HANDLING CROSS-LAYER ATTACKS USING NEIGHBORS MONITORING SCHEME AND SWARM INTELLIGENCE IN MANET

309 views

Published on

The standard MAC protocol widely used for Mobile Adhoc Networks (MANETs) is IEEE 802.11.
When attacks in MAC layer are left as such without paying attention, it could possibly disturb channel access and
consequently may cause wastage of resources in terms of bandwidth and power. In this paper, a swarm based detection
and defense technique is proposed for routing and MAC layer attacks in MANET. Using forward and backward ants,
the technique obtains mean value of nodes between the first received RREQ and RREP packets. Based on this
estimation, the source node decides the node as valid or malicious. Moreover the MAC layer parameters namely
number of neighbors identified by the MAC layer, number of neighbors identified by the routing layer, the number of
recent MAC receptions and the number of recent routing protocol receptions are used to determine the node state. The
source node uses these two node state estimation techniques to construct the reliable path to the destination. This
proposed technique improves the network performance and at the same time prevents attackers intelligently.

Published in: Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
309
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
17
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

HANDLING CROSS-LAYER ATTACKS USING NEIGHBORS MONITORING SCHEME AND SWARM INTELLIGENCE IN MANET

  1. 1. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 41HANDLING CROSS-LAYER ATTACKS USING NEIGHBORSMONITORING SCHEME AND SWARM INTELLIGENCE IN MANETG. IndiraniDepartment of CSEAnnamalai UniversityAnnamalai nagar- 608002,IndiaK.SelvakumarDepartment of CSE,Annamalai University,Annamalai nagar- 608002,IndiaABSTRACT: The standard MAC protocol widely used for Mobile Adhoc Networks (MANETs) is IEEE 802.11.When attacks in MAC layer are left as such without paying attention, it could possibly disturb channel access andconsequently may cause wastage of resources in terms of bandwidth and power. In this paper, a swarm based detectionand defense technique is proposed for routing and MAC layer attacks in MANET. Using forward and backward ants,the technique obtains mean value of nodes between the first received RREQ and RREP packets. Based on thisestimation, the source node decides the node as valid or malicious. Moreover the MAC layer parameters namelynumber of neighbors identified by the MAC layer, number of neighbors identified by the routing layer, the number ofrecent MAC receptions and the number of recent routing protocol receptions are used to determine the node state. Thesource node uses these two node state estimation techniques to construct the reliable path to the destination. Thisproposed technique improves the network performance and at the same time prevents attackers intelligently.KEYWORDS: MANET, MAC, RREQ, RREP, Neighbors monitoring scheme1. INTRODUCTION1.1 Mobile ad-Hoc networks (MANETs)A Mobile ad hoc network is a collection of wirelessmobile nodes that can allow people and devices tocommunicate with each other without the help of anyexisting centralized infrastructure. A MANET is a selfconfiguring network to form an arbitrary andtemporary network. Here each mobile node canfunction as a router or host. Often the topology ofMANET changes as nodes are mobile. Here therouting protocol plays a major role in determining theroutes required for communication between the sourceand destination through the intermediate nodes. TheMANET gets new attractive applications since theyoffer good communication in the changingenvironment. The MANET can be used in theapplications such as rescue operations, tacticaloperations, environmental monitoring, conferences,connecting soldiers in battlefields and social orbusiness application such as Public and Personal AreaNetworks.[1] The weaknesses of ad hoc networks aredynamic topology, lack of infrastructure, exposure ofnodes and channels [2].1.2 General attacks in MANETThe MANETs are more prone to security att ackswhen compared to the wired networks. Due to therestricted features of the MANET such as restrictedprotection of every individual node, uneven behaviourof connectivity, deficit of certification authority,centralized monitoring or administration, security isdifficult to maintain in these networks. In such awireless network, attacks can enter either from insidethe network or from outside. In any case, each node inMANET has to be ready for facing attacks. Inparticular, an attack from a compromised node insidethe network is destructive and difficult to getidentified. [3] Attacks in MANET are generallyclassified as active and passive attacks which aredescribed below.1.2.1 Active attacks:An active attack causes various degrees of damage tothe network depending on the type of attack. It isfurther classified into two categories of attacks suchas internal and external attack.• The internal attacks are performed bythe compromised nodes that belong tothe network.• The external attacks are performed bythe nodes that are not part of thenetwork.Wormhole attack, black hole attack, Byzantine attack,information disclosure and resource consumptionattack are some of the examples of active attacks.
  2. 2. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 421.2.2 Passive attacksIn this attack, the attacker does not interrupt theregular behavior of the network but intrudes the dataexchanged in the network without changing it. Thistype of attack is difficult to identify as the normaloperation of the network is not affected. [3] [4]. Thereis an attack which is specific to the passive attackwhose brief description about it is given below:• Snooping:Snooping refers to the illicit use of anotherperson’s data. This may refer to watching e-mail informally that is displayed onanother’s computer screen or observingother people typing. Also more complicatedsnooping involves a software program toexamine the process of a computer ornetwork device. [5]1.3 Cross layer attacksCross-layer attacks emerge from lack of interactionbetween MAC and routing layers. These attackspropagate from the MAC layer, where they aremanifested as Denial of Service (DoS) attacks, to therouting layer, causing serious degradation of networkperformance in terms of the achieved throughput,latency and connectivity. An attacker can causecongestion in the network by either generating anexcessive amount of traffic or by generating specifictraffic patterns that prevent certain nodes fromcommunicating with other nodes. [6]1.3.1 Effects of cross layer attacks(i) This type of attack exploits thevulnerability of a particular layer(attack point) to launch the attack,but ultimately aspires to disruptthe operations of another layer(target point) [7](ii) By incorporating cross-layerinformation and networkcommunication into the jammingattack, a resource-constrainedadversary can significantlyincrease the efficiency of theattack by targeting specificcommunication channels, helpingto counteract the effect of the anti-jamming systems [9].(iii) Reduces the attacker’s probabilitybeing detected.(iv) Reduce the cost to conduct theattack successfully(v) Achieve the attack goals that maynot be feasible through attackactivities in a single layer.1.3.2 Issues of cross layer attacks(i) It is possible to modify/developanomaly detection in eachindividual layer.(ii) Cross layer defense architecturecan be possible which may bebased on all the layers and alsoindividual layers.(iii) The capability of attackers getseven more strengthened by thepresence of cognitive radio. [9](iv) Due to the anonymization of thenetworks, the cross layer attackershave increased their efficiency[10].1.4 Problem identificationThe security issues in ad hoc routing have beenextensively studied. However, attack strategies thattarget interaction between MAC layer and routinglayer have not been fully addressed. A new class ofattacks, cross-layer attacks, emerges from lack ofinteraction between MAC and routing layers. Theseattacks propagate from the MAC layer, where they areproduced as Denial of Service (DoS) attacks, to therouting layer, causing serious degradation of networkperformance in terms of the achieved throughput,latency and connectivity.In the previous works, only routing attacks considered(i.e) network layer attacks. As an extension work,cross-layer attacks are going to be considered whichinclude both MAC and network layer and provide adetection technique using the same SWARMtechniques.2. LITERATURE REVIEWPatrick Tague et al [8] investigate a class ofcoordinated jamming attacks in which multiplejammers collaboratively apply knowledge about thenetwork layer functionality to efficiently reduce thethroughput of network traffic. They show how aconstrained optimization framework can be used tocharacterize coordinated jamming attacks and allowthe impact of the attack to be quantified from theperspective of the network. Using this network-centricinterpretation of jamming attacks, a network designercan attain a greater understanding of the potentialthreat of jamming. To illustrate their approach, theypropose and evaluate a variety of metrics to model theattack impact, serving both as adversarial objectivefunctions and as network evaluation metrics
  3. 3. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 43Wenkai Wang et al [9] has proposed cross layerattacks and defending the cross layer attacks incognitive radios. The existing research on securityissues in cognitive radio networks mainly focuses onattack and defense in individual network layers.However, the attackers do not necessarily restrictthemselves within the boundaries of network layers. Inthis paper, they design cross-layer attack strategiesthat can largely increase the attackers’ power orreducing their risk of being detected. As a case study,we investigate the coordinated report-false-sensingdata attack (PHY layer) and small-back-off-windowattack (MAC layer). Furthermore, they propose atrust-based cross-layer defense framework that relieson abnormal detection in individual layers and cross-layer trust fusion.John Felix Charles Joseph et al[14] has proposed across-layer based routing attack detection system forad hoc networks. Previous work that uses mostly audittrails collected from the routing protocol suffers frominadequacy of features to construct a reliable modelfor detecting anomalous routing behavior. On theother hand, use of linear detectors lead to very highfalse positives and false negatives because of theinherent on-linear nature of the feature space. In thiswork, these issues are addressed by collating featuresfrom multiple protocols at different layers and using anon-linear detector based on Support Vector Machine(SVM). The consequent problem of computationalexpense of the detection process is addressed by acombination of novel data reduction techniques.Simulation results show that the performance of theproposed CRADS is far superior than conventionalprotocol-specific detection systems.Andriy Panchenko et al [10] have proposed a crosslayer attack on anonymizing networks. Network layeranonymization protects only some of the user’spersonal identification information, namely networkaddresses of the communicating parties. However,even if the lower layers of communication provideperfect protection for the user’s profile, informationleakage on the application layer destroys the wholeeffort. Currently, all widespread implementations ofanonymizing networks do not use a holistic approachand therefore, neither filter nor actively warn usersabout information leakage from the upper layers,which may look innocent to the end user. The extendexisting work on security of anonymizing networks totake into account additional information leakage fromthe application layer. Further they show, under whichconditions and how this kind of information can beused not only to build an extensive user profile at “lowcosts”, but also to speed up traditional attacks that aretargeted at the network layer identification of users’peer partnersLei Guang et al [11] demonstrate a new class ofprotocol-compliant exploits that initiates at the MAClayer but targets ad hoc on-demand routingmechanisms. A misbehaved node implementing thistype of attacks completely follows the specificationsof IEEE802.11 standard and the existing on-demandrouting protocols. However, it can cause routingshortcut attacks or detour attacks. They detail theexploits against two on-demand routing protocols:AODV and DSR. They evaluate the impact of suchattacks on the network performance and proposePrevention from Shortcut Attack and Detour Attack(PSD) to mitigate their impacts.A.Rajaram et al [12] have developed a trust basedsecurity protocol based on a MAC-layer approachwhich attains confidentiality and authentication ofpackets in both routing and link layers of MANETs. Inthe first phase of the protocol, we design a trust basedpacket forwarding scheme for detecting and isolatingthe malicious nodes using the routing layerinformation. It uses trust values to favor packetforwarding by maintaining a trust counter for eachnode. A node is punished or rewarded by decreasingor increasing the trust counter. If the trust countervalue falls below a trust threshold, the correspondingintermediate node is marked as malicious. In the nextphase of the protocol, they provide link-layer securityusing the CBC-X mode of authentication andencryption3. PROPOSED SOLUTION3.1 OverviewIn this paper, a swarm based detection and defensetechnique for cross layer attacks is proposed inMANET. The technique makes use of ant colonybased optimization (ACO) technique to detect attacksin the MANET. During route discovery time, thesource broadcasts RREQ message and the destinationresponds with RREP message. In this broadcasting,each intermediate node stores the time of first receivedRREQ and RREP packets. The source injects forwardant (FA) to compute the mean value between receivedtime of RREQ and RREP packets. The backward ant(BA) updates this information and reaches the sourcenode. While receiving the mean value of nodes, thesource compares mean value with predefinedthreshold value and marks node as valid and maliciousnode. To detect MAC layer attack, each node inMANET calculates Dn using four parameters namelynumber of neighbors identified by the MAC layer,number of neighbors identified by the routing layer,the number of recent MAC receptions and the numberof recent routing protocol receptions. When Dn is zerothe node is identified as the valid node otherwise thenode is identified as the malicious node. When thesource constructs path to the destination, it chooses thepath such that the path contains only the valid nodesby omitting malicious nodes.3.2 Network architectureIn MANET, IEEE 802.11 is used as a standard MACprotocol. The Distributed Coordination Function
  4. 4. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 44(DCF) in IEEE 802.11 combines Carrier SenseMultiple Access/Collision Avoidance (CSMA/CA)with a Request to Send/Clear to Send (RTS/CTS)handshake technique to avoid collisions. Both hiddennode and exposed problems are solved usingRTS/CTS handshake mechanism. At MAC layer, datatransmission channel is divided by inter packet gaps,which are termed as Inter Frame Spaces (IFS).Further, channel access can be provided to the nodesbased on its priority. [13]3.3 Swarm based node monitoring strategyThe MAC and routing layer must support each otherto detect attacker and adversaries during theoperations in MAC layer. It is possible to have moreattacks in MAC layer. The attacker may pretend thechannel as busy such that no node or user transmitstheir data. This attack consequently leads to DoSattack in the network, which drastically reduces thenetwork performance. To detect and prevent such kindof attacks, our technique utilizes swarm based nodemonitoring strategy.When the source has data to be transmitted, itbroadcasts RREQ message and the destinationbroadcasts back the RREP message towards thesource. While receiving RREQ message, eachintermediate node records the time of first RREQpacket it has received. The RREQ packet is kepttacked with its RREQ sequence number. Similarly,each intermediate node stores the time and sequencenumber of first RREP packet it has received. The tablethat contains this information is known as countertable (C- Table). The format of C-table is shown intable – 1.To monitor the network, the source periodically injectsforward ants (FA) in the network. Each FA travelstowards random destination to collect mean timebetween received times of RREQ and RREP packets.While returning from the destination, the backward ant(BA) updates this mean time in its pheromone table.Finally, the BA reaches the destination.Every source has mean table (MN-Table) to store themean times of nodes collected by ants. When the BAreaches the source node, it updates the mean value ofnodes in M-Table. Let Thrd be the route discoverythreshold value. The source compares the mean valueof every node with Thrd. Mean value of nodes lessthan or equal to Thrd are noted as valid nodes. Nodesthat have mean value more than Thrd are noted asmalicious node.Algorithm-11. Let Thrd be the route discovery threshold value2. Consider ni be the mobile node, where i=1, 2…nand mvi be the mean value of node i3. Each node stores time of first received RREQ andRREP packet in C-Table4. FA and BA collect and update mv values ofintermediate nodes in M-Table5. Source compares mvi with Thrd5.1 If (mvi ≤= Thrd) then5.2 Node is considered as valid node5.3 Else if (mvi > Thrd) then5.4 Node is considered as maliciousnode6. End ifWhile constructing path from source to destination,the source considers the valid nodes rather thanmalicious nodes.
  5. 5. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 45Table-1 Format of C-Table3.4 Neighbors monitoring schemeIn this section, four parameters are monitored for eachand every node in the MANET[14]. They are1) Total number of neighborsfound by the MAC layerwhich is denoted as NMAC2) Total number of neighborsfound by the routing layerwhich is denoted as NR3) Total number of receptionsfound by the MAC layerwhich is denoted as RMAC4) Total number of receptionsfound by the routing layerwhich is denoted as RRUsing these four parameters, DN is calculated usingthe formula.RMAcRMACRMACNRRRRNND+−−≈2)(|)((1)Algorithm-21. Let S and D be source and destinationrespectively2. Let DN be the value calculated forevery node in the network.3. If (DN = 0) Then3.1 The node state is a valid node4. Else if (DN not equal to 0) Then4.1 The node state is a maliciousnode5. End ifThis state of node is maintained by each node in MN-Table. The MN-Table has the following format,Table-2 Format of MN-Table3.5 Data transmission through securechannelWhile selecting path, the source uses the two nodestate detection techniques described in section 3.3 and3.4. The source selects the path to the destination suchthat it contains only the valid nodes. Thereby, ourtechnique provides defense against MAC layerattacks.Figure-2 Secure Data transmission4. SIMULATION RESULTS4.1 Simulation model and parametersHere the Network Simulator Version-2 (NS2) is used[14] to simulate our proposed algorithm. In oursimulation, the channel capacity of mobile hosts is setto the same value: 2 Mbps. The distributedcoordination function (DCF) of IEEE 802.11 forwireless LANs as the MAC layer protocol is used. Ithas the functionality to notify the network layer aboutlink breakage.In this simulation, mobile nodes move in a 1000 meterx 1000 meter region for 50 seconds simulation time.The numbers of nodes are varied as 20, 40, 60, 80 and100. It is assumed that each node movesindependently with the same average speed. All nodeshave the same transmission range of 250 meters. Inthis simulation, the node speed is 10 m/s. Thesimulated traffic is Constant Bit Rate (CBR). Thesimulation settings and parameters are summarized intable 3IntermediateNode IDSource ID DestinationIDReceivedTime ofRREQ PacketSequenceNumber ofRREQ PacketReceivedTime ofRREP PacketSequenceNumber ofRREP PacketNode ID Mean Value Node State
  6. 6. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 46No. of Nodes 20, 40, 60, 80 and100.Area Size 1000 X 1000Mac 802.11Radio Range 250mSimulation Time 50 secTraffic Source CBRPacket Size 512Speed 10m/sNo. Of Attackers 1,2,3,4 and 5.Table 3: Simulation Settings4.2 Performance metricsWe evaluate mainly the performance according to thefollowing metrics.Average Packet Delivery Ratio: It is the ratio of thenumber .of packets received successfully and the totalnumber of packets transmitted.Average-end-to-end Delay: It is the total time delaytaken by the nodes to transmit the data to the receiver.Average Packet Drop: It is the average number ofpackets dropped by the misbehaving nodes.Here the Swarm Based Detection and DefenseTechnique using Neighborhood monitoring schemefor Routing and MAC layer Attacks (SBDT-NB) iscompared with Cross-Layer Attack vs. Cross-LayerDefense (CACD) [9].4.3 ResultsA. Based on attackersIn the first experiment, the number of attackersare varied as 1, 2, 3, 4 and 5 in a 100 nodenetwork.Figure 3: Nodes Vs DelayFigure 4: Nodes Vs Delivery RatioFigure 5: Nodes Vs PktsDropFigure 6: Nodes Vs OverheadFrom figure 3, we can see that the delay of ourproposed SBDT-NB is less than the existingCACD technique.From figure 4, we can see that the delivery ratioof our proposed SBDT-NB is higher than theexisting CACD technique.From figure 5, we can see that the packet drop ofour proposed SBDT-NB is less than the existingCACD technique.Nodes Vs Delay024681012141620 40 60 80 100NodesDelay(Sec)CACDSBDT-NBNodes Vs DeliveryRatio00.20.40.60.811.220 40 60 80 100NodesDeliveryRatioCACDSBDT-NBNodes Vs PktsDrop0200040006000800010000120001400020 40 60 80 100NodesPktsDropCACDSBDT-NBNodes Vs Overhead0500010000150002000025000300003500020 40 60 80 100NodesOverheadCACDSBDT-NB
  7. 7. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 47From figure 6, we can see that the overhead ofour proposed SBDT-NB is less than the existingCACD technique.B. Based on nodesIn the second experiment we vary the number ofnodes as 20, 40, 60, 80 and 100.Figure 7: Attackers Vs DelayFigure 8: Attackers Vs Delivery RatioFigure 9: Attackers Vs DropFigure 10: Attackers Vs OverheadFrom figure 7, we can see that the delay of ourproposed SBDT-NB is less than the existing CACDtechnique.From figure 8, we can see that the delivery ratio of ourproposed SBDT-NB is higher than the existing CACDtechnique.From figure 9, we can see that the packet drop of ourproposed SBDT-NB is less than the existing CACDtechnique.From figure 10, we can see that the overhead of ourproposed SBDT-NB is less than the existing CACDtechnique.5. CONCLUSIONIn this paper, a swarm based detection and defensetechnique with neighborhood monitoring scheme isproposed for cross layer attacks in MANET. Usingforward and backward ants, the technique obtainsmean value of nodes, which is the difference betweenfirst received RREQ and RREP packets. Whilereceiving the mean value of nodes, the sourcecompares mean value with predefined threshold valueand marks node as valid and malicious node. Further,using the four MAC layer parameters the node state isidentified. Using, these two node state estimationtechnique, the source constructs path to the destinationby omitting the malicious nodes. The performance ofour technique is proved through simulation results.This Proposed technique prevents attackers wiselyand improves network performance.6. REFERENCES[1] Sevil ¸ Sen, John A. Clark, “A GrammaticalEvolution Approach to Intrusion Detection on MobileAd Hoc Networks”, Proceedings of the second ACMconference on Wireless network security 2009[2] Yian Huang, Wenke Lee, “A CooperativeIntrusion Detection System for Ad Hoc Networks”,International journal of computer applications, 2011Attackers Vs Delay05101520251 2 3 4 5AttackersDelay(Sec)CACDSBDT-NBAttackers Vs DeliveryRatio00.20.40.60.811.21 2 3 4 5AttackersDeliveryRatioCACDSBDT-NBAttackers Vs PktsDrop05000100001500020000250001 2 3 4 5AttackersPktsDropCACDSBDT-NBAttackers Vs Overhead010000200003000040000500001 2 3 4 5AttackersOverheadCACDSBDT-NB
  8. 8. International Journal of Computer Applications Technology and ResearchVolume 2– Issue 1, 41-48, 2013www.ijcat.com 48[3] Sureyya Mutlu, Guray Yilmaz, “A DistributedCooperative Trust Based Intrusion DetectionFramework for MANETs”, IARIA SeventhInternational Conference on Networking and Service,2011[4] N.Shanthi, DR.LGanesan and DR.K.Ramar,“Study of Different Attacks on Multicast Mobile AdHoc Network”, Journal of Theoretical and AppliedInformation Technology, 2009[5] Abhay Kumar Rai, Rajiv Ranjan Tewari, SaurabhKant Upadhyay, Different Types of Attacks onIntegrated MANET-Internet Communication,International Journal of Computer Science andSecurity (IJCSS), 2009[6] Svetlana Radosavac, Nassir Benammar and JohnS. Baras, “Cross-layer attacks in wireless ad hocnetworks”, 38th Conference on Information Sciencesand Systems (CISS), Princeton, March 17-19 2004[7] Kaigui Bian, Jung-Min Park, and Ruiliang Chen,“Stasis Trap: Cross-Layer Stealthy Attacks in WirelessAd Hoc Networks”, K. Bian, J.M. Park and R. Chen,"Stasis Trap: Cross- Layer Stealthy Attacks inWireless Ad Hoc Networks", In Proceedings of IEEEGLOBECOM, 2006.[8] Patrick Tague, David Slater, Guevara Noubir, andRadha Poovendran, “Quantifying the Impact ofEfficient Cross-Layer Jamming Attacks via NetworkTraffic Flows”, Network Security Lab (NSL),University of Washington, Tech.Rep., 2009.[9] Wenkai Wang and Yan (Lindsay) Sun, HushengLi, Zhu Han, “Cross-Layer Attack and Defense inCognitive Radio Networks”, IEEE GlobeCOM, 2010[10] Andriy Panchenko, Lexi Pimenidis, “Cross-LayerAttack on Anonymizing Networks”, IEEEInternational Conference on Telecommunications,(ICT 2008), pp-1-7, 2008.[11] Lei Guang, Chadi Assi, and AbderrahimBenslimane, “Interlayer Attacks in Mobile Ad HocNetworks”, Springer, Mobile Ad-hoc and SensorNetworks Lecture Notes in Computer Science Volume4325, pp 436-448 , 2006[12] A.Rajaram, Dr. S. Palaniswami, “The Trust-Based MAC-Layer Security Protocol for Mobile Adhoc Networks”, International Journal on ComputerScience and Engineering Vol. 02, No. 02, 2010[13] Yihong Zhou, Dapeng Wu and Scott M. Nettles,“Analyzing and Preventing MAC-Layer Denial ofService Attacks for Stock 802.11 Systems”, inproceedings of the Workshop on BWSA,BROADNETS, USA, 2004.[14] John Felix Charles Joseph∗, Amitabha Das∗,Boon-Chong Seet†, Bu-Sung Lee, “CRADS:Integrated Cross Layer approach for DetectingRouting Attacks in MANETs”, WCNCproceedings,2008[15] Network Simulator: http:///www.isi.edu/nsnam/ns

×