Biblio

Advanced Persistent Threat (APT) having focused target along with advanced and persistent attacking skills under great concealment is a new trend followed for cyber-attacks. Threat intelligence helps in detecting and preventing APT by collecting a host of data and analyzing malicious behavior through efficient data sharing and guaranteeing the safety and quality of information exchange. For better protection, controlled access to intelligence information and a grading standard to revise the criteria in diagnosis for a security breach is needed. This paper analyses a threat intelligence sharing community model and proposes an improvement to increase the efficiency of sharing by rethinking the size and composition of a sharing community. Based on various external environment variables, it filters the low-quality shared intelligence by grading the trust level of a community member and the quality of a piece of intelligence. We hope that this research can fill in some security gaps to help organizations make a better decision in handling the ever-increasing and continually changing cyber-attacks.

Machine Learning as a Service (MLaaS) is becoming a popular practice where Service Consumers, e.g., end-users, send their data to a ML Service and receive the prediction outputs. However, the emerging usage of MLaaS has raised severe privacy concerns about users' proprietary data. PrivacyPreserving Machine Learning (PPML) techniques aim to incorporate cryptographic primitives such as Homomorphic Encryption (HE) and Multi-Party Computation (MPC) into ML services to address privacy concerns from a technology standpoint. Existing PPML solutions have not been widely adopted in practice due to their assumed high overhead and integration difficulty within various ML front-end frameworks as well as hardware backends. In this work, we propose PlaidML-HE, the first end-toend HE compiler for PPML inference. Leveraging the capability of Domain-Specific Languages, PlaidML-HE enables automated generation of HE kernels across diverse types of devices. We evaluate the performance of PlaidML-HE on different ML kernels and demonstrate that PlaidML-HE greatly reduces the overhead of the HE primitive compared to the existing implementations.

The huge volume, variety, and velocity of big data have empowered Machine Learning (ML) techniques and Artificial Intelligence (AI) systems. However, the vast portion of data used to train AI systems is sensitive information. Hence, any vulnerability has a potentially disastrous impact on privacy aspects and security issues. Nevertheless, the increased demands for high-quality AI from governments and companies require the utilization of big data in the systems. Several studies have highlighted the threats of big data on different platforms and the countermeasures to reduce the risks caused by attacks. In this paper, we provide an overview of the existing threats which violate privacy aspects and security issues inflicted by big data as a primary driving force within the AI/ML workflow. We define an adversarial model to investigate the attacks. Additionally, we analyze and summarize the defense strategies and countermeasures of these attacks. Furthermore, due to the impact of AI systems in the market and the vast majority of business sectors, we also investigate Standards Developing Organizations (SDOs) that are actively involved in providing guidelines to protect the privacy and ensure the security of big data and AI systems. Our far-reaching goal is to bridge the research and standardization frame to increase the consistency and efficiency of AI systems developments guaranteeing customer satisfaction while transferring a high degree of trustworthiness.

IoT devices introduce unprecedented threats into home and professional networks. As they fail to adhere to security best practices, they are broadly exploited by malicious actors to build botnets or steal sensitive information. Their adoption challenges established security standard as classic security measures are often inappropriate to secure them. This is even more problematic in sensitive environments where the presence of insecure IoTs can be exploited to bypass strict security policies. In this paper, we demonstrate an attack against a highly secured network using a Bluetooth smart bulb. This attack allows a malicious actor to take advantage of a smart bulb to exfiltrate data from an air gapped network.

Although transport agency has employed desensitization techniques to deal with the privacy information when publicizing vehicle license plate recognition (VLPR) data, the adversaries can still eavesdrop on vehicle trajectories by certain means and further acquire the associated person and vehicle information through background knowledge. In this work, a privacy attacking method by using the desensitized VLPR data is proposed to link the vehicle trajectory. First the road average speed is evaluated by analyzing the changes of traffic flow, which is used to estimate the vehicle's travel time to the next VLPR system. Then the vehicle suspicion list is constructed through the time relevance of neighboring VLPR systems. Finally, since vehicles may have the same features like color, type, etc, the target trajectory will be located by filtering the suspected list by the rule of qualified identifier (QI) attributes and closest time method. Based on the Foshan City's VLPR data, the method is tested and results show that correct vehicle trajectory can be linked, which proves that the current VLPR data publication way has the risk of privacy disclosure. At last, the effects of related parameters on the proposed method are discussed and effective suggestions are made for publicizing VLPR date in the future.

Large scale cloud networks consist of distributed networking and computing elements that process critical information and thus security is a key requirement for any environment. Unfortunately, assessing the security state of such networks is a challenging task and the tools used in the past by security experts such as packet filtering, firewall, Intrusion Detection Systems (IDS) etc., provide a reactive security mechanism. In this paper, we introduce a Moving Target Defense (MTD) based proactive security framework for monitoring attacks which lets us identify and reason about multi-stage attacks that target software vulnerabilities present in a cloud network. We formulate the multi-stage attack scenario as a two-player zero-sum Markov Game (between the attacker and the network administrator) on attack graphs. The rewards and transition probabilities are obtained by leveraging the expert knowledge present in the Common Vulnerability Scoring System (CVSS). Our framework identifies an attacker's optimal policy and places countermeasures to ensure that this attack policy is always detected, thus forcing the attacker to use a sub-optimal policy with higher cost.

In this paper, we design NETHCF, a line-rate in-network system for filtering spoofed traffic. NETHCF leverages the opportunity provided by programmable switches to design a novel defense against spoofed IP traffic, and it is highly efficient and adaptive. One key challenge stems from the restrictions of the computational model and memory resources of programmable switches. We address this by decomposing the HCF system into two complementary components-one component for the data plane and another for the control plane. We also aggregate the IP-to-Hop-Count (IP2HC) mapping table for efficient memory usage, and design adaptive mechanisms to handle end-to-end routing changes, IP popularity changes, and network activity dynamics. We have built a prototype on a hardware Tofino switch, and our evaluation demonstrates that NETHCF can achieve line-rate and adaptive traffic filtering with low overheads.

Vehicle ad-hoc network (VANET) is the main driving force to alleviate traffic congestion and accelerate the construction of intelligent transportation. However, the rapid growth of the number of vehicles makes the construction of the safety system of the vehicle network facing multiple tests. This paper proposes an identity-based aggregate signature scheme to protect the privacy of vehicle identity, receive messages in time and authenticate quickly in VANET. The scheme uses aggregate signature algorithm to aggregate the signatures of multiple users into one signature, and joins the idea of batch authentication to complete the authentication of multiple vehicular units, thereby improving the verification efficiency. In addition, the pseudoidentity of vehicles is used to achieve the purpose of vehicle anonymity and privacy protection. Finally, the secure storage of message signatures is effectively realized by using reliable cloud storage technology. Compared with similar schemes, this paper improves authentication efficiency while ensuring security, and has lower storage overhead.

In this paper, we propose the first traffic monitoring and labeling system for iOS devices, named iMonitor, which not just captures mobile network traffic in .pcap files, but also provides comprehensive APP-related and user-related information of captured packets. Through further analysis, one can obtain the exact APP or device where each packet comes from. The labeled traffic can be used in many research areas for mobile security, such as privacy leakage detection and user profiling. Given the implementation methodology of NetworkExtension framework of iOS 9+, APP labels of iMonitor are reliable enough so that labeled traffic can be regarded as training data for any traffic classification methods. Evaluations on real iPhones demonstrate that iMonitor has no notable impact upon user experience even with slight packet latency. Also, the experiment result supports our motivation that mobile traffic monitoring for iOS is absolutely necessary, as traffic generated by different OSes like Android and iOS are different and unreplaceable in researches.

Message Queue Telemetry Transport (MQTT) is widely accepted as a data exchange protocol in Internet of Things (IoT) environment. For security, MQTT supports Transport Layer Security (MQTT-TLS). However, MQTT-TLS provides thing-to-broker channel encryption only because data can still be exposed after MQTT broker. In addition, ACL becomes impractical due to the increasing number of rules for authorizing massive IoT devices. For solving these problems, we propose MQTT Thing-to-Thing Security (MQTT-TTS) which provides thing-to-thing security which prevents data leak. MQTT-TTS also provides the extensibility to include demanded security mechanisms for various security requirements. Moreover, the transparency of MQTT-TTS lets IoT application developers implementing secure data exchange with less programming efforts. Our MQTT-TTS implementation is available on https://github.com/beebit-sec/beebit-mqttc-sdk for evaluation.

Machine learning, specifically deep learning is becoming a key technology component in application domains such as identity management, finance, automotive, and healthcare, to name a few. Proprietary machine learning models - Machine Learning IP - are developed and deployed at the network edge, end devices and in the cloud, to maximize user experience. With the proliferation of applications embedding Machine Learning IPs, machine learning models and hyper-parameters become attractive to attackers, and require protection. Major players in the semiconductor industry provide mechanisms on device to protect the IP at rest and during execution from being copied, altered, reverse engineered, and abused by attackers. In this work we explore system security architecture mechanisms and their applications to Machine Learning IP protection.

As digital microfluidic biochips (DMFBs) make the transition to the marketplace for commercial exploitation, security and intellectual property (IP) protection are emerging as important design considerations. Recent studies have shown that DMFBs are vulnerable to reverse engineering aimed at stealing biomolecular protocols (IP theft). The IP piracy of proprietary protocols may lead to significant losses for pharmaceutical and biotech companies. The micro-electrode-dot-array (MEDA) is a next-generation DMFB platform that supports real-time sensing of droplets and has the added advantage of important security protections. However, real-time sensing offers opportunities to an attacker to steal the biochemical IP. We show that the daisychaining of microelectrodes and the use of one-time-programmability in MEDA biochips provides effective bitstream scrambling of biochemical protocols. To examine the strength of this solution, we develop a SAT attack that can unscramble the bitstreams through repeated observations of bioassays executed on the MEDA platform. Based on insights gained from the SAT attack, we propose an advanced defense against IP theft. Simulation results using real-life biomolecular protocols confirm that while the SAT attack is effective for simple instances, our advanced defense can thwart it for realistic MEDA biochips and real-life protocols.

Field-programmable gate arrays (FPGAs) are widely applied in various fields for its merit of reconfigurability. The reusable intellectual property (IP) design blocks are usually adopted in the more complex FPGA designs to shorten design cycle. IP infringement hence becomes a concern. In this paper, we propose a new pay-per-use scheme using the lock and key mechanism for the protection of FPGA IP. Physical Unclonable Function (PUF) is adopted to generate a unique ID for each IP instance. An extra Finite State Machine (FSM) is introduced for the secure retrieval of PUF information by the FPGA IP vendor. The lock is implemented on the original FSM. Only when the FPGA developer can provide a correct license, can the FSM be unlocked and start normal operation. The FPGA IP can hence be protected from illegal use or distribution. The scheme is applied on some benchmarks and the experimental results show that it just incurs acceptably low overhead while it can resist typical attacks.

Researchers develop bioassays by rigorously experimenting in the lab. This involves significant fiscal and skilled person-hour investment. A competitor can reverse engineer a bioassay implementation by imaging or taking a video of a biochip when in use. Thus, there is a need to protect the intellectual property (IP) rights of the bioassay developer. We introduce a novel 3D multilayer-based obfuscation to protect a biochip against reverse engineering.

The majority of security breaches in cloud infrastructure in recent years are caused by human errors and misconfigured resources. Novel security models are imperative to overcome these issues. Such models must be customer-centric, continuous, not focused on traditional security paradigms like intrusion detection and adopt proactive techniques. Thus, this paper proposes CloudStrike, a cloud security system that implements the principles of Chaos Engineering to enable the aforementioned properties. Chaos Engineering is an emerging discipline employed to prevent non-security failures in cloud infrastructure via Fault Injection Testing techniques. CloudStrike employs similar techniques with a focus on injecting failures that impact security i.e. integrity, confidentiality and availability. Essentially, CloudStrike leverages the relationship between dependability and security models. Preliminary experiments provide insightful and prospective results.

Aiming at the shortcomings of the traditional network active security threat model that cannot continuously control the threat process, a network active security threat model based on offensive and defensive differential game is constructed. The attack and defense differential game theory is used to define the parameters of the network active security threat model, on this basis, the network security target is determined, the network active security threat is identified by the attack defense differential equation, and finally the network active security threat is quantitatively evaluated, thus construction of network active security threat model based on offensive and defensive differential game is completed. The experimental results show that compared with the traditional network active security threat model, the proposed model is more feasible in the attack and defense control of the network active security threat process, and can achieve the ideal application effect.

In recent years the use of security gateways (SG) located within the electrical grid distribution network has become pervasive. SGs in substations and renewable distributed energy resource aggregators (DERAs) protect power distribution control devices from cyber and cyber-physical attacks. When encrypted communications within a DER network is used, TCP/IP packet inspection is restricted to packet header behavioral analysis which in most cases only allows the SG to perform anomaly detection of blocks of time-series data (event windows). Packet header anomaly detection calculates the probability of the presence of a threat within an event window, but fails in such cases where the unreadable encrypted payload contains the attack content. The SG system log (syslog) is a time-series record of behavioral patterns of network users and processes accessing and transferring data through the SG network interfaces. Threatening behavioral pattern in the syslog are measurable using both anomaly detection and graph theory. In this paper it will be shown that it is possible to efficiently detect the presence of and classify a potential threat within an SG syslog using light-weight anomaly detection and graph theory.

The authors carry out the analysis of the process of modeling threats to information security of automated process control systems. Basic principles of security threats model formation are considered. The approach to protection of automated process control systems based on the Shtakelberg game in a strategic form was modeled. An abstract mathematical model of information security threats to automated process control systems was developed. A formalized representation of a threat model is described, taking into account an intruder's potential. Presentation of the process of applying the described threat model in the form of a continuous Deming-Shewhart cycle is proposed.

With the increasing security threats to the information of Industrial Cyber-physical Control Systems, the quantitative assessment of security risk becomes an important basis of information security research. Based on fuzzy hierarchy analysis, this paper constructs the hierarchical model of industrial control system safety risk evaluation, and obtains the exact value of risk. Experimental results show that the proposed method can effectively quantify the control system risk, which provides a basis for industrial control system risk management decision.

In a computer world, to identify anyone by doing a job or to authenticate by checking their identification and give access to computer. Access Control model comes in to picture when require to grant the permissions to individual and complete the duties. The access control models cannot give complete security when dealing with cloud computing area, where access control model failed to handle the attributes which are requisite to inhibit access based on time and location. When the data outsourced in the cloud, the information holders expect the security and confidentiality for their outsourced data. The data will be encrypted before outsourcing on cloud, still they want control on data in cloud server, where simple encryption is not a complete solution. To irradiate these issues, unlike access control models proposed Attribute Based Encryption standards (ABE). In ABE schemes there are different types like Key Policy-ABE (KP-ABE), Cipher Text-ABE (CP-ABE) and so on. The proposed method applied the access control policy of CP-ABE with Advanced Encryption Standard and used elliptic curve for key generation by using multi stage encryption which divides the users into two domains, public and private domains and shuffling the data base records to protect from inference attacks.

KP-ABE mechanism emerges as one of the most suitable security scheme for asymmetric encryption. It has been widely used to implement access control solutions. However, due to its expensive overhead, it is difficult to consider this cryptographic scheme in resource-limited networks, such as the IoT. As the cloud has become a key infrastructural support for IoT applications, it is interesting to exploit cloud resources to perform heavy operations. In this paper, a collaborative variant of KP-ABE named C-KP-ABE for cloud-based IoT applications is proposed. Our proposal is based on the use of computing power and storage capacities of cloud servers and trusted assistant nodes to run heavy operations. A performance analysis is conducted to show the effectiveness of the proposed solution.

Ciphertext-policy attribute-based encryption (CP-ABE) has been proposed to enable fine-grained access control on encrypted data for cloud storage service. In the context of CP-ABE, since the decryption privilege is shared by multiple users who have the same attributes, it is difficult to identify the original key owner when given an exposed key. This leaves the malicious cloud users a chance to leak their access credentials to outsourced data in clouds for profits without the risk of being caught, which severely damages data security. To address this problem, we add the property of traceability to the conventional CP-ABE. To catch people leaking their access credentials to outsourced data in clouds for profits effectively, in this paper, we first propose two kinds of non-interactive commitments for traitor tracing. Then we present a fully secure traceable CP-ABE system for cloud storage service from the proposed commitment. Our proposed commitments for traitor tracing may be of independent interest, as they are both pairing-friendly and homomorphic. We also provide extensive experimental results to confirm the feasibility and efficiency of the proposed solution.

Ciphertext policy attribute-based encryption (CP-ABE) is a promising technology that offers fine-grained access control over encrypted data. In a CP-ABE scheme, any user can decrypt the ciphertext using his secret key if his attributes satisfy the access policy embedded in the ciphertext. Since the same ciphertext can be decrypted by multiple users with their own keys, the malicious users may intentionally leak their decryption keys for financial profits. So how to trace the malicious users becomes an important issue in a CP-ABE scheme. In addition, from the practical point of view, users may leave the system due to resignation or dismissal. So user revocation is another hot issue that should be solved. In this paper, we propose a practical CP-ABE scheme. On the one hand, our scheme has the properties of traceability and large universe. On the other hand, our scheme can solve the dynamic issue of user revocation. The proposed scheme is proved selectively secure in the standard model.

There is a need for new tools and techniques to aid automotive engineers performing cybersecurity testing on connected car systems. This is in order to support the principle of secure-by-design. Our research has produced a method to construct useful automotive security tooling and tests. It has been used to implement Controller Area Network (CAN) fuzz testing (a dynamic security test) via a prototype CAN fuzzer. The black-box fuzz testing of a laboratory vehicle's display ECU demonstrates the value of a fuzzer in the automotive field, revealing bugs in the ECU software, and weaknesses in the vehicle's systems design.

Nowadays, the growing need to connect modern vehicles through computer networks leads to increased risks of cyberattacks. The internal network, which governs the several electronic components of a vehicle, is becoming increasingly overexposed to external attacks. The Controller Area Network (CAN) protocol, used to interconnect those devices is the key point of the internal network of modern vehicles. Therefore, securing such protocol is crucial to ensure a safe driving experience. However, the CAN is a standard that has undergone little changes since it was introduced in 1983. More precisely, in an attempt to reduce latency, the transfer of information remains unencrypted, which today represents a weak point in the protocol. Hence, the need to protect communications, without introducing low-level alterations, while preserving the performance characteristics of the protocol. In this work, we investigate the possibility of using symmetric encryption algorithms for securing messages exchanged by CAN protocol. In particular, we evaluate the using of lightweight ciphers to secure CAN-level communication. Such ciphers represent a reliable solution on hardware-constrained devices, such as microcontrollers.