Visible to the public Biblio

Found 12662 results

Filters: Keyword is pubcrawl  [Clear All Filters]
2022-06-15
Nair, P. Rajitha, Dorai, D. Ramya.  2021.  Evaluation of Performance and Security of Proof of Work and Proof of Stake using Blockchain. 2021 Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV). :279–283.
Storing information in Blockchain has become in vogue in the Technical and Communication Industry with many major players jumping into the bandwagon. Two of the most prominent enablers for Blockchain are “Proof of Work” and “Proof of Stake”. Proof of work includes the members solving the complex problem without having a particular need for the solution (except as evidence, of course), which absorbs a large number of resources in turn. The proof of stake doesn’t require as many resources to enable Blockchain secure information store. Both methodologies have their advantages and their shortcomings. The article attempts to review the current literature and collate the results of the study to measure the performance of both the methodologies and to arrive at a consensus regarding either or both methodologies to implement Blockchain to store data. Post reviewing the performance aspects and security features of both Proofs of Stake and Proof of Work the reviewer attempts to arrive at a secure and better performing blended Blockchain methodology that has wide industry practical application.
Bartoletti, Massimo, Lande, Stefano, Zunino, Roberto.  2021.  Computationally sound Bitcoin tokens. 2021 IEEE 34th Computer Security Foundations Symposium (CSF). :1–15.
We propose a secure and efficient implementation of fungible tokens on Bitcoin. Our technique is based on a small extension of the Bitcoin script language, which allows the spending conditions in a transaction to depend on the neighbour transactions. We show that our implementation is computationally sound: that is, adversaries can make tokens diverge from their ideal functionality only with negligible probability.
Kurt, Ahmet, Mercana, Suat, Erdin, Enes, Akkaya, Kemal.  2021.  Enabling Micro-payments on IoT Devices using Bitcoin Lightning Network. 2021 IEEE International Conference on Blockchain and Cryptocurrency (ICBC). :1–3.
Lightning Network (LN) addresses the scalability problem of Bitcoin by leveraging off-chain transactions. Nevertheless, it is not possible to run LN on resource-constrained IoT devices due to its storage, memory, and processing requirements. Therefore, in this paper, we propose an efficient and secure protocol that enables an IoT device to use LN's functions through a gateway LN node. The idea is to involve the IoT device in LN operations with its digital signature by replacing original 2-of-2 multisignature channels with 3-of-3 multisignature channels. Our protocol enforces the LN gateway to request the IoT device's cryptographic signature for all operations on the channel. We evaluated the proposed protocol by implementing it on a Raspberry Pi for a toll payment scenario and demonstrated its feasibility and security.
Xie, Shuang, Hong, Yujie, Wang, Xiangdie, Shen, Jie.  2021.  Research on Data Security Technology Based on Blockchain Technology. 2021 7th IEEE Intl Conference on Big Data Security on Cloud (BigDataSecurity), IEEE Intl Conference on High Performance and Smart Computing, (HPSC) and IEEE Intl Conference on Intelligent Data and Security (IDS). :26–31.
Blockchain started with Bitcoin, but it is higher than Bitcoin. With the deepening of applied research on blockchain technology, this new technology has brought new vitality to many industries. People admire the decentralized nature of the blockchain and hope to solve the problems caused by the operation of traditional centralized institutions in a more fair and effective way. Of course, as an emerging technology, blockchain has many areas for improvement. This article explains the blockchain technology from many aspects. Starting from the typical architecture of the blockchain, the data structure and system model of the blockchain are first introduced. Then it expounds the development of consensus algorithms and compares typical consensus algorithms. Later, the focus will be on smart contracts and their application platforms. After analyzing some of the challenges currently faced by the blockchain technology, some scenarios where the blockchain is currently developing well are listed. Finally, it summarizes and looks forward to the blockchain technology.
Fan, Wenjun, Chang, Sang-Yoon, Zhou, Xiaobo, Xu, Shouhuai.  2021.  ConMan: A Connection Manipulation-based Attack Against Bitcoin Networking. 2021 IEEE Conference on Communications and Network Security (CNS). :101–109.
Bitcoin is a representative cryptocurrency system using a permissionless peer-to-peer (P2P) network as its communication infrastructure. A number of attacks against Bitcoin have been discovered over the past years, including the Eclipse and EREBUS Attacks. In this paper, we present a new attack against Bitcoin’s P2P networking, dubbed ConMan because it leverages connection manipulation. ConMan achieves the same effect as the Eclipse and EREBUS Attacks in isolating a target (i.e., victim) node from the rest of the Bitcoin network. However, ConMan is different from these attacks because it is an active and deterministic attack, and is more effective and efficient. We validate ConMan through proof-of-concept exploitation in an environment that is coupled with real-world Bitcoin node functions. Experimental results show that ConMan only needs a few minutes to fully control the peer connections of a target node, which is in sharp contrast to the tens of days that are needed by the Eclipse and EREBUS Attacks. Further, we propose several countermeasures against ConMan. Some of them would be effective but incompatible with the design principles of Bitcoin, while the anomaly detection approach is positively achievable. We disclosed ConMan to the Bitcoin Core team and received their feedback, which confirms ConMan and the proposed countermeasures.
Chicaiza, Silvana Abigail Yacchirema, Chafla, Ciro Napoleon Saguay, Álvarez, Luis Fernando Enriquez, Matute, Polo Fabian Iñiguez, Rodriguez, Ramiro Delgado.  2021.  Analysis of information security in the PoW (Proof of Work) and PoS (Proof of Stake)blockchain protocols as an alternative for handling confidential nformation in the public finance ecuadorian sector. 2021 16th Iberian Conference on Information Systems and Technologies (CISTI). :1–5.
Blockchain technology relies on a growing number of globally distributed ledgers known as blockchain. This technology was used for the creation of the cryptocurrency known as bitcoin that allows transactions to be carried out quickly and easily, without the need to use an intermediary "financial institution". The information is sent trough the protocols known as: PoW (Proof of Work) and PoS (Proof of Stake), which must guarantee confidentiality, integrity and availability of the information. The present work shows the result of a bibliographic review on the evolution of the blockchain, the PoW and PoS protocols; as well as the application of these within the framework of Ecuadorian legislation with emphasis on the evolution of risks of the PoW protocol.
Pan, Pengyu, Ma, Xiaobo, Bian, Huafeng.  2021.  Exploiting Bitcoin Mining Pool for Stealthy and Flexible Botnet Channels. 2021 8th International Conference on Dependable Systems and Their Applications (DSA). :741–742.
Botnets are used by hackers to conduct cyber attacks and pose a huge threat to Internet users. The key of botnets is the command and control (C&C) channels. Security researchers can keep track of a botnet by capturing and analyzing the communication traffic between C&C servers and bots. Hence, the botmaster is constantly seeking more covert C&C channels to stealthily control the botnet. This paper designs a new botnet dubbed mp-botnet wherein bots communicate with each other based on the Stratum mining pool protocol. The mp-botnet botnet completes information transmission according to the communication method of the Stratum protocol. The communication traffic in the botnet is disguised as the traffic between the mining pool and the miners in a Bitcoin network, thereby achieving better stealthiness and flexibility.
Tatar, Ekin Ecem, Dener, Murat.  2021.  Anomaly Detection on Bitcoin Values. 2021 6th International Conference on Computer Science and Engineering (UBMK). :249–253.
Bitcoin has received a lot of attention from investors, researchers, regulators, and the media. It is a known fact that the Bitcoin price usually fluctuates greatly. However, not enough scientific research has been done on these fluctuations. In this study, long short-term memory (LSTM) modeling from Recurrent Neural Networks, which is one of the deep learning methods, was applied on Bitcoin values. As a result of this application, anomaly detection was carried out in the values from the data set. With the LSTM network, a time-dependent representation of Bitcoin price can be captured, and anomalies can be selected. The factors that play a role in the formation of the model to be applied in the detection of anomalies with the experimental results were evaluated.
Zou, Kexin, Shi, Jinqiao, Gao, Yue, Wang, Xuebin, Wang, Meiqi, Li, Zeyu, Su, Majing.  2021.  Bit-FP: A Traffic Fingerprinting Approach for Bitcoin Hidden Service Detection. 2021 IEEE Sixth International Conference on Data Science in Cyberspace (DSC). :99–105.
Bitcoin is a virtual encrypted digital currency based on a peer-to-peer network. In recent years, for higher anonymity, more and more Bitcoin users try to use Tor hidden services for identity and location hiding. However, previous studies have shown that Tor are vulnerable to traffic fingerprinting attack, which can identify different websites by identifying traffic patterns using statistical features of traffic. Our work shows that traffic fingerprinting attack is also effective for the Bitcoin hidden nodes detection. In this paper, we proposed a novel lightweight Bitcoin hidden service traffic fingerprinting, using a random decision forest classifier with features from TLS packet size and direction. We test our attack on a novel dataset, including a foreground set of Bitcoin hidden node traffic and a background set of different hidden service websites and various Tor applications traffic. We can detect Bitcoin hidden node from different Tor clients and website hidden services with a precision of 0.989 and a recall of 0.987, which is higher than the previous model.
Fan, Wenjun, Hong, Hsiang-Jen, Wuthier, Simeon, Zhou, Xiaobo, Bai, Yan, Chang, Sang-Yoon.  2021.  Security Analyses of Misbehavior Tracking in Bitcoin Network. 2021 IEEE International Conference on Blockchain and Cryptocurrency (ICBC). :1–3.
Because Bitcoin P2P networking is permissionless by the application requirement, it is vulnerable against networking threats based on identity/credential manipulations such as Sybil and spoofing attacks. The current Bitcoin implementation keeps track of its peer's networking misbehaviors through ban score. In this paper, we investigate the security problems of the ban-score mechanism and discover that the ban score is not only ineffective against the Bitcoin Message-based DoS attacks but also vulnerable to a Defamation attack. In the Defamation attack, the network adversary can exploit the ban-score mechanism to defame innocent peers.
2022-06-14
Tan, Soo-Fun, Lo, Ka-Man Chirs, Leau, Yu-Beng, Chung, Gwo-Chin, Ahmedy, Fatimah.  2021.  Securing mHealth Applications with Grid-Based Honey Encryption. 2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET). :1–5.
Mobile healthcare (mHealth) application and technologies have promised their cost-effectiveness to enhance healthcare quality, particularly in rural areas. However, the increased security incidents and leakage of patient data raise the concerns to address security risks and privacy issues of mhealth applications urgently. While recent mobile health applications that rely on password-based authentication cannot withstand password guessing and cracking attacks, several countermeasures such as One-Time Password (OTP), grid-based password, and biometric authentication have recently been implemented to protect mobile health applications. These countermeasures, however, can be thwarted by brute force attacks, man-in-the-middle attacks and persistent malware attacks. This paper proposed grid-based honey encryption by hybridising honey encryption with grid-based authentication. Compared to recent honey encryption limited in the hardening password attacks process, the proposed grid-based honey encryption can be further employed against shoulder surfing, smudge and replay attacks. Instead of rejecting access as a recent security defence mechanism in mobile healthcare applications, the proposed Grid-based Honey Encryption creates an indistinct counterfeit patient's record closely resembling the real patients' records in light of each off-base speculation legitimate password.
Hataba, Muhammad, Sherif, Ahmed, Elsersy, Mohamed, Nabil, Mahmoud, Mahmoud, Mohamed, Almotairi, Khaled H..  2021.  Privacy-Preserving Biometric-based Authentication Scheme for Electric Vehicles Charging System. 2021 3rd IEEE Middle East and North Africa COMMunications Conference (MENACOMM). :86–91.
Nowadays, with the continuous increase in oil prices and the worldwide shift towards clean energy, all-electric vehicles are booming. Thence, these vehicles need widespread charging systems operating securely and reliably. Consequently, these charging systems need the most robust cybersecurity measures and strong authentication mechanisms to protect its user. This paper presents a new security scheme leveraging human biometrics in terms of iris recognition to defend against multiple types of cyber-attacks such as fraudulent identities, man-in-the-middle attacks, or unauthorized access to electric vehicle charging stations. Fundamentally, the proposed scheme implements a security mechanism based on the inherently unique characteristics of human eye biometric. The objective of the proposed scheme is to enhance the security of electric vehicle charging stations by using a low-cost and efficient authentication using k-Nearest Neighbours (KNN), which is a lightweight encryption algorithm.We tested our system on high-quality images obtained from the standard IITD iris database to search over the encrypted database and authenticate a legitimate user. The results showed that our proposed technique had minimal communication and computation overhead, which is quite suitable for the resource-limited charging station devices. Furthermore, we proved that our scheme outperforms other existing techniques.
Zakharov, E. R., Zakharova, V. O., Vlasov, A. I..  2021.  Methods and Algorithms for Generating a Storage Key Based on Biometric Parameters. 2021 International Russian Automation Conference (RusAutoCon). :137–141.
The theoretical basis made it possible to implement software for automated secure biometric verification and personal identification, which can be used by information security systems (including access control and management systems). The work is devoted to solving an urgent problem - the development of methods and algorithms for generating a key for a storage device based on biometric parameters. Biometric cryptosystems take advantage of biometrics to improve the security of encryption keys. The ability not to store a key that is derived from biometric data is a direct advantage of the method of generating cryptographic keys from biometric data of users over other existing encryption methods.
Pradel, Gaëtan, Mitchell, Chris.  2021.  Privacy-Preserving Biometric Matching Using Homomorphic Encryption. 2021 IEEE 20th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom). :494–505.
Biometric matching involves storing and processing sensitive user information. Maintaining the privacy of this data is thus a major challenge, and homomorphic encryption offers a possible solution. We propose a privacy-preserving biometrics-based authentication protocol based on fully homomorphic en-cryption, where the biometric sample for a user is gathered by a local device but matched against a biometric template by a remote server operating solely on encrypted data. The design ensures that 1) the user's sensitive biometric data remains private, and 2) the user and client device are securely authenticated to the server. A proof-of-concept implementation building on the TFHE library is also presented, which includes the underlying basic operations needed to execute the biometric matching. Performance results from the implementation show how complex it is to make FHE practical in this context, but it appears that, with implementation optimisations and improvements, the protocol could be used for real-world applications.
Vallabhu, Satya Krishna, Maheswari, Nissankararao Uma, Kaveri, Badavath, Jagadeeswari, C..  2021.  Biometric Steganography Using MPV Technique. 2021 IEEE 6th International Conference on Computing, Communication and Automation (ICCCA). :39–43.
Biometric data is prone to attacks and threats from hackers who are professionals in cyber-crimes. Therefore, securing the data is very essential. Steganographic approach, which is a process of concealing data, is proposed as a solution to this. Biometrics are hidden inside other biometrics for safe storage and secure transmission. Also, it is designed to be robust against attacks, and cannot be detected easily. The intention of this paper is to highlight a method of hiding one image in another image by using mid position value(mpv) technique. Here we have to choose the secret biometric on which Arnold transform will be applied resulting in a scrambled version of the secret biometric. This will be enveloped inside cover image which results in a stego-image. Lastly, hidden secret biometric will be decoded from this stego image, which will first result in a scrambled secret biometric. Inverse Arnold Transform will be applied on this to finally result in the decoded secret biometric. The paper further explains the working and processes in detail.
Gvozdov, Roman, Poddubnyi, Vadym, Sieverinov, Oleksandr, Buhantsov, Andrey, Vlasov, Andrii, Sukhoteplyi, Vladyslav.  2021.  Method of Biometric Authentication with Digital Watermarks. 2021 IEEE 8th International Conference on Problems of Infocommunications, Science and Technology (PIC S&T). :569–571.
This paper considers methods of fingerprint protection in biometric authentication systems. Including methods of protecting fingerprint templates using zero digital watermarks and cryptography techniques. The paper considers a secure authentication model using cryptography and digital watermarks.
Vanitha, C. N., Malathy, S., Anitha, K., Suwathika, S..  2021.  Enhanced Security using Advanced Encryption Standards in Face Recognition. 2021 2nd International Conference on Communication, Computing and Industry 4.0 (C2I4). :1–5.
Nowadays, face recognition is used everywhere in all fields. Though the face recognition is used for security purposes there is also chance in hacking the faces which is used for face recognition. For enhancing the face security, encryption and decryption technique is used. Face cognizance has been engaged in more than a few security-connected purposes such as supervision, e-passport, and etc… The significant use of biometric raises vital private concerns, in precise if the biometric same method is carried out at a central or unfrosted servers, and calls for implementation of Privacy improving technologies. For privacy concerns the encoding and decoding is used. For achieving the result we are using the Open Computer Vision (OpenCV) tool. With the help of this tool we are going to cipher the face and decode the face with advanced encryption standards techniques. OpenCV is the tool used in this project
Dhane, Harshad, Manikandan, V. M..  2021.  A New Framework for Secure Biometric Data Transmission using Block-wise Reversible Data Hiding Through Encryption. 2021 Fifth International Conference On Intelligent Computing in Data Sciences (ICDS). :1–8.
Reversible data hiding (RDH) is an emerging area in the field of information security. The RDH schemes are widely explored in the field of cloud computing for data authentication and in medical image transmission for clinical data transmission along with medical images. The RDH schemes allow the data hider to embed sensitive information in digital content in such a way that later it can be extracted while recovering the original image. In this research, we explored the use of the RDH through the encryption scheme in a biometric authentication system. The internet of things (IoT) enabled biometric authentication systems are very common nowadays. In general, in biometric authentication, computationally complex tasks such as feature extraction and feature matching will be performed in a cloud server. The user-side devices will capture biometric data such as the face, fingerprint, or iris and it will be directly communicated to the cloud server for further processing. Since the confidentiality of biometric data needs to be maintained during the transmission, the original biometric data will be encrypted using any one of the data encryption techniques. In this manuscript, we propose the use of RDH through encryption approach to transmit two different biometric data as a single file without compromising confidentiality. The proposed scheme will ensure the integrity of the biometric data during transmission. For data hiding purposes, we have used a block-wise RDH through encryption scheme. The experimental study of the proposed scheme is carried out by embedding fingerprint data in the face images. The validation of the proposed scheme is carried out by extracting the fingerprint details from the face images during image decryption. The scheme ensures the exact recovery of face image images and fingerprint data at the receiver site.
Hofbauer, Heinz, Martínez-Díaz, Yoanna, Kirchgasser, Simon, Méndez-Vázquez, Heydi, Uhl, Andreas.  2021.  Highly Efficient Protection of Biometric Face Samples with Selective JPEG2000 Encryption. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). :2580–2584.
When biometric databases grow larger, a security breach or leak can affect millions. In order to protect against such a threat, the use of encryption is a natural choice. However, a biometric identification attempt then requires the decryption of a potential huge database, making a traditional approach potentially unfeasible. The use of selective JPEG2000 encryption can reduce the encryption’s computational load and enable a secure storage of biometric sample data. In this paper we will show that selective encryption of face biometric samples is secure. We analyze various encoding settings of JPEG2000, selective encryption parameters on the "Labeled Faces in the Wild" database and apply several traditional and deep learning based face recognition methods.
Kuznetsova, Nataliya M., Karlova, Tatyana V., Bekmeshov, Alexander Y., Kirillova, Elena A., Mikhaylova, Marianna V., Averchenkov, Andrey V..  2021.  Mathematical and Algorithmic Prevention of Biometric Data Leaks. 2021 International Conference on Quality Management, Transport and Information Security, Information Technologies (IT&QM&IS). :210–212.
Biometric methods are the most effective and accurate authentication methods. However, a significant drawback of such methods is the storage of authentication information in clear text. The article is devoted to solving this problem by means of symmetric encryption method and the method of dividing the memory space. The method of symmetric encryption ensures confidentiality during storage and transmission of biometric characteristics, the method of dividing the memory space provides an increase of information security level during processing of biometric characteristics.
Su, Liyilei, Fu, Xianjun, Hu, Qingmao.  2021.  A convolutional generative adversarial framework for data augmentation based on a robust optimal transport metric. 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys). :1155–1162.
Enhancement of the vanilla generative adversarial network (GAN) to preserve data variability in the presence of real world noise is of paramount significance in deep learning. In this study, we proposed a new distance metric of cosine distance in the framework of optimal transport (OT), and presented and validated a convolutional neural network (CNN) based GAN framework. In comparison with state-of-the-art methods based on Graphics Processing Units (GPU), the proposed framework could maintain the data diversity and quality best in terms of inception score (IS), Fréchet inception distance (FID) and enhancing the classification network of bone age, and is robust to noise degradation. The proposed framework is independent of hardware and thus could also be extended to more advanced hardware such as specialized Tensor Processing Units (TPU), and could be a potential built-in component of a general deep learning networks for such applications as image classification, segmentation, registration, and object detection.
Zuech, Richard, Hancock, John, Khoshgoftaar, Taghi M..  2021.  Feature Popularity Between Different Web Attacks with Supervised Feature Selection Rankers. 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA). :30–37.
We introduce the novel concept of feature popularity with three different web attacks and big data from the CSE-CIC-IDS2018 dataset: Brute Force, SQL Injection, and XSS web attacks. Feature popularity is based upon ensemble Feature Selection Techniques (FSTs) and allows us to more easily understand common important features between different cyberattacks, for two main reasons. First, feature popularity lists can be generated to provide an easy comprehension of important features across different attacks. Second, the Jaccard similarity metric can provide a quantitative score for how similar feature subsets are between different attacks. Both of these approaches not only provide more explainable and easier-to-understand models, but they can also reduce the complexity of implementing models in real-world systems. Four supervised learning-based FSTs are used to generate feature subsets for each of our three different web attack datasets, and then our feature popularity frameworks are applied. For these three web attacks, the XSS and SQL Injection feature subsets are the most similar per the Jaccard similarity. The most popular features across all three web attacks are: Flow\_Bytes\_s, FlowİAT\_Max, and Flow\_Packets\_s. While this introductory study is only a simple example using only three web attacks, this feature popularity concept can be easily extended, allowing an automated framework to more easily determine the most popular features across a very large number of attacks and features.
Kawanishi, Yasuyuki, Nishihara, Hideaki, Yoshida, Hirotaka, Hata, Yoichi.  2021.  A Study of The Risk Quantification Method focusing on Direct-Access Attacks in Cyber-Physical Systems. 2021 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech). :298–305.
Direct-access attacks were initially considered as un-realistic threats in cyber security because the attacker can more easily mount other non-computerized attacks like cutting a brake line. In recent years, some research into direct-access attacks have been conducted especially in the automotive field, for example, research on an attack method that makes the ECU stop functioning via the CAN bus. The problem with existing risk quantification methods is that direct-access attacks seem not to be recognized as serious threats. To solve this problem, we propose a new risk quantification method by applying vulnerability evaluation criteria and by setting metrics. We also confirm that direct-access attacks not recognized by conventional methods can be evaluated appropriately, using the case study of an automotive system as an example of a cyber-physical system.
Singh, A K, Goyal, Navneet.  2021.  Detection of Malicious Webpages Using Deep Learning. 2021 IEEE International Conference on Big Data (Big Data). :3370–3379.
Malicious Webpages have been a serious threat on Internet for the past few years. As per the latest Google Transparency reports, they continue to be top ranked amongst online threats. Various techniques have been used till date to identify malicious sites, to include, Static Heuristics, Honey Clients, Machine Learning, etc. Recently, with the rapid rise of Deep Learning, an interest has aroused to explore Deep Learning techniques for detecting Malicious Webpages. In this paper Deep Learning has been utilized for such classification. The model proposed in this research has used a Deep Neural Network (DNN) with two hidden layers to distinguish between Malicious and Benign Webpages. This DNN model gave high accuracy of 99.81% with very low False Positives (FP) and False Negatives (FN), and with near real-time response on test sample. The model outperformed earlier machine learning solutions in accuracy, precision, recall and time performance metrics.
Schneider, Madeleine, Aspinall, David, Bastian, Nathaniel D..  2021.  Evaluating Model Robustness to Adversarial Samples in Network Intrusion Detection. 2021 IEEE International Conference on Big Data (Big Data). :3343–3352.
Adversarial machine learning, a technique which seeks to deceive machine learning (ML) models, threatens the utility and reliability of ML systems. This is particularly relevant in critical ML implementations such as those found in Network Intrusion Detection Systems (NIDS). This paper considers the impact of adversarial influence on NIDS and proposes ways to improve ML based systems. Specifically, we consider five feature robustness metrics to determine which features in a model are most vulnerable, and four defense methods. These methods are tested on six ML models with four adversarial sample generation techniques. Our results show that across different models and adversarial generation techniques, there is limited consistency in vulnerable features or in effectiveness of defense method.