Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

A Survey on Schemes to Aid Group Communication in Wireless Sensor’s Network Adnan Sultan School of Computing and Mathematical Sciences’ Liverpool John Moores University Liverpool, United Kingdom [email protected]

Abstract A Wireless Sensor Network can be described as a large collection of densely deployed, selforganized sensor nodes that are left unattended for months or even years in some geographical location inside or very close to the field of interest that co-ordinate with each other to perform some specific task. Nodes in a WSN communicate wirelessly and are selforganizing, the position of each sensor node does not need to be pre-Engineered or predefined which allows random deployment of the Wireless Sensor Nodes either placing one by one in the field of interest, throwing from a vehicle or through air support i.e. helicopter or aero plane. Group communication has attained a particular concern in the context of Wireless Sensor’s Networks (WSN). In the existing literature several schemes have been proposed to support group communication in WSNs. The contribution of this work is to analyse dominant schemes in this area. KeyWords. WSN’s; Group Communication; Cluster Management and Power Efficient Protocol (CUMPE); Adaptive Intra-Cluster Routing (AICR); Low Energy Adaptive Clustering Hierarchy (LEACH) 1. Introduction Development in the field of Wireless Communication and Electronics has enabled the development of small size, low power, distributed and inexpensive devices called Sensor nodes, which normally equipped a sensing device, an on-board processor, Analog-to-Digital converter, transceiver and a battery unit, mobility assistance and location finding systems are optional as required by some applications. These tiny devices are capable of sensing, processing and communicating wirelessly through short distances, also called as MicroElectro-Mechanical system (MEMS), that leverages the idea of Wireless Sensor Networks

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

(WSNs) [6] [10]. A Wireless Sensor node is a device that scans the analog data from its surroundings and then transforms that data into a digital form to send it to the Base Station. A Sensor node consists of a sensing unit, power unit, processing unit and a transceiver.

Sensing Unit Sensor

ADC

Processing Unit Processor

Transceiver

Storage

Power Unit

Figure 1.1: The sensor node hardware architecture

A Wireless Sensor Network can be described as a large collection of densely deployed, selforganized sensor nodes that are left unattended for months or even years in some geographical location inside or very close to the field of interest that co-ordinate co ordinate with each other to perform some specific task. WSNs have many advantages over the traditional traditional wired and wireless ad hoc networks (LAN, WAN, MAN, WLAN) and sensor networks. WSNs are inexpensive, having small size nodes, easily and randomly deployed, self-organizing, self organizing, which can operate unattended for long time, low power consumption, battery powered so need no fixed electrical infrastructure, communicate wirelessly, and can be deployed in remote areas. Nodes in a WSN communicate wirelessly and are self self-organizing, organizing, the position of each sensor node does not need to be pre-Engineered Engineered or predefined predefined which allows random deployment of the Wireless Sensor Nodes either placing one by one in the field of interest, throwing from a vehicle or through air support i.e. helicopter or aero plane etc[1]. Sensor nodes are very much prone to failure, where the vulnerability ulnerability of the Wireless Nodes causes WSNs’ topology very dynamic, therefore protocols and algorithms used in WSNs applications must possess selfself organizing capabilities to maintain connectivity between nodes at all time. [2][3][10-16]. [2][3][10 In this context this work critically analysis existing schemes for group communication in WSNs.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

The rest of this paper is organised as follows. In section 2. Related work is presented. Critically evaluation of the existing schemes is focused in section 3 whilst conclusion and future work is presented in section 4. 2. Literature Review Similar to other communication networks, scalability is one of the major design quality attributes of WSNs. The single-gateway architecture is not scalable for a larger set of sensors covering a wider area of interest, since sensor nodes are typically not capable of long-haul communication; long range communication takes place in multi-hop fashion. To allow the system to cope with additional load and to be able to cover a large area of interest while maintaining dependable services, sensor nodes` grouping (clustering) is usually pursued [16]. Grouping is a technique that superimposes a hierarchy on an otherwise flat network organization. In grouping WSN is divided into small groups of specific nodes that are managed locally by a group leader called Grouping. In each group only one node is elected as the group leader which is responsible for the communication with other group leaders. Each sensor node knows it’s location with respect to the sensor field. All sensor nodes in the same group can communicate with each other directly (via 1 hop) and all the sensors in the same group can communicate with all the nodes in the neighboring groups via group leaders. In the same group only Group Leader, has to be active to maintain connectivity. The active nodes take turn in round robin fashion. Group leader is elected with respect to its location and energy level [18]. Grouping plays a key role in WSNs as it provides support to high level middleware services such as object tracking, security, fault tolerance, power management and so on. It is easy for a user or the base station to query some specific nodes directly in a group deployed in some geographical location instead of the whole network that saves power and communication channels which ultimately increase lifetime of WSN. The duty cycle is inversely proportional to the number of sensors in the same group. Examples of this connectivity approach are given in GAF [17], where they partitioned the whole network into small squares through geographical grids and in each cell only one sensor node is kept active at any point of time which is responsible for connectivity with other active nodes of the neighboring group. Keeping only one node active in each group conserves energy. There are certain issues in grouping techniques which are described as follows.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Group Member Discovery When a WSN is deployed, wireless sensor nodes are divided into groups of specific number of nodes. Static or mobile nodes must be discovered that are eligible for membership according to relevant membership attributes. As the topology changes this process could take place many times, as the changes in topology have a huge effect on the groups in WSNs. The group members for a group are selected according to their positions, energy level, radio range and other constraints. The issue in the group member discovery is when a group leader discovers sensor nodes that are in its range and are also eligible to be included in the group; but that sensor node may be already member of another group, a group overlapping problem will be arise. Another problem is that when a mobile node comes in the range of a group and tries to connect with that group leader to become a member of this group, but that group has already exceeded the limit of nodes that it can accommodate, but that nodes will keep requesting to its nearby groups for the membership which will increase unnecessary load on the group leaders. Group Initialization After wireless sensor nodes have been deployed in the field, they configure themselves in form of groups. At the time of group formation sensor nodes exchange meta-data relevant to groups’ functionality and enforcing global membership constraints. The issue in group initialization is that what would be the assigned range and the optimum number of the group members that would result in maximum performance. Group Dynamics Management In case of node failure, changes in the position of nodes, power level of the nodes, changes in the position of a target or due to some other reasons the nodes are moved or removed from the group, sensor network topology will be changed, in this case new nodes will be discovered that fulfills the criteria of group membership either as a group leader or a common node [9] [19] and the groups may be reconstructed. Maintaining and reconstruction of groups` costs network delay and energy which is crucial for WSNs. 3. Critical analysis Group Management within Wireless Sensor Networks has attained research focus. A good number of schemes have been proposed as a solution to this problem. In this section we present an in-depth critical review of the previously proposed dominant schemes.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Low Energy Adaptive Clustering Hierarchy (LEACH) LEACH is a routing protocol for Wireless Sensor Networks. It is a hierarchical based protocol that follows the cluster based approach. Communication is done via Cluster-Head through TDMA which communicates with non-member nodes. Cluster membership is adaptive in LEACH and data is aggregated at Cluster-Head. Cluster-Head communicate directly with Sink or Base Station. In the light of LEACH specifications the following conclusion could be drawn. The sources and users are stationary and events are monitored; the interest propagation is predetermined and data dissemination mechanism is broadcasting. These features allow LEACH to achieve the desired properties. This protocol is a dynamic cluster based protocol which assumes that the network consists of remote Base Station and a set of homogeneous sensor nodes. In LEACH sensor nodes are in direct communication with the Base Station, but this is expensive in terms of energy usage. LEACH reduces the number nodes which communicate directly with the Base Station through cluster formation. Therefore nodes connected to the Cluster-Head using least amount of power. The ClusterHead communicate directly with the Base Station. LEACH uses direct connection between Cluster-Heads nodes and non-Cluster-Head nodes. Distant nodes from Cluster-Head are not energy efficient, one other main drawback in this scheme is that the Cluster-Head must be turned ON all the time to receive packet from the nodes in its cluster. Each Cluster-Head gives a time slot to each member node within which it communicates with the Cluster-Head. This limits the efficiency of each cluster since the member node can only communicate with the Cluster-Head within a given time. Thus it could also affect the overall Wireless Sensor Network. In addition Cluster-Heads are required to do extensive work which can consume a lot of energy. This can lead to node failure. However LEACH uses rounds to avoid such failure. This however adds further complexity to the structure of LEACH [7]. Adaptive Intra-Cluster Routing (AICR) This scheme has been proposed as an energy aware Intra-Cluster routing solution. In this algorithm keeping Intra-Cluster communication as the main objective, each node is not identical for routing the data. Some nodes are considered in close regions to perform direct routing. The nodes that are outside the region adopt Multi-Hop routing. In this manner the closer nodes are not having additional nodes. In traditional routing the nodes pertain to

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

consume more energy due to additional responsibility of providing routes to other nodes. Thus closer nodes have extra load on them with no concept of load balancing enforced. In this scheme for Intra-Cluster routing Cluster-Head is selected first before the collaboration of Base Station clusters is formed and finally Intra-Cluster routing is carried out. When the Cluster-Head selection phase starts all the deployed nodes send their energy level to the Base Station, on the bases of energy level, geographical area and Least ID the Cluster-Head is selected. Network deployment is considered as manual so the Base station is well informed about the geographical locations of the nodes. Base Station will select the Cluster-Head and will multicast this information to them. Once the Cluster-Head is selected the cluster formation phase starts. Each selected ClusterHead broadcasts its status as cluster-head. The nodes on receiving this message respond to it by showing their willingness to join this cluster. The node sends join request to the respective Cluster-Head and in a response a join confirmation alert is sent by the Cluster-Head.When the routing phase starts the nodes first check their locations, if their location is inside the close region then the mode for these nodes will be a direct routing, while on the other hand nodes not in close region then the mode of routing will be multi-hop routing [8]. In the light of specifications this scheme is dependent on the Base Station to identify the nodes` location. This limits the extendibility of AICR to support some of the related networks i.e. Mobile Ad-Hoc Networks. Selection of Cluster-Head is defined without mentioning as to how the cluster is formed. This scheme regards itself as energy efficient scheme, however formation of Cluster-Head and broadcasting join invitation message are examples of functions that consumes energy. It is due to the fact that specification of AICR is not clear as to how many times this invitation will broadcast. In addition how nodes will decide as to which Cluster-Head they would like to associate with. Intra-Cluster routing has different weaknesses, since it is the combination of both Single and Multi-Hop routing. Nodes are required to identify their location and then to decide whether they need Single-Hop or MultiHop communication. However this scheme fails to explain as to how nodes will make such decision and where the node will gather their location information from. Moreover it could be well understood that assigning such roles to the node could degrade their performance and affects the energy consumption.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Cluster Management and Power Efficient Protocol (CUMPE) The selection of Cluster-Head in CUMPE is based on the same principles as in LEACH. Cluster-Head is chosen based on random number and Node ID. Sensor node with more energy will become the Cluster-Head. Sensor nodes with high connectivity to other nodes will have higher chance to become the Cluster-Head. Routes are setup through flooding cost information and this is done to find the lowest cost link. Cluster-Head also sends packet with fixed signal strength. Any node receiving this packet knows the ID of the Cluster-Head and the signal strength of the packet. This information is use by the node to calculate the cost of the Cluster-Head. Once this calculation is done the node sends a message to tell other nodes about this cost information. Any node receiving the cost information packet will now the node ID of the sender, cost from sender to the Cluster-Head and cost to the sender. The node calculates the cost to the Cluster-Head and compares the current cost. If new cost is higher than the current cost the node ignores cost information packet. If the new cost is lower the node will send out the new information. This scheme follows a complex procedure of maintaining routing information in Wireless Sensor Networks. CUMPE as reported was introduced as an energy conserving protocol, however it could be seen from the specifications that nodes are required to be in Awake state all the time, thus consume more energy. There is higher probability that such energy will consume in obtaining link information rather than the useful communication. In addition nodes are required to fulfill additional responsibility of calculating and passing link cost information that could degrade nodes` individual performance and thereby affects the overall Wireless Sensor Networks efficiency. 4. Conclusions and Future Work This paper evaluates sum of the dominant schemes for group communication in WSNs. In future we will be conducting a comprehensive critical analysis of many of the reported schemes for WSNs. We are committed to share our future research findings with the ongoing research in this area.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

References [1]

Ian F. Akyildiz, Weilian Su, YogeshSankarasubramaniam, ErdalCayirci (2002). “A survey on sensor networks” Communications Magazine, IEEE 40(8): 102- 114.

[2]

Deborah Estrin, R. G., John Heidemann, Satish Kumar (1999). “Next Century Challenges: Scalable Coordination in Sensor Networks” In Proceedings of the ACM/IEEE International Conference on Mobile Computing and Networking: 263270.

[3]

Th. Arampatzis, J. Lygeros, S. Manesis (2005). “A Survey of Applications of Wireless Sensors and Wireless Sensor Networks” Proceedings of the 13th Mediterranean Conference on Control and Automation: 719- 724.

[4]

John A. Stankovic (2004). “Research challenges for wireless sensor networks” Special issue on embedded sensor networks and wireless computing, ACM SIGBED Review 1(2): 9 – 12.

[5]

Mo Li and Baijian Yang, “A Survey on Topology issues in Wireless Sensor Networks”

International Conference on Wireless Networks (ICWN), Las Vegas, Nevada, USA, June 2006. [6]

Jon Agre, L. C. (2000). “An Integrated Architecture for Cooperative Sensing Networks” Computer, IEEE 33(5): 106-108.

[7]

SunayanaSaha, P. B. (2003). “System Design Issues in Single-Hop Wireless Sensor Networks” Proceeding (408) Communications, Internet, and Information Technology - 2003 408-160.

[8]

Mengjie Yu, HalaMokhtar, MadjidMerabti (2006). “A Survey of Network ManagementArchitectureinWirelessSensorNetwork”LJMU Website.http://www.cms.livjm.ac.uk/senman/Papers/2006-093.pdf.

[9]

Mardoqueu Souza Vieira, Nelson Souto Rosa. (2005). “A reconfigurable group management middleware service for wireless sensor networks” Proceedings of the 3rd international workshop on Middleware for pervasive and ad-hoc computing 115: 1-8.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018

[10]

Brett A. Warneke, KristoferS.J.Pister (2002). “MEMS for distributed wireless sensor networks” 9th International Conference on Electronics, Circuits and Systems, 2002.1:291- 294.

[11]

NielsAakvaag, MogensMathiesen, Gilles Thonet (2005). “Timing and Power Issues in Wireless Sensor Networks- An Industrial Test Case” 2005 International Conference on Parallel Processing Workshops 419 – 426.

[12]

AmitSinha, AnanthaChandrakasan (Mar2001). “Dynamic power management in wireless sensor networks” Design & Test of Computers, IEEE 18(2): 62-74.

[13]

Adrian Perig, John Stankovic, David Wagner (2004). “Security in wireless sensor networks” Communications of the ACM 47(6): 53 – 57.

[14]

Juan Liu, Jie Liu, James Reich, Patrick Cheung, Feng Zhao (2004) "Distributed Group Management in Sensor Networks: Algorithms and Applications to Localization and Tracking." Telecommunication Systems 26(17): 235-251.

[15]

Lisa Frye, Liang Cheng (2007). “Network management of a wireless sensor network” Technical Report LU-CSE-07-003. Lehigh University. March 2007.

[16]

LinnyerBeatrys Ruiz, Jose Marcos Nogueira, Antonio. A. F. Loureiro (2003). “MANNA: A Management architecture for wireless sensor networks” Communications Magazine, IEEE 41(2): 116- 125.

[17]

Jinshan Liu, Francoise Sailhan (2005). “Group Management for Mobile Ad Hoc Networks: Design, Implementation and Experiment” Proceedings of the 6th international conference on Mobile data management: 192 – 199.

[18]

Curt Schurgers, ValasiosTsiatsis, SaurabhGaneriwal, Mani Srivastava (2002). “Topology Management for Sensor Networks:Exploiting Latency and Density” Proceedings of the 3rd ACM International Symposium on Mobile Ad Hoc Networking & computing 135 - 145.

[19]

Srinidhi T, Sridhar G. and Sridhar V (2003). “Topology Management in Ad Hoc Mobile Wireless Networks” Real-Time Systems Symposium, Work-in-Progress Session.

Journal of computing and management studies ISSN 2516-2047. Issue 1. Volume 2. January 2018