Biblio

This paper presents an anomalous IP address detection algorithm for network traffic logs. It is based on word embedding techniques derived from natural language processing to extract the representative features of IP addresses. However, the features extracted from vanilla word embeddings are not always compatible with machine learning-based anomaly detection algorithms. Therefore, we developed an algorithm that enables the extraction of more compatible features of IP addresses for anomaly detection than conventional methods. The proposed algorithm optimizes the objective functions of word embedding-based feature extraction and anomaly detection, simultaneously. According to the experimental results, the proposed algorithm outperformed conventional approaches; it improved the detection performance from 0.876 to 0.990 in the area under the curve criterion in a task of detecting the IP addresses of attackers from network traffic logs.

With the growing number of streaming services, internet providers are increasingly needing to be able to identify the types of data and content providers that are being used on their networks. Traditional methods, such as IP and port scanning, are not always available for clients using VPNs or with providers using varying IP addresses. As such, in this paper we explore a potential method using neural networks and Markov Decision Process in order to augment deep packet inspection techniques in identifying the source and class of video streaming services.

Identifying services constituting traffic from given IP network flows is essential to various applications, such as the management of quality of service (QoS) and the prevention of security issues. Typical methods for achieving this objective include identifications based on IP addresses and port numbers. However, such methods are not sufficiently accurate and require improvement. Deep Packet Inspection (DPI) is one of the most promising methods for improving the accuracy of identification. In addition, many current IP flows are encrypted using Transport Layer Security (TLS). Hence, it is necessary for identification methods to analyze flows encrypted by TLS. For that reason, a service identification method based on DPI and n-gram that focuses only on the non-encrypted parts in the TLS session establishment was proposed. However, there is room for improvement in identification accuracy because this method analyzes all the non-encrypted parts including Random Values without protocol analyses. In this paper, we propose a method for identifying the service from given IP flows based on analysis of Server Name Indication (SNI). The proposed method clusters flow according to the value of SNI and identify services from the occurrences of all clusters. Our evaluations, which involve identifications of services on Google and Yahoo sites, demonstrate that the proposed method can identify services more accurately than the existing method.

Named Data Networking (NDN) is a new network architecture that has been projected as the future of internet architecture. Unlike the traditional internet approach which currently relies on client-server communication models to communicate each other, NDN relies on data as an entity. Hence the users only need the content and applications based on data naming, as there is no IP addresses needed. NDN is different than TCP/IP technology as NDN signs the data with Digital Signature to secure each data authenticity. Regarding huge number of uses on IP-based network, and the minimum number of NDN-based network implementation, the NDN-IP gateway are needed to map and forward the data from IP-based network to NDN-based network, and vice versa. These gateways are called Custom-Router Gateway in this study. The Custom-Router Gateway requires a new mechanism in conducting Digital Signature so that authenticity the data can be verified when it passes through the NDN-IP Custom-Router Gateway. This study propose a method to process the Digital Signature for the packet flows from IP-based network through NDN-based network. Future studies are needed to determine the impact of Digital Signature processing on the performance in forwarding the data from IP-based to NDN-based network and vice versa.

Lack of effective accountability mechanisms brings a series of security problems for Internet today. In Next Generation Internet based on IPv6, the system of identity authentication and IP verification is the key to accounting ability. Source Address Validation Improvement (SAVI) can protect IP source addresses from being faked. But without identity authentication mechanism and certain relationship between IP and accountable identity, the accountability is still unreliable. To solve this problem, most research focus on embedding accountable identity into IP address which need either changing DHCP client or twice DHCP request process due to the separate process of user authentication and address assignment. Different from previous research, this paper first analyzes the problems and requirements of combining Web Portal or 802.1X, two main identity authentication mechanism (AAA), with the accountable address assignment in SAVI frame-work. Then a novel Cooperative mechanism for Accountable IP address assignment (CAIP) is proposed based on 802.1X and SAVI, which takes into account the validation of IP address, the authenticity and accountability of identity at the same time. Finally, we build up prototype system for both Fat AP and Thin AP wireless scenarios and simulate the performance of CAIP through large-scale campus networks' data logs. The experiment result shows that the IP addresses and identities in CAIP are protective and accountable. Compared with other previous research, CAIP is not only transparent to the terminals and networks, but also low impact on network equipment, which makes CAIP easy deployment with high compatibility and low cost.

To gain strategic insight into defending against the network reconnaissance stage of advanced persistent threats, we recreate the escalating competition between scans and deceptive views on a Software Defined Network (SDN). Our threat model presumes the defense is a deceptive network view unique for each node on the network. It can be configured in terms of the number of honeypots and subnets, as well as how real nodes are distributed across the subnets. It assumes attacks are NMAP ping scans that can be configured in terms of how many IP addresses are scanned and how they are visited. Higher performing defenses detect the scanner quicker while leaking as little information as possible while higher performing attacks are better at evading detection and discovering real nodes. By using Artificial Intelligence in the form of a competitive coevolutionary genetic algorithm, we can analyze the configurations of high performing static defenses and attacks versus their evolving adversary as well as the optimized configuration of the adversary itself. When attacks and defenses both evolve, we can observe that the extent of evolution influences the best configurations.

Peer-to-Peer botnets have become one of the significant threat against network security due to their distributed properties. The decentralized nature makes their detection challenging. It is important to take measures to detect bots as soon as possible to minimize their harm. In this paper, we propose PeerGrep, a novel system capable of identifying P2P bots. PeerGrep starts from identifying hosts that are likely engaged in P2P communications, and then distinguishes P2P bots from P2P hosts by analyzing their active ratio, packet size and the periodicity of connection to destination IP addresses. The evaluation shows that PeerGrep can identify all P2P bots with quite low FPR even if the malicious P2P application and benign P2P application coexist within the same host or there is only one bot in the monitored network.

Hidden Markov Models (HMM) are probabilistic models that can be used for forecasting time series data. It has seen success in various domains like finance [1-5], bioinformatics [6-8], healthcare [9-11], agriculture [12-14], artificial intelligence[15-17]. However, the use of HMM in cyber security found to date is numbered. We believe the properties of HMM being predictive, probabilistic, and its ability to model different naturally occurring states form a good basis to model cyber security data. It is hence the motivation of this work to provide the initial results of our attempts to predict security attacks using HMM. A large network datasets representing cyber security attacks have been used in this work to establish an expert system. The characteristics of attacker's IP addresses can be extracted from our integrated datasets to generate statistical data. The cyber security expert provides the weight of each attribute and forms a scoring system by annotating the log history. We applied HMM to distinguish between a cyber security attack, unsure and no attack by first breaking the data into 3 cluster using Fuzzy K mean (FKM), then manually label a small data (Analyst Intuition) and finally use HMM state-based approach. By doing so, our results are very encouraging as compare to finding anomaly in a cyber security log, which generally results in creating huge amount of false detection.

This paper focuses on one type of Covert Storage Channel (CSC) that uses the 6-bit TCP flag header in TCP/IP network packets to transmit secret messages between accomplices. We use relative entropy to characterize the irregularity of network flows in comparison to normal traffic. A normal profile is created by the frequency distribution of TCP flags in regular traffic packets. In detection, the TCP flag frequency distribution of network traffic is computed for each unique IP pair. In order to evaluate the accuracy and efficiency of the proposed method, this study uses real regular traffic data sets as well as CSC messages using coding schemes under assumptions of both clear text, composed by a list of keywords common in Unix systems, and encrypted text. Moreover, smart accomplices may use only those TCP flags that are ever appearing in normal traffic. Then, in detection, the relative entropy can reveal the dissimilarity of a different frequency distribution from this normal profile. We have also used different data processing methods in detection: one method summarizes all the packets for a pair of IP addresses into one flow and the other uses a sliding moving window over such a flow to generate multiple frames of packets. The experimentation results, displayed by Receiver Operating Characteristic (ROC) curves, have shown that the method is promising to differentiate normal and CSC traffic packet streams. Furthermore the delay of raising an alert is analyzed for CSC messages to show its efficiency.

During an advanced persistent threat (APT), an attacker group usually establish more than one C&C server and these C&C servers will change their domain names and corresponding IP addresses over time to be unseen by anti-virus software or intrusion prevention systems. For this reason, discovering and catching C&C sites becomes a big challenge in information security. Based on our observations and deductions, a malware tends to contain a fixed user agent string, and the connection behaviors generated by a malware is different from that by a benign service or a normal user. This paper proposed a new method comprising filtering and clustering methods to detect C&C servers with a relatively higher coverage rate. The experiments revealed that the proposed method can successfully detect C&C Servers, and the can provide an important clue for detecting APT.

IP spoofing based DDoS attack that relies on multiple compromised hosts in the network to attack the victim. In IP spoofing, IP addresses can be forged easily, thus, makes it difficult to filter illegitimate packets from legitimate one out of aggregated traffic. A number of mitigation techniques have been proposed in the literature by various researchers. The conventional Hop Count Filtering or probabilistic Hop Count Filtering based research work indicates the problems related to higher computational time and low detection rate of illegitimate packets. In this paper, DPHCF-RTT technique has been implemented and analysed for variable number of hops. Goal is to improve the limitations of Conventional HCF or Probabilistic HCF techniques by maximizing the detection rate of illegitimate packets and reducing the computation time. It is based on distributed probabilistic HCF using RTT. It has been used in an intermediate system. It has the advantage for resolving the problems of network bandwidth jam and host resources exhaustion. MATLAB 7 has been used for simulations. Mitigation of DDoS attacks have been done through DPHCF-RTT technique. It has been shown a maximum detection rate up to 99% of malicious packets.