Search

Malicious node detection in vehicular ad hoc network (VANET) has always been a research hot spot. An efficient misbehavior detection scheme is needed to avoid and reduce the factor of selfishness and maliciousness especially in the case where the selfish beacons exploit the medium. To disseminate honest data over VANET infrastructure, it is very essential for nodes to collaborate with each other during the process of message forwarding and to ensure the successful delivery of honest data over the network. However, using the fake identities in message forwarding process easily results in the dissemination forged data over the network. Therefore, most of the existing techniques for detection of misbehavior in VANET use collaborative-trust–reputation-based approaches to tackle the issue forged and fake data transmission. The objective of this research is to calculate the trust weightages of each vehicle over the network and to reduce the intensity of the malicious vehicular nodes in VANET. The proposed collaboration-based maliciousness detection mechanism comprises data trust module and reputation calculating module, which guarantees honest data communication and reduces the false positive rate of malicious vehicular nodes. The data trust module uses trust evaluation and reputation calculation model to decide whether the vehicle is trustworthy using vehicular behavior vector in the context of packet transmission. The vehicular trust authority uses collaborative approach to integrate several trust evaluation assessments about a particular vehicular node and formulate a complete trust assessment. The performance evaluation shows that the proposed scheme delivers more priority messages with high true positive rate and low false positive rate. Moreover, the experimental results show that Gaussian kernel function best suits our proposed model in comparison with other rationalities. In addition, proposed models are more vigorous in context of true positive rate than the existing schemes, i.e., Dempster–Shafer theory of evidence, majority voted model and Bayesian inference.

A major drawback of signature-based intrusion detection systems is the inability to detect novel attacks that do not match the known signatures already stored in the database. Anomaly detection is a kind of intrusion detection in which the activities of a system are monitored and these activities are classified as normal or anomalous based on their expected behavior. Tree-based classifiers have been successfully used to separate the abnormal behavior from the normal one. Tree pruning is a machine learning technique used to minimize the size of a decision tree (DT) in order to reduce the complexity of the classifier and improve its predictive accuracy. In this paper, we attempt to prune a DT using particle swarm optimization (PSO) algorithm and apply it to the network intrusion detection problem. The proposed technique is a hybrid approach in which PSO is used for node pruning and the pruned DT is used for classification of the network intrusions. Both single and multi-objective PSO algorithms are used in the proposed approach. The experiments are carried out on the well-known KDD99Cup dataset. This dataset has been widely used as a benchmark dataset for network intrusion detection problems. The results of the proposed technique are compared to the other state-of-the-art classifiers and it is observed that the proposed technique performs better than the other classifiers in terms of intrusion detection rate, false positive rate, accuracy, and precision.

During the past few years, Internet has become a public platform for communication and exchange of information online. The increase in network usage has increased the chance of network attacks. In order to detect the malicious activities and threats, several kinds of Intrusion Detection Systems (IDSs) have been designed over the past few years. The goal of IDS is to intelligently monitor events occurring in a computer system or a network and analyze them for any sign of violation of the security policy as well as retain the availability, integrity, and confidentiality of a network information system. An IDS may be categorized as anomaly detection system or misuse detection system. Anomaly detection systems usually apply statistical or Artificial Intelligence (AI) techniques to detect attacks; therefore, these systems have the ability to detect novel or unknown attacks. A misuse detection system uses signature-based detection; therefore, these systems are good at identifying already known attacks but cannot detect unknown attacks.

Network security risks grow with increase in the network size. In recent past, the attacks on computer networks have increased tremendously and require efficient network intrusion detection mechanisms. Data mining and machine-learning techniques have been used for network intrusion detection during the past few years and have gained much popularity. In this paper, we propose an intrusion detection mechanism based on binary particle swarm optimization PSO and random forests RF algorithms called PSO-RF and investigate the performance of various dimension reduction techniques along with a set of different classifiers including the proposed approach. Binary PSO is used to find more appropriate set of attributes for classifying network intrusions, and RF is used as a classifier. In the preprocessing step, we reduce the dimensions of the dataset by using different state-of-the-art dimension reduction techniques, and then this reduced dataset is presented to the proposed PSO-RF approach that further optimizes the dimensions of the data and finds an optimal set of features. PSO is an optimization method that has a strong global search capability and is used here for dimension optimization. We perform extensive experimentation to prove the worth of the proposed approach by using different performance metrics. The standard benchmark, that is, KDD99Cup dataset, is used that contains the information about various kinds of network intrusions. The experimental results indicate that the proposed approach performs better than the other approaches for the detection of all kinds of attacks present in the dataset. Copyright © 2012 John Wiley & Sons, Ltd.

Threats to computer networks are numerous and potentially devastating. Intrusion detection techniques provide protection to our data and track unauthorized access. Many algorithms and techniques have been proposed to improve the accuracy and minimize the false positive rate of the intrusion detection system (IDS). Statistical techniques, evolutionary techniques, and data mining techniques have also been used for this purpose. In this paper, we use a centroid-based technique for network intrusion detection in which the centroid is constructed on the basis of diversity. Diversity of a point is the sum of the distances from a point to all other points in a cluster. The point having minimum diversity is chosen as a centroid. The performance of diversity-based centroid shows significant improvement in the classification of intrusions. Experimental results on the KDDCup99 dataset demonstrate that the proposed method shows excellent performance in terms of accuracy, detection rate, and false positive rate.

A system connected to a network is an open choice for network intrusions unless a powerful intrusion detection or prevention system is implemented. Network security has become a serious issue due to increased unauthorized access and manipulation of network resources. Evolutionary approaches play an important role in identifying attacks with high detection rates and low false discovery rates. In this paper, a binary version of multi-objective particle swarm optimization (PSO) approach is used to detect PROBE attacks in a network. A vector evaluated PSO approach is used in the proposed technique with two objectives i.e., intrusion detection rate and false discovery rate, to guide the process of feature selection. The experiments are performed using the well-known KDD99Cup dataset. Multi-objective PSO approach is used for feature selection from a set of 41 features and Random Forests (RF), a highly accurate and fast algorithm, is used for classification. Empirical results show that the proposed technique outperforms well-known classification and regression techniques in most of the cases.

During the past few years, huge amount of network attacks have increased the requirement of efficient network intrusion detection techniques. Different classification techniques for identifying various attacks have been proposed in the literature. In this paper we propose and implement a hybrid classifier based on binary particle swarm optimization (BPSO) and random forests (RF) algorithm for the classification of PROBE attacks in a network. PSO is an optimization method which has a strong global search capability and is used for fine-tuning of the features whereas RF, a highly accurate classifier, is used here for classification. We demonstrate the performance of our technique using KDD99Cup dataset. We also compare the performance of our proposed classifier with eight other well-known classifiers and the results show that the performance achieved by the proposed classifier is much better than the other approaches.