Wireless Personal Communications (2021) 116:153–176

https://doi.org/10.1007/s11277-020-07709-0

ARIOR: Adaptive Ranking Based Improved Opportunistic

Routing in Wireless Sensor Networks

Premkumar Chithaluru1 · Rajeev Tiwari2 · Kamal Kumar3

Published online: 8 August 2020

© Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract
Adaptive ranking in opportunistic routings is known to significantly improve the perfor-
mance over traditional opportunistic routing protocols in wireless networks using a vol-
unteer node for energy efficiency. Wireless Sensor Networks (WSNs) mainly forward
packets through neighbor nodes, which usually may go to sleep mode (when not in use)
to improve the lifetime of the network. The sleep state of nodes is an infeasible mecha-
nism because the nodes are awake continuously and are able to overhear other broadcasts.
This paper presents a recently researched improved mechanism to determine the Cluster
Head(CH) and set of nodes participating in routing. Node ranking resulted in the improved
routing table for node selection to forward the packets towards the gateway was generated.
By consistent performance modeling and network simulations, ARIOR algorithm exhibits
improvements in terms of Message Success Rate(MSR), Packet Delivery Ratio(PDR) and
Energy Consumption(CE) in comparison to existing Opportunistic Routing(OR) protocols.

Keywords Opportunistic routing (OR) · Wireless sensor network (WSN) · Volunteer node ·
Energy efficiency · Lifetime

1 Introduction

MSR, PDR and CE techniques were explored greatly using OR in Multi-hop WSN com-
pared with the peer routing protocols [1–3]. The fact that wireless mesh- networks have
radio support that is always on, which further allows it to overhear messages with almost

* Rajeev Tiwari
[email protected]; [email protected]
Premkumar Chithaluru
[email protected]
Kamal Kumar
[email protected]
1
Department of Systemics, SCS, UPES, Dehradun, Uttarakhand 248007, India
2
Department of Virtualization, School of Computer Science, University of Petroleum and Energy
Studeis, Dehradun, Uttarakhand 248007, India
3
Computer Science and Engineering, National Institute of Technology, Uttarakhand,
Srinagar (Garhwal), Uttarakhand 246174, India

13
Vol.:(0123456789)
154 P. Chithaluru et al.

no extra charges is put to use [4–6]. This permits multiple receivers to receive a single
transmission where each transmission is specified with particular destination routing
progress [7–10]. The receiver will then forward the packet with the maximum routing
progress it can. In summary, OR’s points of interest is the spatial reuse to acknowledge
two key advantages; it gives higher forwarding scales and limits the effect of dynamic
paths. It results in reasonable enhancement in MSR efficiency relative to traditional OR
protocols. A number of proposals are given for efficiency acheivement in terms of faster
aggregation of data, faster delivery of data through caching as given in [11–13], but
they are not suitable in WSN systems due to their lesser resource capacities. A perfor-
mance comparision and parameters for optimizations can be investigated from proposal
given in [14, 15].
OR protocols for WSNs likewise go for better routing advancement and flexibility to
connect dynamic paths [16, 17], existing peer routing protocols can’t be applied to convert
into opportunistic routing directly. Usually, forwarder nodes are volunteer nodes which are
used to send packets by WSN to assure long network lifetime. The primary improvement in
this research is to incorporate the recognition of volunteer nodes and presents a better OR
protocol tailored to WSNs. By limiting the minimal number of volunteer nodes required for
aggregating packets from source to destination, to essentially improve the network lifetime
and minimize the delay compared with traditional OR protocols in WSNs.
Advance and improved opportunistic routing system based upon Adaptive Rank-
ing for Volunteer Nodes (ARVN) that focuses on energy effectiveness and lower delay
rather than high MSR. ARVN mainly restricts the energy consumption on every node
and by the limiting the active times of sensors during packet traversal from source to
destination. Considering, Expected number of Transmission(ETX) metric based routing,
which limits the maximum number of transmissions needed in relaying a packet from
source to destination, forwarding in ARVN limits the energy consumption in packet
transmission in any sensor network.
The new opportunistic routing is assessed to determine its advantages through imple-
mentation and simulation. plementation and simulation. In this paper, the following
three core contributions are proposed:

I. Adaptive ranking for WSN nodes is presented.

II. Voluntary Node Selection (VNS) algorithm is proposed for scrutinizing volunteer nodes
among multiple forwarders.
III. A new local aspect called spatial node density is incorported to distribute the chances
of becoming forwarders
IV. The routing table is modified to as per new mechanism.

Contributions of this research paper are explained as various sections, Background is dis-
cussed in Sect. 2. Section 3 presents the methodologies and framework of ARVN. Sec-
tion 4 presents the performance of ARVN with its simulation setups, and in Sect. 5 Perfor-
mance Comparison based upon MSR, PDR, and CE. Finally, Sect. 6 concludes this work.

2 Background

The required survey on opportunistic routing in WSNs is discussed in this section.

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 155

2.1 Opportunistic Routing

Opportunistic Routing improves the MSR in comparison to multi-hop and mesh net-
works [18]. While compared to traditional routing protocols, OR takes less time for find-
ing the forwarder in every hop until the packet reaches to CH. In OR [19], every data
packet is received by multiple one-hop sensors, organized by their routing technique. In
this fashion, each sensor in the volunteer set is assigned a scheduled time while sending
packets, which it possibly uses on overhear that it didn’t transmit the packet in question
in an earlier time slot. Due to such routing schemes, OR evades duplicate transmissions
[20]. That’s why, OR is mainly used for wireless applications that are widely spread and
to restrict the number of retransmission paths. This leads to critical MSR improvement
when contrasted with peer routing techniques [21].
Routing schemes in sensor networks comparative objectives: higher routing pro-
gress and flexibility to dynamic links [22]. In any case, OR can’t be actively connected:
WSNs and their applications require few prerequisites, for example, low power sensor
devices and their requirements that restrict those from traditional dense networks [24].
These limits the immediate relevance of OR protocol in the accompanying perspectives:
Consistency and energy efficiency vs MSR: OR is intended to improve the MSR dur-
ing packet transmission [25]. However, WSN routing applications usually request pack-
ets from source to destination with high energy efficiency and not high MSR. How ORs
routing can be intended to enhance the network lifetime in contras to peer routing proto-
cols of WSN [26].
Volunteer nodes in sensor networks: Generally, sensor systems are not to guarantee
for node battery and lifetime of the network. Consequently, all neighbor nodes are in
active states and keep sending an acknowledgment to the volunteer node. It is attributed
to a lack of any such approach which prevents such acknowledgment [27, 28].

2.2 Adapting OR to Wireless Sesnor Networks

Subsequent to presenting the idea of OR and talking about its restrictions with reference
to sensor networks to examine its adjustment to the exceptional prerequisites of WSNs
[29]. The main work targets to identify volunteer participation based on a ranking mech-
anism to transmit packets during routing. In WSNs, the low power sensor nodes, tuning
in to a sender transmits a flow of packets until the planned volunteer awakens and rec-
ognizes it. To incorporate OR into effective routing situations and to advance from this
orthodox unicast-sending plan in one key perspective and the primary source node that
(a) awakens, (b) gets the packet, and (c) gives routing packets to recognize volunteer
based on ranking mechanism.
This work empowers a proficient adjustment of OR protocol to the particular requests
of WSNs as opposed to dense networks, forwarder as volunteer identification focuses
around improving energy efficiency and PDR rather than routing MSR [30–33]. It limits
the radio-on time until a packet is heard/intercepted by a potential volunteer [34–36]. It
coordinates well into wireless routing conditions and guarantees that numerous poten-
tial volunteers can catch a packet in a given wake-up period. Accordingly, the complete
paper focused on spatial and ad-hoc networks to improve the strength of WSNs.

13
156 P. Chithaluru et al.

In this routing plan, the first awoken neighbor that gives a better routing performance
for transmitting packets from source to CH. Subsequently, it leads to improve the MSR,
PDR and decrease the CE as compared to traditional OR protocols.

3 AVRN: Methodologies and Framework

ARIOR exploits Adaptive Ranking based Voluntary Node selection mechanism described
in Sect. 3.3. The important aspects related to ranking exploits residual energy, euclidean
distance and spatial node density as key aspects of underlined WSN. The details of algo-
rithms is defined in sections below.

3.1 Network Model

To mitigate and address the research issue cited above (presented in Sect. 2) , a WSN is
considered to be deployed over spherical or circular area. Network has predefined network
radius of Rmax along with predefined node’s transmission range Nrd . In line with AREOR
[23] , network is clustered into optimal number of clusters based on three spatial node den-
sity, Euclidean Distance and Residual energy where spatial node density is novel aspect
considered in this work. Nodes spreaded in a given radius comes in a cluster. This assures
that network is not partitioned into network segments. ARVN based ARIOR identifies CHs
by utilizing maximum spatial node density, maximum residual energy, as given below. The
simpler computation helps in doing away with computational overhead of cluster forma-
tion. A small recent history about the energy packets received from source node, neigh-
bor nodes along with cluster head positions. This work has considered only static WSN so
nodes attain stable positions and concerns of mobility are done away altogether.

3.2 Ranking Model

The of nodes in ARIOR is done on the basis of important aspects such as residual energy,
Euclidean distance and spatial node density. The equation for computing these parameters
and thereafter rank of the node are listed from Eqs. 1–6. The meaning of various acronyms
and symbols can be obtained from Table 1.
The total number of possible network connections are computed as in Eq. (1). This
gives us space of connections in underlying network.
( )
(N) × (N − 1)
NetConn = (1)
2

The spatial node density is computed in terms of percentage of total connections each
node is involved as a direct node. More neighbors will give higher spatial density and more
fairness in node selections. Lower values of spatial node density marks only and obvious
choices. The values for spatial node density is calculated afresh only when residual node
energy reduces to lower threshold. The spatial node density is computed as below in Eq.
(2).

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 157

Table 1  Acronyms Symbols Meaning

N Count of nodes
Nr Total packets received by node
Nt Total packets transmitted by node
Prp Energy overhead for received packets
Ptp Energy overhead for transmitted packets
Ni Array of energy for sensors network
Ns Source node
Nv Next volunteer
(xi , yi ) Coordinates of source node
(xj , yj ) Coordinates of volunteer node
CH Cluster head
Nre Node’s residual energy
Nme Node’s maximum energy
d Distance between Ns and Nv
SNthr Source node energy threshold
Vh Volunteer node set
Rt Route table
tb Rebroadcast timer
NetConn Number of network connections
Nrd Transmission radus of each node and is
limited to 0.40 ∗ Rmax to 0.60 ∗ Rmax
Nve Highest energy of volunteer node set
Ch Candidate node list
Prs Size of packet
Nst Transmission beigining time of nodes
Nsp Transmission stop time of nodes
n(H) Hop count

NetConn
N [ ]= (2)
sd 𝛽i N[ ]
𝛽i

Further it is required to compute residual energy as below in Eqs. 3 and 4.

Nre [i] = Ni − ((Nr ∗ Prp ) + (Nt ∗ Ptp )). (3)

Nme [i] = max(Nre[i], Nme[i − 1]) (4)

To compute the eucledin distance, following Eq. (5) is designed as below.
√
d(Ns , Nv ) = (xi − xj )2 + (yi − yj )2 (5)

Heuristic fucntion can be computed as given below in Eq. 6.

13
158 P. Chithaluru et al.

𝛽k 𝛽j
j dmin
∑ 1 ∑
NR (𝛽i ) = Nsd[𝛽i ] ∗ Eres[𝛽i ] ∗ 𝛼 ∗ NR (𝛽j )[[ 𝛽𝛽
∗ tmin 𝛽i 𝛽j ] ∗ tmin 𝛽k 𝛽j ] + (1 − 𝛼)
i j
0 dmin 𝛽k ∈Vh
(6)
where 𝛼 is damping factor coefficient.

3.3 ARVN

ARVN an opportunity to identify the volunteer forwarder nodes based on the ranking
mechanism using Eqs. 1–6, define above in Sect. 3.2. This paper, presents one algorithm
each for ranking the nodes and finding the volunteers to aggregate the packets from source
to destination. Also the routing table entries are updated to reflect the most recent residual
energy values besides distance between relay nodes after every transmission cycle.
ARVN algorithm for ranking of volunteer nodes is presented in {Algorithm 1}. Exe-
cution of step 3 to step 13 achieves the distance calculations between current source and
forwarder node, in which an adjacent node with maximum residual energy and maximum
spatial densty is considered as volunteer node and will be selected is shown in Fig. 1.

Fig. 1  Adaptive ranking for volunteer nodes

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 159

AVRN computes/assigns the dyanmic rank of any node by using the node’s residual
energy, spatial node density and the eucledian distance. The generated/assigned rank value
is harnessed for the selection of the volunteer node is discussed ahead in routing table with
examples. For each WSN node, three faspects namely, eucledian distance between nodes, spa-
tial node density Nsd and net residual energy of the WSN node, i.e., Nre gets captured before
any transmission or reciving. Nodes with the minimum rank preffered and assigned as a gate-
way/CH. With identification of CH and subsequent ranking, location of the WSN node is
calculated with respect to the source node. The route table capures node’s energy and their
distances from the source. Spatial node density is a semi-dynamic value and is used in com-
putaion of ranks along with residual energy of nodes and euclidean distance between node and
forwarder node. Based on route table information, the generated/assigned ranks prove useful
in determinting a forward route/path for data packets. Node’s ranks are calculated afresh after
every round of transmission. Spatial node densities are computed afresh at the moment when
net residual energy of node Nre of any node reaches the lower threshold or below lower thresh-
old. Breaching lower energy threshold for a node will affect spatial node density of almost
each surrounding node. The acronym and notations are shown in Table 1.

13
160 P. Chithaluru et al.

Algorithm 2 presents VNS, a method to achieve volunteer node identification. All the
inputs such as SNthr , Vh, where Vh ∈ N, Nve, n(H), Ch, CH, Prs , Nst , Nsp are taken as input
parameters. In line with the main cause of sending routing and data packets from given
source sensor node( Ns ) to the selected Cluster Head (CH) VNS algorithm determines the
most efficient next volunteer Nv. A volunteer set (Vh) is computed/determined from the
candidate list nodes Ch). If SNthr and Message Success Rate (MSR) between source node
and CH is meeting the conditions laid down for the scenario for the transfer of packets,
then packets may be forwarded directly from source node to CH or another sensor node
is selected which succeeds in rank. This selection is applied on Vh. Moreover, the ranks
of each volunteer node are maintained in the routing information table. For each sesor
Nv which belongs to Vh it’s Nve is computed and the given data is broadcasted. A timer is
started after confirming the sender ID, whether node Nv is rank 1 and gets data firstly and
belongs to Vh. In response to received data packet, an acknowledgment is sent to notify the
sender. This also includes n(H). After this, the sent/recieved packet is purged from queue.
Timer is stopped. Sometimes, node Nv expires during or before the time cycle then Nv will
be replaced by an alternate node Nv′ which succeeds Nv in rank value.The expired node is
also removed from Vh. In case, the data packet has not reached to any forward node during
whole duration of retransmission time, then the packet is withdrawn from network and the
location/position of new volunteer is fixed.
It starts with a rebroadcast timer. Step 1 to Step 13 in VNS for optimum energy strategy
{Algorithm 2} selects the volunteer set, Vh, from the candidate node list, Ch. Step 6 checks
the MSR and EC of the Ns to CH is high, then packet transmission is directly between
Ns and CH. otherwise, it picks the next volunteer node from volunteer node set Vh based
on the parameters considered in ARVN {Algorithm 1} along with MSR of the Volunteer
Node, and hop count between the source node and CH. After a volunteer node is selected
for each hop, the source sensor broadcasts the data packet {Step 7}. Transmission of the
data packet through WSN is tracked through steps 9 to 11. Step 9 discusses the acknowl-
edgments of the data packet by each volunteer node. In case the highest ranking Volunteer
Node does not receives data packet {Step 9}, it sends an acknowledgment to the sender,
ARVN checks the next highest volunteer node and sends a data packet to words CH, and
the control goes back to step 9. This repeats until one of the volunteer nodes successfully

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 161

has received the packet. If the timer expires, then the algorithm drops the packet; other-
wise, it sets the positions of the Volunteer Nodes again and tries to send the data packet
again. This phase continues to explore the volunteer nodes until it sends the most optimal
energy sufficient path for data relay. This phase realizes the true volunteer node selection
for optimal energy strategy as shown in Fig. 2.

3.4 Framework of ARVN

The principal goal is to pick up a volunteer node from neighboring nodes of the source
node. The important parameter considered for selection includes a distance between
nodes, spatial node density and residual energy of nodes. Every neighbor of source
node along with the volunteer node of every hop and CH are prioritized/ranked in every

Fig. 2  Volunteer node selection for optimal energy strategy

13
162 P. Chithaluru et al.

Fig. 3  Ranking based volunteer node selection in cluster (round1)

Fig. 4  Ranking based Volunteer node selection in cluster (round2)

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 163

round is shown in Figs. 3 and 4. The node with higher residual energy, maximum spatial
density and minimum distance from the given source sensor node is chosen as the vol-
unteer node.
Route table ARIOR has its own customized routing table which fulfills the require-
ments of routing in ARIOR itself. The routing table is modified from the orthodox OR
mechanisms to suit the improved dynamics adopted in the scrutiny of the next forward-
ing volunteer sensor. A modified routing table maintains node_id (Source node), start
time as braodcast_time , list of ranked nodes for use by VNS as rte_queue and next
hop sensor rte_nexthop and, few more aspects such as sequence number of transmis-
sion cycles, information about route expiry and flags. This helps in encoding the com-
plete route information. An example route route table along with metrices definitions in
Table 2.
Example 1: Table 3, clarifies that ID-1 is a source, Node ID- 3, Node ID-7 are the
highest rank nodes in every hop, and Node ID-12 is CH due to its residual energy, spa-
tial node density and eucledian distance from the source node.
If CH is far from source node and there are chances of packet lose; the volunteers
selected through {Algorithm 1} take opportunity to aggregate the packets and transmits
towards CH within transmission time.
Example 2: It is clear that Node ID-1 is a source, and Node ID- 4, Node ID-6 are
the volunteers, and Node ID-11 is the CH is shown in Table 4. Moreover, based on the
residual energy, spatial node density and euclidean distance from a source node highest
value node is deignated as CH, and CH takes the support from volunteers when CH is
far from source to increase the gain on MSR, PDR.

Table 2  Description of route Parameter Description

table metrics
node_id ID of SensorID
broadcast_time Broadcast begin time
rte_queue Ranked node’s Queuel
rte_nexthop Potential forwarder
rte_cycle_no SeqNo of time cycle
rte_expire Route expiry information
rte_flags Flag is 1 for the success-
ful packets transmis-
sion

Table 3  Route table parameters with reference to Fig. 3 (hop wise)

node_id broadcast_time rte_queue rte_nexthop rte_cycle_no rte_expire rte_flags

1 18.14857 3,7,12 7 1 21.029160 1

Table 4  Route table parameters in reference with Fig. 4 (hop wise)

node_id broadcast_time rte_queue rte_nexthop rte_cycle_no rte_expire rte_flags

1 18.148457 4,6,11 6 2 21.029160 1

13
164 P. Chithaluru et al.

4 ARVN Performance

Considering the ever expanding application of WSNs and IoT, and the increasing significance
of energy concious opportunistic routing in the present wireless communications, this paper
presents a novel opportunistic routing method, to be specific a new ranking algorithm Adap-
tive Ranking for Volunteer Nodes (ARVN) and a Volunteer Node Selection (VNS) algorithm.
This work uses dynamic ranking among the nodes and utilizing it to choose the volunteer node
in OR protocol. Dynamic node ranking depends on three essential variables/factors namely,
spatial node density, position and residual energy. To ease the process, one table for maintain-
ing the rank information and a set of candidate nodes is chosen to decide a volunteer node.
The choice of the volunteer node additionally relies upon a hop count to reach the BS, MSR of
a node, PDR and the residual energy. In this manner, Opportunistic Routing protocol expands
the proficiency of the system and improves energy utilization since it goes for using residual
energy for selection of nodes.

4.1 Volunteer Set Properties of ARVN

Wireless Sensor Networks fits in omni dimensional wireless network model. This work has
considered a multi-hop wireless sensor networks. Network is cosnidered as a collection of
clusters after clustering as per Sect. 3, i.e. CH is created. Nodes maintains prior energy packet
information of neighboring nodes and coordintes of the source node and CH according their
ranks from routing table. Moreover, Nrd = 0.6 ∗ Rmax and d = 0.4 ∗ Rmax are observed as
upper and lower broadcast ranges foor sensors in this work. The network can be considered as
graph G = (N, Vh ), where N is a set of wireless nodes which can relay and Vh is volunteer node
set. The sorted list as per rank is considered as {0, 1, 2, N − 1, N}, and two particular sensors
with the index 0 along with j are source sensor and Cluster Head respectively.

4.2 Simulation Setup

Many proposals in WSNs have implemented OR protocols as simulations in NS2. Besides

many available network tools, here NS2 is used as a simulation tool, as it has most of the
prominent opportunistic routing protocols support which are used for comparision purpose
here. Simulations are done on different times as 25 s, 50 s, 75 s, 100 s, 125 s, 150 s, and 175
s for incorporating stability of simulator. The communication is variable bit rate. The experi-
mentation is carried out using the initial values as mentioned in Table 5.

4.3 Evaluation Criteria

Opportunistic Routing schemes are an effort for improved efficiency in terms of network oper-
tions and energy efficiency. This work has considered Message Success Rate, PDR and Con-
sumed Energy as parameters of interest. Parameters are defines as below:

4.3.1 Message Success Rate‑(MSR)

MSR can be defined as transmission of message success from start to end node(gateway).
It can be measured as total number of data packets received divided by time consumed.
Therefore, MSR can be computed as per Eq. (7).

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 165

Table 5  Parameters used and Parameter name Value

their values
Simulator NS2
Base routing protocol Opportunistic routing protocol(ORP)
Network type Wireless
Simulation time 25 s, 50 s, 75 s, 100 s, 125 s, 150 s, 175 s
Types of tAnteenas Omni antenna
Model of simulation Energy model
Node’s initial energy 20 J
Node count 10–100
Length of queue 75 bits
Data rate 4000 bits
Type of interface Wireless physical interface
MAC type IEEE-802.11
Rx power 0.0831 J/bit
Tx power 0.0462 J/bit
Idle power 0.05 J/bit
Sense power 0.01750 J/bit
Sleep power 0.00048 J/bit

N
∑ Nr
MSR = (7)
i
(Nsp − Nst )

4.4 Packet Delivery ratio(PDR)

PDR is one of the critical parameters and can be defined as success ratio of communication in
the given setting. The PDR ratio can be computed as shown below in Eq. (8).
N ( )
∑ Nt
PDR = ∗ 100 (8)
i
Nr

4.4.1 Consumed Energy (CE)

CE is measured as consumption of energy of every WSn node it can be computed as per Eq.
(9).
N
∑
CE = (Ni − Nre ) (9)
i

While, average energy consumption (AEC) for each node can be computed as given in Eq.
(10).
CE
AEC = (10)
N

13
166 P. Chithaluru et al.

5 Performance Comparison and Discussion

Performance of ARVN is evaluated against the performance of all the OR protocols

such as ExOR, SAOR, EEOR, ENS_OR, ASSORT,EAOR, EFFORT. The performance
is evluated in varying scenrios and different range of values. The parameters of perfor-
mance were identified in Sect. 4.3 which include PDR, CE, and MSR. Performance met-
ric parameters are observed with respect to time (s), packet size, number of nodes, AEC
(J), PDR (%), MSR (bits/s). The results are dicussed in below sections.

5.1 Effect of MSR

It is evident from Fig. 5 that for existing ORs such as ExOR, EEOR, EAOR, SAOR and
the proposed ARVN protocols, MSR have significantly improved to wards stabel and
gurantees transmission success with the increase of time cycles. The performance com-
parison of the proposed w.r.t. existing protocols as shown in the graph in Fig. 5.
The proposed algorithm shows better MSR with a value of 220 bit/s to 242 bit/s at
the timings of 25s to 175s consequently is shown in Table 6. The MSR is high because
the network model is with well-defined topology and specified node selection based on
distance and residual energy. It shows significant gain in MSR with respect to ExOR
is 0.31990453, EEOR is 0.329436128, EAOR is 0.326368924, SAOR is 0.305097378,
EFFORT is 0.295759738, ENS_OR is 0.280813379, ASSORT is 0.27029344 is shown
in Table 7.
In the proposed protocol, the network performance is high because as per network
model WSN topology is divided into clusters/ group. Each cluster has fixed territory
and highest energy node is designated as CH, due to this average gain increased from
0.18% to 0.43%.
it can beconcluded that with the increasing of growth rate of time-cycle limits the
proposed routing offers more stability. Also, stable values of MSR in comparison with
other peer protocols. Overall, the gain w.r.t. ARVN has shown a consideraable average
weighted gain 0.2% from almost all the selected protocols shown in Fig. 6.

Table 6  MSR based performance of routing protocols

MSR
Time (s) MSR(bits/s)
ExOR EEOR EAOR SAOR EFFORT ENS_OR ASSORT ARVN

25 125 131 137 141 147 149 151 220

50 137 139 141 147 149 151 153 225
75 155 157 146 159 161 164 165 231
100 125 157 157 160 163 167 170 234
125 167 169 171 173 167 175 179 239
150 177 177 179 181 184 186 190 241

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 167

Fig. 5  Performance variation of ARVN and other protocols over MSR

5.2 Effect on PDR

PDR of ARVN based ARIOR is computed w.r.t. peer schemes, in terms of percentage of
total data packets that are received at aggrgating point node. The PDR values were also
evaluted in the light of variable time-cycle. The proposed algorithm shows better PDR ini-
tially with a value of 97.01% to 99.78% at the timings of 25 to 175 seconds consequently is
shown in Table 8.
In the following Fig. 7, the PDR performance comaprison of ARVN is plotted with
respect to peer protocols. However, proposed protocol has shown significant improvement
over peer techniques.
ARVN has affetced WSN in such a way that nodes in the network are functionally rley
upon queue nodes. In case, there is some problem affects the volunteer, the next availbe
node in the sorted sorted queue is available and selected, and its effects on the PDR is that
PDR improves during time cycles. It shows significant gain in PDR with respect to ExOR
is 0.49054806, EEOR is 0.435975154, EAOR is 0.426597032, SAOR is 0.34205143,
EFFORT is 0.312830436, ENS_OR is 0.304109471, ASSORT is 0.265926088 is shown
in Table 9.
In addition, effectiveness of ARVN is defined in terms of the gain obtained and the
weighted gain obtained by ARVN technique is 0.22% w.r.t. various protocols as shown in
the following Fig. 8.

5.3 Effect of CE

Due to selection of Ch and sorted queue nodes for communication in WSN the loss of
energy on every node is lesser as compatred to peer protocols. Energy intake obtained from
suggested ARVN protocol is compared with peer protocols by varying node counts, then
ARVN shows better CE initially with a value of 84.45 J/bit to 70.89 J/bit with the number
of nodes increased from 10 to 100 consequently be shown in Table 10 and Fig. 9.
In addition, with the increase in the number of nodes, the amount of CE also reduced. It
shows significant gain in CE with respect to ExOR is 0.295599728, EEOR is 0.256555318,

13
 168

13
Table 7  Gain of other routing protocols Vs ARVN based on MSR
Gain on message success rate
Time (s) ExOR vs ARVN EEOR vs ARVN EAOR vs ARVN SAOR vs ARVN EFFORT vs ARVN ENS_OR vs ARVN ASSORT vs ARVN

25 0.4318182 0.404545 0.377273 0.359091 0.331818 0.322727 0.313636

50 0.3911111 0.382222 0.373333 0.346667 0.337778 0.328889 0.32
75 0.3290043 0.320346 0.367965 0.311688 0.30303 0.290043 0.285714
100 0.465812 0.32906 0.32906 0.316239 0.303419 0.286325 0.273504
125 0.3012552 0.292887 0.284519 0.276151 0.301255 0.267782 0.251046
150 0.2655602 0.26556 0.257261 0.248963 0.236515 0.228216 0.211618
175 0.2561983 0.243802 0.235537 0.22314 0.210744 0.194215 0.181818
Weighted gain 0.3199045 0.329436 0.326369 0.305097 0.29576 0.280813 0.270293
P. Chithaluru et al.
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 169

Fig. 6  Weighted gain variation of ARVN and other protocols over MSR

Table 8  Performance of routing protocols based on PDR

Packet delivery ratio
Time (s) Packet delivery ratio(%)
ExOR EEOR EAOR SAOR EFFORT ENS_OR ASSORT ARVN

25 57.76 54.51 60.4 65.03 66.68 67.01 68.19 97.01

50 55.5 56.02 55.3 66.94 67.26 68.15 69.96 98.15
75 52.7 53.31 59.32 68.65 68.36 68.27 70.13 98.27
100 49.87 55.77 53.53 69.13 68.65 68.89 72.48 98.89
125 47.73 57 57.49 62.56 68.65 68.99 74.05 98.99
150 44.97 56.56 54.91 61.26 68.14 69.45 75.18 99.45

Fig. 7  PDR evalutions of ARVN with respect to peers

13
 170

13
Table 9  Gain of other routing protocols vs ARVN based on PDR
Gain on packet delivery ratio
Time (s) ExOR vs ARVN EEOR vs ARVN EAOR vs ARVN SAOR vs ARVN EFFORT vs ARVN ENS_OR vs ARVN ASSORT vs ARVN

25 0.404597 0.438099 0.377384 0.329657 0.312648 0.309246 0.297083

50 0.434539 0.429241 0.436577 0.317983 0.314722 0.305655 0.287213
75 0.463722 0.457515 0.396357 0.301414 0.304366 0.305281 0.286354
100 0.495702 0.43604 0.458691 0.30094 0.305794 0.303367 0.267064
125 0.51783 0.424184 0.419234 0.368017 0.306496 0.303061 0.251945
150 0.547813 0.431272 0.447863 0.384012 0.314832 0.301659 0.244042
175 0.458809 0.437563 0.451894 0.387352 0.327521 0.300661 0.226398
Weighted gain 0.490548 0.435975 0.426597 0.342051 0.31283 0.304109 0.265926
P. Chithaluru et al.
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 171

Fig. 8  Weighted gain variation of ARVN and other protocols over PDR

EAOR is 0.239779087, SAOR is 0.222817796, EFFORT is 0.359550782, ENS_OR is

0.38399907, ASSORT is 0.380413792 is shown in Table 11.
In addition, the performance of proposed protocol ARVN is defined in terms of the gain
obtained and the weighted gain obtained by ARVN is 0.20% w.r.t. various protocols is
shown in Fig. 10.

6 Conclusion and Future Scope

The work in this paper presents an improved routing technique based on ARVN. ARVN is
an adaptive ranking mechanism based on residual energy, distance and spatial node density.
The new OR namely, ARIOR is result of combining the two techniques ARVN and VNS.
The prinicple takeaway from this work is to discover volunteer nodes in WSN. This paper
framed up the key properties of ARVN and harnessed ARVN as a new energy efficient
routing mechanism. The performance evluation in Sect. 5 has shown that ARVN based
Opportunistic Routing tends to achieve distributed computational overhead, and prompts
a loop-free network topology. ARVN algorithm converges quickly. The performance

Table 10  Performance of routing protocols based on CE

Energy consumption
No. of nodes Average energy consumption (Joules)
ExOR EEOR EAOR SAOR EFFORT ENS_OR ASSORT ARVN

10 68.21 68.12 69.32 69.52 65.45 54.45 54.16 84.45

15 67.23 67.25 68.23 68.21 54.65 53.62 54 83.62
20 63.16 64.25 64.56 65.23 53.89 52.19 52.98 82.19
40 58.42 59.21 60.25 61.45 50.59 49.32 51.19 79.32
50 55.05 56.33 57.59 59.06 48.05 47.69 49 77.69
75 51.17 52.79 54.3 56.28 45.49 44.89 43.1 74.89

13
172 P. Chithaluru et al.

Fig. 9  Performance evaluations in terms of CE

comparison of ARVN is carried out using detailed analytical models built up that ARVN is
an exact estimate of discovering volunteers to transmit the packet towards the destination.
At last, both simulations and initial deployments that ARVN yields altogether improved in
MSR, PDR, and CE. This work may extend to heterogeneous constrained nodes and assess
for validity in heterogeneous WSNs.

13
 Table 11  Gain of peers vs ARVN wrt CE
Gain on energy consumption
No. of nodes ExOR vs ARVN EEOR vs ARVN EAOR vs ARVN SAOR vs ARVN EFFORT vs ARVN ENS_OR vs ARVN ASSORT vs ARVN

10 0.192303 0.193369 0.17916 0.176791 0.224985 0.35524 0.358674

15 0.196006 0.195767 0.18405 0.184286 0.346448 0.358766 0.354221
20 0.231537 0.218275 0.2145 0.206351 0.344324 0.365008 0.355396
40 0.26349 0.25353 0.24042 0.22529 0.362204 0.378215 0.354639
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing…

50 0.291415 0.274939 0.25872 0.239799 0.381516 0.38615 0.369288

75 0.316731 0.295099 0.27494 0.248498 0.392576 0.400588 0.424489
100 0.332769 0.305121 0.28043 0.247143 0.421357 0.423191 0.433912
Weighted gain 0.2956 0.256555 0.23978 0.222818 0.359551 0.383999 0.380414
173

13
174 P. Chithaluru et al.

Fig. 10  Weighted gain variation of ARVN and other protocols over CE

References
1. Biswas, S., & Morris, R. (2005). ExOR: Opportunistic multi-hop routing for wireless networks. ACM
SIGCOMM Computer Communication Review, 35(4), 133–44.
2. Chachulski, S., Jennings, M., Katti, S., & Katabi, D. (2007). Trading structure for randomness in wire-
less opportunistic routing. ACM, 4(5), 67–82.
3. Larsson, P. (2001). Selection diversity forwarding in a multihop packet radio network with fading
channel and capture. ACM SIGMOBILE Mobile Computing and Communications Review, 5(4), 47–54.
4. Choudhury, R. R., & Vaidya, N. H. (2004). MAC-layer anycasting in ad hoc networks. ACM SIG-
COMM Computer Communication Review, 34(1), 75–80.
5. Pimentel, L., Rosário, D., Seruffo, M., Zhao, Z., & Braun, T. (2015). Adaptive beaconless opportunis-
tic routing for multimedia distribution. In Proceedings of International Conference on Wired/Wireless
Internet Communication (pp. 122–135).
6. Coutinho, R. W., Boukerche, A., Vieira, L. F., & Loureiro, A. A. (2014). GEDAR: Geographic and
opportunistic routing protocol with depth adjustment for mobile underwater sensor networks. In Pro-
ceedings of IEEE International Conference on Communications (ICC) (pp. 251–256).
7. Wei, C., Zhi, C., Fan, P., & Letaief, K. B. (2009). AsOR: An energy efficient multi-hop opportunistic
routing protocol for wireless sensor networks over Rayleigh fading channels. IEEE Transactions on
Wireless Communications, 8(5), 2452–2463.
8. Amdouni, I., Adjih, C., Aitsaadi, N., & Muhlethaler, P. (2016) Experiments with ODYSSE: Opportun-
istic Duty cYcle based routing for wirelesS Sensor nEtworks. In Proceedings of IEEE 41st Conference
on Local Computer Networks (LCN) (pp. 232–235).
9. Hai, L., Wang, H., Wang, J., & Tang, Z. (2014). HCOR: A high-throughput coding-aware opportun-
istic routing for inter-flow network coding in wireless mesh networks. EURASIP Journal on Wireless
Communications and Networking, 20(14), 148–161.
10. De Couto, D. S., Aguayo, D., Bicket, J., & Morris, R. (2005). A high-throughput path metric for multi-
hop wireless routing. Wireless Networks, 11(4), 419–34.
11. Tiwari, R., & Kumar, N. (2015). Cooperative gateway cache invalidation scheme for internet-based
vehicular ad hoc networks. Wireless Personal Communications, 85, 1789–1814.
12. Tiwari, R., & Kumar, N. (2016). An adaptive cache invalidation technique for wireless environments.
Telecommunication Systems, 62, 149–165.
13. Tiwari, R., & Kumar, N. (2015). Minimizing query delay using cooperation in ivanet. Procedia Com-
puter Science, 57, 84–90.
14. Chithaluru, Premkumar, Tiwari, Rajeev, & Kumar, Kamal. (2019). Performance analysis of energy
efficient opportunistic routing protocols in wireless sensor network. International Journal of Sensors,
Wireless Communications and Control, 9, 1. https​://doi.org/10.2174/22103​27909​66619​10260​92311​.
15. Tiwari, R. (2010). Gulista khan, “Load Balancing through distributed Web Caching with clusters”. In
Proceeding of the CSNA, 46–54.

13
ARIOR: Adaptive Ranking Based Improved Opportunistic Routing… 175

16. Krishan, T., Alkhawaldeh, R. S., Khawaldeh, S., Al-Ahmad, B., & Al, Smadi A. (2010). Hn-PERP:
Hop by hop-power-efficient routing protocol over underwater wireless sensor networks. ACM, 9(8),
78–98.
17. Eu, Z. A., & Tan, H. P. (2012). Adaptive opportunistic routing protocol for energy harvesting wire-
less sensor networks. In Proceedings of IEEE international conference on communications (ICC) (pp.
318–322).
18. Lukošius, A. (2007). Opportunistic routing in multi-sink mobile ad hoc wireless sensor networks. Uni-
versity of Bremen publishers, pp. 1–296.
19. Mounika, M., & Chinnaswamy, C. N. (2016). Opportunistic routing protocols for wireless sensor
networks: A survey. International Journal of Computer Science and Information Technologies, 7(2),
928–931.
20. Menon, V. G., & Prathap, P. J. (2016). Comparative analysis of opportunistic routing protocols for
underwater acoustic sensor networks. In Proceedings of International Conference on Emerging Tech-
nological Trends (ICETT) (pp. 1–5).
21. Sharma, M., Singh, Y., & Kumar, N. (2014). Opportunistic routing in wireless sensor networks: A
comparative analysis. Journal of Basic Applied Engineering Research (pp. 94–98).
22. Al-Mahdi, H., & Fouad, Y. (2019). Design and analysis of routing protocol for cognitive radio ad hoc
networks in heterogeneous environment. International Journal of Electrical and Computer Engineer-
ing (IJECE), 9(1), 341–351.
23. Chithaluru, Premkumar, Tiwari, Rajeev, & Kumar, Kamal. (2019). AREOR-Adaptive ranking based
energy efficient opportunistic routing scheme in Wireless Sensor Network. Computer Networks, 162,
106863.
24. Shah, R. C., Wietholter, S., Wolisz, A., & Rabaey, J. M. (2005). When does opportunistic routing make
sense? In Proceedings of IEEE international conference on pervasive computing and communications
workshops (pp. 350–356).
25. Ge, Y., Wang, S., & Ma, J. (2018). Optimization on TEEN routing protocol in cognitive wireless sen-
sor network. EURASIP Journal on Wireless Communications and Networking (pp. 1–27).
26. Sanchez-Iborra, R., & Cano, M. D. (2016). JOKER: A novel opportunistic routing protocol. IEEE
Journal on Selected Areas in Communications, 34(5), 1690–1703.
27. Islam, J., & Singh, P. K., (2010). CORMEN: Coding-aware opportunistic routing in wireless mess net-
work, arXiv preprint, pp. 1006–1045.
28. Ghadimi, E., Landsiedel, O., Soldati, P., & Johansson, M. (2012). A metric for opportunistic routing
in duty cycled wireless sensor networks. In Proceedings of 9th Annual IEEE Communications Society
Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON) (pp. 335–343).
29. Zeng, K., Yang, Z., & Lou, W. (2010). Opportunistic routing in multi-radio multi-channel multi-hop
wireless networks. IEEE Transactions on Wireless Communications, 9(11), 3512–21.
30. Yan, Y., Zhang, B., Zheng, J., & Ma, J. (2010). Core: a coding-aware opportunistic routing mechanism
for wireless mesh networks. IEEE Wireless Communications, 17(3), 96–103.
31. Poonkuzhali, R., Sanavullah, M. Y., & Gurupriya, M. R. (2014). A comparative study of novel oppor-
tunistic routing protocols in mobile ad hoc networks. World Academy of Science, Engineering and
Technology, International Journal of Computer, Electrical, Automation, Control and Information
Engineering, 8(8), 1529–1532.
32. Mao, X., Tang, S., Xu, X., Li, X. Y., & Ma, H. (2011). iLight: Indoor device-free passive tracking
using wireless sensor networks. In Proceedings of IEEE INFOCOM (pp. 281–285).
33. Spachos, P., Chatzimisios, P., & Hatzinakos, D. (2012). Energy aware opportunistic routing in wireless
sensor networks. In Proceedings of IEEE Globecom Workshops (pp. 405–409).
34. Hung, C. C., Lin, K. C., Hsu, C. C., Chou, C. F., & Tu, C. J. (2010). On enhancing network-lifetime
using opportunistic routing in wireless sensor networks. In Proceedings of 19th International Confer-
ence on Computer Communications and Networks (pp. 1–6).
35. Luo, J., Hu, J., Wu, D., & Li, R. (2014). Opportunistic routing algorithm for relay node selection in
wireless sensor networks. IEEE Transactions on Industrial Informatics, 11(1), 112–121.
36. Hsu, C. C., Kuo, M. S., Wang, S. C., & Chou, C. F. (2012). Joint design of asynchronous sleep-wake
scheduling and opportunistic routing in wireless sensor networks. IEEE Transactions on Computers,
63(7), 1840–1846.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

13
176 P. Chithaluru et al.

Premkumar Chithaluru is a Research Scholar in Dept. of SoCS at Uni-

versity of Petroleum and Energy Studies (UPES), Dehradun. He com-
pleted his B. Tech from (JNTUH) in CSE and M. Tech from (JNTUH)
in CSE. His area of research are WSN, QoS in Wireless Networks,
Mobile Ad-hoc Networks, and Routing Techniques.

Dr. Rajeev Tiwari is working as an Associate Professor in SCS in

UPES, Dehradun (India). He is a senior IEEE Member. He has done
Ph.D. in CSE from Thapar Univeristy, Patiala (Punjab). He has more
than 13 years of research and teaching experience. His broad area of
research are Cloud Computing, MANET, VANET, QoS in wireless
networks, Cache Invalidation Techniques, Internet of Things (IoT), Big
Data Analytics and Machine Learning. He is also an expert in simula-
tion and design of scenarios on NS2, SUMO and MOVE. He is an
active member of IEEE,ACM,IEI, IACSIT and IAENG. He has total
30+ international publications. His publications are majorly indexed in
SCI, Scopus and DBLP. He has chaired many sessions in international
conferences. He is a lead reviewer in SCI indexed journal of Springer,
IEEE and ScienceDirect and international conferences. He has pub-
lished 8 book chapters also with international publishers. He has con-
tributions in funded projects in field of IoT, Cloud and wireless net-
work QoS.

Kamal Kumar is working as Assistant Professor in National Institute of

Technology, Uttarakhand, India. He has teaching experience of 18
years. He has obtained his Bachelor and Masters from Kurukshetra
University, India. He received his PhD from Thapar University, India.
He has 35 Publication in Journals and International Conferences. He
has served as Technical Program Chair in NGCT’2017, Program Com-
mittee members in multiple international conferences. He chaired an
International Conference, NGCT’2018. His research interest lies in
Wireless Sensor Networks, Security Provisioning, Cloud Computing
and Artificial Intelligence.

13