Scalability Analysis of Depth-Based
Routing and Energy-Efﬁcient
Depth-Based Routing Protocols in Terms
of Delay, Throughput, and Path Loss
in Underwater Acoustic Sensor Networks
Saqib Shahid Rahim, Sheeraz Ahmed, Nadeem Javaid, Adil Khan,
Nouman Siddiqui, Fazle Hadi, and M. Ayub Khan
Abstract In underwater acoustic sensor networks (UWASNs), nodes are either
static or dynamic depending upon the network conﬁguration and type of application.
Direct or multi-hop transmissions are used to forward data toward the sink. Alter-
natively, sinks can also be mobile or static, depending on whether the application is
real time or passive. The variety of nodes and sink deployments greatly affect the
performance of routing protocols. In this chapter, we analyze the effects of node
density and scalability on the performance of routing protocols in UWASNs. Two
S. S. Rahim ()
Career Dynamics Research Centre, Peshawar, Pakistan
Abasyn University, Peshawar, Pakistan
Preston University, Peshawar, Pakistan
S. Ahmed · M. Ayub Khan
Career Dynamics Research Centre, Peshawar, Pakistan
Iqra National University, Peshawar, Pakistan
COMSATS Institute of Information Technology, Islamabad, Pakistan
A. Khan ()
Career Dynamics Research Centre, Peshawar, Pakistan
Abdul Wali Khan University, Mardan, Pakistan
Career Dynamics Research Centre, Peshawar, Pakistan
Higher Education Department, Govt. of Khyber Pakhtunkhwa, Pakistan
© Springer Nature Switzerland AG 2019
M. A. Jan et al. (eds.), Recent Trends and Advances in Wireless and IoT-enabled
Networks, EAI/Springer Innovations in Communication and Computing,
172 S. S. Rahim et al.
popular UWASNs protocols were selected for this purpose: the depth-based routing
protocol (DBR) and energy-efﬁcient depth-based routing protocol (EEDBR). DBR
is a non-cluster-based technique that performs routing using only the depth of nodes,
whereas EEDBR is a location-free scheme that uses both the depth and the residual
energy of nodes to route data. The scalability of node deployment was used to check
the efﬁciency of these schemes in the context of three parameters: packet delivery
ratio, end-to-end delay, and path loss.
One-third of Earth consists of oceanic areas, which are largely unexplored. Several
underwater activities already exist, including underwater mining, water pollution
controls, seismic monitoring, security, and sports. Conventional approaches to
observe underwater environments have several limitations, such as the cost of
devices and extended delays to analyze the examined data . To address these
issues, underwater acoustic sensor networks (UWASNs) are being developed to
explore the underwater environment. UWASNs are composed of sensor nodes (also
known as nodes). These nodes cooperatively examine different activities within a
particular area . A simple architecture of a UWASN is shown in Fig. 16.1.
A UWASN is composed of acoustic sensors with single multi-surface sinks
that are dropped under the water to examine the underwater environment. The
surface sink usually has little power restriction, whereas the sensor nodes have very
restricted energy . Sinks (also known as sonobuoys) exists at the ocean surface.
These sinks contain two modems: an acoustic and a radio. For example, the nodes in
the Sensor Equipped Aquatic (SEA) Swarm architecture observe nearby underwater
events and report them to one of the sinks—a process known as anycasting.The
gathered statistics are unloaded to an examination station through radiowaves for
auxiliary ofﬂine processing [3,4].
UWASNs have potential use in applications such as seismic monitoring, ocean
mine exploration, and disaster prevention. UWASNs provide operational approaches
for routing. This is a necessity for time-critical applications, and hence in the design
of delay-sensitive protocols .
Radio signals are not suitable for transmissions in UWASNs because radio waves
absorb in water rapidly. Hence, acoustic waves are used in this environment ;
for high data rates in underwater communications, acoustic signals are also the best
source . A UWASN achieves this goal with underwater vehicles and sensor nodes.
Underwater transmission has some unique challenges, such as low bandwidth,
limited mobility of nodes, delays, low memory, and battery constraints. Current
research on deep-water activities is based on different technologies. Because
acoustic waves are used in water as a medium of communication and to transmit
data, UWASNs have a greater variety of network designs than do terrestrial
communication systems. Many techniques have been proposed for effective routing
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 173
Fig. 16.1 UWASN architecture
in USWNs, such as cooperation-based routing that transmits from deep water to a
surface sink by relay .
This chapter focuses on the scalability analysis of two well-known schemes:
depth-based routing protocol (DBR)  and and energy-efﬁcient depth-based
routing protocol (EEDBR) . DBR  is a non-cluster-based technique that
perform routing using just the depth of nodes, whereas EEDBR  is a location-
free scheme that uses depth along with the residual energy of nodes to route data.
The scalability of node deployment is applied to examine the efﬁciency of these
schemes in the context of three parameters: packet delivery ratio, end-to-end delay
and path loss. This chapter is arranged as follows: Sect. 16.2 provides the literature
review; Sect. 16.3 discusses the motivation for our work; Sects. 16.4 and 16.5
illustrate the DBR and EEDBR schemes, respectively; and Sects. 16.6 and 16.7
discuss the scalability of these schemes. Finally, Sect. 16.8 describes a cooperative
node technique and the energy harvesting of nodes as future goals for researchers in
174 S. S. Rahim et al.
16.2 Related Work
Researchers have proposed a variety of protocols and techniques to increase
performance and reliable communications in UWASNs. Some of these approaches
are summarized and discussed in this section.
Noh et al.  proposed the Void Aware Pressure Routing for Underwater Sensor
Networks (VAPR) protocol, which features a pressure routing technique. This
scheme uses pressure meters to deliver depths and transmit information data packets
to a surface sink. A series of depth information in periodic beacons and hops are
counted to establish a subsequent hop track, which makes a directional link to a
nearby surface sink.
Javaid et al.  introduced a forwarding equation based on a routing scheme for
UWASNs: Improved Adaptive Mobility of Courier nodes in Threshold-optimized
Depth-based routing Protocol (iAMCTD). This protocol increases the lifetime of an
underwater network through an optimal viable mobility strategy for surface sinks.
The iAMCTD scheme depends on a cost function. Compared with many other
depth-based underwater techniques, this approach exploits the network density for
applications that are time critical. To control the transmission loss, water ﬂooding,
and transmission latency, the authors designed an equation for a depth-dependent
function that also uses signal-to-noise ratio, the signal quality index, optimal holding
time, and energy-cost function routing parameters. The equation computes the prime
energy limit, soft threshold, and hard threshold to provide routing on demand .
In the Cooperative Energy-Efﬁcient for Underwater WSN (Co-UWSN) protocol,
Ahmed et al.  used a cooperation scheme to improve a UWASN’s lifetime,
enhance the delivery ratio of data, and reduce the overall energy tax, which is
speciﬁcally advantageous for time-critical and delay-sensitive applications. Using
cooperative communication, this technique mitigates the effects of noise and
multipath fading. By changing the depth threshold, the number of eligible neighbors
increases; hence, data loss is reduced in delay-sensitive applications. This coopera-
tion technique improves the load balancing of the UWASN and the stability of the
In another study by Ahmed et al. , the Stochastic Performance Analysis
with Reliability and Cooperation (SPARCO) technique was used to increase the
efﬁciency of the network. Cooperative communication was introduced for routing
in UWASNs to effectively consume energy. All nodes of the network were assumed
to consist of a unidirectional antenna. To reduce energy consumption, several nodes
transmitted their data cooperatively to take advantage of the spatial diversity .
In addition, the Adaptive Mobility of Courier nodes in Threshold-Optimized DBR
(AMCTD)  achieved adaptive mobility in special mobile nodes called courier
nodes to improve the life of the network. This protocol calculates the holding time
based on the weight function, which controls the issue of transmission loss .
Javaid et al.  also proposed the Delay-Sensitive Depth-Based Routing
(DSDBR), Delay-Sensitive Energy Efﬁcient Depth-Based Routing (DSEEDBR),
and Delay-Sensitive Adaptive Mobility of Courier nodes in Threshold-optimized
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 175
Depth-based routing (DSAMCTD) schemes. These protocols empower depth-
based routing techniques. The efﬁciency of the proposed protocols was veriﬁed for
UWASNs. These schemes use delay-efﬁcient priority factors and delay-sensitive
holding time to reduce end-to-end delays; however, a minor reduction in the
throughput of the network occurred. All techniques employed an optimum weight
function to compute the speed of the received signal and path loss. Moreover, a
solution for the delay problem was found by forwarding the data efﬁciently, with
nominal relative transmissions in low depth areas and the selection of an optimal
forwarder. Simulation results showed that the protocols greatly reduced end-to-end
delays and enhanced path loss .
Another study by Javaid et al.  proposed a scheme called the Region-
Based Cooperative Routing Protocol (RBCRP). In this protocol, ampliﬁcation
and forwarding occur over Rayleigh worn links in UWASNs. The sender node
transmits data packets, which are sensed by the sensor node to the endpoint and
accessible relays. The bit error rate is tested at the end node, based on negative or
positive retorts to the sender and relays. The authors used mobile sinks with energy
harvesting to improve the packet delivery ratio and network stability. RBCRP was
found to attain enhanced network stability, outage probabilities, and high packet
delivery ratios compared with an incremental best relay scheme .
Ahmed et al.  also proposed cooperative communication to build an energy-
efﬁcient protocol for UWASNs, referred to as Cooperative Energy Efﬁcient routing
for UWSNs (Co-EEUWSN). In this protocol, all nodes of the UWASN contains a
directional antenna, while several nodes coordinate with each other. At the relay, Co-
EEUWSN employed the amplify-and-forward mechanism; however, at the receiving
node, the ﬁxed ratio combining was used. In a comparison of this scheme’s results
with those of EEDBR and cooperative DBR, Co-EEUWSN showed improved
energy efﬁciency, decreased end-to-end delays, and enhanced throughput .
Based on the literature review, most researchers [9,10,14,15] use node depth as a
parameter for data routing. However, they do not address node load balancing and
the distribution of load when the sensor nodes are uneven. Therefore, the efﬁciency
of energy consumption in the nodes is not properly controlled when only depth is
used . Wahid et al.  used both depth and residual energy as a metric to forward
data. DBR attempts to attain a longer network lifetime but has a short stability
period. The reason for this problem is due to the redundant transmission of data
packets with a heavy load on the low-depth sensor nodes of the UWASN. EEDBR
is not a cooperation-based protocol; hence, the packets are led from the source to the
destination using a single route with a multi-hop style. Because of noise and fading
of the multipath environment, many time signals suffer from a high bit error rate.
176 S. S. Rahim et al.
16.4 Description of the DBR Scheme
DBR is based on a greedy approach that attempts to send a data packet from a
node to surface buoys. During the progression of data, the depth of the promoting
nodes shrinks, even as the data reaches the sinks. If the depth of the sending node
is shrunk in each step, then the data can be supplied to a sink at the surface. In
the DBR technique, the node makes the decision to send data based on its own
depth and the preceding sender’s node depth, which is the signiﬁcant concept of
this protocol. According to the DBR scheme, upon receiving the data, the node ﬁrst
takes the depth (dp) of the data of the preceding hop, which resides with the data
packet. Then, the receiving node matches its own depth (dc)todp. If the node is
nearer to the surface of the water (dc <dp), then this node will nominate itself to
send the data. In other cases, the packet will be dropped because the packet came
from a node that is nearer to the sink at the water surface. Multiple links cannot be
fully abolished; hence, DBR use a queue known as a “priority queue.” This queue
decreases the number of sending nodes, as well as controls the number of sending
In DBR, every node also has a packet history buffer. The priority queue is
denoted by Q1 and the packet history buffer by Q2. Q2 consists of a unique
packet identiﬁcation (ID) and packet sequence number. When correctly sending data
packets, the node inserts the ID of the data packet into queue Q2; on overﬂow of
this queue, the least recently accessed mechanism is used. Q1 is composed of data
packets and the scheduled forwarding times for the data packets. The signiﬁcance
of an item in queue Q1 is denoted by the scheduled forwarding time. An item with
a prior forwarding time has greater priority. On reception of a packet, the node ﬁrst
holds the packet for a speciﬁc length of time, which is known as the holding time.
The scheduled sending time of a data packet is calculated based on the received time
of the data packet and the holding time of the data packet .
The holding time is computed by using the linear function of d, where dis the
difference between the depths of the recent node and the preceding node.
Suppose that d1and d2are the depth differences of the n1and n2nodes,
respectively, from the source node (S) as shown in Fig. 16.2.t1is the time that it
takes to receives packet n1from S, whereas t2is the time that it takes to send from
Sto n2. The propagation delay between n1and n2is t12. Two situations can be
expressed by the following:
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 177
Fig. 16.2 Forwarding node
by substituting f(d) with the linear function
where αis non-positive.
As long as |α|≥(t1−t2)+t12/(d1−d2), the conditions of (16.2) and (16.3)
can be met. αdepends on the depth difference of the n1,n2nodes. αcan vary
between 0 and Rfor a node with one-hop neighbors, where Ris the highest range
of transmission of the node. δis a global parameter to replace the depth difference
for holding time computations. Hence α=−2τ/δ. Suppose the node with the least
depth has a holding time of 0. βcan be computed by the following expression:
By substituting αand βin (16.1),
16.5 Description of the EEDBR Scheme
Energy efﬁciency is a major concern in UWASNs because the batteries have very
limited energy and their replacement is costly. Thus, EEDBR is an energy-efﬁcient
scheme in which the major focus is the nodes’ energy .
EEDBR uses two parameters for transferring data: nodes’ depth and their residual
energy. The nodes’ residual energy is used to increase the stability period. A Hello
packet is broadcast by all nodes in the information acquirement phase to its one-hop
neighbors. This packet consists of residual energy and the depth of that node. When
this packet is received by the neighbors, they hold the residual energy and depth
information of only those nodes with lesser depths. Keeping this information from
all nodes is not necessary. The updated information on depth is not so important,
although residual energy is required to be updated from time to time. To solve
this issue, the nodes in an EEDBR scheme check their residual energy with a time
interval-based mechanism .
178 S. S. Rahim et al.
In the data forwarding phase, the data is transferred from a node to a terminus
node or surface sink, depending on the residual energy and depth statistics of the
nodes. In this scheme, all nodes have information on their neighbors’ residual energy
and depth. The forwarding node selects the optimal next hop forwarder node. The
holding time (Tm) is calculated by (16.5):
Tm=(1– (current energy/initial energy)) ∗max holdingtime +pv, (16.5)
In (16.5), max_holding_time is a system metric and pv is the priority value.
This pv is used to avoid multi-forwarding nodes. Hence, to prevent duplicated
transmissions, the pv value is added to the holding time so that the differences
between the holding times of sending nodes have the same residual energies. The
list of forwarder nodes is arranged depending on residual energy. On the reception
of data, forwarder nodes add the priority value to the holding time depending on
the location within the list. The value of the priority is multiplied by two and by
the increase of position in the list index. Hence, the uppermost node of the list has
the maximum priority because it contains the maximum residual energy among all
neighbors; this node will send data immediately on reception .
To balance the energies of nodes, the node that has the maximum energy is
chosen. In a case where more than one node contains the same energy and depth,
any node can be nominated for sending. An abundant suppression of data packet
transmissions highly disturbs the delivery ratio of data, while the delivery ratio in
many applications is more signiﬁcant than the energy. Hence, for these applications,
EEDBR uses an application-based suppression technique. When sending the data,
the source node adds the number of data packets; on reception, the sink node
calculates the delivery ratio. If the delivery ratio is smaller than the anticipated
delivery ratio, then the sink notiﬁes the source by transmitting a packet that consists
of the delivery ratio at the sink. The source also adds the value of the delivery ratio
that is received from the surface sink into the packet. On reception of the data packet,
the sending nodes makes a decision on whether to suppress the packet or transmit it,
depending on the value of the delivery ratio. The forwarding operation of the data is
depicted in Fig. 16.3.
16.6 Scalability Analysis of DBR
In this section, we investigate and depict the scalability of DBR under different
node densities during the stability period. The stability period is deﬁned as the time
period until the ﬁrst network node expires. The following performance metrics for
scalability were selected: throughput, end-to-end delay, and path loss.
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 179
Fig. 16.3 Operation at the forwarding node 
16.6.1 Throughput Analysis
The throughput analysis in terms of the scalability of the DBR scheme is shown in
Fig. 16.4. In ﬁrst 1000 rounds, the delivery ratio was highest when the number of
nodes was 100, 250, 400, or 500 (that is, 100 or near to 100). The lowest delivery
ratio was 17 for 500 nodes at a round number of 5000. The optimal throughput was
found for 250 nodes at each round, with an average delivery that was also better
16.6.2 End-to-End Delay
The end-to-end delay analysis in terms of the scalability of the DBR scheme is
shown in Fig. 16.5. The maximum delay was 133 for 500 nodes at 1000 rounds,
while the minimum delay was 5 for 250 nodes at 5000 rounds. Overall, a low delay
was found from 1000 to 5000 rounds for 100 nodes. The ﬁgure also depicts that
delay was high at each round for 500 nodes.
180 S. S. Rahim et al.
Fig. 16.4 Throughput scalability
Fig. 16.5 End-to-end delay vs rounds
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 181
Fig. 16.6 Path loss vs rounds
16.6.3 Path Loss Analysis
The path loss analysis in terms of the scalability of the DBR scheme is shown in
Fig. 16.6. As the number of nodes increased, the path loss decreased; by decreasing
of the nodes, the loss increases. At round 5000, the path loss was 100 for 100 nodes;
for 250, 400, and 500 nodes, this loss was 90, 88, and 79, respectively.
16.7 Scalability Analysis of EEDBR
In this section, we analyzed and discuss the scalability of EEDBR under different
node densities during the stability period. The stability period is deﬁned as the
time period prior to the total energy drainage of the ﬁrst node in the network. The
following performance metrics for scalability were selected: throughput, end-to-end
delay, and path loss.
182 S. S. Rahim et al.
Fig. 16.7 Throughput scalability of EEDBR
16.7.1 Throughput Analysis
The throughput analysis in terms of the scalability of EEDBR is shown in Fig. 16.7.
The delivery ratio remained constant until 2000 rounds, after which it decreased.
With 400 nodes, the highest average throughput was obtained. With 100, 250, and
500 nodes, the average delivery ratios were 80, 88, and 82. Hence, 100 nodes had
the lowest average throughput.
16.7.2 End-to-End Delay
The end-to-end delay analysis in terms of the scalability of the EEDBR technique is
depicted in Fig. 16.8. The maximum delay was 80 in the case of 500 nodes at round
1000, whereas the minimum delay was 5 for 100 nodes at round 5000. Overall, the
lowest delay from 1000 to 5000 rounds was for 100 nodes. With 500 nodes, the
delay was higher at each round compared to the other node numbers.
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 183
Fig. 16.8 End-to-end delay scalability of EEDBR
16.7.3 Path Loss Analysis
The path loss analysis in terms of the scalability of EEDBR is shown in Fig. 16.9.
The highest average path loss was for 500 nodes. The highest path loss was at round
1000 for 100, 250, 400, and 500 nodes. The lowest path loss was at round 5000 for
100, 250, 400, and 500 nodes.
16.8 Summary of Work and Future Directions
In this chapter, we examined how to increase the stability period and throughput
of UWASNs, as well as how to decrease delay. In mathematical work, we have
also addressed the channel conditions and formulated three different cost functions.
These cost functions were derived using the various layers of water depth. These
cost functions are totally dependent on the path loss occurring in the dense
underwater environment. In future, our focus will be on energy consumption and
implementing energy-harvesting concepts.
184 S. S. Rahim et al.
Fig. 16.9 Path loss scalability of EEDBR
1. Khan, F., Ur Rehman, A., Usman, M., Tan, Z., & Puthal, D. (2018). Performance of cognitive
radio sensor networks using hybrid automatic repeat ReQuest: Stop-and-wait. Mobile Networks
and Applications, 23(3), 1–10.
2. Javaid, N., Jafri, M. R., Khan, Z. A., Qasim, U., Alghamdi, T. A., & Ali, M. (2014).
Iamctd: Improved adaptive mobility of courier nodes in threshold-optimized dbr protocol for
underwater wireless sensor networks. International Journal of Distributed Sensor Networks,
3. Noh, Y., Lee, U., Wang, P., Choi, B. S. C., & Gerla, M. (2013). VAPR: Void-aware pressure
routing for underwater sensor networks. IEEE Transactions on Mobile Computing, 12(5), 895–
4. Ahmed, S., Javaid, N., Khan, F. A., Durrani, M. Y., Ali, A., Shaukat, A., Sandhu, M. M., Khan,
Z. A., & Qasim, U. (2015). Co-UWSN: Cooperative energy-efﬁcient protocol for underwater
WSNs. International Journal of Distributed Sensor Networks, 2015, 75.
5. Javaid, N., Jafri, M. R., Ahmed, S., Jamil, M., Khan, Z. A., Qasim, U., & Al-Saleh, S. S. (2015).
Delay-sensitive routing schemes for underwater acoustic sensor networks. International
Journal of Distributed Sensor Networks, 11(3), 532676.
6. Zidi, C., Bouabdallah, F., & Boutaba, R. (2016). Routing design avoiding energy holes
in underwater acoustic sensor networks. Wireless Communications and Mobile Computing,
7. Pompili, D., & Akyildiz, I. F. (2009). Overview of networking protocols for underwater
wireless communications. IEEE Communications Magazine, 47(1), 97–102.
8. Liaqat, T., Javaid, N., Ali, S. M., Imran, M., & Alnuem, M. (2015). Depth-based energy-
balanced hybrid routing protocol for underwater WSNs. In 2015 International Conference on
Intelligent Networking and Collaborative Systems (INCOS) (pp. 262–267). Piscataway, NJ:
16 Scalability Analysis of Depth-Based Routing and Energy-Efﬁcient Depth-... 185
9. Yan, H., Shi, Z. J., & Cui, J.-H. (2008). DBR: Depth-based routing for underwater sensor
networks. In NETWORKING 2008 ad hoc and sensor networks, wireless networks, next
generation internet (pp. 72–86). Berlin: Springer.
10. Wahid, A., Lee, S., Jeong, H.-J., & Kim, D. (2011). Eedbr: Energy-efﬁcient depth-based
routing protocol for underwater wireless sensor networks. In Advanced computer science and
information technology (pp. 223–234). Berlin: Springer.
11. Ahmed, S., Javaid, N., Ahmed, A., Ahmed, I., Durrani, M. Y., Ali, A., Haider, S. B., & Ilahi,
M. (2016). SPARCO: Stochastic Performance Analysis with Reliability and COoperation for
underwater wireless sensor networks. Journal of Sensors, 2016, 17.
12. Jafri, M. R., Ahmed, S., Javaid, N., Ahmad, Z., & Qureshi, R. J. (2013). Amctd: Adaptive
mobility of courier nodes in threshold-optimized dbr protocol for underwater wireless sensor
networks. In 2013 Eighth International Conference on Broadband and Wireless Computing,
Communication and Applications (BWCCA) (pp. 93–99). Piscataway, NJ: IEEE.
13. Javaid, N., Hussain, S., Ahmad, A., Imran, M., Khan, A., & Guizani, M. (2017). Region based
cooperative routing in underwater wireless sensor networks. Journal of Network and Computer
Applications, 92, 31–41.
14. Ahmad, A., Ahmed, S., Imran, M., Alam, M., Niaz, I. A., & Javaid, N. (2017). On
energy efﬁciency in underwater wireless sensor networks with cooperative routing. Annals
of Telecommunications, 72(3–4), 173–188.
15. Ali, M. T., Rahim, S. S., Jan, M. A., Ishtiaq, A., Ahmed, S., Ahmad, M., Khan, M., &
Ayub Khan, M. (2018). Dist-Coop: distributed cooperative transmission in UWSNs using
optimization congestion control and opportunistic routing. International Journal of Advanced
Computer Science and Applications, 9(6), 356–368.