PhD Forum: Trust and Reputation Aware Geographic Routing Scheme for Wireless Ad Hoc and Sensor Networks P. Raghu Vamsi * and Krishna Kant + Department of Computer Science and Engineering, Jaypee Institute of Information Technology, Noida, India. * prvonline@yahoo.co.in, + k.kant@jiit.ac.in Abstract. This paper presents a Trust And Reputation Aware (TARA) scheme for secure geographic routing in Wireless Ad Hoc and Sensor Networks. This work is inspired from well defined trust and reputation models in the field of social sciences. TARA scheme automatically adjusts the weights associated with the trust metrics and computes the trust value of a node using direct and indirect observations. The adaptive weights and validated trust opinions give the ability for TARA scheme to dynamically identify malicious nodes. The si- mulation results demonstrate the effectiveness of TARA scheme. 1 Introduction Wireless Ad Hoc and Sensor Networks (WASNs) are a special class of wireless net- works which have received a great attention by the researchers in recent years. These networks are composed of autonomous sensing and communication devices (called nodes) with limited resources in terms of energy, processing and memory. Nodes in the network have the capability of establishing ad hoc communication; thereby they are frequently used to setup communication in remote and hostile environments. Dep- loying WASNs in such environments presents vulnerabilities and physical tampering due to unattended deployment and use of insecure wireless links. These vulnerabili- ties can lead to security attacks. These attacks can thwart the network operations such as routing, data aggregation etc. In recent years, trust monitoring has been evolved as a promising method to guard the routing process against such attacks. There exists a number of trust models proposed in the literature for reactive and proactive routing protocols. Trust models for geographic routing protocols are limited. Existing trust models for geographic routing assigns heuristic weights to trust metrics [1]. Such assignment is not suitable for dynamic networks and can imbalance the relative im- portance of other trust metrics. To this end, the objective of this paper is to address the following issues. How to design a self adaptable trust and reputation model using direct and indirect observations to avoid heuristic weight assignments? How to communicate and adopt indirect trust opinions without incurring overhead? How to enhance the dependability and efficiency of a trust and reputation model by combining it with routing protocols?