Scholarly article on topic 'Reliability Analysisof Route RedundancyModel for EnergyEfficient Node Disjoint Multipath RoutinginWireless Sensor Networks'

Reliability Analysisof Route RedundancyModel for EnergyEfficient Node Disjoint Multipath RoutinginWireless Sensor Networks Academic research paper on "Computer and information sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Engineering
OECD Field of science
Keywords
{"Route Maintenance" / "Route Reliability" / "Node Disjoint" / "Multipath Routing" / "Wireless Sensor Networks"}

Abstract of research paper on Computer and information sciences, author of scientific article — Shiva MurthyG, R.J. D'Souza, Varaprasadg

Abstract Theobjectiveofevery sensornodeinasensornetworkistosendthesenseddatainthephenomenatothesinknode. Manyrouting protocols select an optimal path to increase the network lifetime in sensor networks. The consumption bythe nodes along this optimal path is high, thereby causing their early death. In the multipath routing, path reliability ishigh, becauseoftheavailabilityof alternatepathstotheoptimalpaths.Thiswork proposesa theoreticalframework to study the reliability analysis of route redundancy model for node disjoint multipath networks in wireless sensor networks. The reliability of route redundancyin single path and multiple paths with varied number of nodes and its probability values are discussed.

Academic research paper on topic "Reliability Analysisof Route RedundancyModel for EnergyEfficient Node Disjoint Multipath RoutinginWireless Sensor Networks"

Available online at www.sciencedirect.com

SciVerse ScienceDirect

Procedia Engineering 38 (2012) 1487 - 1494

International Conference on Modelling, Optimisation and Computing

(ICMOC 2012)

Reliability Analysis of Route Redundancy Model for Energy Efficient Node Disjoint Multipath Routing in Wireless Sensor Networks

Shiva Murthy G^, R.J.D'Souza^, Varaprasad G%c

$bDepartment of Mathematical and Computational Sciences National Institute of Technology Karnataka, Surathkal Mangalore-575025, India Department of Computer Science and Engineering B.M.S.College of Engineering, Bangalore-560019 India

kgshivam@gmail.com^a,rjd@nitk.ac.in^b, drvaraprasad@gmail.com$c

Abstract

The objective of every sensor node in a sensor network is to send the sensed data in the phenomena to the sink node. Many routing protocols select an optimal path to increase the network lifetime in sensor networks. The consumption by the nodes along this optimal path is high, thereby causing their early death. In the multipath routing, path reliability is high, because of the availability of alternate paths to the optimal paths. This work proposes a theoretical frame work to study the reliability analysis of route redundancy model for node disjoint multipath networks in wireless sensor networks. The reliability of route redundancy in single path and multiple paths with varied number of nodes and its probability values are discussed.

Keywords: Route Maintenance, Route Reliability,Node Disjoint, Multipath Routing, Wireless Sensor Networks.

1. Introduction

Due to advances in wireless communications and electronics over the last few years, the development of networks of low-cost, low-power, multifunctional sensors has received increasing attention. The research in Wireless Sensor Networks (WSN) is multidisciplinary. It contributes to a variety of open issues such as application domain, hardware, communication and networks in order to realise an efficient system. Recent advances in the micro electro mechanical devices with low power, short range transceiver, processor with low capacity and a limited power unit contributed in designing sensor node [1]. WSN is a variant of Mobile Ad hoc NETworks (MANET) consisting of a finite number of tiny, autonomous devices called wireless sensor nodes. A sensor network is designed to detect events or phenomena, collect and process data, and transmit the sensed information to sink node. Basic features of sensor networks are, dense deployment, self-organizing capabilities, short range broadcast communication, co-operative effort of sensor nodes, multi hop routing, frequently changing topology due to fading and node failures, limited energy, transmission power, memory, and computing power. WSN consists of hundreds or thousands of tiny sensor nodes. These sensor nodes have the capability to communicate among themselves to make the information reach the sink node in a single hop or in multiple hops. Sink node is a common destination node for all the sensor nodes in the network. The sink node may

1877-7058 © 2012 Published by Elsevier Ltd. doi:10.1016/j.proeng.2012.06.183

be a fixed node or a mobile node capable of connecting the sensor network to an existing communication infrastructure or to the internet where a user can have access to the reported data [2]. Basically, a sensor node comprises sensing unit, processor,transceiver, mobilizer, position finding system, and power units.

Routing the sensed data from the source to the sink node in a resource constrained environment is still a challenge. An optimal path is selected based on the metrics such as the gradient of information, the minimum hop, minimum transmission cost, high residual energy etc. [3]-[6] to route the data between source and sink. Many routing protocols attempt to reduce the energy usage in the nodes to increase the network lifetime [7]. This provides an optimal path between source and destination. Selecting an optimal path between the source and destination and sending the data through that path may not increase the lifetime of the network. The energy usage in such an approach is not as efficient as that in the multi-path routing approaches.

The multi-path routing protocols [8] select the available multiple paths between the source and sink node. The overhead of route discovery in multi-path routing is much more than that of single-path routing. On the other hand, the frequency of route discovery is much less in a network which uses multi-path routing, since the network can still operate even when one or a few of the multiple paths between a source and destination fail[9]. The route maintenance mechanism enables the alternate path to route the data while an active path is failing.

In this paper, we propose an analytical model for route maintenance in an Energy Efficient Node Disjoint Multipath Routing Protocol (EENDMRP). It focuses on route redundancy in a single node level redundancy over a single path, single node level redundancy through multi node over single path, and single node level redundancy through multiple level multiple nodes in a single path. The rest of this paper is organised as follows. In section 2, we discuss the related works. Overview of the EENDMRP is discussed in section 3. In section 4 reliability analysis of route redundancy model for node disjoint multipath routing for WSN is presented. In section 5, results are discussed. Conclusions are drawn in section 6.

2. Related Works

Marina et al [10]proposed Ad hoc Ondemand Multipath Distance Vector(AOMDV) routing protocol. It is a source initiated,reactive (Node/link) disjoint multipath routing protocol. AOMDV extends the Ad hoc Ondemand Distance Vector(AODV) protocol to discover multiple paths between the source and the destination in every route discovery. Multiple paths are computed to guarantee the network to be loop-free and disjoint. Primary design goal behind AOMDV is to provide efficient fault tolerance in the sense of faster and efficient recovery from route failures. In AOMDV, route maintenance is done by means of Route ERRor (RERR) packets. When an intermediate node detects a link failure (via a link-layer feedback), it generates a RERR packet. The RERR packet propagates towards all traffic sources having a route via the failed link, and erases all broken routes on the way. A source upon receiving the RERR initiates a new route discovery if it still needs the route. Apart from this route maintenance mechanism, AOMDV also has a timer-based mechanism to purge the stale routes. AOMDV uses very small timeout values to avoid stale paths. This may limit the benefit of using multiple paths. AOMDV use a moderate setting of timeout values and additionally use control messages to pro-actively remove stale routes.

Marjan Radi et al [11] proposed a Low-Interference Energy-Efficient Multipath Routing Protocol(LIEMRO) for improving Quality of Service (QoS) in event-based wireless sensor networks. This protocol is mainly designed to improve packet delivery ratio, lifetime, and latency, through discovering multiple interference-minimized node-disjoint paths between source node and sink node. If any node on an active path, after k efforts, could not receive Clear To Send (CTS) packet or ACKnowledgement (ACK) packet from the next hop node, then it sends an error message to the source node (through the reverse path). After the source node receives such error message, it disables the path from which this message has been received and redistributes network traffic over other active paths.

Upadhayaya et al[12] proposed a Node Disjoint Multipath considering Link and Node stability Routing (NDMLNR) protocol. NDMLNR aims to improve the QoS in the network. When the link stability of a node falls below the link stability threshold, it informs its predecessor node by sending the message that, it is not capable of participating in routing. The predecessor sends the message to the source node for disabling that active route. Then the source node sends the data packets through the alternate path. If no alternate path exists, then the source node starts the route discovery procedure again.

The theoretical analysis of k node-disjoint path availability is discussed by Abbasi et al[14] and Bheemalingaiah et al[13]. The probability that there exist k node-disjoint paths is estimated. They fail to give the route maintenance model for node disjoint multipath routing protocol.

3. Overview of the EENDMRP

EENDMRP is a sink initiated,proactive node disjoint multipath routing protocol. In EENDMRP, WSN is assumed to consist of number of stages St, i = 1,2,..., l based on the number of hops between the source and destination. The sink is a stage zero, St0 node. Every node can communicate with sink node is in Sti. We assume that a Sti node can communicate with nodes on the same stage S ti and next stage S ti+1. But, it cannot communicate with S ti-1 nodes. This prevents the formation of paths with loops. Initially, all the nodes in the network have very high hop count value except the sink node. Initially all the nodes have their residual energy level above the threshold energy value. Multiple paths from all the nodes to the sink is generated in route construction phase. During this process Route CONstruction(RCON) packets are exchanged between the nodes. Each sensor node broadcasts the RCON packet once and maintains its own routing table. If there is no route to the sink via the node that received RCON packet, then that node processes the RCON packet. If the route to sink from that node is already available in the node's routing table, then it checks the packet's hop count value. If packet hop count is smaller than node's hop value and its residual energy is above the threshold energy value, then RCON is processed; otherwise it drops the packet. The node that receives the RCON packet, updates the RCON packet. The RCON is updated with incrementing hop count by one,updates the forward node id and appends its node id to the path. The node which receives RCON packet updates its routing table information such as node's hop count and route to the sink node. Similarly, all the nodes in the network receive the RCON packet and update their routing table. Once the all possible multiple paths are generated, the node disjoint multipaths are identified between the source and destination. When the source node needs to send the sensed data to the destination, it spreads the data traffic among the node disjoint multipaths based on its residual energy and filled queue length. If any node disjoint path fails in the routing because of node death or node displacement, it informs the source node through the RERR packet.

3.1. Route Maintenance in EENDMRP

Figure 1 illustrates the operation of route maintenance phase. The link between the nodes 2 and 3 is broken. During the data transfer phase, any link can fail because of reasons like physical misplacement of node or the node's energy is below the threshold energy value during the transfer of data, etc. Then the node 2 sends the RERR packet to its data source. It also sends to the sources which has route to sink node through that node. Node 2 checks its routing table to find the alternate path from node 2 to sink node and selects the best route among the paths to sink node. Then node 2 generates and sends the RERR message to the source node. The RERR message provides the information about the link failure and alternate path between the node which generated the route error packet and sink node. The source node receives the RERR message and updates its routing table by discarding the failed path. Data packets from the source node are redistributed among the rest of the paths.

Figure 1: Route Construction Phase in EENDMRP

4. Reliability analysis of Route Redundancy for Node Disjoint Multipath Routing for WSN

Improving the throughput in the routing protocols for WSN is very much necessary. Once the residual energy of a node falls to zero or its received signal strength is too weak to communicate, the links in the network fail. In some

WSN applications, the communication link fails when the residual energy falls below the threshold energy value or received signal strength is weak in a node, the reliability of the path decreases.

4.1. Single node level redundancy over single path

Suppose that there are M nodes in a path from the source to the sink node. There exist M - 1 hops link between source and the sink node. In a path, if anyone node is redundant or alternate node for a hop between the node i and node j, then the reliability Rsl is the probability that each node is working properly or data through the node reaches to the next hop node. If pm is the probability of node m which is working properly, then

Rsl = 1 - (1 - Pm)

4.2. Single node level redundancy through multi node over single path

Suppose a single node in a path has multi level single node redundancy between the nodes i and j then the reliability Rsml is the probability that data through the multi level node reaches the next hop node without fail and is given by,

Rsml = 1 - EI (1 - Pu)

Where,U is the number of levels and pu is the probability of uth level node working properly.

4.3. Single node level redundancy through multiple level multiple nodes in a single path

Suppose a single node in a path between source and sink node has redundancy in multiple nodes with multiple levels between the nodes i and j then the reliability Rmml is the probability that data through the node reaches the next hop without fail and is given by,

Rmml = 1 - (1 - puv)

u=1 v=1

where,V is the number of nodes in each uth level and puv is the probability of uth level nodes working properly. To make analysis more realistic and to satisfy all the criteria in the node redundancy like, multi hop nodes having different probability values between the present hop to the next in a path, let us assume that, there are M number of nodes in a path between the source and sink node. Every node may have multiple multi hop node redundancy. If the nodes between the hops in any level have different probabilities, then the reliability of a path Rp is,

U V H M-uv

Rp = (1 - (1 - puvh)) pm

u=1 v=1 h=1 m=1

where,H is the number of hops in each uth level and puvh is the probability of uth level nodes working properly. If the node disjoint multipath network has K number of paths between the source and destination, then R, the reliability of node disjoint multipath network that the data sent from the source node reaches the destination without fail is,

K U V H M-uv

R = 1 - n [1 - n (1 - n (1 - n puvh))] U pm

k=1 u=1 v=1 h=1 m=1

5. Results and Discussion

Simulations are conducted using MATLAB R2008a tool to analyze the network reliability. The reliability of redundant paths in a node disjoint multipath network is analyzed here. Initially, reliability of a path is studied in different levels of redundancy. For simplicity, the work is restricted to five levels of redundancy in a path. The number of nodes considered in a path is 2 to 8. The node probability values are taken between 0.5 and 0.9. The reliability of a node disjoint multipath network is analyzed through redundant paths. The number of paths between source and sink node is considered as 2,3 and 4. Again the node probabilities are set between 0.5 and 0.9.

Table 1 show the reliability of a path P; when the node probabilities are 0.5,0.7 and 0.9. In Table 1, the path reliability is high when the number of levels is high and number of nodes is low. When node probability is 0.5 and

number of nodes is 2, the path reliability is 0.25 at redundancy level 1. As the level is increased to 5, the path reliability is 0.7626. There is a 205% increase in path probability as the number of redundancy levels increased from 2 to 5. When the number of nodes in a path is increased from 2 to 8, the % increase in path reliability reduces to 97.45. When the number of nodes in the path is 8 and redundancy level is 5, the path reliability is 0.0193. This is much less compared to redundancy levels 1 or 2 and number of nodes is 4. It indicates that, to get high path reliability, the number of nodes in redundant path should be less and it should have maximum levels of redundancy for the nodes. This is shown in the Figures 2 and 3.

Similarly, Table 1 shows that, the path reliability is high when number of nodes is 2 and number of levels is 5. When the node reliability is 0.7 and 0.9, number of nodes in the path is 2 and redundancy levels is 5 then the path reliability is 0.9654 and 0.9997 respectively.

Table 1: Path Reliability for different node reliability and levels

Node Probability Number of Nodes Number of Levels

1 2 3 4 5

2 0.25 0.4375 0.5781 0.6835 0.7626

0.5 4 0.0625 0.1210 0.1760 0.2275 0.2758

6 0.0156 0.0310 0.0461 0.0610 0.0757

8 0.0039 0.0077 0.0116 0.0155 0.0193

2 0.49 0.7399 0.8673 0.9323 0.9654

0.7 4 0.2401 0.4225 0.5611 0.6665 0.7466

6 0.1176 0.2214 0.3130 0.3938 0.4651

8 0.0576 0.1119 0.1631 0.2114 0.2568

2 0.81 0.9639 0.9931 0.9986 0.9997

0.9 4 0.6561 0.8817 0.9593 0.9860 0.9951

6 0.5314 0.7804 0.8971 0.9517 0.9774

8 0.4304 0.6756 0.8152 0.8947 0.9400

Table 2: Reliability of node disjoint multipath network when,number of levels is 2 for different node probability

Node Probability Number of Nodes Number of Paths

2 0.6835 0.8220 0.8998

0.5 4 0.2275 0.3210 0.4032

6 0.0610 0.0901 0.1183

8 0.0155 0.0232 0.0308

2 0.9323 0.9824 0.9954

0.7 4 0.6665 0.8074 0.8888

6 0.3938 0.5281 0.6326

8 0.2114 0.2997 0.3781

2 0.9986 0.9999 0.9999

0.9 4 0.9860 0.9983 0.9998

6 0.9517 0.9894 0.9976

8 0.8947 0.9658 0.9889

Figures 4, 5, and 6 show the reliability of a node disjoint network when the redundant paths are varied from 2 to 4. In Figure 4, the reliability of the network is shown when the node probability is 0.5. It also shows that, when a network has 4 paths, and the number of nodes in each path is 2, then the network reliability is 0.8998. When number of paths is 4 and the number of nodes in the path is 8, the network reliability is 0.0308. This is much less compared to the case when the number of paths is 2 or 3 and number of nodes is 2 or 4 in a path. It indicates that, to have high network reliability, number of nodes in a path should be less and it should have maximum number of node disjoint multipaths between source and sink node.

Similarly, Figures 5, and 6 show that, the network reliability is high when number of nodes is 2 and number of node disjoint paths is 4. When the node reliability is 0.7 and 0.9, number of nodes in the path is 2 and number of node

Figure 2: Path Reliability when the node probability is 0.5

Number fff Levels

Figure 3: Path Reliability when the node probability is 0.5,07 and 0.9 & number of nodes is 2 and 8

4 1 1 1 1 1 1 -fr- 2 paihs

-■-3 Paths

ч 4\ \ \\ -

■ V v \

4- \ 'Л -

v.......

....... -

Number ol Nod« on ihe palh

Figure 4: Reliability of node disjoint multipath network when, number of levels is 2 and node probability is 0.5

disjoint paths is 4 then the network reliability is 0.9954 and 0.9999 respectively. 6. Conclusion

To improve the reliability through the redundant paths in the network it is suggested to have a maximum number of paths between source and destination. It is necessary to have minimum number of nodes in each redundant path.

\ 1 1 1 2 Pains

- ■ - i Palhs

-A- 4 Pilhs

- \ 7- -

B- v.. X -

5- v* -

4- -

-L i .....

Number fjl Ncries or lire Pa1h

Figure 5: Reliability of node disjoint multipath network when, number of levels is 2 and node probability is 0.7

Figure 6: Reliability of node disjoint multipath network when, number of levels is 2 and node probability is 0.9

The network reliability is increased in node disjoint multipath networks, when each node disjoint path has maximum number of redundant paths and minimum number of nodes in each redundant path. In the node disjoint multipath network, the network reliability is very high, when number of nodes is 2 and number of paths is 4. The network reliability is low, when number of nodes is 8 and number of paths is 2.

7. References

[1] I.F.Akyildiz, W.Su.Y.Sanakrasubramaniam and E. Cayircie, " A survey of sensor networks", IEEE Communications magazine, Vol. 40, No. 8, Pages : 102-116, 2002.

[2] Al-Karaki. J.N. and Kamal. A.E "Routing techniques in wireless sensor networks: a survey",Computer Networks,Vol.11, No.6, pages: 6-28, 2004.

[3] Erdene-Ochir, O, Minier, M., Valois, F. Kountouris, A., "Toward Resilient Routing in Wireless Sensor Networks: Gradient-Based Routing in Focus," In Proceedings of Fourth International Conference on Sensor Technologies and Applications, pages: 478-483, 2010.

[4] Haibo Zhang and Hong Shen," Energy-Efficient Beaconless Geographic Routing in Wireless Sensor Networks" IEEE Transactions on Parallel And Distributed Systems, VOL. 21, No. 6, pages: 881-896, 2010.

[5] Ming-Cai Zheng, Da-Fang Zhang, Jian Luo, "Minimum Hop Routing Wireless Sensor Networks Based on Ensuring of Data Link Reliability," In Proceedings of International Conference on Mobile Ad-hoc and Sensor Networks, pages: 212-217, 2009.

[6] Long Cheng, Sajal K. Das, Jiannong Cao,Canfeng Chen and Jian M,"Distributed Minimum Transmission Multicast Routing Protocol for Wireless Sensor Networks", In Proceedings of International Conference on Parallel Processing, pages: 188-197, 2010.

[7] Hou. T, Jianping. Y, and Midkiff. S.F.," Maximizing the lifetime of wireless sensor networks through optimal single-session flow routing", IEEE TRANSACTIONS ON MOBILE COMPUTING, vol. 5, No. 9, pages:1255-1266, 2006.

[8] Ganesan.D, Govindan.R, Shenker.S, and Estrin. D,"Highly-resilient, energy-efficient multipath routing in wireless sensor networks", Mobile Computing and Communications Review, vol.1, No.2, pages: 1-13, 2002.

[9] Ganjali. Y and Keshavarzian. A, "Load balancing in ad hoc networks: Single-path routing vs. multi-path routing," In in proceeding of EEE INFOCOM 2004, pages 11201125, 2004.

[10] Marina. M. K, and Das. S. R,"Ad hoc on-demand multipath distance vector Routing" WIRELESS COMMUNICATIONS AND MOBILE COMPUTING, pages: 969-988, 2006.

[11] Radi Dezfouli.B, Razak. S. A, and Bakar. K. A," LIEMRO: A low-interference energy-efficient multipath routing protocol for improving QoS in event-based wireless sensor networks," In Fourth International Conference on Sensor Technologies and Applications (SENSORCOMM), pages 551-557, 2010.

[12] Upadhayaya. S, and Gandhi.C,"Node disjoint multipath routing considering link and node stability protocol: A characteristic evaluation," IJCSI International Journal of Computer Science Issues, vol.7, No.2, pages:18-25, 2010.

[13] Bheemalingaiah. M, Naidu, Sreenivasa Rao. D, and Varaprasad G, "Power- aware node-disjoint multipath source routing with low overhead in MANET", International Journal of Mobile Network Design and Innovation, Vol.3, No.1, pages:33-45, 2009.

[14] Abbas. A. M and Abbasi. T. A,"An analytical framework for disjoint multipath routing in mobile ad hoc networks',' In Proceeding of The IFIP International Conference on Wireless and Optical Communications Networks, pages 5-8, 2006.