In this paper, we apply a Vector AutoRegression (VAR) based trust model over the Backpressure Collection Protocol (BCP), a collection mechanism based on dynamic backpressure routing in Wireless Sensor Networks (WSN) and show that the VAR trust model is suited for resource constraint networks. The backpressure scheduling is known for being throughput-optimal. However, it is usually assumed that nodes cooperate with each other to forward the network traffic. In the presence of malicious nodes, the throughput optimality no longer holds and this affects the network performance in collection tree applications of sensor networks. We apply an autoregression based scheme to embed trust into the link weights, making it more likely for trusted links to be scheduled. The novelty in our approach is that the notion of trust can be easily incorporated in a new state of the art distributed and dynamic routing Backpressure Collection Protocol in sensor networks. We have evaluated our work in a real sensor network testbed and shown that by carefully setting the trust parameters, substantial benefit in terms of throughput can be obtained with minimal overheads. Our performance analysis of VAR in comparison with other existing trust models demonstrate that even when 50% of network nodes are malicious, VAR trust offers approximately 73% throughput and ensures reliable routing, with a small trade-off in the end-to-end packet delay and energy consumptions.

No abstract

The focus of this study is to propose a generalised trust-model over routing protocols in mobile ad hoc networks (MANETs). It is observed that the presence of malicious nodes is a critical factor affecting the network performance in an ad hoc network. The novelty in the approach is that the notion of trust can be easily incorporated into any routing protocol in MANETs. The vector auto regression based trust model is introduced to identify malicious nodes that launch multiple attacks in the network. The proposed trust model is incorporated over ad hoc on-demand distance vector (AODV) routing protocol and optimised link state routing (OLSR) protocol in MANETs. The performance evaluations show that by carefully setting the trust parameters, substantial benefit in terms of throughput can be obtained with minimal overheads. The computed trust and confidence values are introduced into the path computation process of the ad hoc routing protocols. It was observed that the nodes in the network were able to learn the malicious activities of their neighbours and hence, alternate trustworthy paths are taken to avoid data loss in the network, with trade-offs in end-to-end packet delay and routing traffic.

Text classification (a.k.a text categorisation) is an effective and efficient technology for information organisation and management. With the explosion of information resources on the Web and corporate intranets continues to increase, it has being become more and more important and has attracted wide attention from many different research fields. In the literature, many feature selection methods and classification algorithms have been proposed. It also has important applications in the real world. However, the dramatic increase in the availability of massive text data from various sources is creating a number of issues and challenges for text classification such as scalability issues. The purpose of this report is to give an overview of existing text classification technologies for building more reliable text classification applications, to propose a research direction for addressing the challenging problems in text mining.