End-to-End Delay and Energy Efficient Routing Protocol for Underwater Wireless Sensor Networks

Wireless Pers CommunDOI 10.1007/s11277-014-1859-z

End-to-End Delay and Energy Efficient Routing Protocolfor Underwater Wireless Sensor Networks

Tariq Ali · Low Tang Jung · Ibrahima Faye

© Springer Science+Business Media New York 2014

Abstract Providing better communication and maximising the communication performancein a Underwater Wireless Sensor Network (UWSN) is always challenging due to the volatilecharacteristics of the underwater environment. Radio signals cannot properly propagateunderwater, so there is a need for acoustic technology that can support better data ratesand reliable underwater wireless communications. Node mobility, 3-D spaces and horizontalcommunication links are some critical challenges to the researcher in designing new rout-ing protocols for UWSNs. In this paper, we have proposed a novel routing protocol calledLayer by layer Angle-Based Flooding (L2-ABF) to address the issues of continuous nodemovements, end-to-end delays and energy consumption. In L2-ABF, every node can calcu-late its flooding angle to forward data packets toward the sinks without using any explicitconfiguration or location information. The simulation results show that L2-ABF has someadvantages over some existing flooding-based techniques and also can easily manage quickrouting changes where node movements are frequent.

Keywords Propagation delays · 3-D spaces · Routing algorithms · UWSN · L2-ABF ·Flooding angle · Horizontal communication

1 Introduction

A reliable Underwater Wireless Sensor Network (UWSN) can provide good services to appli-cations that operate under the constraints of the volatile underwater environments. Some of

T. Ali (B) · L. T. JungCIS Department, Universiti Teknologi PETRONAS, Tronoh, Perak, Malaysiae-mail: [email protected]

L. T. Junge-mail: [email protected]

I. FayeFAS Department, Universiti Teknologi PETRONAS, Tronoh, Perak, Malaysiae-mail: [email protected]

123

T. Ali et al.

these environments may not be feasible for human presence due to unpredictable under-water conditions, such as high water pressure and vast areas of water [1]. The UWSN is abranch of terrestrial-based sensor networks, and it has some similarities to terrestrial-basedwireless sensor networks, like large numbers of nodes and node energy issues. Energy sav-ings is a main concern in UWSNs because sensor nodes are usually powered by batterieswhich are not easy to replace or recharge. Acoustic wave communication is employed asthis is the only viable solution for underwater sensor networks. With this, UWSN has tobear some cost in the form of speed which is five magnitudes less than the radio channel.Moreover, the cost of an underwater sensor node is normally higher than the ground-basedwireless networks. Another issue with underwater networks is that most of the sensor nodestend to passively move with the water currents [2]. These limitations make the terrestrial adhoc network’s protocols not suitable for UWSN. The limitations mentioned above demandsome newly designed protocols for UWSN. Many researchers have been designing new rout-ing protocols for UWSN to support the key characteristics of underwater communications[3,4].

In flooding-based routing techniques, such as [5,6], each node broadcasts a data packettowards its neighbours or throughout the whole network. Therefore, in the event of a nodefailure or data packet losses, other nodes can rebroadcast the same data packets many times.Due to this frequent broadcasting, the result is end to end delay and more energy consump-tion.

In this paper, we therefore, proposed our routing protocol (named L2-ABF) in detail. Theangle-based flooding architecture is used in L2-ABF. In this protocol, we have assumed thateach node can compute its angle-based flooding zone to prevent the flooding of the wholenetwork. Although some angle based flooding techniques already exist, this technique mayhave some advantages like:

• L2-ABF does not require any location information of the sensor nodes.• It does not depend on a special network setup and hardware.• In L2-ABF, nodes use a very simple calculation method to forward data packets.• It is not based on any limitation of deployment of the network.

The L2-ABF protocol is scalable, and delay and energy efficient. It uses an angle-basedflooding architecture in which multi-sinks are anchored on the water’s surface to collect datapackets. The ordinary nodes are deployed on different depth levels from the surface to thebottom in the form of layers. If any node collects the information and tries to send it towardthe upper layer nodes using the angle-based zone and data packet received on one of the sinkson water’s surface, it is considered to be delivered successfully.

2 Related Work

Many routing protocols for UWSNs have been proposed over last ten years. In this sec-tion, we present some related protocols, such as VBF, DFR and DBR. The Vector-BasedForwarding (VBF) protocol has been suggested in order to solve the problem of high errorprobability in dense networks. VBF uses a vector-based forwarding mechanism to forwarddata packets from the source to the destination. A vector is computed from the source towardsthe destination and the nodes inside the computed radius of the vector can only participatein communication. The limitation of VBF lies in the assumption of the localisation of thesensor nodes. The selection of the forwarding nodes are based on the predefined radius of the

123

Underwater Wireless Sensor Networks

vector. Due to this selection, the communication of the nodes can be affected in the sparsenetwork.

In [6], the authors proposed the Directional Flooding-Based Routing (DFR) protocol. InDFR, a packet’s transmission is achieved in a type of scooped flooding in order to providethe high reliability. However, the set of nodes which participate in flooding the packet (thenodes create a so-called flooding zone) is limited to prevent the packet from being floodinginto the whole network. The flooding zone is decided by the angle between the sender andthe destination. The DFR uses two assumptions to complete its transmission round. First,every node in the network knows its location information and second, each node can calculatethe link quality, which is not possible without the geographical location information of thesensors nodes.

Since most routing protocols utilise the concept of geographical routing, they have a keylimitation in that a localisation technique is assumed, which itself is a crucial issue to besolved. In addition, some non-localisation-based protocols such as DBR [3] and H2-DAB[2] were proposed for UWSNs. DBR does not rely on localisation but it suffers from theredundant transmission of packets, void regions and the absence of energy balancing. Thevertical movements are very little and normally ignored [7] in UWSNs, and the sensornodes have the same depth. In DBR, since packets are forwarded based on the depth ofthe sensor nodes, the nodes having smaller depths participate in forwarding the packetsmost of the time. Consequently, the nodes having smaller depths die earlier than the othernodes. Furthermore, the depth sensors increase the overall cost of the network and theyalso consume energy. In H2-DAB, the communication is performed based on the uniqueIDs (called HopIDs) assigned to each sensor node. The consideration of HopID only is notsuitable for UWSNs which are resource constrained networks. Similar to DBR, H2-DABallows the nodes having smaller a HopID to be frequently used for data forwarding. Hence,the nodes having small HopIDs die earlier than the other nodes. In addition, H2-DAB usesRTS/CTS for data forwarding, which is expensive in terms of energy and delay [8]. In theabove mentioned protocols, none of them has taken into account the residual energy and thephysical distance towards the sink node. In particular, DBR uses the depth of the sensor nodesas the metric for forwarding. Hence, it is likely that the packet can be stuck after reachingthe surface, due to the drifting of the sensor nodes away from the sink in the horizontaldirection depending on the water currents. In our proposed protocol, however, each node cancompute its flooding-4based zone towards the sink node. Therefore, it is assured that thereexists a path towards the sink node. Furthermore, the consideration of the residual energybalances the work-load among the sensor nodes. Consequently, our proposed protocol canachieve performance improvements in terms of end-to-end delay, data delivery ratios andenergy consumption.

The main difference in our proposed protocol and other flooding-based protocols is thatL2-ABF does not depend on any type of location information or geographical location.

3 Challenges in UWSNs

When considering UWSNs, a due consideration must be given to the possible challengesthat may be encountered in the subsurface environment. Continuous node movement anda 3D topology are major issues posed by the host conditions. Most of the time, sensornodes are considered to be static but sensor nodes underwater, in fact, can move up to1–4m/s due to different underwater activities [9,10]. Further, radio communications are notsuitable for deep water; accordingly, it requires acoustic communications as a substitute.

123

T. Ali et al.

Due to this substitution, the available propagation speed is five orders of magnitude lessthan the radio frequency as the characteristics of the communication are shifted from thespeed of light to the speed of sound [11]. Moreover, some of the underwater applications,including detection or rescue missions, tend to be ad hoc in nature; some require networkdeployment not only in a short amount of time, but also without any proper planning. Insuch circumstances, the routing protocols should be able to determine the node locationswithout any prior knowledge of the network. Moreover, the network should also be capableof reconfiguring itself with dynamic conditions in order to provide an efficient communicationenvironment.

Other than these, some more challenges in these environments necessarily need to beunder concern. Some of the underwater applications, such as rescue and detection missions,can require deploying the network in a short amount of time, without any proper planning.In this scenario, the routing protocols should be able to determine the node locations withoutany prior knowledge of the network. Additionally, the network should be able to configureautomatically and be capable of reconfiguring itself dynamically to provide an efficientcommunication environment.

Cost is an important issue for Acoustic Networks [12]. A modem for acoustic commu-nications currently costs around $2k. Even, underwater sensors can be more expensive thanthe modem itself. Supporting hardware such as underwater cable connectors also drive upcosts as their price is around $100. Another reason that can increase the cost is the sophisti-cated construction required in order to survive against such a harsh environment. The pressureincreases by an additional atmosphere for every 10m of depth, so even a shallow-water (around100 m) instrument must be able to withstand 10 atmospheres of pressure, while deep-waterinstruments (around 4 km) must be rated to at least 400 atmospheres [13]. Significantly lessexpensive sensors, vehicles, and modems (500 m-range acoustic and very short-range opticaland radio) are being designed and built. These efforts may change the economics for denseunderwater sensor networks in the near future.

Furthermore, a significant issue in selecting a system is to establish what the real rangeand data rate will be for a specific use. A system designed for deep-water may work poorlyin shallow water or even when configured for too high a data rate when reverberation ispresent in the environment [14]. For establishing the upper performance bound, the man-ufacturer’s specifications of the maximum data rates are useful but frequently unachiev-able, particularly in some challenging acoustic environments. The well-funded users haveresorted to purchasing and testing multiple systems in specific environments to determine ifthe systems will meet their needs. An international effort to standardise the tests for acousticcommunication systems is needed [15]. However, to undertake this effort is costly. In otherwords, it seems to be difficult for an impartial body to establish; while, private organisa-tions or government institutes who perform such comprehensive tests tend to not publish theresults.

4 Problem and Network Architecture

We considered the monitoring applications for different objectives and goals like underwateroil/gas. For this purpose, sensor nodes were deployed in the monitoring area in order tocollect the information from the surroundings. As mentioned earlier, our routing protocolbased on a multi-sink architecture was used to increase the reliability of the network andthe data delivery ratios. This architecture was not only helpful to increase the data deliveryratios but it was also able to increase the network lifetime by reducing the energy con-

123


sumption of the nodes around the sinks. The surface sinks were equipped with a radio andan acoustic modem. The RF modem was used to communicate with each other as well aswith the final data processing centre while the acoustic modems were used to communicatewith the sensor nodes deployed underwater. The sensor nodes were deployed at differentdepth levels through a bouncy control mechanism [2,16]. It was considered that the nodescould move freely in a horizontal direction with different water currents while the verti-cal movements of the nodes were neglected due to the very small scale of the movements’variations.

By doing so, we deployed sensor nodes in the form of layers from the surface to the bottom.The numbers of layers depended on the depth of the monitoring area and the communicationrange between the layers. The calculated average depth of the ocean was approximately2.5–3 km [17,18]. However, if we considered this average depth of the ocean and everylayer at 500 m, the maximum 5–9 numbers of layers would have been required to send datapackets from the bottom to the surface. It is important to note, that the performance of ourrouting protocol does not depend on the number of layers. The proposed algorithm can easilysupport more layers. However, if we were to increase the number of layers, it would increasethe cost of network as more nodes would be required to explore the same depth. We coulduse less numbers of layers as acoustic communication can support the maximum of up to5 km but it is not preferred as a long communication range might be a cause for drainingmore energy [19,20]. In our protocol, to save energy and extend the network lifetime, a 500 mtransmission range was used as the distance between the layers. Although, in special cases,nodes could increase this transmission range but in a normal scenario there is no need toincrease the suggested range. To control end-to-end delays, we preferred the vertical anddiagonal communication to forward the data packets toward the upper layers of nodes. Thehorizontal communication between the sensor nodes on the same depth levels is known tocause an increase in the data routing path from the lower layer nodes to the surface sinkswhich are deployed on the water surface. The large data routing path is highly likely toincrease the end to end routing delay in the underwater data transmission and brings alongwith it energy issues.

5 L2-ABF Protocol

The proposed L2-ABF protocol is aimed to reduce the horizontal communication betweenthe peer sensors on the same depth levels in the underwater sensor network. Angle-basedflooding architecture was used in L2-ABF to achieve this goal. L2-ABF completes its taskin two phases. In the first phase, a Layer-ID is assigned to the sensor nodes by the sink nodein the network. In the second phase, the nodes forward sensed data towards the upper layernodes.

5.1 Layer-ID Assigning Mechanism

In this phase, each sensor node is assigned a Layer-ID. The process is as follows. Eachsink node broadcasts a Hello Packet (HP) towards the sensor nodes. Each sensor node willget a Layer-ID from the received HP in order to assign the Layer-ID to each node in thenetwork to rebroadcast the HP. Hence, the nodes rebroadcast the HP by a decrement ofone in the Max-Layer-count field. The HP format is shown in Fig. 1. Upon receiving therebroadcast HP, the nodes will get a Layer-ID by accumulating the Layer-IDs which havebeen received from the previous senders. This process will continue until all nodes are

123

T. Ali et al.

Fig. 1 Layer-ID assigning mechanism

assigned a Layer-ID. After completion of this process, the Layer-ID will be static untilthe sink nodes send the next update after a specific time interval. The vertical movementis ignored in our model. Therefore, there is no need of an update of the Layer-ID veryfrequently by the sinks. The sensor nodes will be considered in the first layer of thosethat receive the HP first from the sink node, and rebroadcast the HP by a decrement ofone in the counter field. The second decrement by the nodes in the counter field is con-sidered as in the second layer and the last decrement in the counter field will be con-sidered the last layer of nodes in the network. When the Max-Layer-ID becomes zero,further rebroadcasting will be stopped. The Layer-ID assigning algorithm is presented inSect. 5.4.

Figure 1 shows that all sink nodes on the water’s surface have a fixed Layer-ID “0” and itstay static throughout the lifetime of the network. The sinks broadcast the HP to the ordinarynodes with the value of the counter field “9”. The ordinary nodes N1 to N6 received the HPdirectly from the sinks and were assigned the Layer-ID “1” and rebroadcast the HP withthe decrement of one in the counter field. The nodes N7–N12 received the HPs from theupper layer nodes and were assigned the Layer-ID “2”. This process continued until thenodes N49–N54 were assigned the Layer-ID “9” and the counter field became zero. Whenthe counter field became zero the node stopped further flooding of the HP and this was lastlayer of the network. Here, it is important to note that when the nodes received multiplecopies of the HPs as N4 in the first layer; N8 in the second layer; N13, N15 and N17 inthe third layer and N51 and N52 in the last layer, these nodes compared the value of thecounter field of the HPs. If the counter field value was same, they took one hello packet into

123


Hello Packet Type Sink-IDs Sk-Layer-ID Max-Layer-count

Fig. 2 Hello Packet format used by surface sinks

Hello Packet Type S-Node-ID Layer-ID Energy-Status

Fig. 3 Hello Packet format used by ordinary nodes

R-Node-ID S-Node-ID Layer-ID Priority

Fig. 4 Hello Reply format used by ordinary nodes

account and discarded the other HP. If the HP had different counter field values, the nodeschose only the packets with the greater value in the counter field and discarded the otherspackets.

5.2 Hello Packet Formats

L2-ABF uses two types of HPs; one is used by the sinks to assign the Layer-ID to the sensornodes and the other is used by the sensor nodes to forward the data packets towards the sinknodes on the water’s surface. The HP generated by the sinks, contains the following fieldsas shown in Fig. 2. Hello Packet Types: will be used to identify the type of HP received;it is either generated by the sinks or from the ordinary sensor nodes. Sink-ID (s) will beused when every sink broadcasts HPs during the first phase then the Sink-ID will help theaccepting nodes to differentiate the HPs received from the multiple sinks. A Sk-Layer-ID isused by the sensor nodes to identify the destination.

Lastly, the Max-Layer-Count field has a value of nine at the start when the sink broad-casts the HP. After receiving it, every node will apply the decrement of one; so, when theninth node receives this HP, it will make this value zero and will not forward it furtherto any other node. The HP generated by the ordinary node contains the following field asshown in Fig. 3. The S-Node-ID is used to identify the sender node and the Layer-ID is theinformation of the current layer of the sender node. On the basis of this information, thereceiving node will decide either the data packets will be forwarded or simply discarded.The Energy-Status is used to identify the healthiness of the sender nodes. Figure 4 showsthe Hello Reply (HR) format; the R-Node-ID is the information of the replying node tothe HP. The purpose of S-Node-ID and the Layer-ID is the same as in the HP. The pri-ority field contains the information which is used by the sender node of the data packetsto give the priority of the replying node to become the next forwarder of the data pack-ets.

5.3 Data Packet Format

L2-ABF uses the Layer-ID to rout the data packets and the Node-ID is the identifier ofthe nodes, separately. The Node-ID is a unique address for every node in the network. Thedata packet format used in L2-ABF is illustrated in Fig. 5. The packet header consists ofthree fields: Sender-ID, Layer-ID, and Packet Sequence Number. The “Sender-ID” is the

123

T. Ali et al.

Sender-IDLayer-ID

Packet Sequence No. Data

Packet header

Fig. 5 Data Packet format

identifier of the source node. The “Packet Sequence Number” is a unique sequence numberassigned by the source node to the packet. Together with the Sender-ID, the Packet SequenceNumber is used to differentiate packets in later data forwarding. The “Layer-ID” is theinformation of the recent forwarder which is updated Layer-by-Layer when the packet isforwarded.

5.4 Algorithm 1: Assigning the Layer-ID to the Deployed Nodes

HP broadcast From all Sinks (Sink-hp) with fixed Layer-ID ‘‘0’’ // HP received1. Layer-ID = 0 // New node, when it has no layer id, it will be considered 02. Layer-count = 13. n= 9 // Initialisation number of layer in the network4. If ( Packet-Type == “Sink-hp”) && (Layer-counter < n)5. If (Layer-ID==0)6. Layer-ID = Layer-counter // Every new node will get Layer-ID 7. Else8. If (layer-ID Layer-counter)9. Discard Data Packet // Nodes already have Layer-ID10. Else11. Layer-ID = Layer-counter12. Layer-counter ++13. Broadcast Sink-hp further14. End IF15. End If16. End If17. Max Layer Count = 9 // In order to stop further broadcast18. No further broadcast for this Sink-hp19. Exit

6 Data Packet Forwarding

The angle-based flooding approach is used in our proposed routing protocol. This routingmechanism is not based on sensor node location information and has been designed fordelay and power-efficient multi-layer communication in underwater acoustic networks. Inthis routing mechanism, there is no need for a sender node to know its own location or thelocation of the final destination (Sink) before transmitting the data packets. Anchor nodesflood the sensed data towards the surface sinks via the upper layer nodes. The forwardernode will define the flooding zone by using the initial angle � = 90 ± 10K as shown inFig. 6. Here, K is a variable and has a finite set of values, K, (1, 2, . . . , 8). After defining theflooding zone, the node will send HP within the defined zone and wait for the HR. If thereis no HR received, the node will increase the value of K in the initial angle, to increase itsflooding zone until the basic condition is met (0 < � < π). Here, we are assuming that

123


Fig. 6 Data Packet forwarding mechanism

after the completion of one round by using the different values of the variable k and thenode did not receive any HR, the nodes can send data packets directly to the sink nodes onthe maximum power. Here, it is important to note that nodes can use random values of thevariable K to increase the size of the flooding zone. The randomness of the K value is morehelpful to control the End-to-End delays as well as the power consumption of the nodes. Theselection of the random values for K depends on the movement of the nodes.

• If a node receives a very small number of HRs, the nodes will be considered to have aslow movement. So, the nodes will use a large value of the variable k to increase theflooding zone.

• If a node receives more HRs, the nodes will be considered to have a fast movement. So,the nodes will use a small value of the variable k to increase the flooding zone.

6.1 Data Packet Forwarding and Priority Calculation

L2-ABF makes use of the flooding-based method as was discussed previously. Becauseof this, at the next layer there may be several nodes that could be suitable forwarders ofa packet. However, high collision and high energy consumption will occur if all of thesesuitable nodes attempt flooding the packets into the network at the same time. Thus, thenumber of forwarding nodes must be managed so that collisions and energy consumptioncan both be reduced. Moreover, L2-ABF has an inherited multiple-path feature (a floodingmanner using an angle-based calculation scheme is used by every sensor node to forwardpackets) so the same packet may be received by several nodes at the same time. Therefore,the packet may be sent more than once. In an ideal situation, to improve energy efficiency,the data packet is flooded by only one node. In order to accomplish this, the concept ofmultiple queues has been introduced. There are two key reasons for needing multiple queuesfor energy savings and the reduction of collisions. The first is that the same data packetsmay be forwarded by several nodes. The second reason is that the same packet may besent by a node more than once [3]. The number of forwarding paths can be controlled byusing a priority queue so that the number of forwarding nodes can be reduced. L2-ABF usesa packet history buffer that guarantees that packet is forwarded by a node only once in acertain period of time; this in turn, solves the second problem. The history buffer can holdthe maximum of 50 data packets. A new packet will take the place of the least used packetwhenever the buffer queue is full. The values necessary to calculate the priority for the next

123

T. Ali et al.

forwarder are the only things contained in the priority queue. These values are based itslayer number and energy status. The algorithm for the data packet forwarding, calculationof the priority and the multiple forwarding controls is presented in Sects. (6.2) and (6.3)respectively.

6.2 Algorithm 2: Data Packet Forwarding

Algorithm {Forward Data Packet (DP), HP, HR}

1. Check the status of DP in Q2 //Q2 is a buffer history2. If (DP in Q2)3. Discard the packet // Packet already sent4. END if5. If (DP not in Q2)6. Calculate the Flooding Zone using = 90±10K // K is variable that has a set of

values [1,2…..8] 7. Send HP // Inside the defined flooding Zone8. Wait for HR9. If (HR received) = yes10. Send DP11. Go to rest mode//Acknowledge Node is Qualified for further flooding12. Else13. If (K 8)14. K++15. Go onto step 6 // increase the size of the flooding zone with increasing the value of “K”16. End If17. End If18. End

6.3 Algorithm 3: Priority Calculation to Become Next Forwarder of Data Packets

MinTh = 8% // Minimum thresholdMaxTh = 60% // Maximum threshold

1. If (Packet-Type == “HP”)&&( L_Rec < L_Snd)2. Get own Layer ID // Here we assume every node knows its own layer number3. MyEng = Int_ Eng – Cons_Eng // Node computes its own energy4. If (MyEng > MaxTh)5. Set_Priority= “Highest”6. Priority = Set_Priority // Send this Priority inside the HR7. Layer-ID= J // Send this Layer-ID inside the HR8. Send HR9. Else If (MinTh < MyEng < MaxTh)10. Set_Priority= “Medium”11. Priority = Set_Priority12. Layer-ID= J13. Send HR 14. Else 15. No Respond 16. End IF 17. Else If (Packet-Type == “HR”)18. Get Layer-ID and Priority // Get layer id and priority from the HR19. for( i=1; i TRP; i++) //TRP= Total number of received packets20. Store[i] Layer-ID21. If( size[store]==1)

123


22. Send DP to this node 23. Else 24. Temp = Min(Store)25. If (Priority of Temp==“Highest” ) 26. Send DP to this node 27. Q2 Copy of DP // Q2 is a buffer history to control multiple sending of packets28. Q1 ID of HR // Q1 is a priority queue to control multiple forwarders 29. Else 30. If (Priority of Temp ==”Medium” ) 31. Send DP to this node 32. Q2 Copy of DP // Q2 is a buffer history to control multiple sending of packets33. Q1 ID of HR // Q1 is a priority queue to control multiple forwarders34. End If 35. End If 36. End

6.4 Waiting Time Calculation

This section describes the average waiting time by the node in five attempts to forward the datapacket. When a forwarding node cannot find the next node from the upper layers after the firstattempt, the source node will make another four attempts by increasing its flooding zone angleto send the current data packets towards the upper layer nodes. We defined the range of thevalues (between 0 and 100) to calculate the maximum waiting time to forward the data packetstoward the upper layers where 0 means no wait and 100 can be the maximum waiting time inthe worst case. We assumed that the node would wait t1 time before going for another attempt.It depends on the number of nodes that replied in the first attempt; t1 can be calculated as:

t1 = K

n1 + 1(1)

Where K is a constant that has the maximum value of the waiting time and n1 is the numberof nodes that replied in the first attempt. If there is no reply from the upper layer nodes, itwill wait t2 time, to go for the second attempt by increasing the angle of the flooding zone.

The t2 time is:

t2 = K

n2 + 2(2)

Where K is a constant that has the maximum value of the waiting time and n2 is the numberof nodes that replied in the 2nd attempt. If it still cannot find any nodes from the upper layers,it will go for a third waiting time, t3:

t3 = K

n3 + 3(3)

Where K is a constant that has the maximum value of the waiting time and n3 is the numberof nodes that replied in the 3rd attempt. If it still cannot find any nodes from the upper layers,it will go for a fourth waiting time of t4, and t4 is:

t4 = K

n4 + 4(4)

Where K is a constant that has the maximum value of the waiting time and n4 is the numberof nodes that replied in the 4th attempt. If it still cannot find any nodes from the upper layers,it will go for a fifth waiting time of t5, and t5 is:

123

T. Ali et al.

t5 = K

n5 + 5(5)

Where K is a constant that has the maximum value of the waiting time and n5 is the numberof nodes that replied in the 5th attempt. Every time depends on two factors: first, the numberof nodes that replied after the other attempt and second, the difference between the numberof nodes in the first and second attempt. So, the average of these times is:

T =[

K|n2−n1+1| + K

|n3−n2+2| + K|n4−n3+3| + K

|n5−n4+4| + Kn5+1

]

5(6)

Equation (6) can be written as:

T = K

⎡⎣

∑4i=1

[1

|ni+1−ni |+1

]+ 1

n5+5

5

⎤⎦ (7)

From Eq. (7), it is clear that the waiting time can be calculated by the availability of neigh-bours’ nodes and the frequency of their changing in location due to the mobility of the nodesby the water current.

7 Mathematical Model for Power Consumption

We assumed N as the total number of sensor nodes to explore area A where these nodes weredeployed in the form of layers from the surface to the bottom. We assumed every node hadan initial power level I−C unit. We considered the power consumptions for the data packetsas included control packets, like the HP and HR. In the scenario, nodes only sent the HRthat was inside the flooding zone. Both control packets were of the same size and consumedvery little power as compared to the data packets. We used some mathematical symbol tocalculate the power consumption in our model as given below with a description.

N Total number of nodes in the network

A Defined monitoring area in the network

I−C Initial power level for each sensor node in the network

P Complete power consumption to for-ward the data packet from the node tothe sink on the water’s surface

Pe Power consumed by the nodes to send data packet from one layer to the next upper layer

Pc Power consumed by control packets including HP and HR

Pı̀ → s Total power consumed when ı̀ data packets are forwarded towards the sink

m Number of layers in the network (divide the depth into layers)

n Number of sensor nodes in each layer

D Total number of data packets generated in the network

K=D/N K is the total number of data packets generated by one node

Pı̀ The power consumption at the ı̀th layer

Tı̀ The lifetime of the layers

Tı̀/n The lifetime of each sensor node belonging to the ı̀th layers

123


We checked the power consumption for both cases, static nodes (Best case) and mobile nodes(Worst case).

7.1 Power Consumption with Static Nodes (Best Case)

In the best case, all of the nodes are static and every node in the network will send only oneHP within the calculated flooding zone and also get a single HR. After receiving the HR,the node will save the ID of the replying node in its routing table to send the remaining datapackets all of the time. In the first attempt, the power consumed by the node to send one datapacket from the lower layer to the next upper layer is:

P = 2pc + pe (8)

Here, pc + pe are the power consumed by the current layer which has data packets. The pcis the power consumed when the node sends the HP and the pe is consumed to forward thedata packets. The reaming pc is power consumed by the upper layer to respond the HP. First,we considered the case of the node at the first layer which has a data packet and can forwardit directly to the sink on the water’s surface and then the second layer node forwards datapackets toward the sink through the first layer and so on. The power consumption at eachlayer can be calculated by the following equations.

P1→s = (pc + pe

)P2→s = (

2pc + 2pe)+

(pc + pe

)P3→s = (2pc + 3pe) + (

2pc + 2pe) + (

pc + pe)

...

Pm−1→s = (2pc + (m − 1) pe) + (2pc + (m − 2) pe) + . . . . . . . . . . . + (2pc + 2pe)

+ (pc + pe

)Pm→s = (2pc + m.pe) + (2pc + (m − 1) pe) + · · · + (2pc + 2pe) + (pc + pe)

(9)

Equation (9) shows that the upper layers consume more power due to their frequent use.When one node at every layer generates one data packet and forwards it to the sink, for “m”layers there will be m data packets. It is clear that the first layer will process all m data packetstoward the sink and consume the power of (2pc + m.pe) more than the other layers. The mthlayer consumes the least power as it will forward only one data packet to the next upper layer,i.e., the power is (pc + pe). When a node generates k data packets on each layer, the powerequation is now given by:

Pk·m→s = (2pc + K · m · pe

) + (2pc + K · (m − 1) pe

) + · · ·+ (

2pc + K · 2pe) + (

pc + K · pe)

(10)

From Eq. (10), the power consumption at the ı̀th layer can be calculated when the layerprocesses its own generated data packets as well as forwards the data packets from all lowerlayers toward the sink on the water’s surface. That is, the power at the ith layer is given as:

Pı̀ = (m − ı̀)K · pe + K · pe + 2pc where ı̀ < m

Or, it can be written as:

Pı̀ = (m − ı̀) β + β +2pc = (m − ı̀ − 1) β +2pc

Where β is to denote the K.pe

123

T. Ali et al.

Now the lifetime of the ith layer can be calculated by :

T = n.I−C(m − ı̀ − 1

)β+2pc

(11)

From Eq. (11), the lifetime of the nodes at the ı̀th layer can be calculated as:

Ti/n = I−C(m − ı̀ − 1) β +2pc

(12)

7.2 Power Consumption with Mobility of Nodes (Worst Case)

We considered the worst case of our scenario as the mobility of the nodes; to calculate thepower consumption, Eq. (8) becomes:

P = pe + (n + 5)pc (13)

In Eq. (13), (pe + 5pc) is the power consumption in the worst case, from the current layer,when the node has to make 5 HP using different values of the variable K, to increase theflooding zone size. It is possible that no node replies after sending the first 4 HP but allreply after the 5th attempt. In such a case, the power consumption will be n · pc in terms ofHPs sent. Here, it should be clear that we only considered the worst case in terms of powerconsumption not in terms of node failure.

Further, we considered the case that the data packets generated are by the first layer andthen the lower layers, consequently. All layers forward data packets toward the surface sinkthrough their respective upper layer. The power consumption can be calculated as:

P1→s = (pe + 5pc)

P2→s = (2pe + npc + 5.2pc) + (pe + 5pc)

P3→s = (3pe + 2.pc + 5.3pc) + (2pe + npc + 5.2pc) + (pe + 5pc)

...

Pm−1→s =((m−1)pe+(m−2) · npc+(m−1)5pc)+· · ·+(2pe+npc+5.2pc) + (pe + 5pc)

Pm→s = (mpe + (m − 1) · npc + (m · 5pc) + · · · + (2pe + npc + 5.2pc) + (pe + 5pc)

(14)

From Eq. (14), it is clear that the mobile node effect is the same as in Eq. (2) for thestatic nodes. Similarly, when one node of each layer generates K data packets, the Eq. (14)is now as:

Pk·m→s = ((2 · K · m · pe − 1

) + K (m − 1) npc + K · m · 5pc)

+ · · · + ((4K · pe − 1

) + K · npc + 2k · 5pc) + ((

2K · pe − 1) + K · 5pc

)

(15)

Equation (15) can be used to calculate the power consumption at the ith layer, as its owngenerated data packets as well as the data packets of the lower layers forwarded through it.

Pi = (2K (m − i + 1) pe − 1

) + K (m − i) npc + K (m − i + 1) 5pc where i < m

Pi =(2K (m −i + 1) pe−1

)+ K (m − i) npc+ K (m − i + 1) 5pc+(2K · pe− 1

)+ K · 5pc

Pi = (m − i)[(

2K · pe − 1) + K · pc + K · 5pc

] + (2K · pe − 1

) + K · 5pc

123


We used β to denote the (2K · pe − 1) + K · pc + K · 5pc. Now we can write the aboveequation as:

Pi = (m −i) β+ (2K · pe − 1

) + K · 5pc (16)

From Eq. (16), we can calculate the lifetime of the i th layer.

Ti = n · I−C(m − ı̀

)β +2K · pi − 1) + K · 5pc

(17)

Here, from Eq. (17), we can calculate the lifetime of a node at the i th layer.

Ti/n = I−C(m − ı̀

)β +2K · pi − 1) + K · 5pc

(18)

Equations (11) and (17) show that increasing the value of i means it is going deeper intothe layer thus making the value of (m − i) β decrease. On the other hand, the upper layer isfacing more power consumption as the number of layers starts to increase in the network.This bottleneck problem in the network can be addressed by using the proposed multi-sinkarchitecture.

8 Performance Evaluations

NS-2 was used for the performance evaluation of L2-ABF. 300 sensor nodes, both floatingand sink nodes, that were deployed in a 3D area of 1,000 m × 1,000 m × 800 m were takenfor use in our simulation. Multiple static sinks at the surface of the water were used with notonly Radio but also Acoustic types of communication which was a luxury. 300 m was themaximum possible distance between the layers of the floating nodes [21]. In this experiment,the sink nodes were considered to be static after the deployment but the remaining nodeswere floating in nature and the vertical movement of the floating nodes was ignored. Thehorizontal movement between the floating nodes with various water currents up to 1–4 m/sat fixed motions was the only movement considered. Table 1 presents the parameters for thesimulation.

Table 1 Simulation setup forL2-ABF

Simulation parameter setup Values

Number of nodes with 8 sinks 300

Deployment area in 3-D 1,000 m × 1,000 m × 800 m

Distance between layer of nodes 300 m

Consider fixed Motions of nodeshorizontally

2 m/s, 4 m/s

Data packet size 1,500 Byte

Transmission range 500 m

Protocol AODV

Antenna type Omni

Channel types Wireless

123

T. Ali et al.

Fig. 7 Simulations results. a average packets delivery ratios, b Average end-to-end delays, c average energyconsumptions

8.1 Performance Metrics

The data delivery ratio, end-to-end delay and energy consumption were the three matricesconsidered for the evaluation of the performance of our routing protocol. The ratio of thepackets that have been received successfully by all of the sinks on the surface of the water asgenerated by all of the sensor nodes in the network is known as the Packet Delivery Ratio.The average delay for all of data packets received successfully at the sinks on the surfaceof the water is known as the End-to-End delay. The measurement of the energy required foreach of the data packets to successfully reach its destination on the surface of the water isknown as the energy consumption.

8.2 Mobility of Nodes

Two different speeds of node movements (2, 4 m/s) and also static nodes were considered forthe data delivery ratio that is presented in Fig. 7a. Using the suggested number of nodes inthe network, the data delivery ratios were at 100 % and remained static. Furthermore, thesedelivery ratios did not even experience any major effect when the density of the nodes wasdecreased. If 30 % of the nodes were unavailable in the network, around 90 % data deliveryratios could be achieved, and even if 50 % of the nodes were unavailable, that amount ofsparseness of the nodes could still see an achievement of 85 % data packets on average nodemovements. Now, as shown in Fig. 7b, c, the end-to-end delays and energy consumptions

123


Fig. 8 Performance with different number of data packets. a Delivery ratios, b end-to-end delays

were analysed in regards to various node movements. In this case, a slight variation in thenode movement results could be seen as compared to the nodes that were static. There was aminor difference initially that began to rise when the number of nodes began to be reduced.However, even this difference was not very high until 50 % of the nodes had become a partof the network.

Throughout the scenario, only a minor difference was noted between the different nodespeeds. This clearly indicated that there was no serious effect on the end-to-end delays orthe energy consumptions caused by the mobility of the nodes. This was the only reason thata complex routing table for information regarding the location of the sensor nodes did notneed to be maintained, even if the position of a node was altered.

8.3 Performance with Different Loads

The end-to-end delays and the delivery ratios were analysed by producing more data packetsin the network so that the performance of L2-ABF with different loads being offered, couldbe checked. A network normally generates 1 packet per second; however, we consideredcases that were not normal. First, 3 packets/s were generated by the network; after that,the network generated 2 packets/s. The delivery ratio with various loads being offered ispresented in Fig. 8a. It shows that the delivery ratios with dense nodes were almost the same.It was only when the number of nodes started to become sparse that a difference began toappear. When the load offered was high and the network had fewer nodes, there were timesthat a next hop could not be found by a node so the number of packets began to increase inthe buffer. The result was that they began to be discarded. The end-to-end delay variationwhen the number of packets in the network was increased is shown in Fig. 8b. The figureindicates that it was easy for the network to handle a situation where even 50 % more packetsbecame a part of the network. Furthermore, when double packets had been generated in thenetwork, even these delays were affordable.

8.4 Analysis with Different K Values

Figure 9a, b shows the data delivery ratios and the end-to-end delays with different valuesof the variable K which were utilised by the nodes so that data packets could be forwardedtoward the sink nodes. These were used for the analysis. Slight variations can be seen in theresults with the various values of the variable K. There is very little effect on either the delays

123

T. Ali et al.

Fig. 9 Performance with different size of flooding zone. a Delivery ratios, b end-to-end delays

or the delivery ratios when a node uses a small value of the variable K to forward the datapackets; this is shown in the results. Initially, this is only a minor difference; however, it thenbegins to rise as the number of nodes begins to be reduced and the node makes use of a largevalue of the variable K. However, it is obvious that this difference is not very high until 50 %of the nodes become a part of the network.

9 Comparison with DBR

Most of the many routing methods that have been proposed are dependent on informationrelated to the location of the sensor nodes in a network. Getting this location informationis quite difficult when there is no GPS system available as in an underwater environmentwhere there is no GPS [22,23]. In DBR [3], location information is not needed as it is insteaddependent on the information regarding the depth of the sensor nodes so that it can forwardthe data packets on to the sinks on the surface of the water. Every node in DBR makes itsdecision based on its current depth in order to forward its data packets. The first thing thata node does when it has a data packet ready to be sent, is compare the embedded depth inthe data packets that have been received with its own current depth. The data packets areforwarded by the node if its current depth is less than the sender’s depth otherwise the datapackets are discarded. Likewise, the current depth is calculated by each node. Then, only thenodes that have a smaller depth level than the depth in the received data packets can forwardtheir data packets. All of the remaining nodes simply discard their packets. As compared toL2-ABF, DBR has some serious problems; two of these problems are as follow: The firstissue is that multiple nodes could possibly have smaller depth levels and the data packetscould be forwarded simultaneously. This would cause collisions in the network and result inthe overhead of power being increased. The second issue is that the problem of a void regionin the network cannot be handled by DBR. This is because there could be a situation whereno nodes qualify to become a forwarder. Such a case would be where the node has a higherdepth level as compared to the sender’s depth level. Because of the greedy fashion of thecurrent node, it would make attempts over and over again although there may be other routesavailable through higher depths [24,25].

Later, L2-ABF was compared with DBR so that their performance could be evaluated.The data delivery ratios with a single sink and multiple sinks were compared first. The results

123


Fig. 10 Performance Comparisons with DBR. a Delivery ratios, b end-to-end delays, c average energyconsumption

are shown in Fig. 10a. Under the condition of multiple sinks, both algorithms were foundto provide nearly identical results with the density of the nodes. A difference between thetwo was noted when the number of nodes was altered. The delivery ratios of DBR started todecrease as the number of nodes started to decrease. However, less effect was noted on thedelivery ratios of L2-ABF. This was caused by the greedy mode of DBR where nodes wereavailable in higher depth levels of the network but they were unable to become involved inthe data packet forwarding.

A surface sink was placed at the centre of the network for the data delivery ratios witha single sink to provide the comparison of both L2-ABF and DBR. The delivery ratios ofL2-ABF were less affected than the DBR delivery ratios as can be seen in the results. Thiswas once again caused by DBR’s greedy mode. In this case, data packets could only beforwarded by the nodes to the surface of the water without any consideration being taken asto the position of the sink node [26].

The comparisons for the end-to-end delays between DBR and L2-ABF are presented inFig. 10b. In this instance, the holding time was used in DBR and was responsible for L2-ABFdelivering data packets with less end-to-end delays than DBR when the network possessedreasonable sensor nodes. In order to control the multiple forwarding of the same packetsby different nodes in DBR, each node would hold the data packet for a short time periodbefore forwarding it. On the other hand, in L2-ABF, each node could forward data packets

123

T. Ali et al.

immediately in a short time period because it only used the angle-based zone to flood thedata packets.

Later the power consumptions of DBR and L2-ABF were compared with a differentnumber of sinks and the results are shown in Fig. 10c. First, fewer nodes were used for thecomparison and the results were almost the same. However, an increase began to be seenin the difference in the power consumptions with the start of the increase in the number ofnodes. Firstly, in DBR, the denseness of the network increased for the same data packetsbecause it used the broadcasting for each of the data packets in a greedy fashion. Secondly,in DBR, each node had to calculate its depth each time it needed to decide on whether tosend or discard the data packets it received. When there was a burst of data packets, thebroadcasting caused more and more power to be consumed during the process of deciding onwhether the data packets should be sent or discarded. The concept of the angle-based zonewas used to carry out the flooding of data packets in L2-ABF. Therefore, the nodes neededonly to calculate the flooding zone; then it flooded that zone with the data packets.

10 Comparison with VBF

The routing method that is the most commonly used for the performance evaluation of routingmethods is VBF [5,14,27]. For this work, it has been used so that the performance of thenew routing methods that have been proposed for sensor networks that are underwater couldbe compared. It is a position-based routing technique that can handle the problem of nodemovements that are continuous. The information about the state of the sensor nodes is notneeded, both, because it is a position-based technique and because only a small number ofnodes participate during the routing of the packets. The problem of packet losses and nodefailures are helped to be handled by the forwarding of data packets along redundant andinterleaved routes from the source to the sink. Each node is already assumed to know itsown location, and the location of all of the nodes involved including the source, forwardingnodes and final destination are carried by each packet. Here, the concept of a vector, such as avirtual routing pipe, has been proposed through which all of the packets would be forwardedfrom the source to the destination. The messages could be forwarded from their source to thedestination only by those nodes closest to this “vector” or pipe. With the use of this concept,there could be a significant reduction in the network traffic; moreover, managing the dynamictopology would be easy.

With VBF, because data packets are able to be forwarded only along a static routingvector, the data delivery ratio depends quite a bit on the routing pipe. Some locations couldhave nodes sparsely deployed or the deployment becomes sparser because of the movementof the water which would make it possible for very few if any nodes to be located in thatvirtual pipe that has the responsibility to forward data; moreover, it is also possible, as canbe seen in Fig. 11a, that some routes may be located outside the pipe which in the end willreduce the delivery ratios. Consequently, the radius threshold of the routing pipe can affectthe routing performance significantly as VBF is very sensitive to this threshold. As a resultof this sensitivity, there is quite a high possibility that there will be no node available forforwarding in this pipe if this threshold is too short. On the other hand, too many nodes couldparticipate in this process if the threshold is too large and this could eventually increase thecommunication overhead.

Finally, the end-to-end delay with various layer counts, 2–9, from the source to the sinkwas measured. As was previously mentioned, a larger end-to-end delay is the result of VBFfollowing a 3-way handshake process before the data packets are transmitted. Additionally,

123


Fig. 11 Comparisons with VBF. a packets delivery ratios, b end-to-end delays

this difference is made more significant by a larger number of layers due to the control packetsbeing required to be exchanged at every layer. On the other hand, as shown in Fig. 11b, theend-to-end delay is increased linearly with L2-ABF according to the number of layers.

11 Conclusion

In this paper, we have proposed the Layer by Layer Angle-Based Flooding (L2-ABF) pro-tocol to handle some critical routing issues in UWSNs. L2-ABF is scalable and efficient forend-to-end delays and energy consumption. Basically, L2-ABF relies on the flooding-basedtechnique to increase the reliability of the network. However, the number of nodes whichflood the data packets is controlled by the calculation of the angle for the flooding zoneto prevent unnecessary flooding of packets over the whole network. The flooding zone isadjusted in layer by layer manner by using the angle-based technique among the upper layernodes. The novelty of the proposed protocol is that it does not depend on location informationand there is no need to maintain complex routing tables. It is very easy to add new nodesin the network at any time and at any location. The real beauty of L2-ABF is that the deliv-ery ratios are not much affected by the density or sparseness of the nodes. The simulationresults show that L2-ABF is good for the long term and real time applications. In future, weare planning to increase the simulation metrics to investigate the relative performance withdifferent perspectives.

Acknowledgments we would like thanks Universti Teknologi PETRONAS for support to complete thisresearch project and also would like to thanks Dr. Muhammad Ayaz Arshad for his technical assistance.

References

1. Wahid, A., Sungwon, L., & Dongkyun, K. (2011). An energy-efficient routing protocol for UWSNs usingphysical distance and residual energy. In OCEANS, 2011 IEEE—Spain

2. Ayaz, M., et al. (2012). An efficient dynamic addressing based routing protocol for underwater wirelesssensor networks. Computer Communications, 35(4), 475–486.

3. Yan, H., Shi, Z.J., & Cui, J.-H. (2008). DBR: Depth-based routing for underwater sensor networks. InProceedings of the 7th international IFIP-TC6 networking conference on AdHoc and sensor networks,wireless networks, next generation internet (pp. 72–86). Springer: Singapore.

4. Basagni, S., et al. (2010). Optimizing network performance through packet fragmentation in multi-hopunderwater communications. In OCEANS 2010 IEEE—Sydney.

123

T. Ali et al.

5. Xie, P., Cui, J.-H., & Lao, L. (2006). VBF: Vector-based forwarding protocol for underwater sensornetworks. In Networking 2006. Networking technologies, services, and protocols; performance of com-puter and communication networks; mobile and wireless communications systems (pp. 1216–1221).Berlin/Heidelberg: Springer.

6. Daeyoup, H., & Dongkyun, K. (2008). DFR: Directional flooding-based routing protocol for underwatersensor networks. In OCEANS 2008.

7. Nicolaou, N., et al. (2007). Improving the robustness of location-based routing for underwater sensornetworks. In OCEANS 2007—Europe.

8. Ayaz, M., et al. (2011). A survey on routing techniques in underwater wireless sensor networks. Journalof Network and Computer Applications, 34(6), 1908–1927.

9. Yunus, F., Ariffin, S.H.S., & Zahedi, Y. (2010). A survey of existing medium access control (mac) forunderwater wireless sensor network (UWSN). In Fourth Asia international conference on mathemati-cal/analytical modelling and computer simulation (AMS).

10. Chirdchoo, N., Soh, S. W., & Chaing, K. C. (2009). Sector-based routing with destination location predic-tion for underwater mobile networks. In International conference on advanced information networkingand applications workshops, 2009. WAINA’09.

11. Ayaz, M., & Abdullah, A., (2009). Hop-by-hop dynamic addressing based (H2-DAB) routing protocol forunderwater wireless sensor networks. In Proceedings of the 2009 international conference on informationand multimedia technology (pp. 436–441). IEEE Computer, Society.

12. Partan, J., Kurose, J., Levine, B. N. (2006) A survey of practical issues in underwater networks, InProceedings of the 1st ACM international workshop on Underwater networks, ACM: Los Angeles, CA.

13. Shang, G.-Y., Feng, Z.-P., & Lian, L. (2011). A low-cost testbed of underwater mobile sensing network.Journal of Shanghai Jiaotong University (Science), 16(4), 502–507.

14. Chitre, M., et al. (2008). Underwater acoustic communications and networking: Recent advances andfuture challenges. In OCEANS 2008.

15. Jiang, Z. (2008). Underwater acoustic networks—issues and solutions. International Journal of Intelligentcontrol and systems, 13, 152–161.

16. Chun-Hao, Y., & Kuo-Feng, S. (2008). An energy-efficient routing protocol in underwater sensor net-works. In 3rd international conference on sensing technology, 2008. ICST 2008.

17. Glenn Elert, H. L. (2006). Depth of the Ocean. http://hypertextbook.com/facts/2006/HelenLi.shtml.18. Anupama, K. R., Sasidharan, A., & Vadlamani, S. (2008). A location-based clustering algorithm for data

gathering in 3D underwater Wireless Sensor Networks. In International symposium on telecommunica-tions, 2008. IST 2008.

19. Jornet, J. M., Stojanovic, M., & Zorzi, M. (2008) Focused beam routing protocol for underwater acousticnetworks. In Proceedings of the third ACM international workshop on underwater networks. 2008(p. 75–82). ACM: San Francisco, CA.

20. Jornet, J. M. & Stojanovic, M. (2008). Distributed power control for underwater acoustic networks. InOCEANS.

21. Chen, K., Zhou, Y., & He, J. (2009). A localization scheme for underwater wireless sensor networks.International Journal of Advanced Science and Technology, 4.

22. Caruso, A., et al. (2008). The meandering current mobility model and its impact on underwater mobilesensor networks. In The 27th conference on computer communications. INFOCOM 2008 IEEE 2008.

23. Younis, O., & Fahmy, S. (2004). HEED: A hybrid, energy-efficient, distributed clustering approach forad hoc sensor networks. IEEE Transactions on Mobile Computing, 3(4), 366–379.

24. Wang, Y., et al. (2012). A simple distributed networking protocol for underwater acoustic networks. TheJournal of the Acoustical Society of America, 131(4), 3487.

25. Mahajan, A., & Walworth, M. (2001). 3D position sensing using the differences in the time-of-flightsfrom a wave source to various receivers. IEEE Transactions on Robotics and Automation, 17(1), 91–94.

26. Ali, T. & Jung, L. T. (2012). Flooding control by using Angle Based Cone for UWSNs. In 2012 interna-tional symposium on telecommunication technologies (ISTT).

27. Ethem, M. Sozer, M. S., & Proakis, J. G. (2000, Janaury). Underwater Acoustic Networks. IEEE Journalof Oceanic Engineering, 25(1), 72–83.

123

http://hypertextbook.com/facts/2006/HelenLi.shtml


Tariq Ali received his M.S. degree in Computer Science fromSZABIST Islamabad Pakistan in 2006. During M.S., his specializationarea belongs to Networks and Communication and passed out with dis-tinction as highest grades in batch (89 %). He worked as a full time lec-turer at computer science department of Gordon College, Rawalpindi,Pakistan from 2007–2009 and as well as he has served more than twoyears as a IT-Manager for IT Department of Govt. Pakistan. In the sameyear, he got PETRONAS fellowship and joined to Universiti TeknologiPETRONAS, Malaysia as a full time Ph.D. candidate. His researchinterests include mobile and sensor networks, routing protocols andunderwater acoustic sensor networks.

Low Tang Jung obtained his Bachelor degree in Computer Technol-ogy from Teesside University, UK in 1989, MSc IT from National Uni-versity of Malaysia in 2001, and Ph.D. in IT from Universiti TeknologiPetronas, Malaysia in 2012. He is currently the Senior Lecturer in theComputer and Information Sciences Department of UTP. He has devel-oped some systems: A remote monitoring and alerting system for oilpipelines corrosion detection using SMS via GSM network, 3D Spa-tial Points Detector—using 2 webcams to detect 3D points, An SMSbased Q&A system for large conference hall, Sign-to-voice and Voice-to-sign language translator for speech impaired people, Various electro-mechanical embedded systems and Underwater Acoustic Communica-tions Data Packet Size Optimization. Low is currently involved in R&Don bio-inspired energy efficient routing in wireless sensor networkunder the ERGS grant (Malaysian government grant), and also involvedin R&D on Sensing Systems for Green IT under the Hitachi (M) Sdn.Bhd. grant. Low has been in the academic line for two decades alreadymainly as lecturer cum researcher in various public and private insti-

tutes of higher learning. He teaches various engineering and ICT courses. His research interest include wire-less technology, embedded systems, wireless sensor network, robotics, green IT.

Ibrahima Faye completed his M.Sc. and Ph.D. in Mathematics at Uni-versity of Toulouse and his Specialized M.Sc. in Engineering of med-ical and biotechnological data at Ecole Centrale Paris. He is currentlyAssociate Professor at the Department of Fundamental and AppliedSciences of Universiti Teknologi PETRONAS. His research interestsinclude Engineering Mathematics, Image Processing and Communica-tion Systems.

123

Documents

End-to-End Delay and Energy Efficient Routing Protocol for Underwater Wireless Sensor Networks