Modified DSR Protocol using Link Prediction Algorithm for Reliable Ad Hoc Network

Modified DSR Protocol using Link Prediction Algorithm for Reliable Ad Hoc Network

Ashutosh

M.Tech, Department of Electronics & Communication Engineering National Institute of Technology, Jamshedpur

Mr. Bhut Nath Singh Munda

Associate Professor, Department of Electronics & Communication Engineering

National Institute of Technology, Jamshedpur

Aman Raj

M.Tech, Department of Electronics & Communication Engineering

National Institute of Technology, Jamshedpur

Abstract The MANET(mobile ad hoc network) is a self- configured, self-organized, infrastructure less network and it consists nodes which are mobile in nature and each node can transfer the data packets with each other over either radio or infrared. Some devices like laptop computers and personal digital assistants that can able to communicate with each other directly can be examples of nodes in the adhoc network. In MANET each device is free to move independently in any direction and will change the links to other devices frequently. Each node must forward the data packets until it reach at destination. MANET have variety range of applications like military battle field, disaster area, commercial use, sensor networks, etc. In those places where a temporary network architecture is required in quick interval of time for the transmission of data. The adhoc network uses no centralized authority. So, there will be no collapse in the network mobile nodes moves away from the transmission range. In this type of network nodes must be able to enter/leave the network according to their wish. Since MANET have limited transmission range of nodes so multiple hops are required for the transmission of the data packets. We have analyzed various routing protocols like DSR(Dynamic Source Routing), DSDV(Destination Sequenced Distance Vector), AODV(Ad-hoc On-demand Distance Vector). The DSR protocol have more restrictions as comparison with other routing protocols like DSDV and AODV routing protocols. The link break is one of the major limitation in DSR protocol. The link break occurs simultaneously in DSR protocol because MANET there is a frequent change of topology which cause continuous mobility of nodes results link break so we have proposed a link prediction algorithm from which DSR protocol link break issue can be resolved. The parameters like energy consumption, throughput, end to end delay has been evaluated in Network Simulator 2(NS2) simulation software.

KeywordsDSR; DSDV; AODV; Link prediction; Topology.

1. 1. INTRODUCTIONMANET(mobile adhoc network) is a type of self- organized, self-configured, infrastructure less network which are connected wirelessly. This network is adhoc because it doesnt depend on a pre-existing infrastructure like access point in wireless network or routers in wired network. An ad hoc network is a self-organizing multi-hop wireless networkthat does not rely on fixed networks or predefined networking. In this network architecture each nodes is used to forward the data packets to other nodes. So, transmission of data depends on network connectivity and the routing algorithm. Since, MANET can reconfigure itself and frequent change of topology makes mobility of the nodes. Wireless mobile adhoc network are self-configuring, dynamic networks in which nodes are free to move. The power consumption in nodes and frequent link break due to continuous mobility of nodes are major limiting factors in it. Consumption of power is always a topic of interest for researchers because in this type of network nodes has very limited amount of battery. When these networks deployed then battery replacement is not possible and due to continuous mobility of nodes there is link break occurs and in MANET for the reliable data transmission proper link management is the most required parameter. The routing protocol is the most important topic and DSR(Dynamic Source Routing), DSDV(Destination- Sequenced Distance-Vector), AODV(Ad-hoc On-demand Distance Vector) are those protocol which are frequently used in MANET. We have analyzed various protocols like DSR, DSDV, AODV routing protocols [4] and DSR protocol has more restrictions compared with DSDV and AODV routing protocols. The major limitation of DSR protocol is flooding and link break due to frequent change of topology. So, we have proposed an algorithm from which proper link management is done and it make the network architecture of the DSR protocol more reliable and the proposed algorithm will have proper link management to make the network architecture of the DSR protocol more efficient than normal DSR protocol.
   2. METHODOLOGYIn MANET(Mobile adhoc network) due to continuous mobility of nodes there is a frequent change in the topology and those mobility of nodes cause link break. In traditional routing algorithms when the link break occurs then new route discovered for the transmission of data packets and that methodology results too much delay and so much costly and ultimately the QoS(Quality of Service) of the network will beaffected. So, we have proposed a methodology which can able to predict the time at which the link will break.So, we have proposed a algorithm to predict the time after which the active link will break termed as link prediction algorithm [7]. The link prediction algorithm work on the basis power of the signal received by the nodes data packets and approximation of time and this works by approximating the time at which power of signal strength drops below the threshold power.

      When threshold power of the signal strength drops below the threshold level then it indicates that two nodes separating from each others transmission range. So, when that signal power falls below certain level then it will alert the source that the link might break. So, source either find an alternative route in advance or it will repair the link.

      We have taken three consecutive power of signal which is received by data packets received by earlier node are used to predict the link using newton divided difference method.

      = 1 + 1 + 1 2 2 1

      Where,

      =Threshold signal strength tp =Predicted time

      Where are signal power strengths for time when packets arrived at ,

      Figure 1. Link prediction used when link break and the routing of data packets done from alternative new path.

      In figure it is mentioned that if and link break then source will find an alternative new path to send data packets from source towards destination.

   3. SIMULATION AND RESULTS

1. Energy consumption:In modified DSR protocol based on link prediction the energy consumption parameter has been analysed in AODV, DSR and DSDV routing protocols Table 1 shows the analysed results in tabular form and figure 1 shows the results in graphical form.TABLE 1: ENERGY CONSUMPTION
   

   No. of nodes	Modified DSR	AODV	DSR	DSDV
   10	52.1894	65.0119	92.396	81.04
   20	52.67	74.7506	97.5785	86.72
   30	54.814	78.65	101.4862	92.99
   40	55.1219	83.5684	104.3964	96.34
   50	56.346	85.4783	109.3	115.04
   60	<>57.7455	89.388	113.22	126.89
   70	59.0557	94.2947	116.12	133.2083
   80	60.3672	98.2026	121.036	146..56
   90	61.6785	104.1157	123.948	163.239
   100	63.989	105.256	128.85	175.933

   No. of nodes	Modified DSR	AODV	DSR	DSDV
   10	52.1894	65.0119	92.396	81.04
   20	52.67	74.7506	97.5785	86.72
   30	54.814	78.65	101.4862	92.99
   40	55.1219	83.5684	104.3964	96.34
   50	56.346	85.4783	109.3	115.04
   60	57.7455	89.388	113.22	126.89
   70	59.0557	94.2947	116.12	133.2083
   80	60.3672	98.2026	121.036	146..56
   90	61.6785	104.1157	123.948	163.239
   100	63.989	105.256	128.85	175.933

    

   2 + + 2

   =

   ± 2 4

   2

   3

   The prediction of link break (tp) alerts the source even before the path breaks, allowing the source to restore the path locally or identify an alternate new route in advance.

   Fig 1: Graphical view of energy consumption

2. Average Throughput Analysis:In modified DSR protocol based on link prediction the Average throughput analysis parameter has been analysed in AODV, DSR and DSDV routing protocols Table 2 shows the analysed results in tabular form and figure 2 shows the results in graphical form.TABLE 2: AVERAGE THROUGHPUT ANALYSIS
   

   No. of nodes	DSDV	Modified DSR	DSR	AODV
   10	562.845	673.0719	626.45	650.3058
   20	565.571	676.1522	627.47	651.257
   30	565.885	676.1989	637.69	655.3657
   40	562.4785	675.079	646.672	651.5789
   50	566.7384	695.5121	648.85	652.7189
   60	588.827	701.1921	647.279	656.9789
   70	594.871	704.7455	656.222	655.2856
   80	597.311	716.478	657.5322	659.8185
   90	597.3956	724.6785	639.251	659.8383
   100	597.3659	727.305	632.0588	675.4573

   Fig 2: Graphical view of average throughput

3. End to end delay analysis:In modified DSR protocol based on link prediction the end to end analysis parameter has also been analysed in DSR, AODV and DSDV routing protocols Table 3 shows the analysed results in tabular form and figure 3 shows the results in graphical form.TABLE 3: END TO END DELAY ANALYSIS
   

   NO.OF NODES	DSR	Modified DSR	AODV	DSDV
   10	1.061	0.2347	0.6814	0.6315
   20	1.026	0.5511	0.7144	0.6364
   30	1.1577	0.6241	0.8744	0.7948
   40	1.1617	0.6125	0.911	0.8134
   50	1.1786	1.2243	0.96552	0.9574
   60	1.667	1.656	1.6828	1.55
   70	1.677	2.2271	2.1862	1.559
   80	1.7	2.2712	2.2693	1.5964
   90	1.77	2.2803	2.647	1.6801
   100	1.9743	2.2977	2.892	1.6158

   Fig 3: Graphical view of end to end delay analysis

   V. CONCLUSION

   The modification in normal DSR protocol shows good performance in end to end delay and throughput as we compared with DSR, AODV, DSDV and modified DSR protocols. The simulation result shows that the proposed approach shows better performance in the transmission of data packets from source to destination. The proposed approach link prediction made the link of the network architecture more reliable. A lots of future scope is available for the this model we have discussed here so we can make some more modification in this protocol to make it much more effective. Further these modifications which are discussed can be implemented in other routing protocols like DSDV and AODV protocols for making these routing protocols much more effective.

   REFERENCES

   1. N.M. Upadhyay, K. Gaurav and A. Kumar, “Modified DSR, an energy conserving approach to DSR protocol in MANET”, IEEE International Conference, Communications and Signal Processing (ICCSP), pp. 1146-1149, April 2014.
   2. P.Khatri, Using identity and trust with key management for achieving security in Ad hoc Networks”, IEEE InternationalConference, Advance Computing Conference (IACC), pp 271-275, February 2014.
   3. Deepti Badal and Rajendra Singh Kushwah, A Energy Efficient Approach to DSR Based Routing Protocol for Ad-Hoc Network, International Journal of Computer Application, Vol.118 No. 4, pp. 14-17, May 2015.
   4. Uma Rathore Bhatt, Abhishek Dangarh, Akansha Kashyap, Aishwarya Vyas, Performance analysis of AODV & DSR Routing Protocol for MANET, IEEE fourth International Conference on Communication Systems and Network technologies, pp.254-258, April 2014.
   5. Pravin Ranjan and R. Leela Velusamy, Route Discovery Mobile Adhoc Network using an optimum angle selection based approach, IEEE fourth International Conference on Computer and Communication technology ,pp. 199-204, September 2014.
   6. Elizabeth MR, Keong TC. A review of current routing protocols for ad hoc mobile wireless networks. IEEE Wireless Personal Communications. 1999 Apr; 6(2):4655
   7. Anita Yadav, Y. N. Singh and R. R. Singh, Improving Routing Performance in AODV with Link Prediction in Mobile Adhoc Networks, Wireless Personal Communication, Volume 81, Issue 1, pp. 603618, July 2015.
   8. Elizabeth M, Belding R, Perkins CE. Evolution and future directions of the ad hoc on-demand distance-vector routin protocol. Ad Hoc Networks Journal. 2003 Jul; 1(1):12550.
   9. Shengming J, Dajiang H, Jianqiang R. A prediction based link availability estimation for mobile ad-hoc networks. In Proceedings of Conference of IEEE Computer and Communications Socities. Infocom; Alaska, USA. 2001 Apr. p. 174552.
   10. Prashant S, Lobiyal DK. DSR with link prediction using Pareto distribution. In IEEE International Conference on Networking and Information Technology; Manila. 2010 Jun. p. 2933.
   11. Adrian JC, Leon PLD, Cormac JS. Link cache extensions for predictive routing and repair in ad hoc wireless networks. In Proceedings of the 4th IEEE Conference on Mobile and Wireless Communication Networks (MWCN); 2002 Sep. p. 537.
   12. Perkins CE. Ad hoc on Demand Distance Vector (AODV) routing. Internet draft; 2003. p. 137. Available from: http:// www.draft- ietfmane-aodv-00.txt