Underwater Wireless Sensor Networks (UWSNs) consist of certain number of sensors and vehicles interacting with each other to collect data. In recent years, the use of Autonomous Underwater Vehicle (AUV) has improved the data delivery ratio and maximized the energy efficiency in UWSNs. Clustering is one of the effective techniques in energy management which increases the lifetime of these networks. One of the most important parameters in creating optimized clusters is the choice of appropriate cluster head (CH), which not only increases the lifetime of the network and the received data in the sink, but also reduces energy consumption. Clustering of networks was primary done via distributed methods in previous researches. It spends too much energy and also involves too many nodes in the clustering process and fades their main functionality, which is gathering data in sensor networks. It also causes more damping of the network. However, in the proposed protocol, instead of having them distributed by the network and the nodes, the stages of clustering and selecting the appropriate CH is the task of the AUV (Autonomous Underwater Vehicle). Since all the necessary measures to cluster in the network will be carried out by the AUV by this method, many control overheads in the process of clustering the network will be removed and energy consumption caused by nodes reduces significantly. With this method, the network scalability will also be manageable and under control. For simulating and implementing our method we mainly used the OPNET software. The results show that energy consumption of nodes in the proposed algorithm has been significantly improved compared to previous results.

As intrusion detection techniques based on malicious traffic signature are unable to detect unknown attacks, the methods derived from characterizing the behavior of the normal traffic are appropriate in case of detecting unseen intrusions. Based on such a technique, one class Support Vector Machine (SVM) is employed in this research to learn http regular traffic characteristics for anomaly detection. First, suitable features are extracted from the normal and abnormal http traffic; then the system is trained by the normal traffic samples. To detect anomaly, the actual traffic (including normal and abnormal packets) is compared to the deduced normal traffic. An anomaly alert is generated if any deviation from the regular traffic model is inferred. Examining the performance of the proposed algorithm using ISCX data set has delivered high accuracy of 89.25% and low false positive of 8.60% in detecting attacks on port 80. In this research, online step speed has reached to 77 times faster than CPU using GPU for feature extraction and OpenMp for parallel processing of packets.