The Internet of Things (IoT) commands an ever-growing population of devices across the nation and abroad. The development of privacy concerns and security goals have not kept pace with the demand for new advances in IoT. We will discuss how the IoT currently functions and why the security in this field is important as the technology grows into every device we touch. This paper will also reference current security implementations and how they expect to cover this growing consumer demand for instant data on many devices at once. With IoT devices using less power and smaller processors, there is major discussion in the computing world on what methods succeed. As standard encryption methods are simply too much for small, low power devices to handle; IoT specific security methods should be highlighted.

A number of determinants predict the adoption of Information Systems (IS) security innovations. Amongst, perceived vulnerability of IS security threats has been examined in a number of past explorations. In this research, we examined the processes pursued in analysing the relationship between perceived vulnerability of IS security threats and the adoption of IS security innovations. The study uses Systematic Literature Review (SLR) method to evaluate the practice involved in examining perceived vulnerability on IS security innovation adoption. The SLR findings revealed the appropriateness of the existing empirical investigations of the relationship between perceived vulnerability of IS security threats on IS security innovation adoption. Furthermore, the SLR results confirmed that individuals who perceives vulnerable to an IS security threat are more likely to engage in the adoption an IS security innovation. In addition, the study validates the past studies on the relationship between perceived vulnerability and IS security innovation adoption.

Network security is an essential element in the day-to-day IT operations of nearly every organization in business. Securing a computer network means considering the threats and vulnerabilities and arrange the countermeasures. Network security threats are increasing rapidly and making wireless network and internet services unreliable and insecure. Intrusion Detection System plays a protective role in shielding a network from potential intrusions. In this research paper, Feed Forward Neural Network and Pattern Recognition Neural Network are designed and tested for the detection of various attacks by using modified KDD Cup99 dataset. In our proposed models, Bayesian Regularization and Scaled Conjugate Gradient, training functions are used to train the Artificial Neural Networks. Various performance measures such as Accuracy, MCC, R-squared, MSE, DR, FAR and AROC are used to evaluate the performance of proposed Neural Network Models. The results have shown that both the models have outperformed each other in different performance measures on different attack detections.

A network in which the vehicular nodes are free to join or leave the network is known as vehicular ad hoc network (VANET). Either vehicle to vehicle or vehicle to infrastructure types of communication is performed in this decentralized type of network. The identification and elimination of Distributed-Denial of Service (DDoS) attacks from VANETs is the major objective of this research. The nodes that can flood victim nodes with large numbers of rough packets are chosen by the malicious nodes in this kind of attack. Identifying such malicious nodes from the network is an important research objective to be achieved. The technique which is proposed in this research is based on the two step verification. In the two steps verification technique, when the network performance is reduced to threshold value then the traffic is monitored that which node is sending data on such high rate. NS2 simulator is used to implement the proposed technique. With respect to various performance parameters, the proposed technique is analyzed. A comparative evaluation of results achieved from proposed and existing techniques is also done to conclude the level of improvement achieved.

This paper aims to provide an intrusion detection system for network traffic that achieves to the low false positive rate with having high attack detection rate. This system will identify anomalies by monitoring network traffic. So, Features extracted from the network traffic by the number of HMM, are modeled as a Classifier ensemble. Then by integrating the outputs of the HMM within a group, probability value is generated. In this system each feature receives a weight and rather than a threshold value, using the fuzzy inference to decide between normal and abnormal network traffic. So at first, the fuzzy rules of decide module are formed manually and based on the value of the security of extraction feature. Then probability output of each HMM groups converted to fuzzy values according to fuzzy rules. These values are applied by a fuzzy inference engine and converted to an output indicating the being normal or abnormal of network traffic. Experiments show that the proposed system in detecting attacks that are the main candidate error is working well. Also, measures recall, precision and F1-measure respectively with 100%, 99.38% and 99.69% will pass. Finally, attack detection rate close to 100% and false positive rate of 0.62%, showing that the proposed system is improved compared to previous systems.

Intrusion Detection is one of the most common approaches used in detecting malicious activities in any network by analyzing its traffic. Machine Learning (ML) algorithms help to study the high dimensional network traffic and identify abnormal flow in traffic with high accuracy. It is crucial to integrate machine learning algorithms with dimensionality reduction to decrease the underlying complexity of processing of huge datasets and detect intrusions within real-time. This paper evaluates 10 most popular ML algorithms on NSL-KDD dataset. Thereafter, the ranking of these algorithms is done to identify best performing ML algorithm on the basis of their performance on several parameters such as specificity, sensitivity, accuracy etc. After analyzing the top 4 algorithms, it becomes evident that they consume a lot of time while model building. Therefore, feature selection is applied to detect intrusions in as little time as possible without compromising accuracy. Experimental results clearly demonstrate that which algorithm works best with/without feature selection/reduction technique in terms of achieving high accuracy while minimizing the time taken in building the model.