Improvement Dynamic Source Routing Protocol by Localization for Ad hoc Networks

Description

IJCSIS is an open access publishing venue for research in general computer science and information security. Target Audience: IT academics, university IT faculties; industry IT departments; government departments; the mobile industry and computing industry. Coverage includes: security infrastructures, network security: Internet security, content protection, cryptography, steganography and formal methods in information security; computer science, computer applications, multimedia systems, software, information systems, intelligent systems, web services, data mining, wireless communication, networking and technologies, innovation technology and management. The average paper acceptance rate for IJCSIS issues is kept at 25-30% with an aim to provide selective research work of quality in the areas of computer science and engineering. Thanks for your contributions in September 2010 issue and we are grateful to the experienced team of reviewers for providing valuable comments.

Shared by: ijcsis

Stats

views:: 472
posted:: 10/10/2010
language:: English
pages:: 6

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Improvement Dynamic Source Routing Protocol by
Localization for Ad hoc Networks
Mehdi Khazaei
Kermanshah University of Technology
Information Technology Engineering Group
Kermanshah, Iran
khazaei84@gmail.com

Abstract-Ad hoc networks are temporary networks with a Issues more important for routing problem in ad hoc network
dynamic topology which don t have any established therefore, each routing protocol should be fault tolerant in
infrastructure or centralized administration. Consequently, in probable route failures [1].
recent years many researchers have focused on these networks.
These networks need efficient routing protocols in terms of Routing protocols in conventional wired networks are
Quality of Services (QOS) metrics. Ad hoc networks suffer from usually based upon either distance vector or link state routing
frequent and rapid topology changes that cause many challenges algorithms as a DSDV [2], CGSR [2] and FSR [2]. These
in their routing. Most of the routing protocols like this proposed algorithms require periodic routing advertisements to be
protocol try to find a route between source and destination broadcast by each router. These conventional routing
nodes and when any route is expired, a new route would be algorithms are clearly not efficient for type of dynamic
formed. Rapid route reconstruction may cause the network changes which may occur in an ad-hoc network [2, 3]. A new
inefficiency. Therefore, we have to decrease this processes. The class of on-demand routing protocols e.g., DSR [4, 5], TORA
proposed protocol as DSR routing protocol build one routes [2], AODV [6, 7]) for mobile ad hoc networks has been
between source and destination but create backup routes during developed with the goal of minimizing the routing overhead.
the route reply process, route maintenance process and use local These Protocols reactively discover and maintain only the
recovery process in order to improve the data transfer and needed routes, in contrast to proactive protocols (e.g., DSDV
attended to QOS. The protocol performance is demonstrated by [2]) which maintain all routes regardless of their usage. The
using the simulation results obtain from the global mobile
key characteristic of an on-demand protocol is the source-
simulation software (Glomosim). The experimental results show
initiated route discovery procedure. Whenever a traffic source
that this protocol can decrease the packet loss ratio and increase
data transfer rather than DSR that, it is useful for the
needs a route, it initiates a Route discovery process by sending
applications that need a high level of reliability. a route request for the destination (typically via a network-
wide flood) and Waits for a route reply. Each route discovery
Keywords; Protocol, Routing, Local Recovery, Mobile Ad-hoc flood is Associated with significant latency and overhead. This
Networks is particularly true for large networks. Therefore, for on-
demand routing to be effective, it is desirable to keep the route
I. Introduction discovery frequency low [8].
Routing in ad hoc networks is a very challenging issue due Single route routing allows the establishment of one route
to nodes mobility, dynamic topology, frequent link breakage, between a source and single destination node. Because of node
limitation of nodes (memory, battery, bandwidth, and mobility, the route may be broken frequently; therefore,
processing power), and limited transmission range of the node having replacement route in cache memory to transmit data
and lack of central point like base stations or servers. On the will improve the fault tolerance and higher aggregate
other hand, there are a lot of performance metrics and quality bandwidth in these networks. Beside of this, by repairing the
services which should be satisfied in an ad hoc network like broken routes locally, the number of route rediscovery
end -to-end data throughput, average end-to-end data delay, processes can be decreased. This paper improves the fault
packet loss ratio, Normalized Routing Load, Packet Delivery tolerance and increase reliability by obtain replacement routes
Ratio, and route optimality. Each protocol can satisfy some of in RREP1 and RRER2 processes and local recovery process
these metrics and has some drawbacks in terms of other together. This optimization is done on the DSR protocol.
metrics. Furthermore, due to the nature of ad hoc networks
(distributed and cooperated routing), even for a fixed metric, The rest of this paper is organized as follows. In section II
each protocol can show a different performance with different the DSR protocol is explained. Section III deals with the
networks features like number of mobile nodes, mobility of related works and Section IV describe the proposed protocol
nodes, pause time and . So by comparing between different mechanism in detail. Performance evaluation by simulation is
ad hoc routing protocols we can extract very important presented in section V and concluding remarks are made in
information about the performance of these protocols in the section VI.
Different situations. In the other hands, the nodes mobility and 1
Route reply
the probability of links failure may cause the fault tolerance 2
Route error

1 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

II. Dynamic Source Routing Protocol (DSR) Reliability is defined as the probability of sending data
DSR is an on-demand routing protocol for ad hoc successfully from the source to the destination node within a
networks. Like any source routing protocol, in DSR the source time window. MP-DSR selects a set of routes that satisfy a
includes the full route in the packets header. The intermediate specific end-to-end reliability Requirement [12].
nodes use this to forward packets towards the destination and MSR is attempts to minimize the end-to-end delay for
maintain a route cache containing routes to other nodes. In sending a data from source to destination by using multi-route
following subsections DSR operation are briefly described [4]. routing and intelligent traffic allocation [13].
A. Route discovery CHAMP is multi-route protocol that uses round-robin
If the source does not have a route to the destination in its traffic allocation to keep routes fresh. It also employs
route cache, it broadcasts a RREQ3 message specifying the cooperative packet caching to improve fault tolerance and
destination node for which the route is requested. The RREQ takes advantage of temporal locality in routing, where a
message includes a route record which specifies the sequence dropped packet is a recently sent packet [9].
of nodes traversed by the message. When an intermediate The local recovery techniques have been used in some
node receives a RREQ, it checks to see if it is already in the routing protocols for route maintenance processes. This
Route record. If it is, it drops the message. This is done to technique aims to reduce the frequency of RREQ floods
prevent routing loops. If the intermediate node had received triggered by nodes that are located in the broken routes [14].
the RREQ before, then it also drops the message. The
intermediate node forwards the RREQ to the next hop SLR is one of these routing protocols. It modifies DSR,
according to the route specified in the header. When the using a new route recovery mechanism called bypass routing.
destination receives the RREQ, it sends back a route reply Bypass routing utilizes both route caches and local error
message. If the destination has a route to the source in its route recovery techniques during failures to reduce the control
cache, then it can send a RREP message a long this route. overhead [15].
Otherwise, the RREP message can be sent along the reverse
LRR is also another routing protocol that uses local
route back to the source. Intermediate nodes may also use their
recovery techniques. In this protocol the information of next-
route cache to reply To RREQs. If an intermediate node has a
to-next (NN) node is stored at each intermediate node along
route to the destination in its cache, then it can append the
the route. After detecting a broken link by an upstream node, it
route to the route record in the RREQ, and send an RREP back
sends out the non-propagating requests to find another node
to the source containing this route. This can help limit
which is in contact with itself and the NN node on the route;
flooding of The RREQ. However, if the cached route is out-
therefore the routes can be repaired locally in the shortest
of-date it can result in the source receiving stale routes [4].
possible time [16].
B. Route maintenance MRFT protocol improves fault tolerance in DSR and SMR
When a node detects a broken link while trying to forward protocols. To achieve the goal of decreasing the packet loss
a packet to the next hop, it sends a RERR message back to the ratio and increasing fault-tolerance, MRFT uses both multi-
source containing the link in error. When an RERR message is route and local recovery techniques together [17].
received, all routes containing the link in error are deleted at
that node [4]. IV. The Proposed Protocol
4
This paper proposes IM-DSR protocol to improve fault
III. Related Works
tolerance and QOS in DSR protocol. To achieve the goal of
Ad hoc routing protocols such as ADOV, DSR, DSDV and decreasing the packet loss ratio and increasing fault-tolerance,
OLSR have been investigated on the ad hoc networks in the IM-DSR uses local recovery techniques and alternate route
past few years. The investigations of the performance of these during route reply and route maintenance that reliability in the
protocols on the ad hoc networks have produced many useful network would be increased. IM-DSR modifying the route
results. However, we have seen very limited findings of how discovery, route reply and route maintenance processes in
these Ad-hoc routing protocols perform on wireless ad hoc DSR. The IM-DSR protocol is including route discovery,
networks. Nonetheless, we can see many attempts at route reply, route maintenance and local recovery processes
developing routing protocols for ad hoc networks under the that discussed in the following subsection.
different deployment of ad hoc networks [8, 9 and 10]. In
following are brought some of these attempts. A. Route Discovery Process
IM-DSR is an on demand routing protocol that builds
SMR is an on demand routing protocol that uses single route using request/reply cycles. When the source node
maximally disjoint routes to transmit data packets. Unlike needs to send data to the destination but no route information
DSR, intermediate nodes do not allow to send RREP packets is known, it floods RREQ packets over the entire network.
back to the source instead, only destination nodes reply to the When an intermediate node receives a RREQ that is not a
RREQ packets and selects maximally-disjoint routes [11]. duplicate, it appends its ID to the packet and rebroadcasts it. In
MP-DSR is a multi-route QOS aware extension to DSR. It IM-DSR all of the duplicate RREQs that are received by
focuses on a QOS metric, end-to-end reliability. End-to-end intermediate nodes are dropped. In IM-DSR, intermediate
4
3
Route Request Improvement-DSR

2 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Nodes are allowed to send RREPs back to the source even If transitional packet is a RRER, it would examine how
when they have route information to the destination in their many times the packet would be saved. If it was the first
route caches. time, the meaning is, the packet would be saved by
examining a route cache and finding alternate route. The
B. Route Reply Process
RRER is sent to destination through that route then the
When receiving the first RREQ, the destination sends a RRER is made and it will report the broken link to the
RREP back to the source. After that, the destination node source of RRER. If it were not first time or if alternate
consumes other RREQs. The Route-Number of the RREP is route were not in the route cache of node, the RRER
one. would be deleted and only a RRER would be sent to the
After receiving RREP packet by the intermediate nodes, if source node. Therefore RRER is saved only for one time
it has not route with same length to destination node, they by the IM-DSR protocol.
store the routes in their route caches. The Route-Number of If transitional packet would be data, it would examine
this routes are zero and used in route maintenance process for how many times the packet would be saved. If this time
improving the break routes also sending data if there is not were less than three, the data packet would be sent by
main route. examining their route cache and alternate route then it
Look at the Fig. 1, suppose that node (H) sends the RREP will send a RRER to the source node. If these times were
to the source node (A), a route is found and sent to node (A) more than three or if alternate route was not in the route
by RREP is A-C-D-G-F-H. Now, suppose that the RREP is cache of node the data packet would be deleted and only a
received by node (C) which is middle node. Node (C) saves RRER would be sent to the source node. Therefore data
the routes to destination (H) which is C-D-G-F-H, additionally packet is saved for three times by the IM-DSR protocol.
node (C) save C-D, C-D-G and C-D-G-F routes in the route If very data packet passed the same route towards
caches. destination node and they faced the broken link (while
When the source node receive the RREP, it will store the sending), the node which recognized an error, for every data
route and use that for transmit data. packet send a RRER to the source node. In order to avoid this
item every node before sending RRER to source node,
C. Route Maintenance and Local Recovery Processes examine this is a first RRER or not. If it was not send, a new
During a transmission session, a problem such as node RRER send to source node.
mobility, or low battery power might be raised, which can lead Every node which recognizes the broken link and makes
to break an existing route and lose route connectivity. This the RRER, examined the route cache in order to find alternate
may force a route rediscovery process by flooding RREQs route and put it in the RRER, which the node that received the
over the network. To avoid this phenomenon, IM-DSR uses RRER, replaces the route in the RRER with the previous
following mechanism that one of them is local recovery invalid route in the route cache.
techniques. Using local recovery techniques is very useful
despite they consume the limited power of each nodes. Fig. 2 shows this matter. The source node (A) sends data to
the destination node (H) through A-C-D-H. When node (D)
Suppose that a node finds a broken link, while sending a sends the data packet, it will find the failure in node (H). By
packet. At first, it seeks the route cache and deletes all routes examine the route cache, it chooses the alternate route D-G-F-
include the broken link, and then according to kind of the H, hence, the data packet is sent to destination node (H)
packet one of the following items is done: through this route, then RRER is made and it is sent to the
If transitional packet would be a RREQ, the node would node (A). This packet includes the alternate route D-G-F-H.
not send RRER to the source node. The node (A) receives the RRER and deletes A-C-D-H from
route cache and replaces A-C-D-G-F-H.
If transitional packet would be a RREP, send RRER to the
node which makes the RREP. Every middle and source node which receives the RRER,
examine those route in route cache which includes the broken
link and should be deleted from cache and if packet included
alternate route, exploited that and saved in route cache with
number two. If in buffer, data packet waiting to send toward
alternate route destination, it will send through that route.
Such as the Fig. 2, while passing the RRER, node (C) adds C-
D-G-F-H to the route cache.
If a node who detected a broken link cannot find any
alternate route in its route cache, so it drops the data packet
and sends a RERR without any repaired route to the source.
After that, because of performing local recovery process by
the node that detects the broken link, the source node does not
trigger the rediscovery process immediately. After detecting
the broken link, node sends a RERR to the source and starts
Figure 1. Routes Structure in an Ad-hoc Network the local recovery process simultaneously. To repair the route

3 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Figure 2. Ad-hoc Network Structure, when occur break link
Figure 3. Ad-hoc Network Structure, when Occur Local Recovery
Locally, node triggers a local query to its neighbors. The
neighbors reply if they have any valid route to the destination. (GLOMOSIM) [18]. The Simulation environment consists of
When node receives the RRP5, it repairs the primary route and 50 numbers of nodes in a rectangular region of size 1500
then sends a RRP back to the source, like shown in the Fig. 3. meters by 1500 meters.
In Fig. 3 when node (D) finds the failure in node (G), First The nodes are randomly placed in the region and each of
examined the route cache for replacement route, if not found them has a radio propagation range of 250 meters. 200
then begun local recovery request which node (F) reply with constant bit rate (CBR) flows are deployed for data
F-G-H as repaired route. When node (D) receives this route transmission. Simulation time is 300 seconds. The random
update its memory cache and sends data from this route to waypoint model is chosen as the node mobility model. All
destination and send RRP to the source node that here is node data packets are 512 bytes. Band width is 2 Mbps and
(A). When node (A) receives the RRP deletes A-C-D-G-H simulation done for 0, 1, 3, 5 and 10 second as stop time.
from route cache and replaces A-C-D-F-G-H. Every middle Minimum and maximum speed for nodes are 0 m/s and 30
node which receives RRP, examine if included repaired route, m/s. IEEE 802.11 selected for MAC layer protocol.
then exploited that and saved in route cache with number two.
In Fig. 3 while passing RRP, node (C) adds C-D-F-G-H to the B. Simulation Results
route cache more node (C) adds C-D, C-D-F and C-D-F-G to Fig. 4 shows packet delivery ratio (PDR) in every two
route cache. protocol. It is defined as ratio of the number of data packets
delivered to the destinations generated by sources. In DSR
If the source node has not primary route to send data, it protocol if sending node of data has not any routes to send
will use the repaired route. If no data packet containing data, it would start route request process by RREQ. In every
repaired route was reported to the source node for a certain request finds a route and if it was invalid, the route request
amount of time, then it sends a new RREQ to the destination process begins again. The IM-DSR protocol against DSR
node. After receiving the new RREQ by the destination node, protocol obtains routes in route reply and Maintenance process
it performs the route rediscovery process that was described in and used them for sending data. This would lead to increase
subsection A. the packet delivery ratio in IM-DSR protocol.
In the IM-DSR protocol, route with Route-Number one is Fig. 5 shows the number of RREQ (NRQ) in every two
main route. Route with Route-number zero are obtain by route protocols. This number is sum of the RREQ in request
reply process and route with Route-Number two are those process. Fig. 5 shows that this number in DSR protocol is
which are obtain in route maintenance process and known as greater than IM-DSR protocol because in every route
repaired route. discovery find only one route and if this route would be
When the source node wants to send data to a destination, invalid begin route discovery again. But IM-DSR finds routes
it tries to use the primary, which has the highest priority (the in route reply, maintenance and local recovery processes,
routes with route-Number one) at first, if not exist, it will use hence reduces this number.
obtain route (the route with Route-Number zero) otherwise
use repaired route (the route with Route-Number two) to send
data.
V. Performance Evaluation
A. Simulation Environment
In order to demonstrate the effectiveness of IM-DSR
protocol, we evaluate our proposed protocol and compare its
performance to the DSR (uni-route). We have implemented
IM-DSR protocol using the Global Mobile Simulation library
5 Figure 4. Packet Delivery Ratio (PDR)
Route Repaired Packet

4 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Figure 5. Number of RREQ (NRQ) Figure 8. Average of Delay (AVD)

Fig. 6 shows sum of the hop-count (SHC) in the network. VI. Conclusion
This number is summation of the all routes hop count that In this paper, we proposed a new routing protocol called
lower value leads to less delay in network. This number is less IM-DSR to provide higher QOS in ad hoc networks. This
in IM-DSR protocol because of less route request process and protocol is an extension of DSR to increase the reliability by
found optimal routes. modifying the route discovery and route maintenance
Fig. 7 shows the number of broken links (NBL) in the processes in DSR also added the local recovery techniques to
network. Problem such as node mobility, low battery power or DSR. The simulation results show that IM-DSR is very
congestion might be raised, which can lead to break an effective in decreasing the packet loss and also increasing the
existing route and lose route connectivity. Number of broken fault tolerance ad hoc networks. In all of the cases, IM-DSR
link in IM-DSR protocol is less than DSR protocol because of has the higher packet delivery ratio than the DSR protocol
used alternate routes that making in route request, reply and while improving the overhead of route maintenance,
maintenance processes and used local recovery process. maintaining acceptable overhead. Therefore the proposed
routing protocol is very useful for the applications that need a
Fig. 8 shows an average of delay (AVD) in two protocols. high level of reliability.
Because of sending data from route with long hop-count and
data packets waiting more time in buffer, delay in DSR REFERENCES
protocol is more than IM-DSR protocol. The IM-DSR [1] V. Nazari Talooki, J. Rodriguez, Quality of Service for Flat Routing
protocol which, between the source and destination, selected Protocols in Mobile Ad hoc Networks , Mobimedia 09, London, UK,
optimal routes and saved multi-route node decreases the delay September, 2009.
in comparisons with DSR protocol. [2] E. M. Royer, A Review of current Routing Protocols for Ad Hoc Mobile
Wireless Networks , IEEE Personal Communications, vol. 6, no. 2, pp.
46-55, 1999.
[3] M. IILyas, The handbook of ad hoc wireless networks , CRC press,
2003.
[4] D. Johnson, Et. al, The Dynamic Source Routing Protocol for Mobile Ad
Hoc Network Internet Draft, www.ietf.org/internetdrafts/ draft-ietf
manet-dsr-10.txt (July 2004).
[5] D. B. Johnson, D. A. Maltz, Dynamic Source Routing in Ad Hoc
Wireless Networks , Mobile Computing, Edited by Imielinski, Korth,
Chapter 5, pp. 153-181, Kluwer Academic Publishers, 1996.
[6] C. Perkins, E. Royer, Ad Hoc On-Demand Distance Vector (AODV)
Routing (Internet Draft), www.ietf.org/internet-drafts/draft-ietfmanet-
aodv-13.txt (Feb 2003).
[7] C. Perkins, E. Royer, Ad-hoc On-demand Distance Vector Routing , 2nd
Figure 6. Sum of Hop-Count (SHC) IEEE Workshop on Mobile Computing Systems and Applications, New
Orleans, LA, United States, pp. 90-100, Feb. 1999.
[8] Q. Jiang, D. Manivannan, Routing protocols for Sensor networks , IEEE,
pages 93 98, 2004.
[9]- S. Mueller, R.P. Tsang and D. Ghosal, Multipath Routing in Mobile Ad
hoc Networks Issues and Challenges , Proceedings of CA
USA,27,USA,2007.
[10] K. Wu, J. Harms, On-Demand Multipath Routing for Mobile Ad Hoc
Networks , 4th European Personal Mobile Communications Conference
(EPMCC 01), Vienna, Austria, Feb. 2001.
[11] S. Ju-Lee, M. Gerla, Split Multipath Routing with Maximally Disjoint
Paths in Ad Hoc Networks , IEEE International Conference on
Communications, vol. 10, pp. 3201-3205, Helsinki, Jun. 2001.
[12] R. Leung, E. Poon, A. C. Chan, B.Li, MP-DSR: A QoS-Aware Multi
Figure 7. Number of Broken Link (NBL) path Dynamic Source Routing Protocol for Wireless Ad-Hoc Networks ,

5 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

26th Annual IEEE International Conference on Local Computer
Networks (LCN2001), pp. 132-141, Tampa, FL, United States, Nov.
2001.
[13] X. Li, L. Cuthbert, On-Demand Node-Disjoint Multipath Routing in
Wireless Ad Hoc Networks , 29th Annual IEEE International
Conference on Local Computer Networks (LCN2004), Tampa, FL,
United States, pp. 419-420, Nov. 2004.
[14] E. Esmaeili, P. Akhlaghi, M. Dehghan and M. Fathi, A New Multi-Path
Routing Algorithm with Local Recovery Capability in Mobile Ad hoc
Networks , Fifth International Symposium, Patras, Greece, July, 2006.
[15] C. Sengul, R. Kravets, Bypass Routing: an On-Demand Local Recovery
Protocol for Ad Hoc Networks , Ad Hoc Networks, In press (Available
online 2005).
[16] R. Duggirala, R. Gupta, Q. Zeng, D. P. Agrawal, Performance
Enhancements of Ad Hoc Networks with Localized Route Repair ,
IEEE Transactions on Computers, vol. 52, no. 7, pp. 854-861, 2003.
[17] M. khazaei, R. berangi A multi-path routing protocol with fault
tolerance in mobile ad hoc networks , Proceedings of IEEE
international CSI, 14th, tehran, iran, Oct, 2009.
[18] UCLA Parallel Computing Laboratory and wireless Adaptive Mobility
Laboratory, GloMoSim: A Scalable Simulation Environment for
Wireless and Wired Network.

Mehdi khazaei received the bachelor s and master s
degrees from Iran University of Science and
Technology (IUST), in 2004 and 2007, respectively.
Currently He is lectureship in Kermanshah University
of Technology (KUT) and his researches focused on
wireless networks, especially ad hoc networks.

6 http://sites.google.com/site/ijcsis/
ISSN 1947-5500