rupesh et al, international journal of computer technology ... · using cooperative contact and...

Using Cooperative Contact and Standing-based Watchdogs Recognizing Selfish Nodes in MANET

Rupesh PG scholar, CSE, Lingraj Appa Engineering

College, Bidar, India Email:[email protected]

Sangameshwar Kawdi Assistant Professor, CSE, Lingraj Appa

Engineering College, Bidar, India Email: [email protected]

Abstract

In mobile adhoc networks, generating and maintaining anonymity for any adhoc node is challenging because of the node mobility, dynamic network topology, cooperative nature of the network.. Existing techniques based on cryptosystem and broadcasting cannot be easily adapted to MANET because of their extensive cryptographic computation and/or large communication overhead. Mobile ad-hoc networks (MANETs) assume that mobile nodes voluntary cooperate in order to work properly. This cooperation is a cost-intensive activity and some nodes can refuse to cooperate, leading to a selfish node behaviour. Thus, the overall network performance could be seriously affected. The use of watchdogs is a well-known mechanism to detect selfish nodes. However, the detection process performed by watchdogs can fail, generating false positives and false negatives that can induce to wrong operations. Moreover, relying on local watchdogs alone can lead to poor performance when detecting selfish nodes, in term of precision and speed.

Keywords: Local Watchdog, MANET, Performance Evaluation, Selfish Node

1. Introduction

MANETs are suitable for many applications, such as creating survivable, powerful interacting for emergency/recovery features, disaster comfort projects, as well as army systems. MANETs include independent collection of mobile phone users that be connected over data transfer usage restricted wi-fi

links. All these issues make protection, executing security, and even node catch important issues. Without comfort security, adversaries can easily learn the identities of the communication parties and the relevant information that two users are communicating adversaries can easily learn the identities of the communication parties and the relevant information that two users are communicating While most prior work in secure MANET routing focused on security issues, less attention has been de- chosen to privacy. Note that, in this perspective, privacy does not mean confidentiality of communication (i.e., data) among MANET nodes. The latter is a fundamental part of secure MANET operation; it is easily attained by encryption, assuming that a appropriate key management solutions are used to set up or distribute cryptographic keys. What we mean by privacy is resistance to tracking. We believe that this narrow interpretation of privacy is well-justified. We need to find a balance between privacy and security: an ideal solution would be tracking-resistant, immune to insider and outsider attacks (and, hopefully, efficient). Security and privacy with respect to outsiders is relatively easy to obtain with standard cryptographic techniques: encryption and authentication of routing information and subsequent data packets. Privacy with respect to insiders is much harder to obtain because it runs counter to security: malicious behaviour by insiders must be traceable however, traceability can violate privacy.

Rupesh et al, International Journal of Computer Technology & Applications,Vol 7(3),443-447

IJCTA | May-June 2016 Available [email protected]

443

ISSN:2229-6093

2. Related Work

[1] A gradual solution to detect selfish nodes in mobile ad-hoc networks. This paper deals with an emergent security problem related to mobile ad hoc network. This new problem is selfishness on packet forwarding is the resource limitation of nodes in the ad hoc network. To save its energy, a node may behave selfishly, thereby it uses the forwarding service of other nodes, but it does not forward packets for them. We propose a new technique called two hops acknowledgement whose performance is improved gradually. A new kind of feedbacks that called two hops ACK; it is an ACK that travels two hops. Two hops of ACK are however delayed which may delay a little bit the detection of the misbehaviour. Detection of the misbehaviour requires many packets lost detections and may take an important time. Extra memory required to hold data structures at the node. A selfish node regarding the packet forwarding process is a node which takes advantage of the forwarding service and asks others to forward its own packets, but it does not participate in this service.

[2] Mitigating routing misbehaviour in mobile ad-hoc network. This paper describes two techniques that improve through put in an ad-hoc network in the presence of nodes that agree to forward packets but fail to do so. This introduces a watchdog that identifies misbehaving nodes and a path rater that helps routing protocol record these nodes. One solution to misbehaving nodes that share on a priori trust relationship A priori trust relationships are based on pre existing relationships built outside the context of the network. The watchdog identifies misbehaving nodes, while the path rate avoids routing packet through these nodes. Watchdog does this by listening promiscuously to the next node’s transmission. The path rater uses this knowledge of misbehaving nodes to choose the network path that most likely to deliver packets.

[3] MADSN: Mobile agent based detection of selfish node in MANET The approach uses a set of mobile agent (MA) that can move from one node to another node within a network. the intrusion. The mobile agents travel through the network, gathering vital information is then processed by the mobile agent itself. As the computation overhead of our algorithm is less, the computation complexity of the mobile agent will be reduced. The computation is done by mobile agent when the source node notices that the destination node does not respond in correct time.

[4] Intrusion detection system for MANET. This paper introduces an enhancement of the watchdog/path rate form of intrusion detection in mobile ad-hoc network. The participating nodes are allowed to listen the nodes that they have conveyed

messages to, in promiscuous mode. If within a certain time frame the message is not relayed, then the node is suggested to be tagged as a misbehaviour node. Watchdogs run on each node when a node forwards the packet. The watchdog does this by listening in promiscuous mode to the next nodes transmission. If the next node does not forward the packet, then it is considered to be the misbehaving and is reported. This is done by sending an alarm message to the other nodes on its friend’s list.

3. System Overview

A selfish node usually denies packet forwarding in order to save its own resources. This behaviour implies that a selfish node neither participates in routing nor relays data packets. A common technique to detect this selfish behaviour is network monitoring using local watchdogs. node’s watchdog consists on overhearing the packets transmitted and received by its neighbours in order to detect anomalies, such as the ratio between packets received to packets being re-transmitted. By using this technique, the local watchdog can generate a positive (or negative) detection in case the node is acting selfishly (or not).

Figure 1. An Example of collaborative watchdog method

a) Initially all nodes have no information about the selfish node.

b) Node 2 detects the selfish node using its own watchdog.

c) Node 2 contacts with node 3 and it transmits the positive about the selfish node.

d) The local watchdog of Node 4 fails to detect the selfish node and it generates a negative detection (a false negative).

The dashed lines describe how the first hand information is collected. The dotted lines describe how second hand information published by the other nodes is handled. The dashed-dotted line describes that a node periodically publishes the reputation



444

ISSN:2229-6093

ratings it has about other nodes in the network. For reputation we are using Bayesian estimation.

Figure 2. CoCoWa Architecture

The Local Watchdog has two functions: the detection of selfish nodes and the detection of new contacts. The local watchdog can generate the following events about neighbour nodes: PosEvt (positive event) when the watchdog detects a selfish node, NegEvt (negative event) when the watchdog detects that a node is not selfish, and NoDetEvt (no detection event) when the watchdog does not have enough information about a node (for example if the contact time is very low or it does not overhear enough messages). The detection of new contacts is based on neighbourhood packet overhearing; thus, when the watchdog overhears packets from a new node it is assumed to be a new contact, and so it generates an event to the network information module. The Diffusion module has two functions: the transmission as well as the reception of positive (and negative) detections. A key issue of our approach is the diffusion of information. As the number of selfish nodes is low compared to the total number of nodes, positive detections can always be transmitted with a low overhead. However, transmitting only positive detections has a serious drawback: false positives can be spread over the network very fast. Thus, the transmission of negative detections is necessary to neutralise the effect of these false positives, but sending all known negative detections can be troublesome, producing excessive messaging or the fast diffusion of false negatives. Consequently, we introduce a negative diffusion factor γ , that is the ratio of negative detections that are actually transmitted. This value ranges from 0 (no negative detections are transmitted) to 1 (all negative detections are transmitted). We will show in the evaluation section that a low value for the γ factor is enough to neutralize the effect of false positives and false negatives. Finally, when the diffusion module receives a new contact event from the watchdog, it transmits a message including this information to the

new neighbour node. When the neighbour node receives a message, it generates an event to the network information module with the list of these positive (and negative) detections.

Figure 3. Block Diagram

The above figure 3 shows the block diagram for how the CoCoWa works.

4. System Modules

4.1 Network Topology

The sensor nodes are randomly distributed in a sensing field. We are using mobile ad hoc network (MANET). This is the infrastructureless network and a node can move independently. In a MANET, each node not only works as a host and also acts as a router. We can find the communication range for all nodes. Every node communicates only within the range. If suppose any node out of the range, node will not communicate those nodes or drop the packets.

4.2 CoCoWa Model The goal of this subsection is to model the behaviour of the different modules of our architecture (see figure 2). The local watchdog is modelled using three parameters: the probability of detection pd, the ratio of false positives pfp, and the ratio of false negatives



445

ISSN:2229-6093

pfn. The first parameter, the probability of detection (pd), reflects the probability that, when a node contacts another node, the watchdog has enough information to generate a PosEvt or NegEvt event. This value depends on the effectiveness of the watchdog, the traffic load, and the mobility pattern of nodes. For example, for Opportunistic Networks or DTNs where the contacts are sporadic and have low duration, this value is lower than for MANETs. Furthermore, the watchdog can generate false positives and false negatives. A false positive is when the watchdog generates a positive detection for a node that is not a selfish node. A false negative is generated when a selfish node is marked as a negative detection. In order to measure the performance of a watchdog, these values can be expressed as a ratio or probability: pfp is the ratio (or probability) of false positives generated when a node contacts a non-selfish node, and pfn is the ratio (or probability) of false negatives generated when a node contacts a selfish node. Using the previous parameters we can model the probability of generating local PosEvt and NegEvt events when a contact occurs. 4.3 Malicious Nodes and Attacker Model Malicious nodes attempt to attack the CoCoWa system by generating wrong information about the nodes. Thus, the attacker model addresses the behaviour or capabilities of these malicious nodes. A malicious node attack consists of trying to send a positive about a node that is not a selfish node, or a negative about a selfish node, with the goal of producing false positives and false negatives on the rest of nodes. In order to do this, it must have some knowledge about the way CoCoWa works. The effectiveness of this behaviour clearly depends on the rate and precision that malicious nodes can generate wrong information. Malicious nodes are assumed to have communications hardware similar to the rest of nodes, so they can hear all neighbour messages in a similar range than the rest of nodes. Nevertheless, the attacker could use high-gain antennas to increase its communications range and thus disseminate false information in a more effective manner. Thus, a specific security measure is needed, such as the one presented in [5].

4.4 The Model for False Positives

Now develop a model for evaluating the effect of false positives. This model evaluates how fast a false positive spreads in the network (the diffusion time). Thus, in this case, a greater diffusion time stands for a lower impact of false positives.

5. Performance Evaluation

Figure 4. Latency

The figure 4 shows the how much the delay is occurred. The green line shows the proposed scheme. The latency is low as shown in above figure.

Figure 5. Energy Consumption

The figure 5 shows the how much the energy is consumed during the packet delivery.



446

ISSN:2229-6093

Figure 6. Packet Delivery Ratio

The figure 6 shows the packet delivery ratio. The figure shows packet delivery ratio is high.

Figure 7. Routing Overhead

The figure 7 shows the routing overhead. The figure shows that the routing overhead is less while transferring the packets from one node to another node.

6. Conclusion

In this paper, an selfish node detection scheme is proposed that enables a routing protocol in MANETs to detect packet dropping attack by a malicious node. In the proposed mechanism, each node independently monitors the packet forwarding behavior of its neighbors. A cooperative mechanism is utilized among the nodes in the same

neighborhood for detection of selfish or malicious nodes. The mechanism is simulated in network simulator and the results show that the scheme is highly robust, efficient and has improved performance mechanisms. When a contact occurs between two collaborative nodes, the diffusion module transmits and processes the positive (and negative) detections. CoCoWa can reduce the overall detection time with respect to the original detection time when no collaboration scheme is used, with a reduced overhead (message cost). This reduction is very significant, ranging from 20% for very low degree of collaboration to 99% for higher degrees of collaboration.

References

[1 ]Djamel DJENOURI, Nadjib BADACHE,”A Gradual Solution to Detect Selfish Nodes in Mobile Ad hoc Networks,”Proc.14th European Conf.Research in computer networks, pp.355-370, 2009 .

[2] S.Marti, T.J. Giuli, K. Lai, and M. Baker, “Mitigating routing misbehavior in mobile ad hoc networks,” in Proc. 2000 MobiCom, pp.255–265.

[3] Debdutta Barman Roy1 and Rituparna Chaki2,” 225MADSN:Mobile Agent Based Detection of Selfish Node in MANET,”in International Journal of Wireless & Mobile Networks (IJWMN) Vol. 3, No. 4, August 2011.

[4] Charlie Obimbo#1, Liliana Maria Arboleda-Cobo,” An Intrusion Detection System for MANET,” ,” IJCSNS International Journal of Computer 258 Science and Network Security, VOL.11 No.5, May 2011.

[5] S. Abbas, M. Merabti, D. Llewellyn-Jones, and K. Kifayat, “Lightweight sybil attack detection in manets,” IEEE Syst. J., vol. 7, no. 2, pp. 236–248, Jun. 2013.



447

ISSN:2229-6093

rupesh et al, international journal of computer technology ... · using cooperative contact and...

Documents