A Competent Ad-hoc Sensor Routing Protocol for Energy Efficiency in Mobile Wireless Sensor Networks

Abstract

In wireless sensor networks (WSNs), energy utilization plays an essential part in the survival time (survivability) of the network. All the network aspects are redesigned to work in the WSN for lesser energy consumption. In this regard, routing protocol and dynamic topology are very crucial aspects to be noted. In this work, we present a competent ad-hoc sensor routing (CASeR) protocol for delay reduction, reliable data communication, and efficient energy usage in mobile WSNs (MWSNs). It is aimed to cope up with challenging requirements of the rising limited battery powered technologies, which requires low energy consumption and end-to-end delay (EED). Further, the CASeR uses reservation based channel allocation using reservation time division multiple access mobility adaptive cross-layer in dynamic networks and cost based multi-hop communication for packet forwarding and gradient maintenance. The MWSN routing protocols, robust ad-hoc sensor routing, mobility adaptive cross-layer routing, and proactive highly ambulatory sensor routing as well as the mobile ad-hoc network protocols, optimized link state routing and ad-hoc on demand distance vector, measure up with the CASeR. The simulation results show improvements over other routing protocols in energy consumption, EED, queuing delay, and reliable data communication. The competence of this protocol makes it highly suitable to minimize time delay in target applications.

Full text

Vol.:(0123456789)
Wireless Personal Communications
https://doi.org/10.1007/s11277-020-07741-0
1 3
A Competent Ad‑hoc Sensor Routing Protocol forEnergy
Eciency inMobile Wireless Sensor Networks
K.ManikandaKumaran1· M.Chinnadurai2
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
In wireless sensor networks (WSNs), energy utilization plays an essential part in the sur-
vival time (survivability) of the network. All the network aspects are redesigned to work
in the WSN for lesser energy consumption. In this regard, routing protocol and dynamic
topology are very crucial aspects to be noted. In this work, we present a competent ad-
hoc sensor routing (CASeR) protocol for delay reduction, reliable data communication,
and efficient energy usage in mobile WSNs (MWSNs). It is aimed to cope up with chal-
lenging requirements of the rising limited battery powered technologies, which requires
low energy consumption and end-to-end delay (EED). Further, the CASeR uses reservation
based channel allocation using reservation time division multiple access mobility adap-
tive cross-layer in dynamic networks and cost based multi-hop communication for packet
forwarding and gradient maintenance. The MWSN routing protocols, robust ad-hoc sen-
sor routing, mobility adaptive cross-layer routing, and proactive highly ambulatory sensor
routing as well as the mobile ad-hoc network protocols, optimized link state routing and
ad-hoc on demand distance vector, measure up with the CASeR. The simulation results
show improvements over other routing protocols in energy consumption, EED, queuing
delay, and reliable data communication. The competence of this protocol makes it highly
suitable to minimize time delay in target applications.
Keywords Routing· Wireless sensor network· Energy efficiency· Dynamic topology·
Ad-hoc network
* K. Manikanda Kumaran
kmkkumaran@egspec.org
M. Chinnadurai
mchinna81@gmail.com
1 Department ofInformation Technology, E.G.S. Pillay Engineering College, Nagapattinam,
TamilNadu, India
2 Department ofComputer Science andEngineering, E.G.S. Pillay Engineering College,
Nagapattinam, TamilNadu, India

K.M.Kumaran, M.Chinnadurai
1 3
1 Introduction
Wireless sensor networks (WSNs) comprise of a huge number of limited capacity micro
electro mechanical systems equipped for estimating physical phenomena and reporting to
the processing node called a sink [1]. The limited ability actually is in terms of processing
capability, power consumption, etc. The deployment area of the WSN makes it so complex
in many aspects such as routing, energy consumption, reliability, and delay. Mostly the
WSNs are placed in remote areas and are intended for surveillance [2] and environmental
monitoring [3] to negate the presence of humans. In remote areas, there will not be any
uninterrupted power supply for distribution. Due to such constraints, sensor networks are
aimed to work with limited power supply. So conservation of energy in the WSN is criti-
cal because the lifetime of the network depends on the limited power supply and the same
is addressed in significant research studies [4]. The responsibility of the sensor node is not
only limited to sensing, transmitting, and receiving, but also to act as the intermediate node
which relays packets during multi-hop transmission. Later, the evolution of the WSN led to
the development of mobile WSNs [MWSNs] [5]. Moreover, most of the components of the
MWSNs are mobile; i.e., they can move their position constantly at different speeds. Along
with limited energy constraints, the key challenges in mobility supported networks are to
maintain the network connectivity alive always and also to ensure reliability in routing
with least delay [6]. The target application of this protocol is in the areas of surveillance
and rescue [7]. This may be carried out by a swarm of unmanned aerial vehicles (UAVs)
equipped with infrared intruder detection sensors, which could indicate the location of the
intruders or victims to the sink. For such type of applications, the protocol requires a least
end-to-end delay (EED), a high packet delivery ratio (PDR), and less energy dissipation to
sustain the lifetime of the network.
In this study, we have developed a competent ad-hoc sensor routing (CASeR) protocol
for reliable communication along with energy efficiency and delay reduction. In a highly
changing topology, the CASeR uses hop count and cost-based metric for network mainte-
nance and data forwarding. To do so, we integrate reservation time division multiple access
(RTDMA) medium access control along with the routing protocol. Using the RTDMA, we
aim to achieve a least EED and high PDA. The RTDMA first allows sequential transmis-
sion so that just a single node can transmit at once. This will mitigate the occurrence of
collision during data transmission. The primary scheme of the RTDMA is a reservation
mechanism, which allows the nodes in the network to reserve medium access for future
transmission. This scheme will be more helpful to eliminate unnecessary queuing delay in
the TDMA scheme, which aids in EED reduction.
In Sect.2, we discuss a part of the current literature related to routing for MWSNs so
as to describe our work. In Sect.3, we present the description of the proposed protocol. In
Sect.4, we derive the proposed network modeling and mathematical analyses on various
metric. In Sect.5, we give points of interest of the simulation result, and in Sect.6 we sum-
marize our conclusions.

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
2 Literature Survey
In general, MWSNs share some common characteristics with the WSNs and mobile ad-
hoc networks (MANETs) but the WSNs require one way data communication from sensor
nodes to a sink. However, the MANETs require two way data communication which charge
the additional overhead routing process. On the contrary, MWSNs require one-way com-
munication and along with this it needs to support the mobile ad-hoc nature of nodes as
like the MANET. So, in order to handle the demanding applications more MWSN routing
protocols have emerged as an extended version of the MANET routing protocols.
The network structure of the MWSN is broadly classified into two blocks, namely,
hierarchical and flat. The roles of the network nodes are defined in hierarchical protocols,
whereas identical tasks are performed by every node in the flat network. Protocols used
in the MANETs are mainly classified as proactive and reactive. Because of the dynamic
nature of the MWSNS, enhanced versions of the MANET’s reactive routing protocols such
as ad-hoc on-demand distance vector (AODV) [8], dynamic source routing (DSR) [9], and
extremely opportunistic routing (OR) [10] are more suitable options for routing.
Some of the MANET protocols which are adapted to use in the MWSNs are AODV ++
[11] and AODV with pre-emptive self-repair [12], OR-RSSI [13], geographically oppor-
tunistic routing [14], and angle-based DSR [15]. Routing protocols used in hierarchical
WSNs are the low energy adaptive clustering hierarchy (LEACH) [16], LEACH-mobile
(LEACH-M) [17], LEACH-M enhanced [18], cluster based routing for MWSNs [19], and
mobility-based clustering [20].
The recently developed protocol, the mobility adaptive cross-layer routing (MACRO)
[21], a routing protocol in view of a new cross-layer interaction approach among the five
reference layers (i.e., application, transport, network, MAC, and physical layers) is dis-
cussed in detail. It essentially misuses the possibility of interaction among the previously
mentioned five layers across the board protocol. In the routing protocols, particularly
intended for mobility supported networks, finding the accessible route as well as guaran-
teeing the route with unwavering quality and experiencing topology changes is of extraor-
dinary significance. The MACRO tries to support the reliable quality of a route by adjust-
ing the conceivable topology changes and channel conditions, for example, the failure of
nodes and genuine congestion. Despite the fact that its essential objective is to build an
end-to-end route quality, it additionally tries to locate the best routes that will encounter
negligible EED and power utilization.
The proactive highly ambulatory sensor routing (PHASeR) protocol [22] is an MWSN
that utilizes a basic hop-count metric to empower the dynamic and vital routing of infor-
mation towards the sink in portable situations. It is roused by the utilization of radiation
mapping by UAVs, which require the solid and opportune conveyance of normal estima-
tion to the sink. Further, the PHASeR keeps up a gradient metric in portable situations by
utilizing a global TDMA MAC layer. It likewise utilizes the strategy of blind forwarding to
go messages through the network in a multipath way.
The robust ad-hoc sensor routing (RASeR) [23] protocol employs a blind forwarding
technique, which serves as a novel method of gradient maintenance. In a changing topol-
ogy like the MWSN, gradient maintenance is done by using a global time division multi-
ple access (GTDMA) MAC. The GTDMA allocates a fixed time slot order by which each
node is assigned to single time slot. A single time slot is large enough to transmit a single
packet which contains the generated data. The order in which the node’s time slots occur
is fixed and loops cyclically, which means generated packets are in a queue till a respective

K.M.Kumaran, M.Chinnadurai
1 3
time slot. It leads to a queuing delay in data forwarding on every forwarding node until
it reaches the sink. The blind forwarding technique used in the RASeR incurs unneces-
sary energy dissipation and give additional overhead to the nodes in data travelling path.
In Sect.3, we address all these limitations and propose a new protocol called the CASeR.
The MWSN protocols discussed in the literature generally give better vision regarding
the PDR and EED performances. Our proposed protocol, i.e., CASeR also considers the
efficient energy utilization, queuing delay as another important performance metric. The
reservation based TDMA MAC and cost based data forwarding technique employed to
ensure minimal energy consumption and computational time are the key contributions of
this study.
3 Protocols Used
The envisioned application of the proposed work is in the MWSNs in which nodes support
a high frequency of mobility. The nodes may be land robots, UAVs, etc. engaged for some
rescue or surveillance applications which will be controlled by a central processing node
called a sink. The primary responsibility of the node is to send the location and time infor-
mation of the victim or intruder to the sink. This information should reach the sink for on-
time rescue of victims or intruder detection and also the energy consumption should be low
to improve the lifetime of the network. The protocol used in this type of network should
accept the inclusion of additional nodes to improve the efficiency on rescuing or searching.
The CASeR, the proposed routing protocol, tries to improve the competence in the PDR
with least delay and less energy consumption. The sink can communicate with the nodes
through the beacon signals, which are time-synchronizing beacon, reservation beacon, and
channel allocation beacon. Based on these considerations, we worked on a competent rout-
ing protocol to obtain best results in the MWSN.
3.1 Reservation TDMA
Initially, the CASeR arranges all the nodes in a predefined order using their corresponding
node IDs. Then, it allows each node to reserve future slots on the next cycle through a bea-
con advertisement. Here, slot is the time taken to transmit one packet and cycle is the time
taken to reserve, assign, and utilize a slot by/for every node in the network. This is demon-
strated in Fig.1, and how each cycle is comprised of tt slots is given as
(1)
tt =r+i+m+n
Fig. 1 RTDMA cycle structure demonstrating a cycle which comprises 4 categories of time slots namely,
reservation slots, R, sink slots, S, reserved slots, RS, and allotted slots, AS

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
where r is the total number of slots allotted for reservation phase, i is the total number of
slots allotted for slot assignment by the sink, m is the total number of slots allocated for
reserved nodes, and n is the total number of slots allotted for unreserved nodes.
A general improvement of the CASeR is achieved by reservation mechanisms and the
combination with TDMAMAC. This scheme typically has two phases: a reservation phase
followed by a transmission phase. During the reservation phase,
Rr
, nodes willing to send
packets can reserve one future slot in the upcoming cycle by sending beacon packet as in
Fig.2 and collisions may occur due to the Aloha scheme, but it will not lead to data loss
because this phase is for reservation purpose only. By placing the reservation phase,
Rr
,
the transmission slots can be first utilized by the nodes having packet to send and also be
accessed without collision. As per reservation, sink broadcasts an access list that contains
the slot allocation details for the upcoming cycle during sink slot,
Si
. In the sink slot, each
reserved node will be allotted onetime slot for forwarding data packet and then the remain-
ing unreserved slots in that particular cycle can be allotted to the remaining nodes which
do not participate in the reservation on the basis of node ID order. Using this time slot,
nodes can share hop count and cost with their neighbors on a regular basis. Thus there is no
need to concern regarding gradient maintenance as it is safeguarded by sharing local topol-
ogy information that too without the need for flooding type transmission. In other words,
the reservation phase can be effectively utilized by the nodes having data packet for send-
ing to the sink. If no slot is reserved during the reservation phase, then the previous cycle’s
order will be followed for the next cycle.
For example, as shown in Fig.3 nodes A, D, and G have reserved their slots 1, 2, and 3
for data transmission during the reservation phase,
Rr
. Therefore, the sink collects all suc-
cessful reservation requests and sends back a reservation list indicating access rights for
Fig. 2 Reservation request packet where n is the number of nodes and
LR
is the total size of the reservation
packet
Fig. 3 Slot reservation and allocation. In the reservation phase, nodes A, D, and G send a request to reserve
a slot each in the next cycle to sink. During the sink slot,
S0
, sink broadcast the slot allotment list in which
slots 1, 2, and 3 are allocated to nodes A, D, and G, respectively on the reservation basis. The remaining
unreserved slots 4, 5, 6, 7, and 8 are allocated to the remaining nodes which do not participate in the reser-
vation as per node ID order

K.M.Kumaran, M.Chinnadurai
1 3
the next cycle of slots through a beacon signal in the slot,
S0
. In the reservation list, the first
three slots are allocated to A, D, and G, respectively for data packet transmission. Then
the remaining slots 4, 5, 6, 7, and 8 are allotted to the unreserved nodes B, C, E, F, and H,
respectively for broadcasting hop count and residual energy (RE) to the neighbors in order
on the basis of node ID. Presently, a time slot later on is saved, and no other node is permit-
ted to transmit amid this slot.
To maintain this scheme, the nodes have to be synchronized [24] with the sink for every
cycle. In general, this scheme may cause a higher delay under a low dense network because
of the reservation phase but allow higher throughput due to fewer collisions in a high dense
network.
The GTDMA is static and entirely contention free, meaning that every node has to
broadcast data in the predefined time slot only. The nodes have to wait for their defined
slot for communication even if they have prior packets to send. The GTDMA is highly
reliable for packet delivery but increasing the number of nodes results in a worstcase of
queuing delay which leads to a high EED. On the other hand, the RTDMA is a dynamic
and reservation-based scheme, and gives priority for data transfer with the least EED and
no collision.
3.2 Dynamic Topology Conguration
In dynamic topology, like the MWSN, the location of the nodes in the network can be var-
ied from time to time but every node has to know its position in the network for successful
transmission of data without any loss or delay. The hop count is a straightforward metric
used for topology configuration, which demonstrates a node’s distance in hops from the
sink. Despite the fact that the nodes will regularly require location awareness for detailing
positions to the sink, it is as yet desirable to over utilize a hop count metric for routing.
Every node has to broadcast its hop count during the allotted time slot, which will be uti-
lized by neighboring nodes to calculate their hop count. The hop counts are then used to
guarantee that the information is constantly sent towards the sink.
Further, by utilizing the DAMA MAC every node has a chance to transmit once in a
cycle. Using the reserved time slot, nodes must transmit a data packet as shown in Fig.4. If
the node has no packet to send, then the beacon packet has to broadcast during the allotted
time slot. The beacon packet contains only three fields of the data packet, i.e., (1) node ID,
(2) hop count and (3) cost (we discuss this further in Sect.3.3) of the node. Using this bea-
con signal, every node can advertise its hop count to neighboring nodes in a single cycle.
Likewise, every node will update the neighboring node’s hop count and later it will be used
Fig. 4 Data packet format where n is the number of nodes,
Lu
is the size of generated data,
Ld
is the total
size of a data packet, and [·] is the ceiling function

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
to determine its own hop count. The hop count of a node can only be determined by adding
one hop count to the lowest hop count of its neighboring node.
The first part of Fig.5 shows the MWSN with a sink and 8 sensor nodes where all the
nodes are labeled with the hop count except node N4 which has newly entered into the net-
work. As discussed earlier, every node will broadcast its hop count using a beacon signal.
Likewise, node N4 will receive all its neighbors’ hop counts 1, 2, 3, 4, 3, and 2 in a single
cycle. Then, node N4 can select the lowest hop count and add one to determine its short-
est path to sink. The second part of Fig.5 shows the updated hop count of node N4 which
is 2. The third part of Fig.5 shows how the highlighted node N8 having hop count 4 has
updated its hop count to 3 on the next cycle by using the newly arrived node N4.
By the cyclic nature of the DAMA MAC, every node can get its hop count through the
neighboring node’s hop count. Likewise, all the nodes in the network can update their hop
count in every cycle. In this way, each node is allowed to overhear the beacon signal of all
neighboring nodes, and it will be helpful for data forwarding before the transmission of data.
This will be helpful for gradient maintenance and topology formation by sharing of hop count
information alone.
3.3 Data Transmission orRouting
The CASeR uses cost- and hop count based forwarding technique to forward data with accu-
rate gradient maintenance towards the sink. Cost is the metric which shows the RE of a node
indicated out of 10. As discussed earlier, the hop count is the node’s distance from the sink
in hops. Each node can broadcast a beacon signal which contains the hop count and cost to
inform its presence to its neighbors. Before data transmission, each node obtains the neighbor
node’s hop count to sink and their RE through the beacon signal broadcasting. During the
Fig. 5 Topology configuration. Node N4 is a new arrival into the network with an indefinite hop count. By
using the hop count broadcast of neighbors it calculates its own hop count. Then, the nodes around it can
redefine their own hop count as well
Fig. 6 Routing table maintained
by node N6
Neighbor
node ID Hop countCost
N5
29
N4
28

K.M.Kumaran, M.Chinnadurai
1 3
data transmission, the sender node can prefer a neighbor node with the least hop count and
high cost to forward packet. This technique is called cost and hop count based forwarding
technique.
For example, assume a network as shown in Fig.7, with 8 nodes where every node is rep-
resented by a square which contains node ID on top, hop count at bottom right, and cost at
bottom left. Initially, data is generated by the node N6. The routing table maintained by node
N6 is shown in Fig.6. This routing table contains the neighbor node ID, cost, and hop count.
During data transmission, node N6 prefers the neighbor with the least hop count for data for-
warding. However, if the hop count of two or more neighbors is the same like in the case of
nodes N4 and N5as shown in Figs.6 and 7, then the node with high cost, i.e., node N5 will
be preferred by node N6 for data forwarding. Now node N5 makes a comparison between
node N1 and N3, and selects N1for data forwarding due to its least hop count. Then, node
N1 forwards the packet to the sink which is in its transmission range. Here, the flooding and
blind broadcasting based transmission is replaced by unicast transmission. This unicast type of
transmission of data mitigates excess computational overhead and unwanted energy consump-
tion of blind forwarding technique [25]. The cost based neighbor selection for data forwarding
leads to even energy dissipation on all the nodes in the network and consequently increases
the lifetime of the network. The blind forwarding of data from one node to another can lead
to unnecessary processing time and energy usage by the neighboring nodes. During blind for-
warding, nodes neighboring the sender will overhear the transmission and the same will be
blindly broadcasted to other nodes. This technique leads to the maximum energy consump-
tion for a single journey of a packet from the sender to the sink. Moreover, it leads to gradual
energy consumption by every node in the geographical region of the packet journey. In other
words, if we consider a packet is broadcasted by the outer edge node of the network, it will be
Fig. 7 Hop count and cost based data forwarding. (1) Initially data is generated at node N6 and then for-
warded to its neighbor node N5 based on cost comparison. (2) Node N5 forwards the data to node N1 based
on the hop count gradient. (3) Then, node N1 forwards the packet which will be heard by the sink

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
overheard by its entire neighboring node. Again one of the neighboring nodes will broadcast
the packet in a favorable gradient towards the sink and this is again overheard by its entire
neighboring node. In this way, the computation will multiply and cause unnecessary overhead
and energy consumption until the packet reaches the sink. One of the major concerns in blind
forwarding is data duplication. Due to blind broadcasting, packets choose multiple paths to
reach the destination (sink) resulting in the generation of duplicate packets. These packets are
also routed to the sink with additional computational time and energy.
The general working process of the CASeR is represented by a flow chart as shown
in Fig.8. Figure5 also shows the basic process and algorithm used during each cycle.
Initially, at the beginning of each cycle, the sink will broadcast a reservation beacon
signal, which invites the nodes to reserve their time slot in the next cycle based on the
R-TDMA. The nodes with pending packets to forward can reserve the time slots by
sending the reservation request packet as shown in Fig.2 to the sink and the remain-
ing nodes will be idle during this phase. Then, the sink assigns slots for the nodes in
that priority and will be first given to the nodes that have reserved their slots. Then the
remaining nodes will be allotted to the unreserved slots in that cycle on the basis of
node ID order. During the reservation phase, the time slots which are reserved by the
nodes are called reserved time slots and the remaining time slots which are allotted to
the idle nodes are called allotted time slots. If it is a node’s reserved time slot, then the
oldest packet in the queue will be forwarded to the neighbor with the least hop count
and high cost using a selection algorithm. Else, if it is a node’s allotted time slot, then it
will broadcast a beacon packet with its current hop count and cost. This process will be
Fig. 8 Flow chart representation of the CASeR sensor node process

K.M.Kumaran, M.Chinnadurai
1 3
repeated by the nodes until the packet reaches the sink. If it is not the node’s time slot,
it will listen to the medium for transmissions. If any data transmission is received by the
node, it will store the data and would look for the reservation beacon to reserve the time
slots in the next cycle. It also will update its hop count (if needed) and the sender’s hop
count and cost in the routing table as shown in Fig.6.
Here, the medley algorithm has a worst case time complexity of O(rq), where rq is
the reserved queue length, and this low complexity will help to reduce energy and time
consumption by decreasing the processing time and queuing delay. On the contrary, in
the RASeR the worst case time complexity of O(q) will be achieved, where q is the
queue length. Hence, in comparison, the RASeR consumes more energy and time than
the CASeR for data transmission. The key aspect of the CASeR, which makes it so
competent, is the use of the RTDMA which enables the reservation phase to reduce
time complexity thus ensuring less computation time and queuing delay. In addition,
the CASeR uses flat network hierarchy that ensures every node will work identically,
which in turn allows the protocol to work in a simple manner, and thus it can easily be
deployed on huge numbers of nodes.
4 Proposed Network Modeling
In our study, the simulation and modeling were done by using the popular OPNET mod-
eler simulation software [26]. In this simulation, we have modeled two network scenarios
utilizing the proposed CASeR protocol and evaluated the performance of the protocol on
various performance metrics such as the PDR, EED, and average energy consumption.
The PDR is defined as the total number of packets received
RP
over the total number of
packets generated
GP
and can be calculated as
The energy consumption is defined as the energy consumed by the nodes to transmit
and receive data packets. The energy consumption, EC, is described in terms of joules per
second per node:
where
Vb
is the battery voltage, It and Ir are the consumption of current by the transceiver
when transmitting and receiving, respectively,
Tt
is the total deployment time of the net-
work,
Rb
is the bit rate,
Bt
is the total number of bits transmitted,
Br
is the total number of
bits received, and n is the number of nodes in the network.
Since the CASeR broadcasts packets to all neighbors,
Br
is given as
where
Nn
is the expected number of neighbor nodes to each node.
This expression needs awareness of the hardware but
Vb
,
It
, and
Ir
can be substituted for
impermanent values in view of potential hardware for comparison purposes.
(2)
PDR
=
R
P
G
P
×100
%
(3)
EC
=

Vb
Rb

⋅

It⋅Bt

+

Ir⋅Br

n⋅Tt

(4)
Br=BtNn

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
The EED,
Dee,
is defined as the time between the generation of the first bit of the packet
in a node and the reception of the last bit of the same packet in the sink.
where
Td
is the time delay caused at each node and
Hn
is the number of hopes expected
during the transmission of packets from source node and the sink.
where
Dq
is the total time a packet is waited in the queue for transmission,
Dt
is the time
taken to transmit a packet into the medium,
Dpc
is the time taken by a node to choose the
receiving node, and
Dpg
is the time taken by a packet to travel from a node to other.
In our simulation, the delay caused by
Dt
,
Dpg
, and
Dpc
is considered to be static values
based on the ideal node capability and the delay caused during queuing is considered to be
the dynamic part of Eq.(6).
5 Simulation
We consider that the target application of this work is a rescue mission that could be per-
formed by a swarm of UAVs or land robots equipped with infrared detection sensors, which
could indicate the location of the victims to the sink which could be a high-powered aerial
or land vehicle. We can deploy the UAV nodes for such an operation without any delay in
waiting for the team to arrange. This will be helpful in finding the victim as early as possi-
ble. Later, the sink would send a beacon signal to synchronize all the nodes in the network.
Then the time slot will be allotted as per the needs of the nodes for packet transmission.
During the working of the network, we can add additional nodes for rescue operations and
these nodes also get synchronized as desired.
Based on this simulation, Memsic IRIS motes [27] were used as modeled nodes with a
transmission radius of 250m and a bit rate of 250 kbps. The energy utilization parameters,
Vb
,
It
and
Ir
, were also modeled on the IRIS node, namely 3V, 16.5mA, and 15.5 mA,
respectively. A packet generation rate of a node is 1pk/s, and was set to give a sensible
traffic level for a swarm detecting numerous targets within the search area. Here, the loca-
tion- and time-related information was considered as the packet data for which 32 bits are
enough to report the position of the node.
5.1 Mobility
The performance comparison of the CASeR with other protocols at different levels of
mobility is shown in Fig.9. The speed of the mobile nodes is varied from 0 to 100 m/s.
The speed of packet creation and network size is kept constant at 1 packet per second,
100 nodes, respectively. Based on Fig. 9a, the RASeR shows a slight improvement in
the PDR than the CASeR due to the blind forwarding technique, whereas the PHASeR
and MACRO show slightly lesser PDR than the CASeR. In Fig.9b, we show the delay
comparison data in which the CASeR outperforms the PHASeR, OLSR, and MACRO in
terms of its less packet queuing time, whereas the RASeR shows much lesser delay than
the PHASeR and MACRO. The increase in mobility speed leads to the frequent change
in topology and therefore sharing of topology information becomes more. As a result, the
(5)
Dee =Td×Hn
(6)
Td=Dq+Dt+Dpg +Dpc

K.M.Kumaran, M.Chinnadurai
1 3
Fig. 9 Performance results of the CASeR in comparison with the RASeR, MACRO, PHASeR, OLSR, and
AODV over increasing maximum speed. a PDR, b average end-to-end delay (AODV’s result was 1.12s and
was detached to get better scale), and c average energy consumption

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
Fig. 10 Performance results of the CASeR in comparison with the RASeR, MACRO, PHASeR, OLSR, and
AODV over increasing numbers of nodes. a PDR, b average end-to-end delay (AODV’s average result was
4.93s and was detached to get better scale), and c average energy consumption

K.M.Kumaran, M.Chinnadurai
1 3
packet forwarding process in a proper gradient becomes an overhead which leads to addi-
tional hop and energy consumption as shown in Fig.9c. Consequently, the CASeR and
RASeR consume much less energy than the MACRO, PHASeR, OLSR, and AODV.
5.2 Scalability
The set of results in Fig. 10 show how protocols respond to changes in the number of
nodes: [10, 20, 30, 40, 50, 60, 70, 80, 90, 100]. With a specific objective to keep a gener-
ally similar node density, the square network zone was likewise altered by changing the
side lengths: [400, 600, 1000, 1200] m. The maximum speed was kept steady at 25m/s
and the data generation rate of the sensor nodes was likewise kept at 1pk/s. This simula-
tion illustrates the scalability of the protocol in terms of the size of the network. The PDR
demonstrates that the CASeR, PHASeR, MACRO, RASeR, AODV, and OLSR assume an
indistinguishable pattern from network measure increments, and all give better outcomes in
smaller network sizes. The RASER demonstrates a little improvement in the PDR; yet, it
watches out for considerably higher delay than the CASeR. One to one forwarding plan of
the CASeR prompts the marginally expanded packet loss yet consumes lesser energy and
less time in delay. With regard to queuing delay, the RASeR indicates considerably higher
delay than the CASeR and demonstrates a superior outcome than the PHASeR, MACRO,
AODV, and OLSR. The energy consumption result demonstrates that the CASeR expend
substantially lesser energy than the RASeR, PHASeR, AODV, OLSR, and MACRO for
packet transmission. In this way, it prompts an appreciable change in the lifetime of the
network. The overall results prove that the increase in the number of nodes leads to perfor-
mance degradation in all the four protocols.
6 Conclusion
In the MWSNs, limited energy and node mobility renders the protocol design very com-
plex. The mobility leads to the frequent change in topology and packet loss. In this study,
we have presented a competent routing protocol, the CASeR, to improve competence in
the PDR with least delay and lesser energy consumption. It uses reservation based slot
allocation using the RTDMA, MAC, and cost based multi-hop communication for packet
forwarding. The reservation scheme helps to eliminate queuing delay which enables the
CASeR to achieve much lesser EED than the RASeR. Routing protocols such as the
RASeR, PHASeR, MACRO, AODV, and OLSR are compared with the CASeR in terms
of simulation. The cost based packet forwarding causes slight lower PDR but leads to a
much lesser energy dissipation. The simulation indicates enhancements over others in
energy consumption, EED, and reliable data communication. The capability of this proto-
col makes it exceptionally effective in minimizing time delay in target applications.
References
1. Chong C. Y., & Kumar, S. P. (2003). Sensor networks: Evolution, opportunities and challenges. In:
Proceedings of IEEE (Vol. 91, No. 8, pp. 1247–1256).

A Competent Ad‑hoc Sensor Routing Protocol forEnergy Efficiency…
1 3
2. Lorincz, K., etal. (2004). Sensor networks for emergency response: Challenges and opportunities.
IEEE Pervasive Computing, 3(4), 16–23.
3. White, B., etal. (2008). Contaminant cloud boundary monitoring using network of UAV sensors.
IEEE Sensors Journal, 8(10), 1681–1692.
4. Chen, Y., & Zhao, Q. (2005). On the lifetime of wireless sensor networks. IEEE Communications
Letters, 9(11), 976–978.
5. Conti, M., & Giordano, S. (2014). Mobile ad hoc networking: milestones, challenges and new
research directions. IEEE Communications Magazine, 52(1), 85–96.
6. Al-Karaki, J. N., & Kamal, A. E. (2004). Routing techniques in wireless sensor networks: A survey.
IEEE Wireless Communications, 11(6), 6–28.
7. Waharte, S., & Trigoni, N. (2010). Supporting search and rescue operations with UAVs. In Pro-
ceedings of international conference on emerging security technologies (EST ’10) (pp. 142–147).
8. Perkins, C. E., & Royer, E. M. (1999). Ad-hoc on-demand distance vector routing. In Proceedings
of 2nd IEEE workshop on mobile computing systems and applications (WMCSA ‘99) (pp. 90–100).
9. Johnson, D. B., & Maltz, D. A. (1996). Dynamic source routing in ad hoc wireless networks. In
T. Imielinski & H. Korth (Eds.), Mobile computing (pp. 153–181). New York: Kluwer Academic
Publishers.
10. Biswas, S., & Morris, R. (2005). ExOR: Opportunistic multi-hop routing for wireless net- works.
ACM SIGCOMM Computer Communication Review, 35(4), 133–144.
11. Ren, S., etal. (2012). An improved wireless sensor networks routing protocol based on AODV. In
Proceedings of IEEE 12th international conference on computer and information technology (CIT
’12) (pp. 742–746).
12. Soliman, H., & AlOtaibi, M. (2009). An efficient routing approach over mobile wireless ad-hoc
sensor networks. In Proceedings of 6th IEEE consumer communications and networking conference
(CCNC ’09) (pp. 1–5).
13. Huo, G., & Wang, X. (2008). An opportunistic routing for mobile wireless sensor networks based
on RSSI. In Proceedings of 4th international conference on wireless communications, networking
and mobile computing (WiCOM ‘08) (pp. 1–4).
14. Han, Y., & Lin, Z. (2012). A geographically opportunistic routing protocol used in mobile wireless
sensor networks. In Proceedings of 9th IEEE international conference on networking, sensing and
control (ICNSC) (pp. 216–221).
15. Kwangcheol, S., Kim, K., & Kim, S. (2011). ADSR: Angle-based multi-hop routing strategy for
mobile wireless sensor networks. In Proceedings of IEEE Asia-Pacific services computing confer-
ence (APSCC) (pp. 373–376).
16. Heinzelman, W. R., Chandrakasan, A., & Balakrishnan, H. (2000). Energy efficient communication
protocol for wireless micro sensor networks. In Proceedings of 33rd Hawaii international confer-
ence on system sciences (HICSS ‘00) ((p. 8020).
17. Kim, D., & Chung, Y. (2006). Self-organization routing protocol supporting mobile nodes for wire-
less sensor network. In Proceedings of 1st international multi-symposiums on computer and com-
putational sciences (IMSCCS ’06) (pp. 622–626).
18. Kumar, G. S., Vinu, M. V., Athithan, P. G., & Jacob, K. P. (2008). Routing protocol enhancement
for handling node mobility in wireless sensor networks. In Proceedings of IEEE region 10 confer-
ence (TENCON) (pp. 1–6).
19. Awwad, S. A. B., Ng, C. K., Noordin, N. K., & Rasid, M. F. A. (2009). Cluster based routing pro-
tocol for mobile nodes in wireless sensor network. In Proceedings of international symposium on
collaborative technologies and systems (CTS ‘09) (pp. 233–241).
20. Deng, S., Li, J., & Shen, L. (2011). Mobility-based clustering protocol for wireless sensor networks
with mobile nodes. IET Wireless Sensor Systems, 1(1), 39–47.

K.M.Kumaran, M.Chinnadurai
1 3
21. Cakici, S., etal. (2014). A novel cross-layer routing protocol for increasing packet transfer reli-
ability in mobile sensor networks. Springer Wireless Personal Communications Journal, 77(3),
2235–2254.
22. Hayes, T., & Ali, F. H. (2015). Proactive highly ambulatory sensor routing (PHASeR) protocol for
mobile wireless sensor networks. Elsevier Pervasive Mobile Computing, 21, 47–61.
23. Hayes, T., & Ali, F. H. (2016). Robust ad-hoc sensor routing (RASeR) protocol for mobile wireless
sensor networks. Ad Hoc Networks, 50, 128–144.
24. Sivrikaya, F., & Yener, B. (2004). Time synchronization in sensor networks: A survey. IEEE Net-
work, 18(4), 45–50.
25. Jurdak, R., etal. (2009). Directed broadcast with overhearing for sensor networks. ACM Transac-
tions on Sensor Networks, 6(1), 31–335.
26. OPNET Technologies Inc, OPNET. (2013). www.opnet .com.
27. Memsic Inc, IRIS. (2014). https ://www.memsi c.com/userfi les/files /Datas heets /WSN/IRIS_Datas
heet.pdf.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
K. Manikanda Kumaran received the B.Tech. degree in Information
Technology from Anna University, India in 2010 and M.E. degree in
Computer Science and Engineering from Annamalai University, India,
in 2012. Presently he is working as an assistant professor in the
Department of Information Technology in E.G.S. Pillay Engineering
College, Nagapattinam, India. His research interests including wireless
sensor network, networking, mobile ad-hoc network and network
security.
Dr. M. Chinnadurai is working as Professor and Head of CSE in E.G.S
Pillay Engineering College Nagapattinam. He completed his Ph.D. in
Anna University, Chennai in the field of VLSI at Faculty of Informa-
tion and Communication Engineering, Anna University, Chennai. He
is professional member in IEEE, CSI, ISTE, etc. and his research work
includes Artificial Intelligence, Network Security, Algorithms and
Cloud Computing.