Portal de Periódicos da CAPES

E 2 MR: energy‐efficient multipath routing protocol for underwater wireless sensor networks

2019; Volume: 8; Issue: 5 Linguagem: Inglês

10.1049/iet-net.2018.5203

2047-4962

Muhammad Khalid, Farah Ahmad, Muhammad Arshad, Waqar Khalid, Naveed Ahmad, Yue Cao,

Water Quality Monitoring Technologies

IET NetworksVolume 8, Issue 5 p. 321-328 Research ArticleFree Access E2 MR: energy-efficient multipath routing protocol for underwater wireless sensor networks Muhammad Khalid, Corresponding Author Muhammad Khalid m.khalid@northumbria.ac.uk orcid.org/0000-0002-2674-2489 Computer & Information Sciences, Northumbria University, Newcastle upon Tyne, NE1 8ST, UKSearch for more papers by this authorFarah Ahmad, Farah Ahmad Computer & Information Sciences, Northumbria University, Newcastle upon Tyne, NE1 8ST, UKSearch for more papers by this authorMuhammad Arshad, Muhammad Arshad Department of Computer Science, Institute of Management Sciences, Peshawar, PakistanSearch for more papers by this authorWaqar Khalid, Waqar Khalid Department of Computer Science, Institute of Management Sciences, Peshawar, PakistanSearch for more papers by this authorNaveed Ahmad, Naveed Ahmad Department of Computer Science, University of Peshawar, Peshawar, PakistanSearch for more papers by this authorYue Cao, Yue Cao orcid.org/0000-0002-2098-7637 School of Computing & Communication, Lancaster University, UKSearch for more papers by this author Muhammad Khalid, Corresponding Author Muhammad Khalid m.khalid@northumbria.ac.uk orcid.org/0000-0002-2674-2489 Computer & Information Sciences, Northumbria University, Newcastle upon Tyne, NE1 8ST, UKSearch for more papers by this authorFarah Ahmad, Farah Ahmad Computer & Information Sciences, Northumbria University, Newcastle upon Tyne, NE1 8ST, UKSearch for more papers by this authorMuhammad Arshad, Muhammad Arshad Department of Computer Science, Institute of Management Sciences, Peshawar, PakistanSearch for more papers by this authorWaqar Khalid, Waqar Khalid Department of Computer Science, Institute of Management Sciences, Peshawar, PakistanSearch for more papers by this authorNaveed Ahmad, Naveed Ahmad Department of Computer Science, University of Peshawar, Peshawar, PakistanSearch for more papers by this authorYue Cao, Yue Cao orcid.org/0000-0002-2098-7637 School of Computing & Communication, Lancaster University, UKSearch for more papers by this author First published: 01 September 2019 https://doi.org/10.1049/iet-net.2018.5203Citations: 13AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract Exploration of underwater resources, oceanographic data collection, tactical surveillance and natural disaster prevention are some of the areas of Underwater Wireless Sensor Network(UWSN) applications. UWSN is different from traditional wireless sensor network. The later uses radio waves for communication between sensors while the former uses acoustic waves for data transmission. Communication through UWSN is more challenging because of the many challenges associated with acoustic channels such as low bandwidth, high transmission delay, usual path loss and intermittent connectivity. In UWSN, some algorithms were introduced to enhance the lifetime of networks, by using a smaller battery and other for critical data transmission. However, data packets flooding, path loss and low network lifetime are few challenges with immediate attention. This study proposes a novel routing scheme referred to as the energy-efficient multipath routing (E2 MR) for UWSN, which is basically designed for long-term monitoring with higher energy efficiency and delivery ratio. The E2 MR establishes a priority table, and the forward nodes are selected based on that priority table. Different experiments are carried out by simulating E2 MR and compared with Depth-Based Routing (DBR), EEDBR and H2-DAB with respect to the number of live nodes, end-to-end delay, packet delivery ratio and total energy consumption. 1 Introduction The ocean covers around a total of more than 70% of the Earth's surface [1]. It is considered to be a major source of nourishment, natural resources, defense, business transportation and adventurous purposes [2]. Ocean plays a vital role in human life; however, we have very little knowledge about the ocean and the resources underneath the ocean [3]. According to existing surveys, almost 10% of the ocean has been explored, while almost 90% of the ocean area is still unexplored [4]. The ocean plays a very important role in human lives, and that is why the unexplored area of underwater resources has received a lot of importance [5]. On one hand, the traditional approaches for ocean monitoring have several demerits, while on the other hand, human presence is not considered to be feasible in an underwater environment [6]. Underwater wireless sensor network (UWSN) is a new technology which provides very efficient and promising methods for underwater exploration of many areas, such as military defense, emergency and business throughout the whole world [7]. Sensors equipped autonomous and unmanned vehicles are specifically designed for underwater monitoring [8]. Autonomous unmanned vehicles are used for the exploration of underwater natural resources [9]. Sensors-equipped unmanned vehicles sense data and send back to sinks. The sinks forward data to the base station for processing; the base station processes that data and takes necessary action [10]. Although WSN is based on radio waves, they cannot be used in deep sea communication which is why underwater sensor nodes rely on acoustic waves for communication [11-13]. Here, when sensor nodes send data to sinks, the sinks forwards sense data to other sinks through radio waves [14]. UWSN is different from traditional terrestrial networks in several aspects. Ambiguities like high propagation delays or lower bandwidth are not usually found in terrestrial networks; however, we usually see these issues in UWSN [15, 16]. There are several issues in underwater communication such as dense salty water, high attenuation (high attenuation signals cannot travel a long distance underwater), absorption effect, electromagnetic and optical signals which do not work well in an underwater environment [17]. Hence to tackle the aforementioned issues, researchers proposed acoustic waves for underwater data communication instead of radio waves [18], and this enhanced the data transfer rate [19]. Although the acoustic channel solved some problems, it created some issues. The speed of the acoustic wave is 1500 m/s [20] which is much less than the speed of light. Owing to this slow speed, it will increase the propagation delay and packet delivery time [20]. As mentioned earlier in this paper, the bandwidth of the acoustic channel is very limited, e.g. almost <100 kHz [20, 21]. Sensor nodes are considered to be static in an underwater environment; however, the topology of underwater sensor nodes dynamically changes, because sensor nodes move (1–3 m/s) due to the flow of water [22]. Sensor nodes are battery-operated in the ocean, and that is why it is very difficult to replace the battery [23, 24]. In underwater multi-hop and multipath networks, the network topology is essential as data are forwarded through multiple nodes towards sink nodes [14]. When data are successfully delivered to one sink node, the sink node forwards that data to the intended receiver through radio waves [25]. In addition, the lower bandwidth is a major issue in acoustic communication. That is why routing protocols of terrestrial networks (which require high bandwidth) are not suitable in acoustic communication, due to the high energy consumption and end-to-end delays. Also in underwater networks, the topology does not remain the same because sensor nodes are dynamic due to the flow of water [26]. Routing protocols of TCP/IP, delay-tolerant networks [27], mobile ad-hoc networks, vehicular ad-hoc networks [21] and WSNs cannot be directly applied to underwater networks. Routing protocols that are used in other networks cannot be directly applied to underwater networks. To date, many protocols have been proposed for underwater sensor networks. These are mainly divided into two types which are localisation-based and localisation-free protocols [28]. Localisation-free protocols do not require any prior geographic or network information. Most of these protocols are used in underwater networks [10]. Researchers proposed various routing protocols for UWSN. These protocols are mainly divided into two categories, which are localisation-based and localisation-free routing protocols [29]. In localisation protocols, geographic information (that is the location of every node) and networks information (topological information of networks) is required in advance. Unlike localisation-based protocols, localisation-free routing protocols do not require location and network information in advance [30]. Most of the researchers proposed localisation-free routing protocols for UWSN [10], due to its less energy consumption and suitability in underwater network scenarios [31-33]. Summarised contribution of the proposed algorithm: The goal of the proposed algorithm, namely energy-efficient multipath routing (E2 MR) protocol, is to avoid flooding-type routing and enhanced energy consumption of every node. The following are some of the major contributions of this paper. Energy efficiency: E2 MR can reduce energy consumption (which will be proved in the simulation section of this paper). Avoidance of multiple message copies: E2 MR avoids all those types of routing which leads to a large number of messages in the networks, which obviously enhances network lifetime. The algorithm avoids multiple copies for both sensing nodes and sinks. Avoid flooding: E2 MR avoids flood strategies in which one node forwards the received packet to all nodes in the networks. Packet-holding time: E2 MR enhances the holding time of packets because the holding time depends on the residual energy (RE). This algorithm enhances the holding time. The rest of the paper is organised as follows. In Section 2, the architecture of UWSN has been discussed. In Section 3, a literature review of existing routing protocols is discussed. Section 4 is related to our design. In Section 5, a comparison of our design with literature falling into the localisation-free routing protocol is made. Finally, Section 6 concludes our work. 2 Architecture of the UWSN This section will briefly introduce the architecture of UWSN. Fig. 1 shows the general architecture of UWSN. There are generically five different components in the architecture of UWSN, which are the data sensing unit, energy management unit, processing unit, data communication unit and the depth measuring unit [34]. Data sensing unit: The data sensing unit is responsible for the sensing of any sort of data in UWSN. Various kinds of sensor nodes (used for flooding, underwater resources and movement etc.) are used for various purposes. In particular, the data sensing unit senses data even for nodes in the sleep mode. Fig. 1Open in figure viewerPowerPoint Node architecture Processing unit: The processing unit is one of the most important components of UWSN. It is responsible for any sort of processing. Data communication unit: The data communication unit is responsible for data transfer between various sensing nodes. It sends data from the sensing nodes to sink nodes, and also exchanges data between sink nodes and the base station. Depth measuring unit: The depth measuring unit is used for measuring the depth of every node. It plays a vital role because the position layer of the node is very important for routing in UWSN. Here, the position layer refers to which depth of ocean a node is located. Energy management unit: UWSN is operated on battery, and is almost impossible to replace the battery. Therefore, the operation of UWSN depends on availability of energy. In the case of energy exhaustion, the node will be shut down, and this influences the network service. The energy management unit is responsible for two tasks, it manages the remaining energy of the nodes (energy must be maintained for lasting network lifetime) and also manages consumption of the node in run time (to consume less energy in run time). 3 Background on routing in UWSN In the past decade, a lot of research has been conducted and certain results to minimise energy consumption in UWSNs were achieved. Khalid et al. [35] proposed a multimedia cross-layer protocol. The contents of the protocol are as follows: (i) study of the interaction of key components of the underwater communication system, such as forward error correction, modulation, media access control and routing; (ii) based on the design of a distributed cross-layer communication method, sensor nodes can share the network bandwidth efficiently. The protocol confirms improving energy efficiency and network throughput through experimental results. Khalid et al. [36] proposed an energy optimisation path unconscious hierarchical routing protocol called E-PULRP. The E-PULRP consists of a layered and communication phase, proposing a layered structure, utilises a gathering node as the centre and other nodes located on concentric circles. By considering the width of each layer and node transmission loss, the success probability of the nodes to send data and to avoid node transmission loss is improved. In the communications phase, an alternative energy optimal relay node algorithm transmits data to the sink node. Experiments on a comparative analysis with other algorithms display the validity of the E-PULRP protocol for energy efficiency. Carlson et al. proposed a network of underwater acoustic communication fading channel de-multiplexing asymmetric communication protocol called the AMDC. The protocol takes into account the uneven distribution of underwater noise and the actual underwater propagation environment with noise attenuated [5]. The underwater communication space is divided to build a tree-based multi-path transmission channel to improve the network energy efficiency and reliability of data transmission. Routing protocols can be divided into two categories [37], i.e. location-based routing (LBR) and location-free routing (LFR). LFR protocols do not rely on any pre-network geographical information. These type of routing protocols perform their operations without having any location information of other nodes in the network [36]. Most of the LFR protocols use the flooding phenomenon for a faster packet delivery ratio. While in LBR, the geographic information of the network must be known to every node in the network [35]. In LBR protocols, path calculation and the node's geographic information are pre-requisites for the network, which results in high end-to-end delay and energy consumption. 3.1 Location-based routing Vector-based forwarding (VBF) is a LBR scheme and maintains its routing path frequently. VBF is primarily a position-based scheme, where a very small number of nodes are involved in the data forwarding process. As a specified number of nodes are involved in sending data packets, it usually sends the packet in a single direction towards the sink. In VBF, every node knows the location of other nodes and their respective information. The sending node knows the final location of the data packet that is being sent by the node. VBF uses the idea of developing a virtual pipe in routing process. In a virtual pipe, a few numbers are involved in the routing procedure and their combination develops a routing pipe. The data packets are forwarded with the help of nodes that lie in the area of the virtual pipe. The enhanced version of VBF is presented as hop-by-hop vector-based forwarding (HH-VBF). HH-VBF focuses on robustness, energy efficiency, path loss and higher delivery of data packets. VBF used a single virtual pipe for packet forwarding while HH-VBF proposed the use of multiple virtual pipes for data forwarding. HH-VBF involves a larger number of nodes in the data-forwarding process and it develops multiple virtual pipes, through which the packet can be delivered to its final destination. 3.2 Location-free routing 3.2.1 Depth-based routing (DBR) DBR is a LFR scheme and does not need any pre-network node location information. The DBR primarily takes sensor depth into consideration while forwarding a data packet. In data packet forwarding, a node compares its depth with that of the proposed receiver node. It only forwards data when the depth of the receiver node is lower than the sender itself. Sometimes it is unable to find a node with defined parameters, and, as a result, it simply drops the packet or sends it back to a node at a higher depth. It starts sending data to all nodes whose depth is lower than that of the sender node. On one hand, it is beneficial for decreasing end-to-end delay, but, on the other hand, it generates a sort of flooding which produces higher energy consumption. This flooding process in DBR continues until the packet is received by any of the sinks installed onshore. Most of the time, this process produces multiple copies at the sink level. DBR analyses only the depth information while performing data-forwarding operations. DBR leads to a few drawbacks as short network life, flooding and higher energy consumption. It mostly sends data to multiple nodes of the same depth level. DBR has no proper mechanism defined for path selection, and the protocol generates a random path for every data packet generated. 3.2.2 Energy-efficient DBR Energy-efficient DBR (EE-DBR) is an enhanced form of DBR. It has more capabilities as compared to DBR. When a node in EE-DBR forwards a data packet, it takes the depth of the receiving node, RE and distance from the sink. In the first step the depths are compared, just like in the case of DBR. While in the second step, it checks for RE and compares it with the set threshold. The nodes with a higher RE than the threshold and a lower depth than the sender node are selected as data forwarders. Every node in the network usually has information on the depth and RE of their neighbour nodes. The drawback of EE-DBR is that it is not flexible in the long term and in a few cases, it floods the data packets as well. Sometimes a node might forward a packet to another node, which is far away from the sender node. Similarly, no mechanism is defined for analysing the shortest and efficient path selection. 3.2.3 Hop-by-hop dynamic address-based scheme Hop-by-hop-dynamic address-based routing (H2-DAB) is a LFR scheme. The scheme dynamically assigns addresses to nodes. The address ‘0’ is assigned to the sink as it is on the uppermost portion. This address is lower for the nodes near the sink while higher for nodes distant from the sink. In this scheme, every node is allotted two kinds of addresses called node-ID and hop-ID. Node-ID is the physical address of the node which stays the same throughout the network lifetime whereas hop-ID changes when the node moves from one place to another. Hop-ID starts from the top level or sink. It moves downwards in an increasing manner. Similarly, the node with higher depth has the highest hop-ID. H2-DAB supports multi-sink architecture. The scheme assigns the same ‘0’ ID to all sinks. On having the same hop-ID, the data packet received by any sink is considered as received. After receiving at a sink, it is easy to forward it to other sinks. Sometimes due to the random movement of nodes it is not possible to find out a node with suitable hop-ID. In this situation either a sender node has to wait for an appropriate next hop-ID or send the data packet backward. 3.2.4 Energy-efficient dynamic address-based routing Energy-efficient dynamic address-based routing (EE-DAB) does not require any network-related information for data forwarding. In this routing scheme, every node is provided with two kinds of basic ID's. The first ID type is called s-ID. This ID remains fixed for a node throughout the network's lifetime while the other type is call the c-ID. The second type of ID is also known as the next-hop ID. Both of these two IDs consist of two digits. 3.2.5 Mobile delay-tolerant routing Acoustic communication uses more energy than radio communication and wireless sensor nodes are battery operated and consume higher energy which lead to a serious problem. Thus energy efficiency has become a major problem with UWSNs. A delay-tolerant protocol is proposed which is called the delay-tolerant data dolphin scheme. This proposed scheme is designed for delay-tolerant systems and applications. In this protocol all the sensing nodes stay static and the data sensed by static nodes are passed on to data dolphin which acts a courier node. In this methodology high energy consumed hop by hop communication is avoided. Data dolphins which act as courier nodes are provided with continuous energy. In the architecture all the static nodes are deployed in the sea bed. These static sensors go into sleep mode if there is no data to sense and it periodically wakes up when it senses some data. After sensing some kind of desired data it simply forwards this data to courier nodes which are also called data dolphins. These data dolphins take this data and deliver it to a base station or sink. The number of dolphin nodes depends upon the kind of network and its application and the number of nodes deployed in the network. 3.2.6 Energy-efficient-multipath grid-based geographic routing Energy-efficient multipath grid-based geographic routing (EMGGR) protocol divides the whole network area into 3D grids where XYZ coordinates are used to identify each grid. In EMGGR, nodes are deployed randomly in the network area. Certain nodes are used as gateways for forwarding data packets. Gateways are selected through an appropriate selection procedure and at most one gateway is selected in each cell. The selection is carried out on multiple parameters like distance from the other node, sink and RE. 4 Proposed scheme This section discusses the proposed routing model for UWSN in detail. 4.1 Node architecture The general architecture of an underwater wireless sensor node is composed of five main elements which are the energy management unit, data sensing unit, depth measuring unit, communication unit and the central processing unit [6]. The processing unit is responsible for all kind of data processing while energy management unit has the responsibility to manage the remaining energy of node. Data sensing unit is used for sensing data in forward relevant data to the node ahead. Data sensing unit is always in active mode even when node itself is in sleep mode. Communication unit is responsible for overall data communication whereas depth measuring unit is used for measuring depth of water when deployed in sea. 4.2 Network architecture The proposed protocol will be able to avoid the phenomenon of data flooding and the creation of multiple copies. It will be able to take advantage of having underwater sensor network architecture with multiple sinks. This kind of network will have multiple equipped sinks both with acoustic and radio-frequency modems. These sinks are deployed at the surface of the water. The static sensor nodes are deployed in the desired underwater area. These nodes are capable of collecting data and forwarding it to sinks in multi-hop fashion or to courier nodes, if any. Courier nodes are provided with continuous power and they are only capable of receiving data from static sinks and forwarding it to sinks. A sink can easily communicate with another sink through radio channels. We can easily validate this assumption by the fact that sound propagates almost at the speed of ∼ m/s in water, five of orders of magnitude slower than that of radio waves which is a propagation of m/s in air. In our scenario, we have assumed that a packet reaches its destination as soon as it is successfully delivered to any of the sinks. 4.3 E2 MR protocol for UWSNs The main problems of the already proposed routing protocols are high energy consumption, fewer nodes and smaller lifetime. Most of the existing routing protocols in UWSN are based on the flooding phenomenon. It forwards multiple messages to nodes. This type of routing strategy is not efficient in any kind of network and especially in UWSN where the RE is much limited as it forwards multiple copies without a pre-defined efficient manner. This type of strategy leads to high energy consumption of both forwarder and receiver nodes. Also, this type of routing scheme degrades the overall network performance. This paper proposes an ‘energy-efficient multipath routing protocol’ for UWSN. The main goal of this routing algorithm is energy efficiency, higher packet delivery ratio and longer network lifetime. The proposed scheme has enhanced energy consumption and improved network lifetime. It avoids sending multiple packets to the next node in the network. The proposed routing algorithm of this paper takes advantage of the UWSN architecture of multiple sinks [20]. In this type of UWSN architecture, multiple sinks are deployed in the desired environment. Multiple sinks operate both on acoustic frequency modem and on radio frequency. In this type of architecture, underwater sensors nodes are randomly deployed into a specific location. A sink node communicates with another sink node through a radio frequency channel. The proposed routing algorithm of this paper assumes when sensor nodes forward packets to multiple sink nodes. If the packet is successfully delivered to any of the sinks among multiple sinks, it will be considered successful delivery. This paper also assumes that every node knows the depth information of all network nodes, which is also called vertical distance. Depth information of nodes is easily calculated through a depth sensor. The proposed scheme is called an E2 MR protocol for UWSNs. The goal of this routing protocol is to provide energy-efficient communication and it will run a node for a longer lifetime. The proposed protocol will also contribute towards the lifetime of the network. It will avoid flooding by forwarding data packet to a pre-selected node. It will also lead to avoid generating multiple copies. The E2 MR protocol is designed to provide energy-efficient communication, high network lifetime and higher packet-delivery ratio. The E2 MR protocol will be able to take advantage of having underwater sensor network architecture with multiple sinks [20]. This kind of network will have multiple equipped sinks both with acoustic and radio-frequency modems. These sinks are deployed at the surface of water. Underwater sensor nodes are deployed inside the water in the desired area of interest. These nodes are able to collect data, and also help forward the data to sinks. A sink can easily communicate with another sink through radio channels. In our scenario, we have assumed that a packet reaches its destination as soon as it is successfully received by any of the sinks deployed on the water surface. In the E2 MR algorithm, a small table is created at every node. In this paper, this table is called the priority table. The proposed priority table calculates a priority value for all eligible nodes in the networks. During the initial set-up phase, the proposed algorithm broadcasts the hello packet to every surrounding node in the networks. The proposed algorithm forwards a request packet to every surrounding node for basic information. The request packet includes RE, depth information, node ID and distance from the sender node. When nodes receive the hello packet, they will send back the required information with a REPLY–HELLO packet which will include all the requested information. After receiving the replying packet, all nodes with a higher depth than that of the sender node are dropped. The nodes having a lower depth are entered into a small priority table after calculating their values with the formula discussed in a later section. Nodes with higher values will be placed higher in the table. They will be used as forwarding nodes for sending out data when received. Fig. 2 shows the network architecture of E2 MR. Fig. 2Open in figure viewerPowerPoint Network architecture The E2 MR protocol basically focuses on energy efficiency and will avoid all those phenomena which lead to more energy consumption like the avoidance of a flooding mechanism where a node sends a received packet to all nodes which lie in its range of communication, avoiding the creation of multiple copies where multiple copies of the same data packet are received by the sink and the holding time in which a node holds a packet for a certain amount of time depends on their RE. Protocol Design: In protocol design, this paper briefly explains the working condition, network set-up phase and communication phase of E2 MR. 4.4 Set-up phase In the set-up phase, all nodes are randomly deployed in the ocean. In step one, the nodes initially broadcast some control message; in this paper this control message is called a hello packet. There are two different scenarios for communication of nodes in UWSN, that is soft communication and hard communication. Nodes forward the hello packet to those nodes which lie in the 25-m range, which is known by the name soft communication. In hard communication, range nodes can directly communicate with sink nodes. When receiver nodes receive the control message (hello packet) from the sender, it will reply with response hello packets. Receiver nodes reply with those parameters which the sender inquired (which is already mentioned in a previous section of this pape