A Misbehavior Detection Scheme Establishment in Delay Tolerant Networks

Delay Tolerant Network(DTNs) are a class of unique network characterized like lack of guaranteed connectivity ,typically low frequency between DTN nodes and long propagation delay within the networks. Existing routing algorithms for DTN assumes that nodes are willing to forward packets for others but in real word selfish and malicious behaviors occurs while forward packets for nodes. Due to unique characteristics the message propagation process DTNs follows a Store-Carry and Forward manners. In this paper, we propose iTrust, probabilistic misbehavior detection schemes for secure and to improve the efficiency of DTN routing towards efficient trust establishment. The basic idea of iTrust is introducing Trusted Authority (TA) to judge the nodes behavior based on the collected routing evidences and probabilistically checking. To further improve the performance of the proposed probabilistic inspection schemes, we introduce a reputation system. The extensive analysis and simulations result shows that the proposed schemes substantiate the effectiveness and efficiency of the proposed schemes. KeywordsDTN, Selfish nodes, iTrust, Trusted Authority, Store –Carry and Forward B.Sivakumar et al, International Journal of Computer Science and Mobile Computing, Vol.3 Issue.5, May2014, pg. 76-82 © 2014, IJCSMC All Rights Reserved 77 INTRODUCTION Delay Tolerant Networks (DTNs) have the unique feature of intermittent connectivity, which makes routing quite different from other wireless networks. Since an end-to-end connection is hard to setup, store-carry-and-forward is used to deliver the packets to the destination. In the real world, most people are selfish; we have two observations from the social perspective. First, a selfish user is usually willing to help others with whom he has social ties (e.g., friends, coworkers, roommates) will be referred to as social selfishness. Second, for those with social ties, a selfish user may give different preferences will be referred to as individual selfishness. While forwarding packets [4,6] if connectivity is interrupted, then routing protocols would provide an alternative path after at most a transient outage. This is also assumed for emerging wireless Mobile Ad-hoc NETworks (MANETs). For wireless networks with intermittent connectivity, also called Delay or Disruption Tolerant Networks (DTNs), lack of continuous connectivity, network partitioning and very long delays are actually the norm, not the exception. For example, the in-transit messages in DTNs, also called bundles, as shown in fig.1,could only be forwarded when two DTN nodes (N1, N2) move within each other’s transmission range and contact with each other during a period of time. If no other DTN node is within the transmission range of DTN node N1, N1will buffer the c u r r e n t bundles and carry them until other DTN node appears within its transmission range. Therefore, the bundle propagation process in DTNs follows a “store-carry-andforward" manner and the bundles are opportunistically routed toward the destinations by intermittent connections. Fig. 1. Bundle store-carry-and-forward in DTNs. A misbehavior detection and mitigation protocol is highly desirable to assure the secure DTN routing as well as the establishment of the trust among DTN nodes in DTNs. Mitigating routing misbehavior in traditional mobile ad hoc networks use neighborhood monitoring or destination acknowledgement to detect packet dropping , and exploit credit-based and reputationbased incentive schemes. The existing misbehavior detection schemes work well for the traditional wireless networks, the unique network characteristics including lack of contemporaneous path, high variation in network conditions, difficult to predict mobility patterns, and long feedback delay. B.Sivakumar et al, International Journal of Computer Science and Mobile Computing, Vol.3 Issue.5, May2014, pg. 76-82 © 2014, IJCSMC All Rights Reserved 78 Recently, there are quite a few proposals for misbehaviors detection in DTNs , most of which are based on forwarding history verification (e.g., multi-layered credit , three-hop feedback mechanism , or encounter ticket ), which are costly in terms of transmission overhead and verification cost. The basic idea of iTrust is introducing a periodically available Trusted Authority (TA) to judge the node’s behavior based on the collected routing evidences and probabilistically checking. In this paper, we propose iTrust, a Probabilistic Misbehavior Detection Scheme for DTN, to adaptively detect misbehaviors in DTN and achieve the tradeoff between the detection cost and the detection performance. The proposed iTrust schemes are inspired from the Inspection Game, which is a game theory model in which an inspector verifies if another party, called inspectee, adheres to certain legal rules. Furthermore, the inspector could check the inspectee with a higher probability than the Nash Equilibrium points to prevent the offences, as the inspectee must choose to comply the rules due to its rationality. To further improve the performance of the proposed probabilistic inspection scheme, we introduce a reputation system, in which the inspection probability could vary along with the target node’s reputation. Under the reputation system, a node with a good reputation will be checked with a lower probability while a bad reputation node could be checked with a higher probability. The contributions of this paper can be summarized as follows. • Firstly, we propose a general misbehavior detection framework based on a series of newly introduced data forwarding evidences. The proposed evidence framework could not only detect various misbehaviors but also be compatible to various routing protocols. • Secondly, we introduce a probabilistic misbehavior detection scheme by adopting the Inspection Game. A detailed game theoretical analysis will demonstrate that the cost of misbehavior detection could be significantly reduced without compromising the detection performance. We also discuss how to correlate a user’s reputation (or trust level) to the detection probability, which is expected to further reduce the detection probability. • Thirdly, we use extensive simulations as well as detailed analysis to demonstrate the effectiveness and the efficiency of the iTrust. PERLIMINARY In this section, we formulate the system model, routing model, threat model and design requirement. A. System Model We consider a normal DTN consisted of mobile devices owned by individual users. Each node B.Sivakumar et al, International Journal of Computer Science and Mobile Computing, Vol.3 Issue.5, May2014, pg. 76-82 © 2014, IJCSMC All Rights Reserved 79 i is assumed to have a unique ID Ni and a corresponding public/private key pair. We assume that each node must pay a deposit C before it joins the network, and the deposit will be paid back after the node leaves if there is no offend activity of the node. Similar to [10], we assume that a periodically available TA exists so that it could take the responsibility of misbehavior detection in DTN. For a specific detection target Ni, TA will request Ni’s forwarding history in the global network. Therefore, each node will submit its collected Ni’s forwarding history to TA via two possible approaches. In a pure peer-to-peer DTN, the forwarding history could be sent to some special network components (e.g., roadside unit (RSU) in vehicular DTNs or judge nodes in via DTN transmission. In some hybrid DTN network environment, the transmission between TA and each node could be also performed in a direct transmission manner (e.g., WIMAX or cellular networks). We argue that since the misbehavior detection is performed periodically, the message transmission could be performed in a batch model, which could further reduce the transmission overhead. B. Routing Model We adopt the single-copy routing mechanism such as First Contact routing protocol, and we assume the communication range of a mobile node is finite. Thus a data sender out of destination node’s communication range can only transmit packetized data via a sequence of intermediate nodes in a multihop manner. Our misbehaving detection scheme can be directly used but not limited in metric-based routing algorithms, such as MaxProp and ProPHET.