Wireless Personal Communications

, Volume 100, Issue 4, pp 1569–1583 | Cite as

A Vector-Based Routing Protocol in Underwater Wireless Sensor Networks

  • Sayyed Majid Mazinani
  • Hadi Yousefi
  • Mostafa Mirzaie


One major concern shared by many researchers about underwater wireless sensor networks (UWSNs), with respect to the limitations and particularities of underwater environment, is the problem of routing. These limitations include three-dimensional topology, limited bandwidth, node movement, long delay, limited energy, and construction costs. The new routing protocols for underwater networks have been developed on the basis of voracious routing systems. The main problem with UWSNs is finding an efficient route between the source and the target to send more packets to the target with lower levels of energy consumption. In this research, by improving VBF algorithm, which is dependent on the radius of the routing pipe, an algorithm is introduced which considers pipe radius as a function of the environment’s dimensions and of the range and the number of nodes. Consequently, by changing one of these parameters, the radius of the routing pipe changes. However, to control the energy consumed by the nodes, there exists a function that, if the recipient node’s energy to receive the packet is much lower than that of the sender node, the proposed method reduces the size of the routing pipe’s radius to lessen its chance of being selected as the guiding node so that other nodes are able to have the chance of getting the packet’s guiding node. The proposed algorithm has been compared with VBVA, HHVBF, and VBF protocols; the simulation results obtained from NS-2 simulator indicate that the proposed protocol could cut back on energy consumption, especially in networks with high number of nodes, by relying on changing the width of the routing pipe in proportion to network density. It was also successful in delivering more packets in non-dense networks.


Underwater sensor networks Routing algorithms NS-2 Routing pipe 

1 Introduction

In recent years wireless sensor networks have attracted a lot of attention among researchers and a lot of progress has been made in the field. At the end of twentieth-century wireless sensor networks moved to center stage. At first, these networks covered ground practical programs only. But later, and due to the importance of processing data on a planet whose surface was covered with 70% water, the underwater application of these networks was reasonably justified [1, 2]. The vast number of researches conducted into this field point to its numerous functions. Several such fields include health, military operations, urban management, environment, construction, retrofitting colossal structures such as bridges and tunnels, marine inspection, oil field identification, pollution detection, navigation, prevention of natural disasters such as tsunamis, and aggregating data on oceanology [3, 4, 5, 6, 7, 8]. Routing protocols play a significant role in network model. In UWSNs, it is routing that particularly takes on the duty to transmit data from the source to the destination [9]. The presence of an efficient routing algorithm to deliver packets to the destination is so crucial that it ought to be chosen by the designer with utmost care. On the other hand, underwater sensor networks enjoy some unique features that makes designing an efficient routing algorithm difficult and challenging. These features include: (1) replacing radio signals with acoustic ones due to the quick weakening of the former. (2) Dynamic topology of the network that is down to the movement of sensor nodes caused by water movements. Of course, it is possible to deploy some fixed nodes both on the surface and at the bottom of the water [10, 11]. Designing protocols which focus on reducing energy consumption is vital for these kinds of networks because sensor nodes run on batteries whose replacement or recharging proves no easy task. Reducing energy consumption is a cause for concern in UWSNs [7]. In this article, a vector-based routing strategy is proposed that is able to monitor data transmission procedures and prevent useless transmissions that cause energy dissipation by benefiting from the width of the routing pipe as a dynamic variable. This algorithm has been compared with VBF, HH-VBF, VBVA protocols.

The rest of the article is organized as follows: Sect. 2 deals with analyzing different routing methods. The proposed algorithm is discussed in Sect. 3. The analysis of the proposed method and its comparison with other ones appear in Sect. 4. And, finally, Sect. 5 provides the results.

2 Related Works

2.1 Different Routing Types

In sensor networks routing fit into three categories of reactive, proactive, and geographic. From among these categorizations, the geographic routing is the most suitable one for underwater wireless sensor networks [12].

2.1.1 Reactive Protocols

These protocols try to minimize the delay of data, which is caused by finding a new path, by keeping route data transmitted from each node to others updated. This process is carried out by broadcasting of the control packets which contain the table for routing data (as in routing vectors). To establish routes, these protocols incur overheads in the network. This results from the constant broadcasting of the updated data in the network which is caused by a change in the topology of the network brought about by the movement of at least one node. In these protocols, each node has the capacity to establish a path to every other node. Consequently, reactive routing protocols do not suit underwater networks [8, 12].

2.1.2 Proactive Protocols

In these protocols, a node will start the routing process only when it needs to start routing to a particular destination. These are more suitable for dynamic environments, but they have a high delay rate. Moreover, the source node must broadcast a lot of control packets in the network to establish the path. Consequently, due to high delay rate, the peculiarities of the seabed, and the changeability of acoustic channels, proactive protocols are not suitable for underwater environments [8, 13].

2.1.3 Geographic Protocols

These protocols determine the path from the source to the destination based on position information, meaning that each node takes the next step based on the position of his and of the destination’s neighboring nodes. Although this method is promising, it is not clear whether position information is accurate. To achieve accurate position information, it is necessary to have accurate synchronicity between the nodes, and this, because of propagation delay of variables in underwater environments, proves almost unattainable. On the other hand, global positioning system (GPS) does not work well either in underwater networks. In fact, GPS uses waves in the 1.5 GHZ band which do not work underwater [8, 13]. In geographic routing protocols, the position of the destination node is the most important piece of information. Accordingly, the transmitter must be aware of the position of the destination node. These pieces of information on nodes’ position are gathered using a positioning service [13].

2.2 Routing Protocols in Underwater Wireless Networks

This section deals with analyzing some routing algorithms in underwater sensor networks.

2.2.1 Analyzing VBF Protocol

The first recommended routing protocol for underwater sensor networks is VBF [7]. This algorithm is a geographical one which requires a complete positioning. Each node’s position is estimated by AOA technique or by the strength of the signal. The position information related to the sender node, forwarder node, and the target node is carried in the packets. The transference path is determined by a vector, which is located inside a routing pipe, from the sender node to the target node. Every node inside this pipe acts as a candidate to send packets. If the node which receives the packet is inside the routing pipe, it will send the packet; otherwise it will delete it [7, 14]. Figure 1 provides a high-level view of this algorithm. In this figure S0 is the sink node located in top of the figure.
Fig. 1

VBF protocol performance [15]

Vector-based forwarding protocol or VBF is among the first protocols proposed for underwater environment. It was proposed to solve the problem of node mobility, to be deployed in environments with dynamic topology, and to be efficient with respect to energy consumption. This algorithm is position-based protocol. In each packet the position or the coordinates of the sender, target and the forwarder nodes are included so that they can be used in routing operations [7, 10].

In VBF, the position or the coordinates of the nodes are obtained by positioning techniques. To perform the routing process, the two concepts of ‘vector routing’ and ‘pipe routing’ are used. The vector routing is a straight line from the source node to the destination node, while the routing pipe is cylinder with adjustable radius and with the centrality of vector routing. The route of a packet from the source to the destination is determined by the vector routing [7].

Upon receiving a packet, the receiver node calculates its position in relation to the former forwarder node from where it received the packet. Whenever the node can determine that it is close enough to the vector routing, then that node is considered as a new forwarder node and will put its coordination inside the packet and will forward it to the next node. Otherwise, it will remove the packet. In this protocol, all forwarding nodes are inside the routing pipe. Those nodes which are not close to the routing vector will not forward and remove the received packets [7, 16].

These protocols do not need the information regarding the position of all the nodes. VBF is a source routing protocol. This means that routing begins from the source node. Each packet carries simple routing data, and inside each packet there exists three field coordinates of SP, TP, and FP which indicate the position of the sender, target, and forwarder nodes, respectively. To support node mobility, each packet contains a field called Range. When a packet reaches the region specified by TP field, it will determine the transmission limit by using the value inside the range field and then floods it to the network. In fact, by estimating its distance from the routing vector and Radius field, a node will be able to determine whether it can play a role as guiding the packet or not [7].

In VBF protocol, nodes which are close enough to the routing vector or which are in fact inside the routing pipe will guide the packet. In those networks in where nodes are close to each other, this will result in an increase in energy consumption, the amount of the unnecessary packet transmission, and, finally, protocol overhead. Therefore, an appropriate policy should be adopted with respect to dense networks. One method to analyze and determine the degree of node density and to adjust the radius of the routing pipe is to determine node density. The VBF protocol uses a self-adapting algorithm to solve this problem [7, 15].

This algorithm uses desirableness factor to evaluate the suitability of a node to guide a packet. In S0S1 vector, where S1 is the sender node, S0 is the target node, and F is the former forwarder node, the desirableness factor for node A is calculated as follows:
$$a = \frac{p}{w} + \frac{{\left( {R - d*cos\theta } \right)}}{R}$$
where p is the projection of node A on S0S1 vector, d is the distance between node A and node F, θ is the angle between fS0 vector and fA vector, R is the transmission range of node F, and w is the radius of the routing pipe. Figure 2 shows the concept of desirableness factor and its parameters. Based on the definition of desirableness factor, it can be observed that for each node which is close to routing vector, that is, 0 < p < w, each node’s desirableness factor falls in the range of [0, 3]. The closer the desirableness factor is to 0, the closer the node is to the suitable position [7, 15].
Fig. 2

Node desirableness factor [15]

If the guiding node in VBF protocol finds a node to guide in the vicinity of the routing vector, it retains that packet for a period of time to maintain adaptability. This time period is referred to Tadaptation. In other words, any qualified node delays packet transmission by a time period which is shown in the following equation.
$$Tadaption = \sqrt a T_{delay} + \frac{R - d}{{V_{0} }}$$
where T delay is the predefined maximum delay, V0 is the propagation speed of acoustic signals in the water and normally equals 1500 M/s, and d is the distance between the node and the forwarder node. Basically, the self-adaptive algorithm gives a higher priority to the suitable node to maintain packet transmission [7, 15, 16].

Vector-based routing protocols are very sensitive to routing pipe radius. Due to their condition inside the routing pipe, some nodes are constantly exposed to data transmission which results in premature node die. All position-based protocols such as VBF are scalable, but they have low performance in end-to-end delay. Moreover, protocol’s performance in non-dense networks is very low [17].

2.2.2 Evaluating HH-VBF Protocol

To resolve VBF’s difficulties, HH-VBF algorithm has been proposed. This is in fact a phase by phase algorithm. With respect to features such as requiring the positioning of all the nodes and using routing pipe, the HH-VBF is similar to VBF algorithm. The difference, however, is that in VBF algorithm a singular virtual pipe is created between the source and the sink. But in HH-VBF algorithm a virtual routing pipe is created in each phase. Accordingly, a step-by-step approach is used in routing process. Figure 3 yields a high-level view of the implementation of this algorithm [7, 13, 15].
Fig. 3

HH-VBF protocol performance

In HH-VBF algorithm, each node, immediately after receiving a packet, calculates the vector from the sender to the sink and estimates its distance to that vector. If the distance is less than the radius routing pipe, this node is qualified to guide the packet, and thus, becomes a candidate to guide a packet.

In non-dense networks, HH-VBF protocol is able to find many routes and is less sensitive to the radius of the routing pipe [17]. As a result of adopting a step-by-step approach and creating a pipe in each route, HH-VBF algorithm has a better packet transmission rate than VBF, and if there is an empty route in the network, it is sure to find it. But its energy consumption is high, and in networks with high node mobility, it is not as useful as VBF [15].

Definition of the pipe in HH-VBF differ from VBF and has less dependent to pipe radius. By using a proper definition of pipe radius in each step, presence of unnecessary intermediate nodes that do not participate in routing can be decreased. Therefore, total energy of the network can be preserved.

2.2.3 Evaluating VBVA Protocol

Routing protocols such as VBF and HH-VBF use a greedy method to select the node in the next step. Adopting a greedy policy is not successful everywhere. For example, it is possible that a node cannot find any of its neighbors qualified to guide the packet and, therefore, cannot guide the packet to the next step. In routing protocols this is referred to as routing hole. VBVF is the first protocol which was proposed to avoid holes in underwater networks. This protocol uses two mechanisms to avoid the holes: Vector-Shift and Back-Pressure mechanisms. In VBVBF protocol, when a node detects a hole on a packet’s way, it tries to avoid the hole by shifting the vector. If a node does not hear any packet guiding action in the new vector, it is considered as the last node. Vector-Shift cannot find an alternative path for the last node, and instead it must use Back-Pressure mechanism. In Back-Pressure mechanism, when a node detects that it is the last node, it sends the node backwards. This will continue until a packet arrives at a node to be able to use Vector-Shift mechanism [16, 17, 18].

This method could solve the problem of holes in underwater networks. Hole avoidance mechanism creates several vectors that results in network improvement and network strength [17]. The occurrence of holes and using hole avoidance mechanism increases energy consumption. Besides, extra overhead is produced as a result of hole avoidance mechanism, while the existence of holes creates long delays in the network.

As we said, VBVA can solve hole problem but to do this, not only wastes the energy of the nodes but also leads to delay in the network. So dynamic programming methods can be applied for selecting the next node instead of back-tracking methods.

3 The Proposed Protocol

VBF is one of the routing algorithms in underwater wireless sensor networks. In VBF protocols, those nodes which are located around the radius of the routing pipe are used to guide packets. Therefore, when the radius of the routing pipe is big, the number of nodes participating in guiding the packet increases. Thus, more energy is consumed, and if the radius of the routing pipe is small, fewer nodes are used to guide packets, while it is possible that fewer packets are received by the sink. In our proposed method, we have considered pipe radius as being dependent on the environment, node range, and the total number of nodes. To manage the energy consumed by the nodes, we have taken into consideration the remaining energy level of a node. The proposed algorithm is similar to VBF but has the following differences: (1) the remaining energy and (2) the changeable width of the pipe.

In the proposed method, each packet contains the position of the source and the destination and the remaining energy. The transmission route is determined by a vector from the source to the destination. Each node knows its position. When a node receives a packet, it reads position of the sender node from the packet. Distance between receiver and sender nodes can be calculated by using Euclidean formula. If a node determines that it is close enough to the routing vector (if it is lower than the threshold), it will put its calculated position and the remaining energy in the packet and will send it. Otherwise, it will remove the packet. Therefore, the transmission path is a virtual pipe from the source to the destination, and sensor nodes inside the pipe are suitable to transmit data. As in VBF algorithm, utility factor is calculated based on Eq. (1). If the value of utility factor ‘a’ has a large value, the node is not suitable to send a packet. If utility factor ‘a’ is closer to zero, then the node is close to the best position. When a node receives a packet, first it calculates its position and determines whether it is inside the routing pipe or not. If it is inside the routing pipe, the node will retain the packet for a specific Tadaptation duration. As in VBF algorithm, this value is calculated using Eq. (2).

In T adaptation time, if a node receives other packets transmitted from other nodes, it will calculate utility factor as mentioned in Eq. (1) for all of them. If utility factor is between (0 < α < 3), it transmits the packet; otherwise it will remove it. When a node receives a packet, it first tries to figure out whether it is inside the calculated radius or not. In the following, the proposed algorithm is presented in the form of a pseudo-code.
The width of the pipe is defined as follows:
$${\text{Width}}\_1 \, = 2\sqrt {\frac{{{\text{x}}*{\text{y}}*{\text{z}}}}{{{\text{R}}*{\text{n}}}}}$$
where x, y, z are the size of the environment, n in the number of nodes, and R is the node transmission range. All these variables are fixed and never change.

A 3-dimension environment including length (X), width (Y) and height (Z) has been considered for simulation.

When both the number of nodes and range of packet transmission are low, routing pipe radius should be increased to have more nodes in the pipe and conversely when both the number of nodes and range of packet transmission are high, radius should be decreased. By using this, unnecessary and repeated transmissions can be avoided.

In the Eq. (3), the radius of the routing pipe increases in proportion to the size of the environment, the number of nodes, and the range of packet transmission. If the number of nodes is larger and the range of packet transmission is wider, the width, which is the same as the radius of the routing pipe, should decrease. As a result, contrary to VBF algorithm, in the proposed method, the radius of the routing pipe does not have a constant value.
$${\text{Width}} = {\text{Width}}\_1*{\text{pr}}$$
Based on the pseudo-code, width value is calculated in two phases. First, width_1 is calculated based on Eq. (3). The obtained value is multiplied by pr variable. In the first line of the pseudo-code, the calculation procedure for pr variable is referred to where r_residual_energy is the amount of residual energy of the receiver node, and f_residual_energy is the guiding node’s residual energy. For example, in Fig. 4, Node F has transmitted a packet and node A received it. Now node A should see whether it is inside the routing pipe or not. Let us assume that the residual energy of node A is less than node F. If the packet is transmitted using VBF algorithm, then the node A will be inside the routing pipe and will transmit the packet, without paying attention to the residual energy of node A. In the second line of the pseudo-code, width_1 is multiplied by pr and the final width is calculated based on the following equation:
$${\text{Width}} = 2\sqrt {\frac{{{\text{x}}*{\text{y}}*{\text{z}}}}{{{\text{R}}*{\text{n}}}} } *{\text{pr}}$$
Fig. 4

Performance of the proposed algorithm

Because the residual energy of node A is less than node F, then pr will be less than one. Therefore, node A may not send the packet in our proposed method because its residual energy is less than that of the Forwarder node. The third line of the pseudo-code examines whether the node is inside the routing pipe or not. In algorithm 1, we mean PROJECTION function as distance determination and PKT is the candidate node that should calculate its distance by using Euclidean formula.

4 Performance and Evaluation

Before dealing with simulation details, examining the assumptions of the model system is essential. These assumptions are as follows:
  • The nodes are aware of their position.

  • All the nodes are equal and enjoy similar structure.

  • The source and the destination nodes are fixed in the position

  • Nodes do not use solar energy to recharge their batteries, and a node whose energy is depleted will be removed from the network.

The proposed algorithm is analysed and compared with VBF, HH-VBF, and VBVA protocols. NS-2 software is used to perform simulation. Simulation parameters are presented in Table 1.
Table 1

Simulation parameters

Simulation software

NS2 version 2.30 (Aqua-sim)

Analyzed area

1000 m × 1000 m × 500 m

Number of nodes


Transmission range

100 m

Width of the routing pipe

100 m

Packet size

50 Bytes

Simulation duration

1000 s

Initial energy


Transmission energy consumption

2 w

Energy consumed when the packet is received

0.75 w

Energy consumption when the network is at rest


Source address

(900, 900, 500)

Destination address

(100, 100, 0)

The first analyzed factor is the average energy consumption of the entire network. The values are presented in Table 2 and Fig. 5 show its fluctuation graph.
Table 2

Results obtained from simulation (average energy consumption)

Number of nodes

Average of energy consumption






























Fig. 5

Energy consumption in different protocols

At first when it is non-dense, the network’s energy consumption is relatively equal to that of other methods. However, with the increase in the number of nodes and the network becoming denser, it boosts its performance so much that even with more than 2500 nodes, it achieves lower energy consumption compared with other methods. As explained in detail with respect to the proposed method, less number of nodes means that the radius of the routing needs to be enlarged so that more packets could be delivered to the destination. On the other hand, when pipe radius is enlarged, nodes consume more energy to guide the packets. Accordingly, in non-dense networks, the level of energy consumption in the proposed method is more than in other ones. Conversely, by adding to the number of nodes, the routing pipe radius decreases and consumes less energy to guide the packets as a result. Table 3 shows the average number of received packets while Fig. 6 presents their fluctuation graph.
Table 3

Results obtained from simulation (average of number of received packets)

Number of nodes

Average of number of received packets






























Fig. 6

The number of received packets in different protocols

Energy consumption in the proposed algorithm is approximately equal to that of VBF protocol while the proposed algorithm delivers more packets to the sink and improves the performance of VBF and VBVA protocols. In the proposed protocol, the nodes which have more power than their sender stand a much better chance of receiving and transmitting packets. With the increase and the decrease of the radius of the routing pipe, the number of steps to reach the sink decreases. Accordingly, not only is the energy consumption in the proposed algorithm less than other methods, but also the number of received packets has improved, especially in non-dense networks. The HH-VBF algorithm, which has the highest number of delivered packets, imposes high level of energy consumption on the network. In underwater sensor networks, where energy maintenance is of great importance, this can cause some problems.

With respect to the fact that the criterion of energy consumption in the proposed method and VBF protocol are roughly the same, it is necessary to conduct a careful analysis by introducing some changes in the initial energy in both methods.

Figures 7 and 8 show energy fluctuations and the number of delivered packets in both methods with the initial energy of 100 J, while Figs. 9 and 10 show energy fluctuations and the number of delivered packets in each of the methods with the initial energy of 10 J.
Fig. 7

The amount of consumed energy with the initial energy of 100 J

Fig. 8

The number of received packets with the initial energy of 100 J

Fig. 9

The amount of consumed energy with the initial energy of 10 J

Fig. 10

The number of received packets with the initial energy of 10 J

With the decrease of the initial energy of the nodes, the performance of the proposed methods has improved. Compared with VBF protocol, the proposed method delivers more packets to the sink by maintaining its energy consumption.

When compared with other methods, the proposed method has obviously succeeded in reducing energy consumption, particularly in networks with high number of nodes, by changing the width of the routing pipe in proportion to network density. In non-dense networks, it could deliver more packets. Therefore, the proposed algorithm is highly recommended to the network designer as a routing protocol for applications in which energy consumption and the number of delivered packets are important, the number of nodes is high, and the network enjoys lesser density.

5 Conclusion

In this article, imitating VBF algorithm, a vector-based traffic protocol has been proposed to deal with routing challenges in underwater sensor networks. VBF is considered as a constant reference point for routings in underwater sensor networks. Compared with other protocols, it enjoys powerful scalability and efficient energy consumption. The efficiency of this protocol is due to the fact that only those nodes which are in the path of the routing pipe are involved in data traffic, hence the efficient energy consumption. In this research, by improving on VBF algorithm, which is dependent on the radius of the routing pipe, a routing algorithm has been proposed which has considered pipe radius as a function of the environment’s dimensions, range, and the number of nodes. Thus, by changing one of these parameters, the radius of the routing pipe changes accordingly. However, to manage nodes’ energy consumption, there is a function that if the receiving node’s residual energy is much lower than that of the sender node, it reduces the chance of being selected as the guiding node by reducing the size of the routing pipe radius so that other nodes can have the chance of becoming the packet guiding node. The proposed algorithm has been compared with VBF, HH-VBF, and VBVA ones, and simulation results indicate that, by relying on changing the width of routing pipe in proportion to network density, it could achieve lower energy consumption and higher packet deliverance in networks with large number of nodes and in networks with lower density, respectively.


  1. 1.
    Zenia, N. Z., Aseeri, M., Ahmed, M. R., Chowdhury, Z. I., & Shamim Kaiser, M. (2016). Energy-efficiency and reliability in MAC and routing protocols for underwater wireless sensor network: A survey. Journal of Network and Computer Applications, 71, 72–85.CrossRefGoogle Scholar
  2. 2.
    Mohamad, M. M. (2013). Greedy routing in underwater acoustic sensor networks: A survey. International Journal of Distributed Sensor Networks, 9(7).
  3. 3.
    Singh, A. K., Purohit, N., Goutele, S., & Verma, S. (2012). An energy efficient approach for clustering in WSN using fuzzy logic. International Journal of Computer Applications, 44(18), 8–12.CrossRefGoogle Scholar
  4. 4.
    Rashid, B., & Rehmani, M. H. (2016). Applications of wireless sensor networks for urban areas: A survey. Journal of Network and Computer Applications, 60, 192–219.CrossRefGoogle Scholar
  5. 5.
    Salam, H. A., & Khan, B. M. (2016). Use of wireless system in healthcare for developing countries. Digital Communications and Networks, 2(1), 35–46.CrossRefGoogle Scholar
  6. 6.
    Li, F., Li, Y., Zhao, W., Chen, Q., & Tang, W. (2006). An adaptive coordinated MAC protocol based on dynamic power management for wireless sensor networks. In IWCMC (pp. 1073–1077).Google Scholar
  7. 7.
    Xie, P., Cui, J.-H., & Lao, L. (2006) VBF: Vector-based forwarding protocol for underwater sensor networks. In Networking 2006. Networking technologies, services, and protocols; performance of computer and communication networks; mobile and wireless communications systems (pp. 1216–1221). Springer.Google Scholar
  8. 8.
    Khosravi, M. R., Rostami, H., & Basri, H. (2015). Enhanced VBF for delay-aware routing in UWSNs. International Journal of Advanced Research in Computer Science and Software Engineering, 5(5), 100–104.Google Scholar
  9. 9.
    Khan, T., Ahmad, I., Aman, W., Azam, I., Khan, Z. A., Qasim, U., Avais, S., & Javaid, N. (2016). Clustering depth based routing for underwater wireless sensor networks. In IEEE 30th international conference on advanced information networking and applications.
  10. 10.
    Xie, P., Zhou, Z., Nicolaou, N., See, A., Cui, J.-H., & Shi, Z. (2010). Efficient vector-based forwarding for underwater sensor networks. EURASIP Journal on Wireless Communications and Networking. Scholar
  11. 11.
    Akyildiz, I. F., Pompili, D., & Melodia, T. (2004). Challenges for efficient communication in underwater acoustic sensor networks. ACM SIGBED Review—Special Issue on Embedded Sensor Networks and Wireless Computing, 1(2), 3–8.Google Scholar
  12. 12.
    Jiang, J., Han, G., Guo, H., Shu, L., & Rodrigues, J. J. P. C. (2016). Geographic multipath routing based on geospatial division in duty-cycled underwater wireless sensor networks. Journal of Network and Computer Applications, 59, 4–13.CrossRefGoogle Scholar
  13. 13.
    Souiki, S., Feham, M., Feham, M., & Labraoui, N. (2014). Geographic routing protocols for underwater wireless sensor networks: A survey. International Journal of Wireless & Mobile Networks (IJWMN), 6(1), 69–87.CrossRefGoogle Scholar
  14. 14.
    Namesh, C., & Ramakrishnan, B. (2015). Analysis of VBF protocol in underwater sensor network for static and moving nodes. International Journal of Computer Networks and Applications, 2, 20–26.Google Scholar
  15. 15.
    Nicolaou, N., See, A., Xie, P., Cui, J.-H., & Maggiorini, D. (2007). Improving the robustness of location-based routing for underwater sensor networks. In OCEANS 2007-Europe.
  16. 16.
    Pouryazdanpanah, K. M., Anjomshoa, M., Salehi, S. A., Afroozeh, A., & Moshfegh, G. M. (2014). DS-VBF: Dual sink vector-based routing protocol for underwater wireless sensor network. In: Control and system graduate research colloquium (ICSGRC) (pp. 227–232).Google Scholar
  17. 17.
    EL-Rabaie, S., Nabil, D., Mahmoud, R., & Alsharqawy, M. A. (2015). Underwater wireless sensor networks (UWSN), architecture, routing protocols, simulation and modeling tools, localization, security issues and some novel trends. Networking and Communication Engineering, 7, 335–354.CrossRefGoogle Scholar
  18. 18.
    Xie, P., Zhou, Z., Peng, Z., Cui, J.-H., & Shi, Z. (2009). Void avoidance in three-dimensional mobile underwater sensor networks. In Wireless algorithms, systems, and applications (pp. 305–314).Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Imam Reza International UniversityMashhadIran
  2. 2.Islamic Azad University, Neyshabur BranchNeyshaburIran

Personalised recommendations