2014 Fourth International Conference on Advanced Computing & Communication Technologies
Trust Based Secure on Demand Routing Protocol (TSDRP) for MANETs Akshai Aggarwal Gujarat Technological University, Ahmedabad - 380009, Gujarat, India
[email protected]
Savita Gandhi
Nirbhay Chaubey
Keyurbhai A Jani
Computer Science Department, Gujarat University, Ahmedabad – 380009, Gujarat, India
[email protected]
Computer Science Department, Institute of Science & Technology for Advanced Studies & Research, Vallabh Vidyanagar, Gujarat, India
[email protected]
Department of Information Technology, Ganpat University, Kherva, Gujarat, India
[email protected]
Abstract—Mobile Ad Hoc Networks (MANETs) are collections of mobile nodes that can communicate with one another using multihop wireless links. MANETs are often deployed in the environments, where there is no fixed infrastructure and centralized management. The nodes of mobile ad hoc networks are susceptible to compromise. In such a scenario, designing an efficient, trustworthy and secure routing protocol has been a major challenge over the last many years. In this paper, we propose a Trust Based Secure On Demand Routing Protocol called “TSDRP”. Ad hoc On-demand Distance Vector (AODV) routing protocol has been modified to implement TSDRP for making it secure to thwart attacks like Blackhole attack and DoS attack. To evaluate the performances, we have considered Packet Delivery Fraction (PDF), Average Throughput (AT) and Normalized Routing Load (NRL).
framework to provide security for preventing blackhole attack (BH) and Denial of Service (DoS) attacks . Our paper is divided into several sections as follows: Section II describes the fundamental working of AODV. In section III, a description of the blackhole attack (BH) and Denial of Service (DoS) attack are given. Section IV describes the previous work in this area. Section V provides complete understanding of our proposed Trust Based Secure On Demand Routing Protocol “TSDRP”. Section VI and VII provide the simulation set up and result analysis respectively. We have concluded this paper in Section VIII followed by the references. II. FUNDAMENTAL WORKING OF AODV Ad hoc On Demand Distance Vector (AODV) [8 - 10] is a pure on-demand route acquisition protocol which has been developed especially for MANETs and is being considered by the Internet Engineering Task Force (IETF) working group for standardization. AODV does not maintain up-to-date information about the network topology as is done by the proactive routing protocols like Destination Sequenced Distance Vector (DSDV) [11] and Optimized Link State Routing Protocol (OLSR) [12], but it creates routes on demand. As its name suggests, AODV protocol performs Route Discovery using control messages Route Request (RREQ) and Route Reply (RREP). In AODV, routes are set up by flooding the network with RREQ packets which, however, do not collect the list of the traversed hops. Rather, as RREQ traverses the network, the traversed mobile nodes store information about the source, the destination, and the mobile node from which they received the RREQ. The later information is used to set up the reverse path from destination to the source. When the RREQ reaches a mobile node, that knows a route to the destination or is the destination itself, the mobile node responds to the source with a RREP packet which is routed through the reverse path set up by the RREQ. This sets the forward route from the source to the destination. To avoid overburdening the mobiles with information about routes which are no longer (if ever) used, nodes discard this information after some time called timeout. Whenever either destination or intermediate node moves, a Route Error (RERR) is sent to the affected source nodes. When source node receives the RERR, it can reinitiate route discovery if the route is still needed. Neighbourhood information is obtained by periodically broadcasting Hello packets [9].
Keywords: MANETs, AODV; Routing Protocol, blackhole,; Denial of Service; PDF; AT; NRL.
I. INTRODUCTION Past few years, have witnessed a rapid escalation in the field of mobile computing due to proliferation of inexpensive, widely available wireless devices. Thus, it has opened vast opportunity for researchers to work on Ad Hoc Networks. In a Mobile Ad Hoc Network (MANET), nodes within one another’s wireless transmission range can communicate directly; however, nodes outside one another’s range have to rely on some other nodes to relay messages [1]. Thus, a multi-hop scenario occurs, where several intermediate hosts relay the packets sent by the source host to make them reach the destination node. MANET is one that comes together as needed, not necessarily with any support from the existing infrastructure or any other kind of fixed stations [2, 3, 4, 5]. Various routing protocols in MANETs have been proposed and these protocols can be classified into the three categories: Table-driven (or Proactive), on-demand (or Reactive) and Zone based (or Hybrid) [5,6,7]. Many experiments have been done to find the best protocol out of the available protocols. Almost all experiments lead to the most used and reliable protocol namely Ad hoc Ondemand Distance Vector (AODV), which is a reactive routing protocol. But the AODV protocol has no security measures in-built in it. Thus, it is vulnerable to many types of attacks. In this paper, we demonstrate that the blackhole attack (BH) and Denial of Service (DoS) attack are possible in AODV routing protocol and propose a trust based 978-1-4799-4910-6/14 $31.00 © 2014 IEEE DOI 10.1109/ACCT.2014.95
432
IV. PREVIOUS WORK Quite a few established works have been carried out for securing AODV routing protocol which are discussed in this section, however, no single standard protocol captures common security threats and provides guidelines to make routing protocol secure. Thus, there is still a room for continuous improvements of existing AODV routing protocol. Trust and security are tightly interrelated concepts. Using trust, tight security can be enhanced in the network and establishment of trust in MANETs is still open and challenging research field. L. Xiaoqi and M. R. Lyu et al. in [16] proposed Trusted AODV (TAODV) protocol in which he proposed modification of AODV routing protocol with node trust value. They introduced two new control packets Trust Request (TREQ) and Trust Reply (TREP) and extended routing table with four new fields; positive events, negative events, route status, opinion. Using this approach, secure route can be established by calculating trust value of each participating node in the route establishment process from source to destination. This protocol relies completely on trust value of nodes. H. S. Jassim and S. S.Yussof et al. in [17] proposed Reliant Ad hoc On demand Distance Vector Routing (RAODV). It uses modified scheme called direct and recommendations trust model which is incorporated inside AODV. This scheme assures that packets are not handed over to malicious nodes. Based on this trust value a node is selected to perform packet transfer. This protocol results in higher percentage of successful data delivery compared to normal AODV. R. S. Mangrulkar and M. Atique in [18] proposed a routing algorithm that adds a field in request packet which stores trust value indicating node trust on Neighbour. Based on level of trust factor, the routing information will be transmitted. This not only save the node's power but also save transmission bandwidth. A trusted path irrespective of its length is used for communication in the network. D. Umuhoza, J. I. Agbinya etc. in [19] proposed a framework for estimating the trust between nodes in an ad hoc network based on quality of service parameters: transit time variation, deleted, multiplied and inserted packets, processing delays. It has been shown that only two end nodes need to be involved and thereby overheads are reduced. R. Ferdous, V. Muthukkumarasamy etc. in [20] introduced a Node-based Trust Management (NTM) scheme in MANETs. NTM is based on the assumption that individual nodes themselves are responsible for their own trust level. Mathematical framework of trust in NTM along with some algorithms for trust formation in MANET is developed, based on experience characteristics offered by nodes. In 2013, R. Feng, S. Che, X. Wang and N. Yu et. al. [21] proposed a novel trust mechanism named TDSAODV. TDS-AODV can establish trusted route with minimum hops and maximum path trust based on trust mechanism denoted by TDS-AODV. In this protocol, a node makes a routing decision according to the trust values of its neighbour nodes. Finally, two routes are built: the main route with highest route trust value in the candidate routes and a backup route.
For the maintenance of the routes, two methods can be used: a) ACK messages in MAC level or b) HELLO messages in network layer. The fundamental route discovery process of AODV is depicted in Figure 1.
Figure 1. Fundamental Route Discovery Process of AODV
In general, AODV is efficient and scalable but has no inherent security mechanisms. It is full of security vulnerabilities making it easy for attacker to carry out routing attacks. The major vulnerabilities present in AODV routing protocol are due to (i) Decreasing hop count in RREQ/RREP (ii) Increasing sequence number in RREQ/RREP (iii) Forging the RERR message (iv) Possibility of attacker in impersonating as a source node S by forging a RREQ with its IP address as IP address of source node S (v) Possibility of attacker impersonating as a destination node D by forging a RREP with its IP address as IP address of the destination node D. III. DESCRIPTION OF THE BLACKHOLE ATTACK (BH) AND DENIAL OF SERVICE (DOS) ATTACK Blackhole (BH) Attack: In blackhole attack, the malicious node waits for the neighbours to initiate a RREQ packet. As the node receives the RREQ packet, it will immediately send a false RREP packet with a modified higher sequence number. So, that the source node assumes that node is having the fresh route towards the destination. The source node ignores the RREP packet received from other nodes and starts to send the data packets over malicious node. A malicious node takes all the routes towards itself. It does not allow forwarding any packet anywhere thus swallows all the data packets [13-15]. Denial of Service (DoS) attacks: Denial of Service (DoS) attacks have become one of the most worrying problems for MANET managers. In a military environment, a successful DoS attack is extremely dangerous. In DoS attack, a malicious node can override the restriction put by RREQ_RATELIMIT [9] (limit of initiating/forwarding RREQs in AODV) by increasing it or disabling it. An adversary node may choose to set the value of parameter RREQ_RATELIMIT to a very high number which allows it to flood the network with fake RREQs and lead to a kind of DoS attack. In DoS attack a genuine node cannot fairly serve other nodes due to the network-load imposed by the fake RREQs. The effects of this attack could be deadly, as the entire service discovery process of the network could be crippled.
433
Simulation results reveal that TDS-AODV can eliminate malicious nodes when building the route. The revised Dempster-Shafer (D-S) evidence theory is used to combine multiple recommended pieces of evidence and obtain the recommended trust value. In this protocol, a source establishes a main path and a backup path which are evaluated by two aspects: hop counts and trust values. The above listed works throw light on efforts to introduce and enhance reliability in mobile ad hoc network by using trust schemes. V.
The objective of carried out research work is to propose routing protocols and techniques for secure route discovery and maintain it by preventing different attacks, thus ensuring the transfer of data packets over the network safely. Framework of TSDRP routing protocol to secure route discovery and its maintenance is illustrated in Figure 2.
OVERVIEW OF TSDRP
This section describes details of our proposed Trust based secure routing protocol (TSDRP) including its architecture, framework, route discovery and its maintenance and attack prevention. TSDRP has two phases. Route discovery phase and route maintenance phase. Our proposed Framework Architecture mainly has three modules (i) Direct Observation (ii) Promiscuous Mode Observation and (iii) Trust Module for Secure Route Discovery Establishment, its Maintenance and Attack Prevention. TSDRP to prevent malicious actions like Blackhole attack and DoS attack by calculating trust value for their neighbour based on our proposed framework. AODV routing protocol has been modified to implement trust model of TSDRP by introducing Node Trust Table (NTT) and Packet Buffer (PB). NTT is used to store information for neighbours & malicious nodes. Each node stores node ID of its neighbour & calculates trust value for that node based on the packet observation through following equation 1. ntv= max(0,min(1,C*Txyi+(1-C)*(Txyi+A))) Where,
Eq. (1)
C (Constant) = 0.93, A = RQC or RPC or DC or BC, RQC (RREQ Constant) = 0.3 (Success), -0.3(failure), RPC(RREP Constant) = 0.3(Success), -0.3(failure), DC (Data Constant) = 0.4(Success), -0.4(failure). Initial, Txyi (Trust of node x for y at ith Event/time) = 0.5(default) & gets updated on Successful & failed Transmission, BC (Blackhole Constant ) = -7.2 and minimum value is set as 0 whereas maximum trust value is set as 1. Threshold value set as 0.5 (i.e. 50% of 0-1) and if any node has less than 0.5 trust value then packet coming from that node is simply dropped. Packet Buffer (PB) contains three different types of PB namely PB_RREQ, PB_RREP and PB_DATA to store control packets & data packets sent by node itself or received from other node and forwarded, based on the algorithm used in promiscuous mode & PBtimer. Each node while sending or forwarding RREQ, RREP and DATA to next node store controls packets/data packets in its respective PB buffer. PB has functionality to insert, delete, search, update, print table & also access some entries of NTT to update trust for Neighbours based on observations. Function PB_Purge( ) deletes all expired packet from the buffers when called at predefined interval. Here, network security enhancement is completely performed in the lime light of trust value.
Figure 2. Framework of TSDRP routing protocol
VI.
SIMULATION SETUP AND NETWORK SCENARIO This section explains the complete evaluation methodology along with simulation environment and network scenario in detail.
434
Simulation was performed using Network Simulator (NS-2) [22-23] for 100 sec. to measure the performance of proposed TSDRP and to be able to compare it with that of AODV. The simulation setup and network scenario are summarized in Table 1 and Table 2 respectively.
below. To test and compare the performance of TSDRP against AODV, we used NS-2.34 [22] and also developed a set of tools viz. Traffic file, Mobility Files, TCL scripts [24] and AWK programs to post-process the output trace files.
TABLE 1. SUMMARY OF SIMULATION SETUP Parameter
Value
Simulator
Ns-2(ver.2.34)
Simulation Time
100 s
Number of Nodes
70
Routing Protocol
AODV, TSDRP
Traffic Model
CBR(UDP)
Pause Time
2s
Node Mobility
60 m/s
Number of Sources
4
Terrain Area
1000m x 1000m
Mobility Model
Random Waypoint
Packet Size
512 byte
Packet Rate
4pkt/s
MAC Protocol
IEEE 802.11 with RTS/CTS
Propagation Model
Two-Ray Ground Model
Propagation Model
Two-Ray Ground Model
Antenna Type
Omni Antenna
1) Impact of number of malicious nodes (1-7 malicious node) on the routing path in the Network
a)
Result of Blackhole attack
Figure 3 (a).
PDF vs. Number of Malicious Nodes.
TABLE 2. SUMMARY OF NETWORK SCENARIOS Sr.
Network
No.
Scenario
1.
Malicious nodes
Numbers of malicious nodes are varied from 1 to 7.
2.
Traffic Load
Numbers of sources communicating in network are varied from 1 to 4.
Description
Figure 3 (b). AT vs. Number of Malicious Nodes.
A. Performance Metrics Following are the performance metrics considered to demonstrate the performance of our proposed TSDRP and AODV under different environment network scenario. 1.
Packet Delivery Fraction (PDF): This is the ratio of the number of data packets successfully delivered to the destinations to those generated by sources.
2.
Average Throughput (AT): It is the rate of successfully transmitted data packets in a unit time in the network during the simulation.
3.
Normalized Routing Load (NRL): The number of routing packets transmitted per data packet delivered at the destination.
Figure 3 (c).
NRL vs. Number of Malicious Nodes.
Figures 3(a), 3(b) and 3(c) represent the impact under blackhole attack on PDF, AT and NRL of our proposed TSDRP and that of AODV while number of malicious nodes in the network is varied from 1 to 7. In the normal scenario, when there is no malicious node in the network, performance of both routing protocols are almost similar.
VII.
RESULT ANALYSIS OF TSDRP UNDER MALICIOUS ACTIONS LIKE BLACKHOLE ATTACK AND DOS ATTACK
The simulation results of compared protocols according to the network scenario described in Table 2 are presented
435
PDF of our proposed TSDRP is consistently maintained between 80 to 85 % whereas that of AODV falls down to 5%. AT of our proposed TSDRP is always higher and it consistently maintained (above 30), whereas AT of AODV falls down drastically to below 5 even in the presence of 1 malicious node. NRL of our proposed TSDRP is always small and that of AODV is fluctuating between 50 and 1000.
consistently maintained (above 32) while that of AODV falls down drastically. Moreover, NRL of TSDRP is always small but that of AODV is monotonically increasing with increase in number of malicious nodes. 2) Impact of Traffic Load (number of sources communicating in network is 1-4) in the Network
a) Result of Blackhole attack
b) Result Of Denial Of Service (DoS) Attack
Figure 5 (a).
PDF vs. Number of Connections.
AT vs. Number of Malicious Nodes
Figure 5 (b).
AT vs. Number of Connections.
NRL vs. Number of Malicious Nodes
Figure 5 (c).
Figure 4 (a).
PDF vs. Number of Malicious Nodes
Figure 4 (b).
Figure 4 (c).
NRL vs. Number of Connections.
Figures 5(a), 5(b) and 5(c) represent the effect under blackhole attack on PDF, AT and NRL of our proposed TSDRP and that of AODV while number of connections in the network is varied from 1 to 4. It is evident from the graph that the performance of TSDRP is much better than the AODV with respect to PDF, AT and NRL.
Figures 4(a), 4(b) and 4(c) represent the effect under DoS attack on PDF, AT and NRL of our proposed TSDRP and that of AODV while number of malicious nodes in the network are varied from 1 to 7. PDF of our proposed TSDRP is maintained above 80% whereas that of AODV continuously keeps falling. AT of our proposed TSDRP is
436
b) Result of Denial of Service (DoS) Attack
delivering packets to the destinations even in the presence of increasing number of malicious node and increasing number of traffic connections in the MANETs. We have compared performance of TSDRP with that of AODV with respect to different performance metrics. Simulation has been carried out under various scenarios and after observation of performance analysis, it can be concluded that in case of blackhole attack and DoS attack TSDRP has performed very well in almost all parameters: PDF, AT and NRL as compared to AODV. REFERENCES [1] [2]
Figure 6 (a).
[3]
PDF vs. Number of Connections.
[4] [5] [6]
[7] [8]
[9] Figure 6 (b).
[10]
AT vs. Number of Connections.
[11]
[12]
[13]
[14] Figure 6 (c).
NRL vs. Number of Connections.
[15]
Figures 6(a), 6(b) and 6(c) represent the effect under DoS attack on PDF, AT and NRL of our proposed TSDRP and that of AODV when number of connections in the network is varied from 1 to 4. TSDRP is found to be outperforming AODV with respect to all the performance metrics.
[16]
[17]
[18]
VIII. CONCLUSION Our proposed TSDRP is a robust, secure on-demand routing protocol enables the secure route discovery and its maintenance in MANETs. TSDRP protocol is capable of
437
L. Zhou and Z. J. Haas, “Securing ad hoc networks”, IEEE Network Magazine, Vol. 13, pp.24-30, November/December 1999 D. P. Agrawal and Q.A. Zeng, “Introduction to Wireless and Mobile Systems”, Brooks/Cole Publishing, August 2002 C. M. Cordeiro and D. P. Agrawal, “Mobile Ad Hoc Networking,” Tutorial Presented in the 2nd Brazilian Symposium on Computer Networks” pp. 125-186, May 2002 C. M. Cordeiro, D.P.Agrawal, “Ad Hoc & Sensor Networks”, Theory and Applications, World Scientific Publishing Ltd. 2006 C. Perkins, “Ad Hoc Networking”, Pearson Education, First Impression, 2008 S. Priyadarsini, “Trust based Secured Route Discovery with improved Route Lifetime algorithm (TSRD-RL)”, International Journal of Electronics and Electrical Engineering, Vol. 1 No.1, 2009 A. M. Pushpa, “Trust Based Secure Routing In Aodv Routing Protocol”, IEEE-2009 C. E. Perkins, “The Ad Hoc On-Demand Distance-Vector Protocol (AODV)” Ad Hoc Networking, Addison-Wesley, pp. 173–219, 2001 C. Perkins, E Royer and S. Das, “Ad hoc On-demand Distance Vector (AODV) Routing”, RFC 3561, July 2003 C. Perkins and E. Royer “Ad hoc On-Demand Distance Vector (AODV) Routing”, Proceedings of the 2nd IEEE Workshop of Mobile Computing Systems and Applications, pp. 90-100, February 1999 C. E. Perkins and Pravin Bhagwat, “Highly Dynamic DestinationSequenced Distance-Vector Routing (DSDV) for Mobile Computers”, Proceedings of ACM Conference on Communications Architectures, Protocols and Applications, pp. 234 – 244, August 1994 T. Clausen, P.Jacquet, “Optimized Link State Routing Protocol (OLSR)”, Project Hipercom,INRIA, IETF Internet Draft, July 2003 http://tools.ietf.org/html/draft-ietf-manet-olsr-11 Payal N. Raj and Prashant B. Swadas, "DPRAODV: A Dynamic Learning System against Blackhole Attack in AODV based MANET", IJCSI International Jtheynal of Computer Science Issues, Vol. 2, 2009. Alem, Y. F. ; Zhao Cheng Xuan; , "Preventing black hole attack in mobile ad-hoc networks using Anomaly Detection," Future Computer and Communication (ICFCC), 2010 2nd Dokurer, Semih.”Simulation of Black hole attack in wireless Adhoc networks”. Master's thesis, AtılımUniversity, September 2006 X. Li, M. R. Lyu, J. Liu, “A Trust Model Based Routing Protocol for Secure Ad Hoc Networks”, In the Proceedings of IEEE Aerospace Conference (IEEEAC- 2004), Pp. 1286-1295, 2004 H. S. Jassim, S.Yussof et al., “A Routing Protocol based on Trusted and shortest Path selection for Mobile Ad hoc Network”, IEEE 9th Malaysia International Conference on Communications 2009 R. S. Mangrulkar, M. Atique: “Trust based secured adhoc On demand Distance Vector Routing protocol for mobile adhoc network”. In: Sixth International Conference on Wireless Communication and Sensor Networks (WCSN), 2010 .
[19] D. Umuhoza, J. I. Agbinya, C.W. Omlin: “Estimation of Trust Metrics for MANET Using QoSParameter and Source Routing Algorithms”. In: The 2nd International Conference on Wireless Broadband and Ultra Wideband Communications, 2007. [20] R. Ferdous, V. Muthukkumarasamy, A. Sattar: “Trust Management Scheme for Mobile Ad-Hoc Networks”. In: IEEE 10th International Conference on Computer and Information Technology (CIT), 2010. [21] R. Feng, S. Che, X. Wang, and N. Yu et. al., “A Credible Routing Based on a Novel Trust Mechanism in Ad Hoc Networks”, International Journal of Distributed Sensor Networks Volume 2013, Article ID 652051, 12 pages, 2013
[22] “The Network Simulator-NS-2”, Home http://www.isi.edu/nsnam/ns/ns-build.html
page,
[Online]
[23] NS-2 Documentation, [Online] http://www.isi.edu/nsnam/ns/nsdocumentation.html [24] http://elmurod.net/?p=196 “Adding Malicious Node to AODV [25] Marc Greis’ Tutorial for the UCB/LBNL/VINT Network Simulator “ns”; http://www.isi.edu/nsnam/ns/tutorial/ [26] http://www.ece.ubc.ca/~teerawat/NS2.htm [27] Network Simulator - 2 (NS-2); 2.html
438
http://mohit.ueuo.com/NS-
