Vol-7, Special Issue-Number5-July, 2016, pp1044-1049 http://www.bipublication.com
Article search
Re
A Scheduling Method Along with the General Consensus Estimation for
Coverage Optimization in the Wireless Sensor Networks
Sobhan Dehghani1, Sarkhosh Seddighi Chaharborj2 and Mortaza Zolfpour Arokhlo3
1,2
Department of Computer Sciences, Bushehr branch, Islamic Azad University, Bushehr, Iran 3
Department of Computer Sciences, Sepidan branch, Islamic Azad University, Sepidan, Iran
1
netanalyze@gmail.com 2
ss27782007@gmail.com 3
zamortaza2@live.utm.my
ABSTRACT—
Wireless Sensor Network (WSN) consists of many small devices called sensor nodes that work together to perform specific task(s). Wireless sensor networks have unique features that differentiate them from other wireless networks and create special challenges. Non-rechargeable, non-replaceable and limited power supply of the sensor nodes are the main challenges of these networks such that with complete discharge of the node power supply, it will become useless, part of the information may be out of reach and the performance of the network may fall into danger. As coverage is of great importance in the Wireless sensor network, a solution is proposed in this paper. In this method each node is placed in a particular location due to the importance of overall network coverage in WSN and also in order to reduce energy consumption of the system, all the nodes are on standby except one. To establish a duty cycle and prevent the imposition of too much load on a node, the active node that is responsible for sensing and maintaining the coverage in the area, may be replaced by another active node in the next cycle. In such cases that a part of the network is free of nodes and its information are inaccessible, that area can be determined by estimating the consensus between other neighboring nodes. Moreover, the wireless network’s overall coverage is expected to improve by zoning the network area, and using consensus-based estimation and duty cycles.
Keywords— Consensus Estimation, Coverage, Duty Cycle, Wireless Sensor Network, , Zoning, ..
I. INTRODUCTION
Recent advances in electronics and wireless communications have created the capabilities of design and manufacturing tiny devices called sensor nodes with low power consumption, small size, reasonable prices and various applications. These nodes which are able to receive various kinds of data due to their sensors, processing and transferring feature, lead to establish and develop network which have been
non-rechargeable. Thus, the algorithms try to minimize energy consumption of the system[1]. There are also various applications where coverage is an important parameter measured throughout the network lifetime. As the sensor network has a limited lifetime, some of the nodes may die in a certain region making a void zone where sensing doesn’t takes place[2]. In order to improve network coverage, reduction of energy consumption using some structures like clustering will be useful. Another approach is to divide network into different zones.
Clustering is an effective technique in reducing the energy consumption. This means that instead of sending data from each individual node, some nodes are elected as cluster head and other nodes called cluster members, choose the nearest as its cluster head. This process constructs clusters. Each cluster member sends its data to its cluster head. The cluster heads send cluster members’ data to the sink after fusion, aggregation and compression. Clustering algorithms, which LEACH [3] is the most famous and most popular of them, increase the lifetime of the network.
The death of nodes terminates some regions of the network become with no coverage and their data be unavailable. It means that coverage of the network is closely related to energy consumption of the system. There are some efficient techniques in order to manage energy consumption such as zoning and implementing duty cycles which is the aim of this paper. The network partitioned in several zones called area. In each area several nodes placed and corporate with each other in order to improve network coverage.
In each area only one of the nodes is active and the others are asleep. Active node selection strategies will be discussed in the following sections of the paper.
The active nodes sensed data of the area and transmit it to the database. Due to considerable load on active nodes, the active node is changed in regular intervals.
After a certain time, some regions are empty due to the death of them. Consume estimation technique would be useful in this case as it estimates the lost data and improves network coverage.
The rest of this paper is organized as follows; Section II briefly reviews some existing protocols in WSN. Section III describes proposed method. Section IV presents simulation results, and finally Section V concludes this paper.
II. RELATED WORKS
LEACH [3] is one of the most famous clustering approaches in WSN. Main techniques of LEACH protocol include algorithms for distributing cluster forming, adaptive cluster forming, and cluster header position changing. The technique of distributing cluster forming ensures higher level of self-organization of target nodes. The adaptive cluster forming and cluster header position changing algorithms ensure sharing the energy dissipation fairly among all nodes and prolong the lifetime of the whole system as a result. LEACH protocol runs with many rounds. Each round contains two states: clustering setup state and steady state. The former forms the clusters in self-adaptive mode and the latter transfers data. The time of the second state is usually longer than the time of the first one for saving the protocol payload. The process is shown in Figure 1.
Figure1. Operation time of LEACH
their data loss rate and activate their neighbours based on these local measurements; it cannot be too low to hamper data delivery, but it cannot be too high either since neighbouring nodes might interfere with each other leading to a high collision rate.
In [5], Kumar et al. adopt the Randomized Independent Scheduling (RIS) mechanism to extend network lifetime while achieving asymptotic K-coverage. RIS assumes that time is divided into cycles based on a time synchronization method. At the beginning of a cycle, each sensor independently decides whether to become active with probability p or go to sleep with probability 1 − p. Thus the network lifetime is increased by a factor close to 1/p (i.e. p determines the network lifetime). An idea to increase the coverage ratio as well as the lifetime of the system is suggested in this paper by introducing mobility in heterogeneous WSN, where the heterogeneity is considered in energy of nodes. In [6] authors have chosen the deployment in heterogeneous WSN with loco-mobility capability nodes with different energies. The simulation results showed that in the algorithm called Enhancing Coverage Ratio using Mobility (ECRM), maximum area is covered, and in addition, some nodes are in off state to conserve the energy. It is considered the nodes to cover large area while being reliable in sensing by saving energy. The algorithm’s target is to schedule the sensor nodes in such a way that they can monitor a region efficiently. Random Backoff Sleep Protocol (RBSP) [7], a probe based protocol, uses a dynamic sleeping window for neighbouring nodes, based on the amount of residual energy at an active node. In RBSP, the neighbouring nodes wake-up very frequently when the residual energy of the current active node is very less. In order to improve energy-effectiveness in wireless sensor network, in practice some sensors in observation points are selected not to gather data. In this case, the insufficient data gathered by the rest of sensors have to cover the total network so that
the complete information of the whole environment could be estimated rationally, which is similar to compressive sensing. However, the process of estimation has to cost a lot of energy, which is a crucial problem. The [8] proposes a practical and effective information coverage approach in which an actual constrained condition is considered for consensus estimation to reduce unnecessary energy cost reasonably. In [8] experiments, the method has been proved valuable and feasible. III.PROPOSED METHOD
As mentioned above, the main challenges in WSN is limited nodes’ power supply which have robust relative to maintain maximum coverage. Thus various schemes have introduced for reducing energy consumption which clustering is one of the most effective of them. In the proposed method network is divided into some areas as in figure 2. This figure is a deployment schema of 100 sensor nodes in 100x100 flat environments. The blue cycles are shown sensor nodes. At first, base station is located at (50, 0) and broadcast a signal in 35 degree according to figure 1. Nodes that get this signal marked themselves as first zone. Then base station broadcast signal in different degree to coverage all the environment. In this point, each node knows about its zone. Then, base station start layered network with send signal this time by fixed length and no restricted degree. Therefore layered and zoning network will build. Each node assigns itself to a layer and a zone called area (figure 3).
Figure 3. Zoning network
In addition to clustering, zoning and duty cycle can improve network energy consumption and coverage. In clustering amazing load in on cluster head nodes and a percent of redundant nodes are in active mode. After zoning, in each area one node with maximum value is selected as active node. The active node sense environmental data and forward it to base station. All other nodes in order to energy consumption kept in sleep mode. The value of node i in order to determine active node is defined as:
Centralityi=MD-DB
MD = Max_Distance
DB =Distance_Between (middle of area and nodei)
) max_
( ) _
( 4
centrality centrality energy
initial
remaining node
Value i i
i
The active node in each area in order to send sensed data to base station may use intermediate active nodes due to base station is located far aware of active nodes in upper layers. The active node selects next node in one lower layer which
have same zone number or one different number zone. The routing example is shown in figure 4.
Figure 4. Routing Example
During network process, some nodes will die and network coverage will go under risk. It might that some areas become empty in terms of alive nodes. In order to have total network coverage, the base station uses other active nodes’ data to estimate void regions according to this concept that nodes closer to each other have same sensed data. The estimate data of empty node is obtained from following equation.
_ _
1
area empty of center area_node, 1
area empty of center area_node,
k
area k
area
area
Dist Dist Data Sensed Data
Estimated
and steady state called round. In setup phase active nodes are selected and middle active nodes set. Then in steady states sensed data transmit to base station. The rounds run one after another till the end of the network (figure 1).
IV.SIMULATION
To evaluate proposed schema a simulator is required that implements the proposed schema. We in this paper use Matlab software. In order to start simulation on the first step initial parameters should be determined. Initial parameters, which are shown in table 1, contain number of nodes, deployment manner, nodes’ initial energy, network environment size and etc.
Table 1. Initial Parameters
Values Parameters
300
Number of Nodes
100*100
Network Size
0.5 Joule
Initial Energy
50*0.000000001 Joule
Eelec
0.0013*0.000000000001 Joule
Emp
10*0.000000000001 Joule
Efs
4000 bit
Data Packet
32 bit
Control Packet
Most existing methods and articles in the field of wireless sensor network to prove the superiority of their methods, the longevity of the network are examined. In order to measure network lifetime usually three criterions are used. First node die (FND) is considered as one of the main criterions. FND means the time interval between network starts and first node depletes of energy. The next criterion is Half Node Dies (HND) means the time interval between network starts and death of half of the nodes. The last criterion is considered as the time interval between network starts and death of all nodes or Last Node Dies (LND). Figure 5 illustrates the obtained FND, HND and LND values of LEACH, ECRM and proposed zoning approaches. As is obvious, the proposed zoning approach in terms of FND, HND and LND has better performance than LEACH and ECRM.
Figure 5. FNDs, HNDs and LND in proposed, LEACH and ECRM
Figure 6. Energy consumption in proposed, LEACH and ECRM
The diagrams that illustrate in figures 6 and 7 show amount of energy consumption and dead nodes in each network respectively. From this graph can be concluded that number of dead nodes in each round in the proposed method is less than LEACH and ECRM. Also amount of energy consumption in LEACH and ECRM in each round is more than the proposed method.
V. CONCLUSION
In this paper we have proposed the protocol which elects the sensing node depending on the energy of the nodes, and provide more coverage in long lifetime using the concept of mobility of sensor nodes. The simulation results have shown that it has better lifetime and coverage than LEACH and ECRM under same conditions. The proposed protocol is highly scalable as if the additional nodes are added, then they can be shifted evenly. According to the simulation results, the proposed method has better performance than LEACH and ECRM. The performance of proposed method was examined in terms of energy consumption, delay and trend of dead nodes and was shown that the proposed method has better performance.
REFERENCES
1. Zhang, P. J., Guo, G. Z., & Yu, Z. Z., “Application of Wireless Sensor Network in Embedded Smart Home System”, Applied Mechanics and Materials, 2015, Volume 738, pp. 74-78.
2. Hirani, P. K. & Singh, M., “A Survey on Coverage Problem in Wireless Sensor Network”, International Journal of Computer Applications, 2015, Volume 116, Issue 2, pp. 1-3.
3. P. Saxena, P. Prasenjit, N. Kumar, Energy Aware Approach in Leach Protocol for
Balancing the Cluster Head in Setup Phase: An Application to Wireless Sensor Network, Journal of Information Assurance & Security; 2015, Volume 10 Issue 1, pp. 40-47.
4. Cerpa, A., & Estrin, D., “ASCENT: Adaptive Self-Configuring sEnsor Networks Topologies”, IEEE Transactions on Mobile Computing, 2004, Volume 3, Issue 3, pp. 272–284.
5. Kumar, S., Lai, T.H., & Balogh, J., “On K-Coverage in A Mostly Sleeping Sensor Network” In Proceedings of the tenth Annual International Conference on Mobile Computing and Networking (MobiCom), 2012, pp. 144–158.
6. Nudurupati, D.P., & Singh, R.K., “Enhancing Coverage Ratio using Mobility in Heterogeneous Wireless Sensor Network”, Procedia Technology, 2013, Volume 10, pp. 538 – 545.
7. More, A., & Raisinghani, V., “Random Backoff Sleep Protocol for Energy Efficient Coverage in Wireless Sensor Networks”, Advanced Computing, Networking and Informatics, 2014, Volume 28, pp. 123-131. 8. Yang, H., Tang, K., Yu, J., & Zhu, L., “A
