Certificates are the foundation of secure communication over the internet as of today. While certificates can be issued with long validity periods, there is always a risk of having them compromised during their lifetime. A good practice is therefore to use shorter validity periods. However, this limits the certificate lifetime and gives less flexibility in the timing of certificate replacements. In this thesis, we use publicly available network logs from Rapid7's Project Sonar to provide an overview of the current state of certificate usage behavior. Specifically, we look at the Let's Encrypt mass revocation event in March 2020, where millions of certificates were revoked with just five days notice. In general, we show how this kind of datasets can be used, and as a deeper exploration we analyze certificate validity, lifetime and use of certificates with overlapping validity periods, as well as discuss how our findings relate to industry standard and current security trends. Specifically, we isolate automated certificate services such as Let's Encrypt and cPanel to see how their certificates differ in characteristics from other certificates in general. Based on our findings, we propose a set of rules to help improve the trust in certificate usage and strengthen security online, introducing an Always secure policy aligning certificate validity with revocation time limits in order to replace revocation requirements and overcoming the fact that mobile devices today ignore this very important security feature. To round things off, we provide some ideas for further research based on our findings and what we see possible with datasets such as the one researched in this thesis.

La détection tardive des menaces à la sécurité entraîne une augmentation significative du risque de dommages irréparables, invalidant toute tentative de défense. En conséquence, la détection rapide des menaces en temps réel est obligatoire pour l'administration de la sécurité. De plus, la fonction de virtualisation de la fonction réseau (NFV) offre de nouvelles opportunités pour des solutions de sécurité efficaces et à faible coût. Nous proposons un système de détection de menaces rapide et efficace basé sur des algorithmes de traitement de flux et d'apprentissage automatique. Les principales contributions de ce travail sont : i) un nouveau système de détection des menaces de surveillance basé sur le traitement en continu, ii) deux ensembles de données, d'abord un ensemble de données de sécurité synthétiques contenant à la fois du trafic légitime et malveillant, et le deuxième, une semaine de trafic réel d'un opérateur de télécommunications à Rio de Janeiro, au Brésil, iii) un algorithme de pré-traitement de données, un algorithme de normalisation et un algorithme de sélection de caractéristiques rapides basé sur la corrélation entre des variables, iv) une fonction de réseau virtualisé dans une plate-forme Open Source pour fournir un service de détection des menaces en temps réel, v) placement quasi-optimal des capteurs grâce à une heuristique proposée pour positionner stratégiquement les capteurs dans l'infrastructure du réseau, avec un nombre minimal de capteurs, et enfin vi) un algorithme glouton qui alloue à la demande une séquence de fonctions de réseau virtuel. / The late detection of security threats causes a significant increase in the risk of irreparable damages, disabling any defense attempt. As a consequence, fast real-time threat detection is mandatory for security administration. In addition, Network Function Virtualization (NFV) provides new opportunities for efficient and low-cost security solutions. We propose a fast and efficient threat detection system based on stream processing and machine learning algorithms. The main contributions of this work are i) a novel monitoring threat detection system based on streaming processing, ii) two datasets, first a dataset of synthetic security data containing both legitimate and malicious traffic, and the second, a week of real traffic of a telecommunications operator in Rio de Janeiro, Brazil, iii) a data pre-processing algorithm, a normalizing algorithm and an algorithm for fast feature selection based on the correlation between variables, iv) a virtualized network function in an Open source Platform for providing a real-time threat detection service, v) near-optimal placement of sensors through a proposed heuristic for strategically positioning sensors in the network infrastructure, with a minimum number of sensors, and finally vi) a greedy algorithm that allocates on demand a sequence of virtual network functions.

An important part of maintaining network security is collecting and processing information about cyber threats, both from network operator's own detection tools and from third parties. A commonly used type of such information are lists of network entities (IP addresses, domains, URLs, etc.) which were identified as malicious. However, in many cases, the simple binary distinction between malicious and non-malicious entities is not sufficient. It is beneficial to keep other supplementary information for each entity, which describes its malicious activities, and also a summarizing score, which evaluates its reputation numerically. Such a score allows for quick comprehension of the level of threat the entity poses and allows to compare and sort entities. The goal of this work is to design a method for such summarization. The resulting score, called Future Maliciousness Probability (FMP score), is a value between 0 and 1, assigned to each suspicious network entity, expressing the probability that the entity will do some kind of malicious activity in a near future. Therefore, the scoring is based of prediction of future attacks. Advanced machine learning methods are used to perform the prediction. Their input is formed by previously received alerts about security events and other relevant data related to the entity. The method of computing the score is first described in a general way, usable for any kind of entity and input data. Then a more concrete version is presented for scoring IPv4 address by utilizing alerts from an alert sharing system and supplementary data from a reputation database. This variant is then evaluated on a real world dataset. In order to get enough amount and quality of data for this dataset, a part of the work is also dedicated to the area of security analysis of network data. A framework for analysis of flow data, NEMEA, and several new detection methods are designed and implemented. An open reputation database, NERD, is also implemented and described in this work. Data from these systems are then used to evaluate precision of the predictor as well as to evaluate selected use cases of the scoring method.

With the emergence of cloud-native applications, the need to secure networks and services creates new requirements concerning automation, manageability, and scalability across data centers. Several solutions have been developed to overcome the limitations of the conventional and well established IPsec suite as a secure tunneling solution. One strategy to meet these new requirements has been the design of software-based overlay networks. In this thesis, we assess the deployment of a traditional IPsec VPN solution against a new secure overlay mesh network called Nebula. We conduct a case study by provisioning an experimental system to evaluate Nebula in four key areas: reliability, security, manageability, and performance. We discuss the strengths of Nebula and its limitations for securing inter-service communication in distributed cloud applications. In terms of reliability, the thesis shows that Nebula falls short to meet its own goals of achieving host-to-host connectivity when attempting to traverse specific firewalls and NATs. With respect to security, Nebula provides certificate-based authentication and uses current and fast cryptographic algorithms and protocols from the Noise framework. Regarding manageability, Nebula is a modern solution with a loosely coupled design that allows scalability with cloud-ready features and easier deployment than IPsec. Finally, the performance of Nebula clearly shows an overhead for being a user-space software application. However, the overhead can be considered acceptable in certain server-to-server microservice interactions and is a fair trade-off for its ease of management in comparison to IPsec. / Med framväxten av molninbyggda applikationer skapar behovet av säkra nätverk och tjänster nya krav på automatisering, hanterbarhet och skalbarhet över datacenter. Flera lösningar har utvecklats för att övervinna begränsningarna i den konventionella och väletablerade IPsec-sviten som en säker tunnellösning. En strategi för att möta dessa nya krav har varit utformningen av mjukvarubaserade överläggsnätverk. I den här avhandlingen bedömer vi implementeringen av en traditionell IPsec VPN-lösning mot ett nytt säkert överläggsmeshnätverk som kallas Nebula. Vi genomför en fallstudie genom att bygga upp ett ett experimentellt system för att utvärdera Nebula inom fyra nyckelområden: tillförlitlighet, säkerhet, hanterbarhet och prestanda. Vi diskuterar styrkan i Nebula och dess begränsningar för att säkra kommunikation mellan tjänster i distribuerade molnapplikationer. När det gäller tillförlitlighet visar avhandlingen att Nebula inte uppfyller sina egna mål om att uppnå värd-tillvärd- anslutning när man försöker korsa specifika brandväggar och NAT. När det gäller säkerhet tillhandahåller Nebula certifikatbaserad autentisering och använder aktuella och snabba kryptografiska algoritmer och protokoll från Noise-ramverket. När det gäller hanterbarhet är Nebula en modern lösning med en löst kopplad design som möjliggör skalbarhet med molnklara funktioner och enklare distribution än IPsec. Slutligen visar prestandan hos Nebula tydligt en overhead för att vara en användarutrymme-programvara. Dock kan kostnaderna anses vara acceptabla i vissa server-till-server-mikroserviceinteraktioner och är en rättvis avvägning om vi tar i betraktande dess enkla hantering jämfört med IPsec.

Capabilities-based networks present a fundamental shift in the security design of network architectures. Instead of permitting the transmission of packets from any source to any destination, routers deny forwarding by default. For a successful transmission, packets need to positively identify themselves and their permissions to the router. A major challenge for a high performance implementation of such a network is an efficient design of the credentials that are carried in the packet and the verification procedure on the router. A network protocol that implements data path credentials based on Bloom filters is presented in this thesis. Our prototype implementation shows that there is some connection setup cost associated with this type of secure communication. However, once a connection is established, the throughput performance of a capabilities-based connection is similar to that of conventional TCP.

Today's Internet is seeing a massive shift from traditional client-server applications towards real-time, context-sensitive, and highly immersive applications. The fusion between Cyber-physical systems, The Internet of Things (IoT), Augmented/Virtual-Reality (AR/VR), and the Tactile Internet with the Human-in-the-Loop (TaHIL) means that Ultra-Reliable Low Latency Communication (URLLC) is a key functional requirement. Mobile Edge Computing (MEC) has emerged as a network architectural paradigm to address such ever-increasing resource demands. MEC leverages networking and computational resource pools that are closer to the end-users at the far edge of the network, eliminating the need to send and process large volumes of data over multiple distant hops at central cloud computing data centers. Multiple 'cloudlets' are formed at the edge, and the access to resources is shared and federated across them over multiple network domains that are distributed over various geographical locations. However, this federated access comes at the cost of a fuzzy and dynamically-changing network security perimeter because there are multiple sources of mobility. Not only are the end users mobile, but the applications themselves virtually migrate over multiple network domains and cloudlets to serve the end users, bypassing statically placed network security middleboxes and firewalls. This work aims to address this problem by proposing adaptive network security measures that can be dynamically changed at runtime, and are decoupled from the ever-changing network topology. In particular, we: 1) use the state of the art in programmable networking to protect MEC networks from internal adversaries that can adapt and laterally move, 2) Automatically infer application security contexts, and device vulnerabilities, then evolve the network access control policies to segment the network in such a way that minimizes the attack surface with minimal impact on its utility, 3) propose new metrics to assess the susceptibility of edge nodes to a new class of stealthy attacks that bypasses traditional statically placed Intrusion Detection Systems (IDS), and a probabilistic approach to pro-actively protect them.:Acknowledgments Acronyms & Abbreviations 1 Introduction 1.1 Prelude 1.2 Motivation and Challenges 1.3 Aim and objectives 1.4 Contributions 1.5 Thesis structure 2 Background 2.1 A primer on computer networks 2.2 Network security 2.3 Network softwarization 2.4 Cloudification of networks 2.5 Securing cloud networks 2.6 Towards Securing Edge Cloud Networks 2.7 Summary I Adaptive security in consumer edge cloud networks 3 Automatic microsegmentation of smarthome IoT networks 3.1 Introduction 3.2 Related work 3.3 Smart home microsegmentation 3.4 Software-Defined Secure Isolation 3.5 Evaluation 3.6 Summary 4 Smart home microsegmentation with user privacy in mind 4.1 Introduction 4.2 Related Work 4.3 Goals and Assumptions 4.4 Quantifying the security and privacy of SHIoT devices 4.5 Automatic microsegmentation 4.6 Manual microsegmentation 4.7 Experimental setup 4.8 Evaluation 4.9 Summary II Adaptive security in enterprise edge cloud networks 5 Adaptive real-time network deception and isolation 5.1 Introduction 5.2 Related work 5.3 Sandnet’s concept 5.4 Live Cloning and Network Deception 5.5 Evaluation 5.6 Summary 6 Localization of internal stealthy DDoS attacks on Microservices 6.1 Introduction 6.2 Related work 6.3 Assumptions & Threat model 6.4 Mitigating SILVDDoS 6.5 Evaluation 6.6 Summary III Summary of Results 7 Conclusion 7.1 Main outcomes 7.2 Future outlook Listings Bibliography List of Algorithms List of Figures List of Tables Appendix

<p>Dockerizing web servers has gained significant popularity due to its lightweight containerization approach, enabling rapid and efficient deployment of web services. However, the security of web server containers remains a critical concern. This study proposes a novel approach to enhance the security of Docker-based web servers using bpftrace to trace Nginx and Apache containers under attack, identifying abnormal syscalls, connections, shared library calls, and file accesses from normal ones. The gathered metrics are used to generate tailored AppArmor profiles for improved mandatory access control policies and enhanced container security. BPFtrace is a high-level tracing language allowing for real-time analysis of system events. This research introduces an innovative method for generating AppArmor profiles by utilizing BPFtrace to monitor system alerts, creating customized security policies tailored to the specific needs of Docker-based web servers. Once the profiles are generated, the web server container is redeployed with enhanced security measures in place. This approach increases security by providing granular control and adaptability to address potential threats. The evaluation of the proposed method is conducted using CVE’s found in the open source literature affecting nginx and apache web servers that correspond to the classification system that was created. The Apache and Nginx containers was attacked with Metasploit, and benchmark tests including ltrace evaluation in accordance with existing literature were conducted. The results demonstrate the effectiveness of the proposed approach in mitigating security risks and strengthening the overall security posture of Docker-based web servers. This is achieved by limiting memcpy and memset shared library calls identified using bpftrace and applying rlimits in 9 AppArmor to limit their rate to normal levels (as gauged during testing) and deny other harmful file accesses and syscalls. The study’s findings contribute to the growing body of knowledge on container security and offer valuable insights for practitioners aiming to develop more secure web server deployments using Docker. </p>

This thesis project aimed to design and implement a secure system for handling and safeguarding personal data. The purpose of the work is to prevent unauthorized actors from gaining access to systems and data. The proposed solution is a Zero Trust architecture which emphasizes strong security measures by design and strict access controls. The system must provide minimal access for users and should be integrated with the existing cloud-based infrastructure. The result is a system that leverages Keycloak for identity management and authentication services, GitLab to provide a code hosting solution, GPG for commit signing, and OpenVPN for network access. Through the utilization of Gitlab, Keycloak and OpenVPN the system achieved a comprehensive design for data protection, user authentication and network security. This report also highlights alternative methods, future enhancements and potential improvements to the completed system. / Målet med denna rapport är att designa och implementera ett säkert system för hantering och skydd av personlig data. Syftet med arbetet är att förhindra obehöriga att få tillgång till system och data. Den föreslagna lösningen är en Zero Trustarkitektur som betonar skärpta säkerhetsåtgärder genom design och strikta åtkomstkontroller. Systemet måste ge minimal åtkomst för användare som brukar det och integreras med den befintliga molnbaserade infrastrukturen. Resultatet är ett system som använder Keycloak för hantering av identiteter och autentisering, GitLab för att tillhandahålla ett kodarkiv där användare kan ladda upp sin kod, GPG för att signera commits, och OpenVPN för nätverksåtkomst. Genom användning av GitLab, Keycloak och OpenVPN uppnådde systemet en omfattande design för dataskydd, användarautentisering och nätverkssäkerhet. Denna rapport nämner också alternativa metoder, framtida och potentiella förbättringar av det färdiga systemet.

<p>The progressive integration of microcontrollers into various domains has transformed traditional mechanical systems into modern cyber-physical systems. However, the beginning of this transformation predated the era of hyper-interconnectedness that characterizes our contemporary world. As such, the principles and visions guiding the design choices of this transformation had not accounted for many of today's security challenges. Many designers had envisioned their systems to operate in an air-gapped-like fashion where few security threats loom. However, with the hyper-connectivity of today's world, many CPS find themselves in uncharted territory for which they are unprepared.</p> <p><br></p> <p>An example of this evolution is the Controller Area Network (CAN). CAN emerged during the transformation of many mechanical systems into cyber-physical systems as a pivotal communication standard, reducing vehicle wiring and enabling efficient data exchange. CAN's features, including noise resistance, decentralization, error handling, and fault confinement mechanisms, made it a widely adopted communication medium not only in transportation but also in diverse applications such as factories, elevators, medical equipment, avionic systems, and naval applications.</p> <p><br></p> <p>The increasing connectivity of modern vehicles through CD players, USB sticks, Bluetooth, and WiFi access has exposed CAN systems to unprecedented security challenges and highlighted the need to bolster their security posture. This dissertation addresses the urgent need to enhance the security of modern cyber-physical systems in the face of emerging threats by proposing a proactive vulnerability identification and defense construction approach and applying it to CAN as a lucid case study. By adopting this proactive approach, vulnerabilities can be systematically identified, and robust defense mechanisms can be constructed to safeguard the resilience of CAN systems.</p> <p><br></p> <p>We focus on developing vulnerability scanning techniques and innovative defense system designs tailored for CAN systems. By systematically identifying vulnerabilities before they are discovered and exploited by external actors, we minimize the risks associated with cyber-attacks, ensuring the longevity and reliability of CAN systems. Furthermore, the defense mechanisms proposed in this research overcome the limitations of existing solutions, providing holistic protection against CAN threats while considering its performance requirements and operational conditions.</p> <p><br></p> <p>It is important to emphasize that while this dissertation focuses on CAN, the techniques and rationale used here could be replicated to secure other cyber-physical systems. Specifically, due to CAN's presence in many cyber-physical systems, it shares many performance and security challenges with those systems, which makes most of the techniques and approaches used here easily transferrable to them. By accentuating the importance of proactive security, this research endeavors to establish a foundational approach to cyber-physical systems security and resiliency. It recognizes the evolving nature of cyber-physical systems and the specific security challenges facing each system in today's hyper-connected world and hence focuses on a single case study. </p>

<div>The exponential growth of Internet of Things (IoT) and cyber-physical systems is resulting in complex environments comprising of various devices interacting with each other and with users. In addition, the rapid advances in Artificial Intelligence are making those devices able to autonomously modify their behaviors through the use of techniques such as reinforcement learning (RL). There is thus the need for an intelligent monitoring system on the network edge with a global view of the environment to autonomously predict optimal device actions. However, it is clear however that ensuring safety and security in such environments is critical. To this effect, we develop a constrained RL framework for IoT environments that determines optimal devices actions with respect to user-defined goals or required functionalities using deep Q learning. We use anomaly based intrusion detection on the network edge to dynamically generate security and safety policies to constrain the RL agent in the framework. We analyze the balance required between ‘safety/security’ and ‘functionality’ in IoT environments by manipulating the exploration of safe and unsafe benefit state spaces in the RL framework. We instantiate the framework for testing on application layer control in smart home environments, and network layer control including network functionalities like rate control and routing, for SDN based environments.</div>