Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7148
Browse
2 results
Search Results
Conference Object Citation - Scopus: 5Nfa Based Regular Expression Matching on Fpga(IEEE, 2021) Sert, Kamil; Bazlamaçcı, CüneytString matching is about finding all occurrences of a string within a given text. String matching algorithms have important roles in various real world areas such as web and security applications. In this work, we are interested in solving regular expression matching hence a more general form of string matching problem targeting especially the field of network intrusion detection systems (NIDS). In our work, we enhance a non-deterministic finite automata (NFA) based method on FPGA considerably. We propose to use a matching structure that processes two consecutive characters instead of one in order to yield better memory utilization and provide a novel mapping of this new architecture onto FPGA. The amount of digital circuitry needed to represent the NFA is reduced due to having less number of states and less number of LUTs in the devised 2-character regex matching process. An evaluation study is performed using the well-known Snort rule set and a sizable performance improvement is demonstrated.Article Citation - Scopus: 1Performance Analysis and Feature Selection for Network-Based Intrusion Detection With Deep Learning(Türkiye Klinikleri, 2022) Caner, Serhat; Erdoğmuş, Nesli; Erten, Yusuf MuratAn intrusion detection system is an automated monitoring tool that analyzes network traffic and detects malicious activities by looking out either for known patterns of attacks or for an anomaly. In this study, intrusion detection and classification performances of different deep learning based systems are examined. For this purpose, 24 deep neural networks with four different architectures are trained and evaluated on CICIDS2017 dataset. Furthermore, the best performing model is utilized to inspect raw network traffic features and rank them with respect to their contributions to success rates. By selecting features with respect to their ranks, sets of varying size from 3 to 77 are assessed in terms of classification accuracy and time efficiency. The results show that recurrent neural networks with a certain level of complexity can achieve comparable success rates with state-of-the-art systems using a small feature set of size 9; while the average time required to classify a test sample is halved compared to the complete set.