Trust Evaluation in Wireless Ad hoc Networks using Fuzzy System Pallavi Khatri Department of CSE ITM Gwalior, INDIA +9179801028 pallavi.khatri.csit@itmuniverse.in S.Tapaswi Department of IT ABV-IIITM, Gwalior, INDIA +9165763468 stapaswi@iiitm.ac.in U.P.Verma School Of Studies, Physics JU, Gwalior, INDIA 0751-2442782 upv.udai@gmail.com ABSTRACT Machines can be made intelligent using fuzzy systems. Proposed work presents a scheme for developing Trust among nodes of an Ad hoc Network using Fuzzy system. Ad hoc Networks are infrastructure less, self organizing networks. Nodes are totally dependent on themselves for taking any decision as there is no Central Authority (CA) in Ad hoc networks. Trust evaluation using Fuzzy system helps to make routing decision for secure data transmission. In proposed approach the Fuzzy system is incorporated using MATLAB and it is found that the total network performance is enhanced using this approach. Categories and Subject Descriptors D.2.2 [Software Engineering]: Design Tools and Techniques, Flowchart. I.2.3 [Computing Methodologies]: Artificial Intelligence, Deduction and Theorem Proving, Fuzzy Reasoning. General Terms Algorithms, Security, Design, Experimentation, Performance. Keywords Trust, Fuzzy system, Ad hoc network. 1. INTRODUCTION Ad hoc networks are infrastructure less networks with each node acting as a CPU and router for other nodes. Nodes in such networks rely on other nodes for routing the information from one end to the other. As Ad hoc network are wireless and all transmissions are done on the fly, they become more susceptible to various kinds of attacks. Communication in the network depends on the nodes and at any point of time the risk of a node being compromised exists [5]. The operation of Ad hoc network depends on the nodes and cooperation among them for providing connectivity and communication routes. Some nodes in network may behave maliciously and result in degrading the performance of the network. For attack free transmission some trust among the nodes can be established so that the communication can be done through a node having highest trust level. To reduce the effect of non trusted nodes and achieve a higher level of security this work proposes a fuzzy system which calculates the trust level of a node and helps to maintain a secure network by choosing the best and most reliable routes between the nodes of an Ad hoc network. Paper is organized in the following subsections. Section 2 briefs about the motivation for choosing the fuzzy system for evaluating trust, related work is being discussed in section 3. Proposed scheme is explained in section 4. Section5 briefs about the fuzzy system which has been used for the study. Section 6 analyses the simulation environment and results followed by conclusion in section 7 2. MOTIVATION The basic routing protocols such as DSDV (Destination- Sequenced Distance Vector Routing), TORA (Temporally- Ordered Routing algorithm), DSR (Dynamic Source Routing Protocol), AODV [2] (Ad hoc On Demand Distance Vector) etc. are not capable of fighting the routing attacks. They need to be modified with a security module embedded to deal with routing attacks, maintaining the efficiency of the protocol should not be lost. Most of the proposed solutions for secure protocol are PKI (Public Key Infrastructure) based. The problem of such approaches is that the protocols rely on a central authority or some distributed trusted third party for enforcing cryptographic mechanisms [11] .The requirement of trust authorities goes against the basic nature of ad hoc networks. The network even fails to detect the misbehavior and internal attacks done by a compromised CA. In addition the computation and network overhead of PKI mechanisms are not suitable for MANETs in terms of throughput, availability, battery consumption, and robustness [10]. PKI based secure routing protocols have some constraints, key generation and distribution which are essentials of PKI security are difficult to be implemented within MANET which lacked trust authorities. Therefore the notion of relative security protection may be more suitable for MANET. Here we propose a Fuzzy based model which does not rely completely on PKI and evaluate trust level of a node to designate it as good or bad node. A node is permitted to remain in the network based on the negative or positive trust of the node. This trust level is computed on the basis of various parameters that can affect the total performance of the network. 3. RELATED WORK The proposed trust evaluation model is adapted from [3] which is based on AODV protocol [2]. Hallani and Shahrestani in [3] use a fuzzy based approach to evaluate the trust of a node and Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. CUBE 2012, September 3 5, 2012, Pune, Maharashtra, India. Copyright 2012 ACM 978-1-4503-1185- 779