A Robust Zone Disjoint Multipath Routing Protocol

A Robust Zone Disjoint Multipath Routing Protocol M. Arvind 1, Dr. M. Sugumaran 2

Department of Computer Science and Engineering, Pondicherry Engineering College, India (1) [email protected]

Abstract

A new multipath routing protocol which is disjoint at the zone level is proposed. The proposed routing protocol is based on the hybrid Zone based Hierarchical Link State Routing Protocol (ZHLS). The proposed work establishes multiple zone disjoint paths between the source and the destination zones. The establishment of multiple paths between the source and the destination zones helps in increasing the network lifetime and also in reducing the network congestion and overhead. The proposed protocol is simulated using the Network Simulator (ns-2.34).

Keywords: multipath routing protocol, Zone based hierarchical link state routing protocol (ZHLS), hybrid routing protocols. 1. Introduction

A mobile ad-hoc network is a rapidly deployable and self organising network which does not depend upon any fixed infrastructure. The network consists of nodes which are mobile and which are equipped with wireless interfaces. The nodes in the network act as servers, clients as well as routers. The nodes communicate with each other directly if they are within the transmission range of one and another. The nodes if they are not within each other's transmission range make use of the intermediate nodes to transmit the data packets to the destination node. The intermediate nodes act as routers which forward the data packets to the destination. Mobile ad hoc networks are characterised by dynamic topology, limited bandwidth and scarce wireless resources. The nodes that comprise the network operate by making use of limited battery power. If certain nodes in a network are used extensively their battery power is depleted and hence they are not available for future use. This reduces the lifetime of the network. Single path routing protocols make extensive use of the nodes that comprise the path between the source and the destination nodes to transfer packets and hence their battery power is quickly depleted and this results in reduction of the lifetime of the network. Moreover as a result of the rapidly changing topology the establishment of a single path between the source and the destination nodes results in frequent route discovery which increases

the overhead associated with the protocol. Multipath routing protocols establish multiple paths between the source and the destination nodes. The establishment of multiple paths between the source and the destination increases the lifetime of the network, ensures load balancing, reduces congestion in the network. Also if one of the paths established between the source and the destination is broken, the other available paths can be made use of to route data. This reduces the frequency of the route discovery process. The proposed routing protocol establishes multiple zone disjoint paths between the source and the destination zones thereby overcoming the drawbacks associated with the single path routing protocols. 2. Related Work 2.1. Zone Routing Protocol (ZRP)

Zone Routing Protocol (ZRP) [1] is a hybrid ad hoc routing protocol which limits the proactive nature to the nodes local neighbourhood. It is composed of an IntrAzone Routing Protocol (IARP) [2], IntErzone Routing Protocol (IERP) [3] and Bordercast Resolution Protocol (BRP) [4]. IARP is any proactive routing protocol such as Optimized Link state Routing [5], Destination Sequenced Distance Vector (DSDV) [6] routing protocol where the proactive route updates are limited to the local neighbourhood of the node. The scope of the IARP is defined by the routing zone radius: the distance in hops that IARP route updates are relayed. The IARP makes use of the Neighbour Discovery Protocol (NDP) to obtain information about a node's neighbours. The routing zone for a node is defined as the set of nodes whose minimum distance in hop is no greater than the zone radius. Peripheral nodes are those nodes whose distance in number of hops from the node is exactly equal to the radius of the zone. All the other nodes in the zone whose distance from the node is less than the radius of the zone are called interior nodes. Every node will broadcast its local routing information within its own routing zone. Nodes maintain a routing table which contains paths to all the nodes within its local neighbourhood. A route within a nodes routing zone can be found out immediately without any delay. Any existing proactive routing protocol can be adapted to IARP. IERP is the global reactive routing component of ZRP. Any reactive routing protocol like Ad hoc on demand

M.Arvind,Dr.M.Sugumaran, Int. J. Comp. Tech. Appl., Vol 2 (3), 559-564

559

ISSN:2229-6093

mailto:[email protected]�

routing protocol (AODV) [7], Dynamic Source Routing (DSR) [8] can be used as the IERP routing protocol. When a node needs a route to a node which is not within it routing zone, IERP will help to find it. IERP initiates a route discovery process; IERP make use of the Bordercasting Routing Protocol (BRP) to route the query outwards to the peripheral nodes. Any existing reactive routing protocol can be adapted to IERP. The peripheral nodes perform the bordercasting again if they cannot reply to the query. Finally, the query will be spread throughout the network. If the destination node is present in the zone of a node a route reply is sent back to the source node. The path to a destination node consists of a list of peripheral nodes. A new route discovery is required whenever a peripheral node changes. As a result of the dynamic topology of the network there is a need for frequent route discovery process. This increases the overhead associated with the routing protocol.

2.1. Zone based Hierarchical Link State Routing Protocol (ZHLS)

Zone-based Hierarchical Link State Routing Protocol (ZHLS) [9] is a hybrid routing protocol which incorporates location information into a novel peer-to-peer hierarchical approach. ZHLS limits the proactive behavior within a zone and maintains a reactive approach outside the zone. The network is divided into non overlapping disjoint zones. Each node make use of the Global Positioning System (GPS) or other geo-location mechanisms as proposed in [10] to identify its location in the network and then maps its current location onto the zone map which is loaded into every node before deployment. Two types of Link State Packets namely the node Link State Packets (node LSP) and the zone Link State Packets (zone LSP) are used to maintain a two level topology Information about the node level topology is obtained by making use of the node LSP and information about the zone level topology is obtained by making use of the zone LSP. Each node sends out a link request and those nodes in its communication range reply with link responses. When all the link responses from the neighboring nodes are received the node generates its node LSP which contains the node ID's and the zone ID's of the neighboring nodes and zones. This process is repeated asynchronously by all the nodes in the zone. The node LSP's are propagated locally only within a zone. When all the node LSP's are received the nodes know how the nodes are interconnected to each other in a zone and which zones to which their zone is connected to. The Intra Zone Routing table is constructed by applying a variation of the shortest path first algorithm on the received node LSP's. When all the node LSP's are received each node knows to which zones its zone

is connected to and by making use of this information generate a zone LSP which is globally flooded throughout the network. When all the zone LSP's are received the nodes know the zone level topology. The zone LSP's of all the nodes of a particular zone are identical and only one zone LSP from one zone is globally broadcasted and hence this reduces the overhead associated with the protocol. In order to identify the current zone in which the destination is present a location search mechanism is used. A location request is sent to all the zones in the network. The location request is routed by making use of the inter zone routing protocol to all the zones in the network. A location reply is then sent back to the source node. The inter zone routing table is made use of by all the zones in the network to route the data to the destination node whereas only the destination zone makes use of the intra zone routing table to route the data packets to the destination node. 3. Proposed Work The proposed work is based on the hybrid Zone based hierarchical link state routing protocol (ZHLS). The proposed work establishes multiple zone disjoint paths between the source and the destination zones. The process of establishing the node level topology and the zone level topology is identical to the ZHLS routing protocol. 3.1. Node Level Topology

The node level topology is obtained by a node by sending out a location request. Nodes which are within the communication range of the node in turn reply with link responses. When all the link responses are received the node generates a node LSP to all the nodes in its zone. The above process is repeated asynchronously by all the nodes in the zone. At the end of this process all the nodes know the node level topology. The node level topology and the zone level topology is shown in the Figure 1 and 2 respectively.

Figure 1. Node Level Topology

a

b

d

c

e

f

j

g

h

i

1 2

3

4

5


560

ISSN:2229-6093

Figure 2. Zone Level Topology The process of obtaining the node level topology is shown in the Figures 3(a) - (d).

(a)

(b)

(c)

(d)

Figure 3 Obtaining Node Level Topology (a) Node a broadcasts link request to its neighbors. (b) Node a receives link responses from its neighbors. (c) Node a generates its node LSP which is broadcasted within the zone. (d) All the nodes perform the process asynchronously. At the end of this stage the nodes in the zone contain a list of all the node LSP's of all the nodes in the network as shown in the Table 1. A variation of the shortest path algorithm is used to build the intra zone routing table as shown in Table 2.

Table 1. Node LSP's in Zone 1

Source Node LSP

a b, c, d, 4

b a, e

c a, 3

d a

e b, f, 2

f e,2

3.2. Zone Level Topology After obtaining the node level topology, the zone

level topology is obtained by making use of the information present in the received node LSP's to generate the Zone LSP. Every node in the network makes use of the information obtained from the node LSP's to generate the zone LSP. All the nodes in a particular zone generate the same zone LSP. So only one zone LSP from one particular zone are broadcasted throughout the network. Since all the zone LSP's generated from a zone are the same duplicate zone LSP's are dropped and not broadcasted throughout the network. This reduces the overhead associated with the protocol to some extent. After obtaining all the zone LSP's the interzone routing table is obtained by making use

1 2

3

4

5 6

7 8

9

a

b

d

c

g

1 2

3

4

5

a b

d

c

g

1 2

3

4

5

a b

d

c

g

1 2

3

4

5

a

b

d

c

e

f

j

g

h

i

1 2

3

4

5


561

ISSN:2229-6093

of a variant of the shortest path first routing algorithm as shown in Table 4. Table 3 shows the received zone LSP's.

Table 2. Intra Zone Routing Table of Node a

Destination Next Node

b b

c c

d d

e b

f b

2 b

3 c

4 g

Table 3. Zone LSP's

Source Zone LSP

1 2,3,4

2 1,6

3 1,7,8

4 1,9

5 6,9

6 2,5

7 3

8 3

9 4,5

3.3. Location Search and Routing Mechanism

Whenever a node wants to send data to the destination node it has to identify the current zone ID of the destination node in order to successfully route data to the destination node. As the nodes are mobile they cannot be associated with a fixed zone ID and hence it is mandatory for the source to initiate a location search mechanism to identify the current zone ID of the destination node. Whenever a source node wants to send data to the destination node it checks its intra zone routing table to see if the destination node is present in the table if so it routes the packets to the destination node through the path maintained in the intra zone routing table. If the source node does not have a path to the destination node in the intra zone routing table it initiates a location search mechanism to identify the zone in which the destination node is present in. A location request of the following format is unicasted to all the zones in the network <Source node ID, Source Zone ID, Destination Node ID, Dest. Zone ID>. Whenever a gateway node receives the location request it checks its zone to

see if the destination node is present in its zone if so, a location reply of the following format is sent back to the source node. <Dest. Node ID, Dest. Zone ID, Source Node ID, Source Zone ID>. The source node then makes use of its inter zone routing table to route the packets to the destination node. All the nodes in the intermediate zones route the data packets according to its inter zone routing table. Only the destination zone nodes make use of the intra zone routing table to route the packets to the destination node.

Table 4. Inter Zone Routing Table of Node a

Destination Next Zone Next Node

2 2 b

3 3 c

4 4 g

5 4 g

6 2 b

7 3 c

8 3 c

9 4 g

3.4. Multipath Formation

In order to establish multiple zone disjoint paths between the source and the destination zones whenever a destination node receives a route request it starts up a Route Request (RREQ) collection timer. As long as the timer runs, the destination node receives and processes duplicate route requests originating from the same source node. It receives and processes only the route requests which are disjoint at the zone level (i.e.) the paths between the source and the destination nodes have no common zones between each other. the only exception is that the source and the destination zones can be the same. When the timer runs out the route reply is sent back to the source node through the various zone disjoint paths established between the source and the destination nodes. The source node receives the requests and stores them and sends the data through the various zone disjoint paths. This results in the establishment of multiple zone disjoint paths between the source and the destination zones.

4. Simulation

The proposed work is simulated in the Network Simulator version 2.34 (ns 2.34). The network is split into nine non overlapping disjoint zones each of size 300x300 units. The simulation has been designed with a total of 100 nodes distributed over the nine disjoint zones. The nodes are mobile and are free to move within a zone as well as from one zone to another.

Because the proposed work establishes multiple paths between the source and the destination nodes,


562

ISSN:2229-6093

the packet delivery ratio increases because multiple paths are established and they can be used to route packets to the destination node simultaneously as shown in Fig. 4.

Fig. 4. Comparison of Packet Delivery Ratio between the existing ZHLS and the proposed

work.

As our proposed work establishes multiple zone

disjoint paths between the source and the destination nodes, even if interference spreads across a particular area since the paths established are disjoint with each other it does not affect all the paths and hence data can still be routed to the destination node and this greatly reduces the packet loss associated with the proposed work and also if an attacker wants to disrupt the delivery service he has to compromise multiple nodes present in multiple paths to be successful and this also greatly reduces the packet loss associated with the proposed work which is shown in Figure 5.

Figure 5. Comparison of packet loss between the existing ZHLS and the proposed work.

5. Conclusion and Future Work In this paper a zone disjoint multipath routing protocol which establishes multiple zone disjoint paths between the source and the destination nodes is proposed. The proposed work helps in load balancing, increasing the life time of the network and increasing the packet delivery ratio and reducing the loss of packets associated with the proposed protocol. The simulation results verify our claims and prove that the proposed work reduces the packet loss and increases the packet delivery ratio when compared to the existing ZHLS protocol. As an area of focus in future, the possibility of applying dedicated security mechanisms to the proposed work may be carried out to further enhance the security of the proposed protocol. 6. References [1] Z. J. Haas, “The zone routing protocol (ZRP) for ad hoc networks”, Internet Draft, July 2002. [2] Haas, Zygmunt J., Pearlman, Marc R., Samar, P.: “Intrazone Routing Protocol (IARP)”, IETF Internet Draft, draft-ietf-manet-iarp-01.txt, June 2001. [3] Haas, Zygmunt J., Pearlman, Marc R., Samar, P.: “Interzone Routing Protocol (IERP)”, IETF Internet Draft, draft-ietf-manet-ierp-01.txt, June 2001. [4] Haas, Zygmunt J., Pearlman, Marc R., Samar, P.: “The Bordercast Resolution Protocol (BRP) for Ad Hoc Networks”, IETF Internet Draft, draft-ietf-manet-brp-01.txt, June 2001. [5] P. Jacquet, P. Muhlethaler, A. Qayyum, “Optimized Link State Routing Protocol”, Internet Draft, draft-ietf-manet olsr-00.txt, November 1998. [6] C. E. Perkins and P. Bhagwat, “Highly Dynamic Destination-Sequenced Distance-Vector Routing (DSDV) for Mobile Computers”,Computer Communications Review, pp. 234–44, Oct 1994. [7] C. E. Perkins and E. Royer, “Ad-hoc on-demand distance vector routing”, in Proceedings of the Second IEEE Workshop on Mobile Computing Systems and Applications, (New Orleans LA), pp. 90–100, Feb. 1999. [8] D.B. Johnson, D.A. Maltz, “ Dynamic source routing in adhoc wireless networks”, in T.Imielinski, H. Korth (Eds.), Mobile Computing, Kluwer Academic Publishers, Dordrecht, pp. 153–181,1994. [9] M. Joa-Ng and I.-T. Lu, “A peer-to-peer zone-based two-level link state routing for mobile ad hoc networks,” IEEE Journal on Selected Areas in Communications, vol. 17, pp. 1415–1425,1999. [10] M. Joa-Ng, “Routing protocol and medium access protocol for mobile adhoc networks,” Ph.D. dissertation, Dept. Elect. Eng., Polytechnic Univ., Brooklyn, NY, 1999. [11] Takashi Hamma, Takashi Katoh, Bhed Bahadur Bista, Toyoo Takata, "An Efficient ZHLS Routing Protocol for Mobile Ad Hoc Networks",in


563

ISSN:2229-6093

Porc. of 17th International Conference on Database and Expert Systems Applications (DEXA'06), pp.66-70, 2006. [12] Kentaro Aburada, Naonobu Okazaki and Mirang Park,"Proposal of a zone disjoint multi-path routing for ad hoc networks", in Proc. of 8th Asia Pacific Symposium on Information and Telecommunication technology (APSITT), pp.1-4, 2010. [13] Siguang Chen and Meng Wu, "Anonymous Multipath Routing Protocol Based on Secret Sharing in Mobile Ad Hoc Networks", in Proc. of International Conference on Measuring Technology & Mechatronics Automation, pp. 582-585, 2010. [14] Siguang Chen and Meng Wu, "Secure Multipath Routing Based On Secret Sharing in Mobile Ad Hoc Networks", in Proc. of IEEE International Conference on Network Infrastructure and Digital Content, pp. 539-542, 2009.


564

ISSN:2229-6093

Documents

A Robust Zone Disjoint Multipath Routing Protocol