The Comparison of Enhanced Fault-Tolerance Mechanism of AODV with PRRTD-AODV Routing Protocol in Wireless Sensor Network

Mrs. Sharonpriya1, Mr.Jagathguru.G2 Assistant Professor, Department of Computer Science & Engineering, B. S. Abdur Rahman University Chennai-48,sharonpriya2004@gmail.com, 2 Post Graduate Student, Department of Computer Science & Engineering, B. S. Abdur Rahman University Chennai-48,jagathguru.g@gmail.com
1

Abstract
The wireless sensor networks suffer from resource limitations, high failure rates and faults caused by the defective nature of wireless communication and the wireless sensor characteristics. In this paper, the drawbacks of enhanced fault-tolerance mechanism of Ad hoc On-Demand Distance Vector (AODV) routing protocol for WSN called the ENFAT-AODV is addressed. In order to overcome the drawbacks of ENFAT-AODV the PRRTD-AODV routing protocol is proposed. The proposed protocol is to find the route which has the highest end-to-end reliability from the source to the destination based on hop-by-hop packet reception probability. The end-to-end reliability of a route is considered by the product of hop-by-hop packet reception probability of the links in the route. For reliable delivery, the protocol chooses the route which has the highest end-to-end reliability. The experiment shows that the proposed protocol is superior in terms of packet delivery ratio, reduce dropped packets, end-to-end delay, and the delay jitter.

1. Introduction
A wireless sensor network is a wireless network consisting of spatially distributed autonomous devices using sensors to cooperatively monitor physical or environmental conditions, such as temperature, sound, vibration, pressure, motion or pollutants, at different locations.

Fig. 1 Wireless sensor nodes scattered in a sensor field

Key words:
Wireless sensor network, Fault-tolerance, AODV, Backup path,Packet Reception Rate
.

WSNs are prone to failure due to physical damage, communication link errors, environmental interference, software bugs, malicious attack, and so on. It results in low reliability of performance of sensor nodes. In Section 2,The ENFAT- AODV routing protocol is depicted, In Section 3,the proposed PRRTD-AODV routing protocol is discussed. In Section 4,the results from comprehensive simulations are presented, analyzed and evaluated. Finally, the conclusion and future work in Section 5.

2.

The Enhanced Fault-Tolerant AODV Routing Protocol

ENFAT-AODV allows nodes on a main path of data delivery to obtain a backup route, which is used when their main path gets failed. The fault-tolerance mechanism of AODV routing protocol is to, when the main path gets failure, the node immediately utilizes its backup route to deliver the next coming data packets, instead of the previously broken main route, without an interruption of data packet transmission. As a result, it increases more reliability and availability of the network. ENFAT-AODV is a routing protocol utilizing also a distance vector algorithm. Each node in the system contains two separate routing tables called Main Routing Table and Backup Routing Table. The Backup Routing Table is new added from the original AODV for backup route management.

fresh enough main route to the destination, it will generate a main Route Reply (main RREP) packet. Then, the main RREP is unicasted in a hop-by-hop fashion to the source. As the main RREP is forwarded back to the source, every intermediate node which processes the main RREP creates a forward route (main route) to the destination. When the source receives the main RREP, it records the main route to the destination in its main routing table. If multiple RREPs are received by the source, the route with the shortest hop count is chosen.

2.2 Backup Route Construction

The backup route establishment is during the phase of forwarding back the main RREP message to the source. During main route reply phase, nodes on a main path (including a source node) which receive a main RREP to create a backup route towards a destination node (run a Backup Route Discovery process) by broadcasting a route request packet with Backup flag set (backup RREQ). After broadcasting the backup RREQ, the node waits for a route reply packet with Backup flag set (Backup RREP) from the destination itself or an intermediate node which has only active short backup route information for the destination in its backup routing table and is not a node on the main path of data delivery from the source to the destination. In ENFAT-AODV, the hop count is used to determine the excellent backup route as well. For another condition, if a backup RREQ reaches at a node which is on the main path of data delivery from the source to the destination, except the destination node, it discards the received backup RREQ silently. As for the destination, if it receives a backup RREQ directly from the backup route requesting node which is the destination nodes next hop along the main

Main Route Discovery

ENFAT-AODV does not require any modification to the AODV's RREQ propagation process. When a source node determines that it needs a path to a destination node for transmitting data packets and does not have one available, it broadcasts a main Route Request (main RREQ) packet for that destination node. At each intermediate node, when a main RREQ is received, a reverse route to the source is created. If the receiving node has never received this main RREQ before (by checking the Flooding ID and Source IP Address in the main RREQ packet), Also it is not the destination node and does not know a fresh enough main route to the destination, it will rebroadcast the main RREQ to its neighbors; otherwise, it will silently discard the received packet. If the receiving node is the destination or has a

path towards the source node, it also discards the received backup RREQ. The reason behind these conditions is to prevent unnecessarily wide dissemination of the backup RREQ and establishment of useless backup route.

2.3 Route Maintenance

Every time when a node forwards a data packet through the main route successfully, it checks whether it has active backup route to the destination. If such active backup route cannot be found, it runs Backup Route Discovery process again to obtain a backup route to the destination. In case it can find an active backup routing table entry for the destination in its backup routing table, it will check the expiry time of the active backup routing table entry, if the entry lifetime is almost expired, the node will update its backup route by unicasting a small message to the destination through its backup route before its backup route will be inactive and wait for a reply. 2.4 Drawbacks of ENFAT-AODV The ENFAT-AODV produces extra control packets because of backup route establishment and updating, it provides a little bit more network energy consumption as compared to the results of original AODV. However, the ENFAT-AODV performs well only in the static or very low movement scene. In a strong WSN (no failure occurred on a main path of data delivery), AODV is more suitable than ENFAT-AODV because the backup route may be useless the data packets are delivered through the same main path without utilizing a backup route. However, ENFAT-AODV is appropriate to be deployed in wireless sensor networks, especially for high failure rate systems, which are prone to a lot of failures.

Fig.2 Backup route establishment: An intermediate node generates a backup RREP.

Fig. 3 Backup route establishment: An intermediate node generates a backup RREP.

3. Packet Reception Rate Destination (PRRTD) AODV

to

In PRRTD- AODV, the Packet Reception Rate (PRR) is used as the metric, and consider the product of each PRR in a route.

The best reliable route is the one which has the highest product of each PRR. Generally, PRR is measured and defined by successful packet delivery ratio. Packet Reception Rate between node x and y is defined as equation below:
PRR(x,y) =
TotalNumberOfPacketTransmissions ----------------------------------------------------TotalNumberOfSuccessfulTransmissions

From a node x to source there might be several different paths, then there are several PRRTS of node x, but the PRRTS node x holds will be the highest. If there is another path from node x to the source of which the PRRTS is higher than the known one, node x will update the route, and also update its PRRTS. So, the proposed reliable routing protocol can be described in the following algorithms: When (node x receives an RREQ message from node y) { If (PRRTS(x) < PRRTS(y) * PRR(x, y)) { Updates route to the source from x through y. Updates PRRTS(x) = PRRTS(y) * PRR(x, y); } Forwards RREQ to the next node; } When (node x receives an RREP message from node y) { If (PRRTD(x) < PRRTD(y) * PRR(x, y)) { Updates route to the destination from x through y; Updates PRRTD(x) = PRRTD(y) * PRR(x, y); } } Relays RREP back to the next node; } By applying this algorithm, a certain node will store the highest PRRTD and PRRTS. So, the route which the source chooses to transmit data will be the route having highest packet reception rate to the destination.

That is for example shown in Fig 4 the PRR of node D from node G is 0.4. According to our protocol, node A will have 3 values of PRR in each route: 0.105 of the route A-BC-F-J; 0.252 of A-D-G-J; 0.1944 of A-E-HI-J. Then the route which has the maximum PRR will be selected to transmit data. In this case, it is A-D-G-J. An example of WSN with PRR strength Packet Reception Rate between node x and y will be changed dynamically every time x and y exchange packets.

Figure 4. An example of WSN with PRR strength

PRRTD-AODV Details
In order to gain the route which has the highest product of PRR from the source to the destination? From a node x to destination there might be several different paths, then there are several PRRTD of node x, but the PRRTD node x holds will be the highest. If there is another path from node x to the destination of which the PRRTD is higher than the known one, node x will update the route, and also update its PRRTD.

4. Simulation Results
Performance of the proposed protocol has been evaluated through simulation using

NS2. We have 2 kinds of simulations: the first is the comparison between ENFATAODV and PRRTD-AODV routing protocol For the first type of simulation, the number of nodes varies from 5, 10, 15, 20, 25. The average packet delivery ratio, the average End-To-End delay, and the average jitter delay will be considered. The following table shows the detailed setup for our simulation: Simulation time Dimension Transmission power Packet size 30minute 500x500m2 5Mbps 512bytes

ENFAT-AODVs End-to-End delay increases in contrast with PRRTD- AODV.

1.5 1 0.5 0 5 10 15 20 25 PRRTDAODV ENFATAODV

Figure 6. End-To-End Delay vs. Node Density

Table 1. Simulation parameters

Figure 7 presents a considerable difference between these two routing protocols average jitter delays.
1.2 1 0.8 0.6 0.4 0.2 0 5 10 15 20 25 PRRTDAODV ENFATAODV2

In Figure 5, PRRTD-AODV shows a better performance of average packet delivery ratio compared with ENFAT-AODV. The figure shows that the packet delivery ratio of PRRTD-AODV is more stable than ENFAT-AODV. When the number of node is lower or equal 25, the average packet delivery ratios of two protocols are approximately equal. After the number of node reaches 25.
5 4 3 2 1 0 5 10 15 20 25 ENFATAODV PRRTDAODV

Figure 7. Jitter Delay vs. Node Density

5. Conclusion
Reliability is one of the most interesting topics in wireless sensor networks. In order to increase the reliability of the network and overcome the drawbacks of ENFAT-AODV, the Packet Reception Rate is take it as a metric to find the best route one which has the highest product of Packet Reception Rate of each hop in it. The simulation result shows that the packet delivery ratio, the average End-To-End delay and the average jitters can be enhanced when compared with ENFAT-AODV

Figure 5. Packet Delivery Ratio vs Node Density

Figure 6 shows that ENFAT-AODV and PRRTD-AODV have the same average Endo-End delay when the network density is low. But when the network density goes up,

REFERENCES
[1] Che-Aron, Z., Al-Khateeb, W., and Anwar, F., The Enhanced Fault tolerant mechanism of AODV Routing Protocol for Wireless Sensor Network, International Journal on Computer Science and Network Security, 2010. HonaiPhong, N., and Myung-kyun, K., Enhancing Reliability on Wireless Sensor Network by AODVER Routing Protocol, IEEE International Conference On Computer Engineering and Technology, 2010. Kocharia, R., and Sanjeev, S., An Implementation in AODV Based on Active Route Timeout between Sensor nodes in Wireless Sensor Networks, International Journal on Computer Science and Engineering, Mar 2010. Daewon, J., Jaeseon, H., and Hyuk, L., Adaptive Contention Control For Improving End-to-End Throughput Performance of Multihop Wireless Networks, IEEE transaction On Wireless Communication, Feb 2010.

[2]

3]

[4]