collection. The trust 
warning for pathways is always active if it is less than the trust 
value. 

3.2. Multipath Routing (MR) Process 

In order to ensure the source-destination route, the routing 
protocols must show significant energy efficiency levels. In this 
respect, there are now a number of energy-efficient routing 
algorithms. The Protocols in this document propose that a 
considerable effort be made to depart from origin to endpoint 

 

Figure 1. Proposed model Schematic diagram.  

 

Figure 2. Fuzzy logic process.  



 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 4 

node on the most energy-efficient path. A new routing method 
for MANET is developed called AODV-BR that identifies the 
least node’s residual energy in the paths and classifies the 
descending path in the descendant order of nodal residual 
energy. 

Figure 2 provides an ideal Fuzzy Logic (FL) method block 
diagram. After several interactions, the connection quality of a 
route is likely to be significantly diminished. Thus, the HBAT 
algorithm, which works admirably efficiently, selects the best way 
among the traditional route in the network. 

3.2.1. AODV-BR Protocol  

The response messages and data packets are checked, and 
local repair is conducted in the AODV-BR by a backup route. 
The source and endpoint nodes have the ability to change the 
data in the AODV-BR. Thus, with the assistance of this strategy, 
the finest overheard backup route is taken up. When the data 
packet is eradicated, it is transmitted via a substitute path in the 
AODV-BR. 

3.3. Routing Process on Optimal Fuzzy Logic 

In this study, QoS is improved by providing the shortest route 
using the Optimal Fuzzy Logic (OFL) algorithm, where HBAT 
is developed for optimal rule generation in OFL algorithm. 
Initially, the description of OFL is provided, where FL consists 
of four phases: flushing, generating a rule, membership, and 
defuzzifying. The values of a number of rules are properly 
translated into linguistic variables which in fluid logic describe 
the strength of settlement with the arithmetic value of the targets. 
Three separate kinds of values of trust like friends, acquaintance 
and strangers are characterized correspondingly by the three 
fuzzy as (high, medium and low). 

3.3.1. Fuzzification 

The hard value of any fuzzy collection is subjected to 
fluctuation to describe the value of the relationship. It also 
demonstrates the ability of a higher system to adjust the real 
scalar quality to a fluffy value. A floating X-collector partition 
represents a job A: x an L, where L reaches [0, 1], otherwise 
known as the membership feature. 

3.3.2. Membership Function 

This function is educated by many forms in order to evaluate 
the very comfortable Member Company, by put on the easiest 
separate Member Function from one another, through straight 
lines. 

3.3.3. Rule Generation Process 

The collection of IF-THEN rules builds on the data of a 
human expert to detect the preferred performance of the 
scheme. The rules can be maintained with the aim of achieving 
the entire functionality of the system. Under Table 1 below are 
presented sample fuzzy rules. 

3.4. Data Transmission  

After receipt of the message from the destination, the data 
packet is sent to the destination node. There are several 
transmissions in preferred pathways, and the link quality is likely 
to be lost. Thus, because of loss of the connection value, the 
source node does not accept any recognitions from the target 
node. The network opts for an ideal way from the basis routes. 
The best way to reach the goal should be energy-effective and 
must be the shortest. Therefore, in order to identify the shortest 
route in an optimal way, the research study uses HBAT algorithm 
in OFL, which is explained as follows: 

3.4.1. Bat Algorithm 

Bat algorithm (BAT), which was projected by Xin-She Yang 
in 2010, is a swarm-intelligence algorithm. In the algorithm, the 
search process is motivated by the fats, by the ability to search 
and locate the proofs. 

The bats emit noise and can make a difference between the 
food and beasts, and estimate the distance, using the reflected 
sound waves. The following equation (2) is used to update the 
location of the bats (solution) during the optimization procedure: 

𝑥𝑖
𝑡 = 𝑥𝑖

𝑡−1 + 𝑣𝑖
𝑡  , (2) 

where the current solution is signified by 𝑥𝑖
𝑡−1 and the new, 

updated position at iteration 𝑖 − 𝑡ℎ solution is represented by 𝑥𝑖
𝑡 . 

The velocity is represented by 𝑣𝑖
𝑡 . The velocity at time step 𝑡 is 

designed as follows in (3): 

𝑣𝑖
𝑡 = 𝑣𝑖

𝑡−1 + (𝑥𝑖
𝑡−1 − 𝑥∗)𝑓𝑖  , (3) 

where 𝑥∗, denotes the current best global location, and 𝑓𝑖   
specifies the 𝑖 − 𝑡ℎ bat frequency. The solution frequency is 
consistently taken from the specified range from the least to the 
maximum frequency and is evaluated as follows in (4): 

𝑓𝑖 = 𝑓min + (𝑓max − 𝑓min)𝛽, (4) 

where 𝑓min and 𝑓max are represented as the minimum and 
maximum frequency, correspondingly and 𝛽 is represented as a 
random number, 𝛽 ∈  [0, 1, which is defined as follows in (5): 

𝑥new = 𝑥old + 𝜖𝐴
𝑡  , (5) 

where 𝛽 is a scaling factor with random values among 0 and 1, 
when the average loudness of all bats is given by 𝐴𝑡 . The latch is 
updated as follows, after the prey is found by a bat and that are 
shown in (6) and (7): 

𝐴𝑖
𝑡 = 𝛼𝐴𝑖

𝑡−1, 𝑟𝑖
𝑡 = 𝑟𝑖

0[1 − e−𝛾 𝑡 ] (6) 

𝐴𝑖
𝑡 → 0, 𝑟𝑖

𝑡 → 𝑟𝑖
0, while 𝑡 → ∞ , (7) 

where 𝐴𝑖
𝑡  designates to the loudness of 𝑖 − 𝑡ℎ bat, at iteration 𝑡, 

and 𝑟 is the pulse emission level. The values of 𝛼 and 𝛾 are 
constant.  

3.4.2. Hybridized Bat Algorithm 

The process of exploitation is strong in the unique bat 
algorithm, but it is caught in local optimum when exploring the 
search space. In this paper, the hybridized bat algorithm is used 
to improve the exploration phase of the Bat Algorithm with the 
search of bees from the algorithm for an Artificial Bee Colony, 
in this way, it is prevented to get trapped to the local optima. 
Randomly inside the lower and lower limits are generated the 
initial solution populations as in (8); 

Table 1. Fuzzy Rules. 

S.No Mobility Trust value Delay Route 

1 Low Low Low Below optimal 

2 Medium Low Medium Below optimal 

3 High High Low Sub optimal 

4 High Medium Low Sub optimal 

5 Low High Low Low optimal 

6 Low High Low Optimal 

7 Medium Medium Low Optimal 



 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 5 

𝑥𝑖,𝑗 = 𝑙𝑏𝑗 + rand ∙ (𝑢𝑏𝑗 − 𝑙𝑏𝑗 ) , (8) 

where 𝑙𝑏𝑗  and 𝑢𝑏𝑗  are referenced to both the lower and the upper 

bound; rand is a random number of uniform distribution, 
rand ∈  [0, 1]; and 𝑥𝑖,𝑗  is a solution where the items are 

referenced by 𝑗. The fitness value is calculated as shown in (9) 
for each individual in the population after the generation of the 
starting population. 

𝐹𝑥𝑖 = {

1

𝑓𝑥𝑖
𝑖𝑓 𝑓𝑥𝑖 ≥ 0

1 + |𝑓𝑥𝑖 | otherwise,

 (9) 

where the fitness function of 𝑖 − 𝑡ℎ individual is denoted by 𝐹𝑥𝑖 , 

and 𝑓𝑥𝑖  is the objective function of the 𝑖 − 𝑡ℎ individual. Table 2 
shows the parameters used in proposed algorithm. 

Two alternative methods improve the exploration of search 
space. The counter (t) determines if the search procedure of the 
bat algorithm or the observer bee mechanism will be used. At 
each step, the persons move and update the position in 
accordance with the Eq. (2), however the observer technique 
uses the following equation (10) if the value of t is odd: 

𝑥𝑖,𝑗
𝑡 = 𝑥𝑖,𝑗

𝑡−1 + rand ∙ (𝑥𝑖,𝑗
𝑡−1 − 𝑥𝑘,𝑗

𝑡−1), (10) 

where the novel solution at time step 𝑡 is denoted by 𝑥𝑖,𝑗
𝑡 . The 

𝑗 − 𝑡ℎ element of the 𝑖 − 𝑡ℎ individual is denoted by 𝑥𝑖,𝑗
𝑡−1and 

𝑥𝑘,𝑗
𝑡−1 denotes the 𝑘 − 𝑡ℎ neighbor solution, rand ∈  [0, 1]. 

The random walk of the solution according to (5) will utilize 
the promising area. 

The selection probability that is expressed as follows in (11) 
determines whether the newly produced solution is established 
or discarded: 

𝑝𝑖 =
𝐹avg

∑ 𝐹𝑥𝑖
𝑛
𝑖=1

 , (11) 

where 𝑝𝑖  means the selection possibility and 𝐹 as the fitness 
value.  

4. RESULTS AND DISCUSSION 

In a series of simulated SAPTV scenarios, testing is 
performed with the proposed SAPTV-HBAT algorithm, existing 
SAPTV particle swarm optimization, Whale optimization 
Algorithm (WOA) and BAT algorithms. A C++-based NS-2 
simulator implements the planned algorithm and uses the 
amount of node, node mobility, traffic load and halt time for 
several simulation scenarios. Table 3 lists the simulation 

parameters. Nodes are randomly positioned within a 1000-meter 
radius in all circumstances. Every mobile node's maximum 
transmission range is 250 m. Nodes often move with speed in 
the range [0, 10] m/s, based on the random mobility point model. 
In this concept of mobility, all nodes migrate to a new destination 
point and remain there for a certain duration termed a pause. 

The channel is 2 Mbps and the IEEE 802.21 power save 
mode is used by the MAC protocol. Window size and beacon 
interval ad hoc traffic indication mode are 0.05s and 0.25s. Each 
simulation is performed for 900 s. Two ray floor models are 
employed to propagate. Average control messages, power 
consumption, packet delivery ratio (PDR), network service life 
and delay are involved in the investigation. Simulation outcomes 
were obtained after 10 runs to achieve stable status. 

Table 2. Hybridized bat algorithm control factors. 

Notation Parameter Value 

N Population size 20 

𝛾 Constant parameter 0.9 

𝐴𝑚𝑖𝑛 
Constant minimum 

loudness 
1 

α Constant parameter 0.9 

𝐴0 
Maximum initial 

loudness 
100 

𝑓min Minimum frequency 0 

𝑓max Maximum frequency 1 

MaxIter Maximum iteration 30 

Algorithm 1. hybridized bat algorithm 

𝑂𝑏𝑗𝑒𝑐𝑡𝑖𝑣𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑓(𝑥) 
𝐼𝑛𝑖𝑡𝑖𝑎𝑙𝑖𝑧𝑒 𝑡ℎ𝑒 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑏𝑎𝑡𝑠, 𝑡ℎ𝑒 𝑣𝑎𝑙𝑢𝑒𝑠 𝑜𝑓 𝑣𝑖 , 𝑟𝑖   
𝑎𝑛𝑑 𝐴𝑖, 𝑑𝑒𝑓𝑖𝑛𝑒 𝑡ℎ𝑒 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 
𝑝𝑢𝑙𝑠𝑒 (𝑓𝑖 )𝑎𝑡 𝑥𝑖 , 𝑡ℎ𝑒 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑖𝑡𝑒𝑟𝑎𝑡𝑖𝑜𝑛  
(𝑀𝑎𝑥𝐼𝑡𝑒𝑟), 𝑎𝑛𝑑 𝑠𝑒𝑡 𝑡ℎ𝑒 𝑖𝑡𝑒𝑟𝑎𝑡𝑖𝑜𝑛 
𝑐𝑜𝑢𝑛𝑡𝑒𝑟 (𝑡) 𝑡𝑜 0 
𝒘𝒉𝒊𝒍𝒆 𝑡 <  𝑀𝑎𝑥𝐼𝑡𝑒𝑟 𝒅𝒐 
𝒇𝒐𝒓 𝑖 

=  1 𝑡𝑜 𝑁 (𝑒𝑎𝑐ℎ 𝑁 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛) 𝒅𝒐 
𝒊𝒇 𝑡 𝑖𝑠 𝑒𝑣𝑒𝑛 𝒕𝒉𝒆𝒏 
     𝐶𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒 𝑡ℎ𝑒 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦 𝑎𝑛𝑑 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑣𝑎𝑙𝑢𝑒 
     𝑢𝑠𝑖𝑛𝑔 (3) 𝑎𝑛𝑑 (4) 
     𝑃𝑒𝑟𝑓𝑜𝑟𝑚 𝑡ℎ𝑒 𝑏𝑎𝑡 𝑠𝑒𝑎𝑟𝑐ℎ 𝑝𝑟𝑜𝑐𝑒𝑑𝑢𝑟𝑒 𝑢𝑠𝑖𝑛𝑔 (2) 
𝒆𝒍𝒔𝒆 
     𝑃𝑒𝑟𝑓𝑜𝑟𝑚 𝑡ℎ𝑒 𝑜𝑛𝑙𝑜𝑜𝑘𝑒𝑟 𝑠𝑒𝑎𝑟𝑐ℎ 𝑝𝑟𝑜𝑐𝑒𝑑𝑢𝑟𝑒 𝑢𝑠𝑖𝑛𝑔 (11) 
𝒆𝒏𝒅 𝒊𝒇 
𝒊𝒇 𝑟𝑎𝑛𝑑 > 𝑟𝑖 𝒕𝒉𝒆𝒏 
     𝑆𝑒𝑙𝑒𝑐𝑡 𝑡ℎ𝑒 𝑓𝑖𝑡𝑡𝑒𝑠𝑡 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 
     𝑃𝑒𝑟𝑓𝑜𝑟𝑚 𝑡ℎ𝑒 𝑟𝑎𝑛𝑑𝑜𝑚 𝑤𝑎𝑙𝑘 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 𝑢𝑠𝑖𝑛𝑔 (5) 
𝒆𝒏𝒅 𝒊𝒇 
𝒊𝒇𝑝𝑖 < 𝐴𝑖  𝑎𝑛𝑑 𝑓(𝑥𝑖 )  <  𝑓(𝑥∗) 𝒕𝒉𝒆𝒏 
     𝑇ℎ𝑒 𝑛𝑒𝑤𝑙𝑦 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 𝑖𝑠 𝑎𝑐𝑐𝑒𝑝𝑡𝑒𝑑 
     𝑅𝑒𝑑𝑢𝑐𝑒 𝐴𝑖  𝑎𝑛𝑑 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒 𝑟𝑖  𝑢𝑡𝑖𝑙𝑖𝑧𝑖𝑛𝑔 (6) 
𝒆𝒏𝒅 𝒊𝒇 
𝒆𝒏𝒅 𝒇𝒐𝒓 
     𝐹𝑖𝑛𝑑 𝑎𝑛𝑑 𝑠𝑎𝑣𝑒 𝑡ℎ𝑒 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑏𝑒𝑠𝑡 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 𝑥 ∗ 
𝒆𝒏𝒅 𝒘𝒉𝒊𝒍𝒆 
𝑅𝑒𝑡𝑢𝑟𝑛 𝑡ℎ𝑒 𝑏𝑒𝑠𝑡 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛. 

Table 3. Simulation structure. 

Parameter Value 

Packet Size 256 bytes 

Pause Time 0–600 s 

Transmission Range (R) 250 m 

Data Size 5 Mbytes 

Data Rate 2 Mbits/sec 

Simulation Period 900 s 

Traffic Load 1 to 5 packets 

Traffic Type CBR 

Sum of Nodes 100–150 

Initial Energy 180 J 

Maximum Node Speed 10 m/s 

Network Area 1000 m × 1000 m 



 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 6 

Initially, the energy consumption of each model is test and the 
graphical results are shown in Figure 3. 

The Particle Swarm Optimization (PSO) algorithm, WOA, 
BAT and the proposed HBAT of SAPTV achieved 4.9 J, 3.7 J, 
3.5 J and only 2.4 J, when the node is 120. However, the energy 
consumption is increased, when the number of nodes upsurges. 
For instance, when the node is 150, the PSO, WOA, BAT and 
proposed HBAT of SAPTV achieved 5.8 J, 4.8 J, 3.9 J and only 
2.8 J. This decrease is due to a decrease in energy, hops and 
traffic consumption in data transmission over the defined path. 
Moreover, hop-proportional energy is also available. Traffic 
loads hence reduce energy consumption when hop numbers are 
optimized. The packet size for the proposed SAPTV-HBAT 
protocol is shown in Figure 4.  

It is experiential that the proposed SAPTV-HBAT has better 
PDR than SAPTV-PSO SAPTV-WOA and SAPTV-BAT. The 
PDR increases when the number of nodes upsurges. When the 
node is 100, the PSO, WOA, BAT and proposed HBAT with 

SAPTV achieved PDR of 65, 79, 87 and 96. In addition, these 
techniques achieved PDR of 72, 80, 88 and 97, when the node is 
130. The increased sum of nodes makes sure the optimal route 
from the source to the destination is selected. The amount of 
backup and duration, i.e., path life, is also increased. Figure 5 
displays the average packet latency in diverse network sizes for 
the proposed system. 

When the node is 110, the average packet latency of SAPTV-
PSO, SAPTV-WOA, SAPTV-BAT and SAPTV-HBAT is 0.66, 
0.61, 0.47 and 0.42. The same techniques achieved 0.91, 0.81, 
0.67 and 0.54 of average packet latency, when the node reaches 
130th level. But, the existing technique PSO and WOA with 
SAPTV reaches high packet latency i.e. nearly 1.2 to 1.3, where 
BAT achieves 0.93 and proposed HBAT with SAPTV achieved 
0.61 of average packet latency, when the node is 150. Compared 
to other approaches, the proposed SAPTV-HBAT is observed 
to be less delay. The growing number of paths between nodes 
results in the selection of the best path with fewer hop counts 
and node sharing. Therefore, this approach has a reduced latency 
and uses bandwidth better. Figure 6 displays the average sum of 

 

Figure 3. Graphical Representation of Proposed Model in terms 
of Energy Consumption. 

 

Figure 4. Graphical Representation of Proposed SAPTV-
HBAT on the basis of Packet delivery ratio. 

 

Figure 5. Graphical Representation of Proposed SAPTV-HBAT in terms of 
Average Packet Latency. 

 

Figure 6. Graphical Representation of Proposed SAPTV-HBAT in terms of 
Average number of control packet. 

0

1

2

3

4

5

6

100 110 120 130 140 150

E
n

e
r
g

y
 C

o
n

su
m

p
ti

o
n

 

Number of Nodes 

SAPTV-

PSO

SAPTV-

WOA

SAPTV-

BAT

SAPTV-

HBAT

0

20

40

60

80

100

1
0
0

1
1
0

1
2
0

1
3
0

1
4
0

1
5
0

P
a

c
k

e
t 

d
e
li

v
e
r
y

 r
a

ti
o

 

Number of Nodes 

SAPTV-PSO

SAPTV-WOA

SAPTV-BAT

SAPTV-HBAT

0

0,2

0,4

0,6

0,8

1

1,2

1,4

100 110 120 130 140 150

A
v

e
r
a

g
e
 P

a
c
k

e
t 

L
a

te
n

c
y

 

Number of Nodes 

SAPTV-

PSO

SAPTV-

WOA

SAPTV-

BAT

SAPTV-

HBAT

0

5

10

15

20

25

1
0
0

1
1
0

1
2
0

1
3
0

1
4
0

1
5
0

N
e
tw

o
r
k

 L
if

e
ti

m
e
 

Number of Nodes

SAPTV-PSO

SAPTV-WOA

SAPTV-BAT

SAPTV-HBAT



 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 7 

control packets for the SAPTV-HBAT design for various 
network sizes. 

When the node is 100, the average sum of control packets of 
PSO, WOA, BAT and HBAT with SAPTV is 1150, 1020, 950 
and 830. The same techniques achieved 1250, 1158, 1059 and 
950 of average number of control packets, when the node is 120. 
But, the existing technique PSO and WOA with SAPTV reaches 
high control packets i.e. nearly 1400 to 1500, where BAT 
achieves nearly 1200 to 1300 and proposed HBAT achieved 1120 
and 1210 of average control packets, when the node is 140 and 
150. It shows that control packets are increasing with the growth 
in the sum of nodes. The HBAT optimization approach used to 
locate paths from source to destination is employed in a multi-
path network. Figure 7 shows the study of network life.  

The lifetime of the network shall be measured as any of the 
nodes in the track are determined to be dead among the start of 
the data transmission into a particular path. When the node is 
100, the average network lifetime of PSO, WOA, BAT and 
HBAT with SAPTV is 7, 9, 12 and 14. The same techniques 
achieved 8, 11, 13 and 15 of average network lifetime, when the 
node is 110. The average network lifetime of SAPTV-PSO, 
SAPTV-WOA, SAPTV-BAT and SAPTV-HBAT is 13, 15, 17 
and 19, when the node is 130. But, the existing technique PSO 
and WOA reaches less network life time i.e., nearly 14 to 17, 
where BAT achieves nearly 18 and 20 and proposed HBAT 
achieved 21 and 23 of average path life time, when the node is 
140 and 150. The lifetime of the network is better than everyone 
in the projected SAPTV-HBAT. It is because the path selection 
is based on a multi-target function. The data transmission will 
not involve the node with excessive electricity consumption and 
traffic backlog. High traffic losses lead to packet loss and 
needless energy use. The HBAT in SAPTV also displays better 
network lifetime than PSO, WOA and BAT due to the 
consideration of trustworthy routes. 

5. CONCLUSION 

MANET is an autonomous mobile node approach. QoS 
protocols which can be adapted to quick topology modifications 
would assist high-speed MANET network applications. A route 

may lose its link quality after the number of broadcasts. In this 
work, a multi-path routing system for the dependable exchange 
of data was suggested for QoS. OFL is developed to find the 
shortest route in multipath routing, where the optimal rules are 
generated by HBAT. The performance of the proposed model is 
compared with existing optimization approaches in terms of 
network life, PDR, energy consumption, average control 
packages and latency. The proposed protocol can be employed 
as an effective high-speed MANET solution with QoS 
necessities and limitations on resources such as multimedia 
communication between car and vehicle, the mobile video 
surveillance system etc. In the context of the previous energy-
sensitive routing protocols the energy usage of the suggested 
routing protocol was quite small. As a future effort, the author 
will further examine the development of high-speed routing 
protocols that take into account both trajectory stability and 
node density. Furthermore, the next research will create the way 
forward by implementing effective protocols and maximizing the 
performance, the QoS and lowering communication delays. 

REFERENCES 

[1] L. Ciani, A. Bartolini, G. Guidi, G. Patrizi, A hybrid tree sensor 
network for a condition monitoring system to optimise 
maintenance policy, Acta IMEKO, 9(1) (2020), pp. 1-7 
DOI: 10.21014/acta_imeko.v9i1.732  

[2] R. Desai, B.P. Patil, D.P. Sharma, Routing Protocols for Mobile 
Ad Hoc Network: A Survey and Analysis, Indonesian Journal of 
Electrical Engineering and Computer Science, 7(3) (2017), pp.795-
801. 
DOI: 10.11591/ijeecs.v7.i3.pp795-801  

[3] S. Yan, Y. Chung, Improved ad hoc on-demand distance vector 
routing (AODV) protocol based on blockchain node detection in 
ad hoc networks, International Journal of Internet, Broadcasting 
and Communication, 12(3) (2020), pp.46-55. 
DOI: 10.7236/IJIBC.2020.12.3.46  

[4] A. Sulthana, S. S. Mirza, An efficient Kalman noise canceller for 
cardiac signal analysis in modern telecardiology systems, IEEE 
Access, 6 (2018), pp.34616-34630. 
DOI: 10.1109/ACCESS.2018.2848201  

[5] D. Peters, P. Scholz, F. Thiel, Software Separation in Measuring 
Instruments through Security Concepts and Separation Kernels, 
Acta IMEKO, 7(1) (2018), pp.13-19. 
DOI: 10.21014/acta_imeko.v7i1.510  

[6] S. Putluri, S. Y. Fathima, Cloud-based adaptive exon prediction 
for DNA analysis, Healthcare technology letters, 5(1) (2018), pp. 
25-30. 
DOI: 10.1049/htl.2017.0032  

[7] A. Boukerche, B. Turgut, N. Aydin, M. Z. Ahmad, L. Boloni, D. 
Turgut, Routing protocols in ad hoc networks: a survey, Computer 
Networks, 55(13) (2011), pp. 3032–3080. 
DOI: 10.1016/j.comnet.2011.05.010  

[8] F. Noorbasha, M. Manasa, R. T. Gouthami, S. Sruthi, D. H. Priya, 
N. Prashanth, M. Z. Ur Rahman, FPGA Implementation of 
Cryptographic Systems for Symmetric Encryption. Journal of 
Theoretical and Applied Information Technology, 95(9) (2017), 
pp. 2038-2045. 

[9] S. Chakraborty, S. Chakraborty, S. Nandi, S. Karmakar, Fault 
resilience in sensor networks: distributed node-disjoint multi-path 
multi-sink forwarding, Journal of Network and Computer 
Applications, 57, pp. 85–101, 2015. 
DOI: 10.1016/j.jnca.2015.07.014  

[10] P. V. V. Kishore, A. S. C. S. Sastry, Double technique for 
improving ultrasound medical images. Journal of Medical Imaging 
and Health Informatics, 6(3) (2016), pp. 667-675 
DOI: 10.1166/jmihi.2016.1743   

Figure 7. Graphical Representation of Proposed SAPTV-HBAT on the basis of 
Network Lifetime. 

0

200

400

600

800

1000

1200

1400

1600

1
0
0

1
1
0

1
2
0

1
3
0

1
4
0

1
5
0

A
v

e
r
a

g
e
 n

u
m

b
e
r
 o

f 
c
o

n
tr

o
l 

p
a

c
k

e
t 

Number of Nodes 

SAPTV-PSO

SAPTV-WOA

SAPTV-BAT

SAPTV-

HBAT

https://doi.org/10.21014/acta_imeko.v9i1.732
https://doi.org/10.11591/ijeecs.v7.i3.pp795-801
https://doi.org/10.7236/IJIBC.2020.12.3.46
https://doi.org/10.1109/ACCESS.2018.2848201
https://doi.org/10.21014/acta_imeko.v7i1.510
https://doi.org/10.1049/htl.2017.0032
https://doi.org/10.1016/j.comnet.2011.05.010
https://doi.org/10.1016/j.jnca.2015.07.014
https://doi.org/10.1166/jmihi.2016.1743


 

ACTA IMEKO | www.imeko.org March 2022| Volume 11 | Number 1 | 8 

[11] M. Sedrati, A. Benyahia, Multipath routing to improve quality of 
service for video streaming over mobile ad hoc networks, Wireless 
Pers. Communication, 99(2018), pp. 999–1013. 
DOI: 10.1007/s11277-017-5163-6  

[12] H. A. Ali, M. F. Areed, and D. I. Elewely, An on-demand power 
and loadaware multi-path node-disjoint source routing scheme 
implementation using NS-2 for mobile ad-hoc networks,’Simul. 
Model. Pract. Theory, 80 (2018), pp. 50–65. 
DOI: 10.1016/j.simpat.2017.09.005  

[13] M. B. Channappagoudar, P. Venkataram, Performance evaluation 
of mobile agent-based resource management protocol for 
MANETs, Ad Hoc Netw., (2016), 36, pp. 308–320. 
DOI: 10.1016/j.adhoc.2015.08.008  

[14] E. Petritoli, F. Leccese, Precise takagi-sugeno fuzzy logic system 
for UAV longitudinal stability: An industry 4.0 case study for 
aerospace., ACTA IMEKO, 9(4) (2020), pp.106-113. 
DOI: 10.21014/acta_imeko.v9i4.723  

[15] E. Benoit, Fuzzy scales for the measurement of color. ACTA 
IMEKO, 3(3) (2014), pp.57-62. 
DOI: 10.21014/acta_imeko.v3i3.89  

[16] S. S. Mirza, M. Z. Ur Rahman, Efficient adaptive filtering 
techniques for thoracic electrical bio-impedance analysis in health 
care systems, Journal of Medical Imaging and Health Informatics, 
7(6) (2017), pp.1126-1138. 
DOI: 10.1166/jmihi.2017.2211  

[17] D. Chander, R. Kumar, QoS enabled cross-layer multicast routing 
over mobile ad hoc networks, Procedia Computer Science, 
125(2018), pp. 215–227. 
DOI: 10.1016/j.procs.2017.12.030  

[18] A. Taha, R. Alsaqour, M. Uddin, M. Abdelhaq, T. Saba, Energy 
efficient multipath routing protocol for mobile ad-hoc network 
using the fitness function, IEEE Access, 5 (2017), pp. 10369–
10381. 
DOI: 10.1109/ACCESS.2017.2707537  

[19] W. Wang, X. Wang, D. Wang, Energy efficient congestion control 
for multipath TCP in heterogeneous networks, IEEE Access, vol. 
6, pp. 2889–2898, 2018. 
DOI: 10.1109/ACCESS.2017.2785849  

[20] N. Muruganantham, H. El-Ocla, Routing using genetic algorithm 
in a wireless sensor network, Wireless Pers. Communication, 
111(4) (2020), pp. 2703–2732. 
DOI: 10.1007/s11277-019-07011-8  

[21] W. Jabbar, W. Saad, M. Ismail, MEQSA-OLSRv2: A multi criteria 
based hybrid multipath protocol for energy-efficient and QoS-
aware data routing in MANET-WSN convergence scenarios of 
IoT, IEEE Access, 6 (2018), pp. 76546–76572. 
DOI: 10.1109/ACCESS.2018.2882853  

[22] T. K. Saini, S. C. Sharma, Prominent unicast routing protocols for 
mobile ad hoc networks: Criterion, classification, and key 
attributes, Ad Hoc Netw,, 89(2019), pp. 58–77. 
DOI: 10.1016/j.adhoc.2019.03.001  

[23] J. Deepa, J. Sutha, A new energy based power aware routing 
method for MANETs, Cluster Comput., 22(S6) (2019), pp. 13317–
13324. 
DOI: 10.1007/s10586-018-1868-x  

[24] M. Farsi, M. Badawy, M. Moustafa, H. A. Arafat, Y. Abdulazeem, 
A congestion-aware clustering and routing (CCR) protocol for 
mitigating congestion in WSN, IEEE Access, 7(2019), pp. 
105402–105419.  
DOI: 10.1109/ACCESS.2019.2932951  

[25] P. Pal, S. Tripathi, C. Kumar, Bandwidth estimation in high 
mobility scenarios of IEEE 802.11 infrastructure-less mobile ad 
hoc networks, International Journal of Communication System, 
32(15) (2019), p. e4080. 
DOI: 10.1002/dac.4080  

[26] M. Ahmad, A. Hameed, F. Ullah, I. Wahid, S. U. Rehman, H. A. 
Khattak, A bio-inspired clustering in mobile ad-hoc networks for 
internet of things based on honeybee and genetic algorithm, 
Journal of Ambient Intelligence and Humanized Computing, 
11(1) (2020), pp. 4347–4361. 
DOI:  10.1007/s12652-018-1141-4  

[27] S. Amutha, B. Kannan, M. Kanagaraj, Energy‐ efficient cluster 

manager‐ based cluster head selection technique for 
communication networks, International Journal of 
Communication Systems, 34(5) (2021), p. e4741. 
DOI: 10.1002/dac.4427  
 

 

https://doi.org/10.1007/s11277-017-5163-6
https://doi.org/10.1016/j.simpat.2017.09.005
https://doi.org/10.1016/j.adhoc.2015.08.008
https://doi.org/10.21014/acta_imeko.v9i4.723
https://doi.org/10.21014/acta_imeko.v3i3.89
https://doi.org/10.1166/jmihi.2017.2211
https://doi.org/10.1016/j.procs.2017.12.030
https://doi.org/10.1109/ACCESS.2017.2707537
https://doi.org/10.1109/ACCESS.2017.2785849
https://doi.org/10.1007/s11277-019-07011-8
https://doi.org/10.1109/ACCESS.2018.2882853
https://doi.org/10.1016/j.adhoc.2019.03.001
https://doi.org/10.1007/s10586-018-1868-x
https://doi.org/10.1109/ACCESS.2019.2932951
https://doi.org/10.1002/dac.4080
https://doi.org/10.1007/s12652-018-1141-4
https://doi.org/10.1002/dac.4427