Federated Learning (FL) enables multiple users to collaboratively train a machine learning model without sharing raw data, making it suitable for privacy-sensitive applications. However, local model or weight updates can still leak sensitive information. Secure aggregation protocols mitigate this risk by ensuring that only the aggregated updates are revealed. Among these, single-setup secure aggregation protocols, where key generation and exchange occur only once, are the most efficient due to reduced communication and computation overhead. However, existing single-setup secure aggregation protocols often lack support for dynamic user participation and do not provide strong privacy guarantees such as forward and backward secrecy.

Cyber threats have evolved in complexity, aiming at a wide range of sectors using advanced methods and tools. This evolving threat landscape challenges existing cybersecurity frameworks, many of which lack the adaptability to counteract the complex tactics of sophisticated adversaries. Developing robust cyber defense strategies requires simulating dynamic interactions between attackers and defenders across high, moderate, and low-impact scenarios. The Flip-It cyber game serves as an intelligent framework for simulating these interactions, enabling the analysis of adaptive strategies in cybersecurity. This paper aims to address the problem of mitigating malware prevalence in full consideration of attack/defense capabilities in arbitrary network topologies. This paper proposes a sophisticated discrete-time epidemic model to characterize security state transitions over time for all three scenarios within the Flip-It game framework. On this basis, the original problem is modeled as a closed-loop control problem to seek the optimal containment strategy. Deep Reinforcement Learning (DRL) is then used to tackle the problem, generating efficient defense strategies that are well-adapted to changing cybersecurity environments. Numerical simulations based on small-world networks, scale-free networks, and router networks are then carried out to generate corresponding strategies. Additionally, we have evaluated the performance of the proposed method against the State-Of-The-Art (SOTA) in terms of attack/defense objective function, control actions, number of devices under the control of the attacker and defender, stability, execution time, and scalability. This comprehensive approach integrates epidemiological modeling, game theory, and advanced machine learning to effectively tackle the complexities of contemporary cybersecurity threats. • Mitigates malware across low, medium, and high-impact cyberattacks. • Integrates the Flip-It game for attacker-defender dynamic interactions. • Employs DRL to enable adaptive and optimized defense strategies. • Evaluates defense evolution across diverse network topologies.

Realistic network traffic simulation is critical for evaluating intrusion detection systems, stress-testing network protocols, and constructing high-fidelity environments for cybersecurity training. While attack traffic can often be layered into training environments using red-teaming or replay methods, generating authentic benign background traffic remains a core challenge-particularly in simulating the complex temporal and communication dynamics of real-world networks. This paper introduces TempoNet, a novel generative model that combines multi-task learning with multi-mark temporal point processes to jointly model inter-arrival times and all packet- and flow-header fields. TempoNet captures fine-grained timing patterns and higher-order correlations such as host- pair behavior and seasonal trends, addressing key limitations of GAN-, LLM-, and Bayesian-based methods that fail to reproduce structured temporal variation. TempoNet produces temporally consistent, high-fidelity traces, validated on real-world datasets. Furthermore, we show that intrusion detection models trained on TempoNet-generated background traffic perform comparably to those trained on real data, validating its utility for real-world security applications.

This study provides a comprehensive examination of AtomBombing, a stealthy fileless code injection technique that leverages the Windows Global Atom Table for covert payload storage and execution. Unlike traditional injection strategies that rely on memory manipulation or file-based artifacts, AtomBombing avoids direct memory writes and operates entirely through legitimate Windows APIs, making it exceptionally evasive against modern endpoint detection and response tools. Through our proof-of-concept (PoCs) implementations, we demonstrate how adversaries could exploit atom-based payload fragmentation, double layer encoding, and time-based triggers to execute malicious tasks while minimizing forensic visibility. Building on those PoCs, we introduce ABOMB-FOD, a $\mathbf{1. 2 ~ M B}$ multistage loader that survives reboots, bypasses user account controls, and still fits comfortably within the Atom Table capacity limits ($\leq$ 65,535 entries $+\mathbf{2 5 5} \mathbf{B} \approx \mathbf{1 6} \mathbf{~ M B}$). The AtomBombing Process Orchestrator illustrates the ability to securely store encrypted payloads and orchestrate parallel execution using PowerShell, while the Atom Table Backdoor showcases persistent command-and-control behavior activated under specific system conditions. Our findings underscore the inadequacy of current security solutions in monitoring Atom Table interactions, i.e., critical API functions remain largely overlooked in behavioral analysis, despite their potential for stealthy data injection. Consistent with this blind spot, our evaluation confirms that standard defenses, including Windows Defender with cloud protection enabled, fail to detect or flag AtomBombing activity, even when ABOMBFOD or other payloads are executing and network interactions are in progress. We note that a straightforward heuristic, such as flagging any process that issues more than a certain number of GlobalAddAtomW invocations (e.g., 500) within a short time window (e.g., one minute) and stores high-entropy data, can be effective in identifying candidates for further investigation.

Device authentication in Low Earth Orbit (LEO) satellite-based Internet of Things (IoT) networks is critical for enabling secure and reliable communication between remote IoT devices and satellites. It prevents unauthorized access and security breaches. State-of-the-art authentication methods for terrestrial networks, such as Authentication and Key Management for Applications (AKMA), are inadequate when directly applied to such networks because IoT devices have constrained communication and computational capabilities. Further, the satellite environment is highly dynamic, with frequent handovers and variable latency, leading to vulnerabilities like man-in-the-middle (MITM) and spoofing attacks. To address these challenges, we propose a modified AKMA framework for decentralized and continuous authentication in LEO satellite-based IoT networks. Our proposed modification utilizes local key refreshment for seed generation, seed update, and seed refreshment in a decentralized manner, enabling tailored transmission patterns for IoT devices. This reduces the need for repeated authentication attempts with satellites and effectively mitigates handoff-associated threats. We examine the authentication performance of the system in the presence of an illegitimate Unmanned Aerial Vehicle (UAV) above the legitimate IoT devices. Our results through simulations and emulation show improvement in the authentication rate of legitimate IoT devices and a reduction in the misdetection rate of illegitimate UAVs compared to state-of-the-art physical channel-based authentication schemes. Our proposed modified AKMA enables its application in LEO satellite-based IoT networks.

Recent advancements have sparked a growing interest in integrating spatiotemporal analysis with large-scale language models. However, spatiotemporal data often contains sensitive information, making it unsuitable for open third-party access. To address this challenge, we propose a Graph-GAN-based model for generating privacy-protected spatiotemporal data. Our approach incorporates spatial and temporal attention blocks in the discriminator and a spatiotemporal deconvolution structure in the generator. These enhancements enable efficient training under Gaussian noise to achieve differential privacy. Extensive experiments conducted on three real-world spatiotemporal datasets validate the efficacy of our model. Our method provides a privacy guarantee while maintaining the data utility. The prediction model trained on our generated data maintains a competitive performance compared to the model trained on the original data.

Detecting lateral movement threats in large-scale system logs is a critical challenge due to the scarcity of labelled attack data, the presence of imbalanced datasets, and the sophisticated nature of modern adversaries. To address these issues, we propose ConceptUML , a semantic-driven, fully unsupervised threat detection framework designed to automatically identify anomalies related to lateral movement in heterogeneous log data. ConceptUML is structured around a three-phase architecture. In Phase 1 (Latent Semantic Learning) , contextualized embeddings generated by Sentence-BERT are combined with Non-negative Matrix Factorization to extract abstract concepts from system logs and external threat intelligence sources such as MITRE ATT&CK and CAPEC. In Phase 2 (Unsupervised Threat Detection) , a Hidden Markov Model is applied to cluster logs based on learned concepts, and each cluster is scored according to its semantic similarity to known adversarial techniques. Phase 3 (Decision Refinement) uses topic modelling to further isolate malicious event log subsets from within suspicious clusters, enabling high-precision triage. We evaluate ConceptUML using four real-world event log datasets, including Windows Event Logs and multiple subsets of the LMD-23 dataset, encompassing attacks such as exploitation of hashing techniques and remote services. The enhanced model with topic modelling achieves up to 92.54% detection quality and reduces detection error to as low as 8.14%, outperforming several baseline approaches including AutoEncoder, LogAnomaly, LOF, and DBScan. Our results confirm that ConceptUML delivers interpretable, scalable, and highly effective detection of lateral movement threats without requiring labelled training data or extensive manual feature engineering.

The escalating prevalence of Advanced Persistent Threats (APTs) necessitates the development of more robust solutions capable of effectively thwarting these attacks by monitoring system activities across individual hosts. Existing cloud-native security applications utilize a combination of rule-based and machine learning-based detection techniques to protect digital assets . However, these approaches have limitations. Rule-based detection depends on predefined rules to identify specific attack patterns. Persistent attackers can often evade detection by carefully ensuring that their behavior circumvents these rules. In contrast, machine learning-based detection techniques, which learn attack patterns from data, rely heavily on the availability of labeled data for training. However, labeled data is often unavailable and can be labor-intensive and costly to obtain. In this paper, we address the challenge of detecting APT attacks more holistically by leveraging attackers’ behavior during communication with Command and Control (C2) servers, a critical phase observed in most APT attacks. We aim to reduce false positive alerts for threat hunters by analyzing system network logs to detect potential network beaconing, a common attribute of various malware . We introduce a novel hybrid approach, called NetSpectra Sentinel , which employs a Continuous Time Hidden Markov Model (CT-HMM) to detect hidden states underlying observed patterns within the network logs and Time Series Decomposition (TSD) to model temporal patterns. We evaluate the effectiveness of our approach using 14 benchmark datasets and one synthetic dataset , comparing our method with other state-of-the-art statistical-based and botnet detection techniques. The results demonstrate that our technique achieves significantly higher accuracy in most cases, and even when existing techniques fail, our approach can still detect beaconing post-initial compromise with up to 90% accuracy. Additionally, we achieve up to four times better performance in terms of precision compared to existing statistical-based techniques.

Protecting sensitive information on data streams is a pivotal challenge for modern systems. Current approaches to providing privacy in data streams can be broadly categorized into two strategies. The first strategy involves transforming the stream into a private sequence of values, enabling the subsequent use of non-private methods of analysis. While effective, this approach incurs high memory costs, often proportional to the size of the database. Alternatively, a compact data structure can be used to provide a private summary of the stream. However, these data structures are limited to predefined queries, restricting their flexibility. To overcome these limitations, we propose a lightweight synthetic data generator, PrivHP, that provides differential privacy guarantees. PrivHP is based on a novel method for the private hierarchical decomposition of the input domain in bounded memory. As the decomposition approximates the cumulative distribution function of the input, it serves as a lightweight structure for synthetic data generation. PrivHP is the first method to provide a principled trade-off between accuracy and space for private hierarchical decompositions. It achieves this by balancing hierarchy depth, noise addition, and selective pruning of low-frequency subdomains while preserving high-frequency ones, all identified in a privacy-preserving manner. To ensure memory efficiency, we employ private sketches to estimate subdomain frequencies without accessing the entire dataset. Central to our approach is the introduction of a pruning parameter k , which enables an almost smooth interpolation between space usage and utility, and a measure of skew tail k , which is a vector of subdomain frequencies containing all but the largest k coordinates. PrivHP processes a dataset X using M = O (k log 2 | X |)) space and, on input domain Ω = [0,1] d , while maintaining ε-differential privacy, produces a synthetic data generator that is at distance O ( M (1-1/d) /ε n + ||tail k ( X )|| 1 /M 1/d n ) from the empirical distribution in the expected Wasserstein metric. Compared to the state-of-the-art, PMM, which achieves accuracy O ((ε n) -1/d ) with memory O (ε n), our method introduces an additional approximation error term of O (||tail k ( X )|| 1 /(M 1/d n)), but operates in significantly reduced space. Additionally, we provide interpretable utility bounds that account for all error sources, including those introduced by the fixed hierarchy depth, privacy noise, hierarchy pruning, and frequency approximations.

AI-enabled critical infrastructures (ACIs) integrate artificial intelligence (AI) technologies into various essential systems and services that are vital to the functioning of society, offering significant implications for efficiency, security and resilience. While adopting decentralized AI approaches (such as federated learning technology) in ACIs is plausible, private and sensitive data are still susceptible to data reconstruction attacks through gradient optimization. In this work, we propose Compressed Differentially Private Aggregation (CDPA), a leakage-resilient, communication-efficient, and carbon-neutral approach for ACI networks. Specifically, CDPA has introduced a novel random bit-flipping mechanism as its primary innovation. This mechanism first converts gradients into a specific binary representation and then selectively flips masked bits with a certain probability. The proposed bit-flipping introduces a larger variance to the noise while providing differentially private protection and commendable efforts in energy savings while applying vector quantization techniques within the context of federated learning. The experimental evaluation indicates that CDPA can reduce communication cost by half while preserving model utility. Moreover, we demonstrate that CDPA can effectively defend against state-of-the-art data reconstruction attacks in both computer vision and natural language processing tasks. We highlight existing benchmarks that generate 2.6x to over 100x more carbon emissions than CDPA. We hope that the CDPA developed in this paper can inform the federated AI-enabled critical infrastructure of a more balanced trade-off between utility and privacy, resilience protection, as well as a better carbon offset with less communication overhead.

Training Deep Neural Networks (DNNs) can be expensive when data is difficult to obtain or labeling them requires significant domain expertise. Hence, it is crucial that the Intellectual Property (IP) of DNNs trained on valuable data be protected against IP infringement. DNN fingerprinting and watermarking are two lines of work in DNN IP protection. Recently proposed DNN fingerprinting techniques are able to detect IP infringement while preserving model performance by relying on the key assumption that the decision boundaries of independently trained models are intrinsically different from one another. In contrast, DNN watermarking embeds a watermark in a model and verifies IP infringement if an identical or similar watermark is extracted from a suspect model. The techniques deployed in fingerprinting and watermarking vary significantly because their underlying mechanisms are different. From an adversary's perspective, a successful IP removal attack should defeat both fingerprinting and watermarking. However, to the best of our knowledge, there is no work on such attacks in the literature yet. In this paper, we fill this gap by presenting an IP removal attack that can defeat both fingerprinting and watermarking. We consider the challenging data-free scenario whereby all data is inverted from the victim model. Under this setting, a stolen model only depends on the victim model. Experimental results demonstrate the success of our attack in defeating state-of-the-art DNN fingerprinting and watermarking techniques. This work reveals a novel attack surface that exploits generative model inversion attacks to bypass DNN IP defenses. This threat must be addressed by future defenses for reliable IP protection.

Deep learning system components are vulnerable to backdoor attacks. Detectors are no exception. Detectors, in contrast to classifiers, possess unique characteristics, architecturally and in task execution; often operating in challenging conditions, for instance, detecting traffic signs in autonomous cars. But, our knowledge dominates attacks against classifiers and tests in the "digital domain".To address this critical gap, we conducted an extensive empirical study targeting multiple detector architectures and two challenging detection tasks in real-world settings: traffic signs and vehicles. Using diverse, methodically collected videos captured from driving cars and flying drones, incorporating physical object trigger deployments in authentic scenes, we investigated the viability of physical object-triggered backdoor attacks in application settings.Our findings revealed 7 key insights. Importantly, the prevalent "digital" data poisoning method for injecting backdoors into models does not lead to effective attacks against detectors in the real world, although proven effective in classification tasks. We construct a new, cost-efficient attack method, dubbed Morphing, incorporating the unique nature of detection tasks; ours is remarkably successful in injecting physical object-triggered backdoors, even capable of poisoning triggers with clean label annotations or invisible triggers without diminishing the success of physical object triggered backdoors. We discovered that the defenses curated are ill-equipped to safeguard detectors against such attacks. To underscore the severity of the threat and foster further research, we, for the first time, release an extensive video test set of real-world backdoor attacks. Our study not only establishes the credibility and seriousness of this threat but also serves as a clarion call to the research community to advance backdoor defenses in the context of object detection. Our dataset—DriveByFlyBy—release, demo videos and code is at https://BackdoorDetectors.github.io.

Vision Transformers (ViTs) have emerged as a compelling alternative to Convolutional Neural Networks (CNNs) in the realm of computer vision, showcasing tremendous potential. However, recent research has un-veiled a susceptibility of ViTs to adversarial attacks, akin to their CNN counterparts. Adversarial training and randomization are two representative effective defenses for CNNs. Some researchers have attempted to apply adversarial training to ViTs and achieved comparable robustness to CNNs, while it is not easy to directly apply randomization to ViTs because of the architecture difference between CNNs and ViTs. In this paper, we delve into the structural intricacies of ViTs and propose a novel defense mechanism termed Random entangled image Transformer (ReiT), which seamlessly integrates adversarial training and randomization to bolster the adversarial robustness of ViTs. Recognizing the challenge posed by the structural disparities between ViTs and CNNs, we introduce a novel module, input-independent random entangled self-attention (II-ReSA). This module op-timizes random entangled tokens that lead to “dissimilar” self-attention outputs by leveraging model parameters and the sampled random tokens, thereby synthesizing the self-attention module outputs and random entangled tokens to diminish adversarial similarity. ReiT incorporates two distinct random entangled tokens and employs dual randomization, offering an effective countermeasure against adversarial examples while ensuring comprehensive deduction guarantees. Through extensive experiments conducted on various ViT variants and benchmarks, we substantiate the superiority of our proposed method in enhancing the adversarial robustness of Vision Transformers.

Large machine-learning training datasets can be distilled into small collections of informative synthetic data samples. These synthetic sets support efficient model learning and reduce the communication cost of data sharing. Thus, high-fidelity distilled data can support the efficient deployment of machine learning applications in distributed network environments. A naive way to construct a synthetic set in a distributed environment is to allow each client to perform local data distillation and to merge local distillations at a central server. However, the quality of the resulting set is impaired by heterogeneity in the distributions of the local data held by clients. To overcome this challenge, we introduce the first collaborative data distillation technique, called CollabDM, which captures the global distribution of the data and requires only a single round of communication between client and server. Our method outperforms the state-of-the-art one-shot learning method on skewed data in distributed learning environments. We also show the promising practical benefits of our method when applied to attack detection in 5G networks.

Abstract Named Data Networking (NDN) is an emerging network architecture. An important characteristic of NDN is its in-network cache, which enables Data packets to be available from multiple locations on the Internet. Hence the enforcement of access control mechanisms becomes even more critical in the NDN. This paper proposes a novel access control scheme referred to as Role-Based Encryption for NDN (NDN-RBE), which uses a broadcast encryption mechanism to achieve secure data access control. Our scheme uses the role inheritance property of the traditional Role-Based Access Control (RBAC) model to achieve efficient data access control over hierarchical content. This makes our scheme particularly suitable for large-scale real-world content-centric services like Netflix. Our scheme also supports additional design features such as anonymous signature-based authentication, batch signature verification and two types of privilege revocations. In addition, our formal security analysis demonstrates that our scheme is provably secure against Chosen Plaintext Attacks. Our performance and functionality comparison show that our scheme outperforms other notable existing works in terms of security, functionality, computation, communication and storage overhead. Furthermore, our experimental results show an improvement in content delivery time of the order of 15 percent compared with the other closely related works.

The digitisation of agriculture, integral to Agriculture 4.0, has brought significant benefits while simultaneously escalating cybersecurity risks. With the rapid adoption of smart farming technologies and infrastructure, the agricultural sector has become an attractive target for cyberattacks. This paper presents a systematic literature review that assesses the applicability of existing cyber threat intelligence (CTI) techniques within smart farming infrastructures (SFIs). We develop a comprehensive taxonomy of CTI techniques and sources, specifically tailored to the SFI context, addressing the unique cyber threat challenges in this domain. A crucial finding of our review is the identified need for a virtual Chief Information Security Officer (vCISO) in smart agriculture. While the concept of a vCISO is not yet established in the agricultural sector, our study highlights its potential significance. The implementation of a vCISO could play a pivotal role in enhancing cybersecurity measures by offering strategic guidance, developing robust security protocols, and facilitating real-time threat analysis and response strategies. This approach is critical for safeguarding the food supply chain against the evolving landscape of cyber threats. Our research underscores the importance of integrating a vCISO framework into smart farming practices as a vital step towards strengthening cybersecurity. This is essential for protecting the agriculture sector in the era of digital transformation, ensuring the resilience and sustainability of the food supply chain against emerging cyber risks.

Physical-layer authentication is a popular alternative to the conventional key-based authentication for Internet of Things (IoT) devices due to their limited computational capacity and battery power. However, this approach has limitations due to poor robustness under channel fluctuations, reconciliation overhead, and no clear safeguard distance to ensure the secrecy of the generated authentication keys. In this regard, we propose a novel, secure, and lightweight continuous authentication scheme for IoT device authentication. Our scheme utilizes the inherent properties of the IoT devices’ transmission model as its source for seed generation and device authentication. Specifically, our proposed scheme provides continuous authentication by checking the access time slots and spreading sequences of the IoT devices instead of repeatedly generating and verifying shared keys. Due to this, access to a coherent key is not required in our proposed scheme, resulting in the concealment of the seed information from attackers. Our proposed authentication scheme for IoT devices demonstrates improved performance compared to the benchmark schemes relying on physical channels. Our empirical results find a near threefold decrease in the misdetection rate of illegitimate devices and close to zero false alarm rate in various system settings with varied numbers of active devices up to 200 and signal-to-noise ratio from 0 to 25 dB. Our proposed authentication scheme also has a lower computational complexity of at least half the computational cost of the benchmark schemes based on support vector machine and binary hypothesis testing in our studies. This further corroborates the practicality of our scheme for IoT deployments.

Understanding the attributes of critical data and implementing suitable security measures help organisations bolster their data-protection strategies and diminish the potential impacts of ransomware incidents. Unauthorised extraction and acquisition of data are the principal objectives of most cyber invasions. We underscore the severity of this issue using a recent attack by the Clop ransomware group, which exploited the MOVEit Transfer vulnerability and bypassed network-detection mechanisms to exfiltrate data via a Command and Control server. As a countermeasure, we propose a method called Buffer-Based Signature Verification (BBSV). This approach involves embedding 32-byte tags into files prior to their storage in the cloud, thus offering enhanced data protection. The BBSV method can be integrated into software like MOVEit Secure Managed File Transfer, thereby thwarting attempts by ransomware to exfiltrate data. Empirically tested using a BBSV prototype, our approach was able to successfully halt the encryption process for 80 ransomware instances from 70 ransomware families. BBSV not only stops the encryption but also prevents data exfiltration when data are moved or written from the original location by adversaries. We further develop a hypothetical exploit scenario in which an adversary manages to bypass the BBSV, illicitly transmits data to a Command and Control server, and then removes files from the original location. We construct an extended state space, in which each state represents a tuple that integrates user authentication and system components at the filesystem level.

In the rapidly changing cybersecurity landscape, threat hunting has become a critical proactive defense against sophisticated cyber threats. While traditional security measures are essential, their reactive nature often falls short in countering malicious actors’ increasingly advanced tactics. This paper explores the crucial role of threat hunting, a systematic, analyst-driven process aimed at uncovering hidden threats lurking within an organization's digital infrastructure before they escalate into major incidents. Despite its importance, the cybersecurity community grapples with several challenges, including the lack of standardized methodologies, the need for specialized expertise, and the integration of cutting-edge technologies like artificial intelligence (AI) for predictive threat identification. To tackle these challenges, this survey paper offers a comprehensive overview of current threat hunting practices, emphasizing the integration of AI-driven models for proactive threat prediction. Our research explores critical questions regarding the effectiveness of various threat hunting processes and the incorporation of advanced techniques such as augmented methodologies and machine learning. Our approach involves a systematic review of existing practices, including frameworks from industry leaders like IBM and CrowdStrike. We also explore resources for intelligence ontologies and automation tools. The background section clarifies the distinction between threat hunting and anomaly detection, emphasizing systematic processes crucial for effective threat hunting. We formulate hypotheses based on hidden states and observations, examine the interplay between anomaly detection and threat hunting, and introduce iterative detection methodologies and playbooks for enhanced threat detection. Our review encompasses supervised and unsupervised machine learning approaches, reasoning techniques, graph-based and rule-based methods, as well as other innovative strategies. We identify key challenges in the field, including the scarcity of labeled data, imbalanced datasets, the need for integrating multiple data sources, the rapid evolution of adversarial techniques, and the limited availability of human expertise and data intelligence. The discussion highlights the transformative impact of artificial intelligence on both threat hunting and cybercrime, reinforcing the importance of robust hypothesis development. This paper contributes a detailed analysis of the current state and future directions of threat hunting, offering actionable insights for researchers and practitioners to enhance threat detection and mitigation strategies in the ever-evolving cybersecurity landscape.

Vehicle platooning systems are vulnerable to malicious attacks that exploit vehicle-to-vehicle (V2V) communication, causing potential instability and increased collision risks. Conventional machine learning (ML) detection methods show promise but can be circumvented by intelligent adversaries. In this paper, we present a novel, end-to-end attack detection and mitigation approach that uniquely incorporates advancements in (adversarial) machine learning, control theory, and game theory. We employ a non-cooperative security game with imperfect information to model complex attack/defense interactions. This aids in making informed decisions regarding detector deployment and attack mitigation, even amidst possibly misleading attack detection reports. We model our control system reconfiguration attack mitigation approach as a switched system and provide a n in-depth stability analysis. The simulations conducted in a sophisticated simulator demonstrate our approach’s potential for real-world online deployment. Our game-based defense formulation significantly improves inter-vehicle distance and defense utilities against both cyber-physical and adversarially-masked attacks while reducing the distance disturbance caused by the ambient traffic by up to 87% compared to baseline defense approaches.

Deep neural networks (DNNs) have achieved state-of-the-art performance on face recognition (FR) tasks in the last decade. In real scenarios, the deployment of DNNs requires taking various face accessories into consideration, like glasses, hats, and masks. In the COVID-19 pandemic era, wearing face masks is one of the most effective ways to defend against the novel coronavirus. However, DNNs are known to be vulnerable to adversarial examples with a small but elaborated perturbation. Thus, a facial mask with adversarial perturbations may pose a great threat to the widely used deep learning-based FR models. In this paper, we consider a challenging adversarial setting: targeted attack against FR models. We propose a new stealthy physical masked FR attack via adversarial style optimization. Specifically, we train an adversarial style mask generator that hides adversarial perturbations inside style masks. Moreover, to ameliorate the phenomenon of sub-optimization with one fixed style, we propose to discover the optimal style given a target through style optimization in a continuous relaxation manner. We simultaneously optimize the generator and the style selection for generating strong and stealthy adversarial style masks. We evaluated the effectiveness and transferability of our proposed method via extensive white-box and black-box digital experiments. Furthermore, we also conducted physical attack experiments against local FR models and online platforms.

We present a new algorithm to train a robust malware detector. Malware is a prolific problem and malware detectors are a front-line defense. Modern detectors rely on machine learning algorithms. Now, the adversarial objective is to devise alterations to the malware code to decrease the chance of being detected whilst preserving the functionality and realism of the malware. Adversarial learning is effective in improving robustness but generating functional and realistic adversarial malware samples is non-trivial. Because: i) in contrast to tasks capable of using gradient-based feedback, adversarial learning in a domain without a differentiable mapping function from the problem space (malware code inputs) to the feature space is hard; and ii) it is difficult to ensure the adversarial malware is realistic and functional. This presents a challenge for developing scalable adversarial machine learning algorithms for large datasets at a production or commercial scale to realize robust malware detectors. We propose an alternative; perform adversarial learning in the feature space in contrast to the problem space. We prove the projection of perturbed, yet valid malware, in the problem space into feature space will always be a subset of adversarials generated in the feature space. Hence, by generating a robust network against feature-space adversarial examples, we inherently achieve robustness against problem-space adversarial examples. We formulate a Bayesian adversarial learning objective that captures the distribution of models for improved robustness. To explain the robustness of the Bayesian adversarial learning algorithm, we prove that our learning method bounds the difference between the adversarial risk and empirical risk and improves robustness. We show that Bayesian neural networks (BNNs) achieve state-of-the-art results; especially in the False Positive Rate (FPR) regime. Adversarially trained BNNs achieve state-of-the-art robustness. Notably, adversarially trained BNNs are robust against stronger attacks with larger attack budgets by a margin of up to 15% on a recent production-scale malware dataset of more than 20 million samples. Importantly, our efforts create a benchmark for future defenses in the malware domain.

Moving Target Defenses (MTD) are proactive security countermeasures that change the attack surface in a system in ways that make it harder for attackers to succeed. These techniques have been shown to be effective, and their application in software-defined networking (SDN) against simple automated attacks is growing in popularity. However, with the increased knowledge of and ease of access to Artificial Intelligence (AI) techniques, AI is starting to be used to enhance cyber attacks, which are becoming increasingly complex. Hence, the evaluation of MTDs against simple automated attacks is no longer enough to demonstrate their effectiveness in increasing system security.

Radio frequency fingerprint identification (RFFI) is a lightweight device authentication technique particularly desirable for power-constrained devices, e.g., the Internet of things (IoT) devices. Similar to biometric fingerprinting, RFFI exploits the intrinsic and unique hardware impairments resulting from manufacturing, such as power amplifier (PA) nonlinearity, to develop methods for device detection and classification. Due to the nature of wireless transmission, received signals are volatile when communication environments change. The resulting radio frequency fingerprints (RFFs) are distorted, leading to low device detection and classification accuracy. We propose a PA nonlinearity quotient and transfer learning classifier to design the environment-robust RFFI method. Firstly, we formalized and demonstrated that the PA nonlinearity quotient is independent of environmental changes. Secondly, we implemented transfer learning on a base classifier generated by data collected in an anechoic chamber, further improving device authentication and reducing disk and memory storage requirements. Extensive experiments, including indoor and outdoor settings, were carried out using LoRa devices. It is corroborated that the proposed PA nonlinearity quotient and transfer learning classifier significantly improved device detection and device classification accuracy. For example, the classification accuracy was improved by 33.3% and 34.5% under indoor and outdoor settings, respectively, compared to conventional deep learning and spectrogram-based classifiers.

Maintaining data confidentiality at the asymmetric-resource devices across emerging technologies needs varying cryptographic algorithms. Quantum computing makes preserving data confidentiality across asymmetric infrastructure more difficult. However, exploiting the architecture of classical cryptographic schemes to integrate the post-quantum constructs could be used to maintain post-quantum level confidentiality over the Internet. This paper presents a post-quantum secure classical ABE-IBE proxy re-encryption scheme (L ABE-IBE PRE) that utilizes the classical ABE-IBE proxy re-encryption capabilities at the end nodes in a system and raises the data confidentiality to post-quantum secure level over the Internet. The proposed L ABE-IBE PRE transforms a ciphertext of the classical ABE scheme to a post-quantum secure ciphertext and from a post-quantum secure ciphertext to a ciphertext of the classical IBE scheme. We compare our proposed L ABE-IBE PRE scheme with classical ABE-IBE proxy re-encryption schemes, including Encryption Switching ABE-IBE (ES.ABE-IBE) scheme (He et al., 2019). We discuss the security and efficiency of our proposed scheme.

Numerous open-source and commercial malware detectors are available. However, their efficacy is threatened by new adversarial attacks, whereby malware attempts to evade detection, e.g., by performing feature-space manipulation. In this work, we propose an explainability-guided and model-agnostic testing framework for robustness of malware detectors when confronted with adversarial attacks. The framework introduces the concept of Accrued Malicious Magnitude (AMM) to identify which malware features could be manipulated to maximize the likelihood of evading detection. We then use this framework to test several state-of-the-art malware detectors' ability to detect manipulated malware. We find that (i) commercial antivirus engines are vulnerable to AMM-guided test cases; (ii) the ability of a manipulated malware generated using one detector to evade detection by another detector (i.e., transferability) depends on the overlap of features with large AMM values between the different detectors; and (iii) AMM values effectively measure the fragility of features (i.e., capability of feature-space manipulation to flip the prediction results) and explain the robustness of malware detectors facing evasion attacks. Our findings shed light on the limitations of current malware detectors, as well as how they can be improved.

Existing integrity verification approaches for deep models are designed for private verification (i.e., assuming the service provider is honest, with white-box access to model parameters). However, private verification approaches do not allow model users to verify the model at run-time. Instead, they must trust the service provider, who may tamper with the verification results. In contrast, a public verification approach that considers the possibility of dishonest service providers can benefit a wider range of users. In this paper, we propose PublicCheck, a practical public integrity verification solution for services of run-time deep models. PublicCheck considers dishonest service providers, and overcomes public verification challenges of being lightweight, providing anti-counterfeiting protection, and having fingerprinting samples that appear smooth. To capture and fingerprint the inherent prediction behaviors of a run-time model, PublicCheck generates smoothly transformed and augmented encysted samples that are enclosed around the model's decision boundary while ensuring that the verification queries are indistinguishable from normal queries. PublicCheck is also applicable when knowledge of the target model is limited (e.g., with no knowledge of gradients or model parameters). A thorough evaluation of PublicCheck demonstrates the strong capability for model integrity breach detection (100% detection accuracy with less than 10 black-box API queries) against various model integrity attacks and model compression attacks. PublicCheck also demonstrates the smooth appearance, feasibility, and efficiency of generating a plethora of encysted samples for fingerprinting.

The digital age has brought a world of opportunity to children. Connectivity can be a game-changer for some of the world’s most marginalized children. However, while legislatures around the world have enacted regulations to protect children’s online privacy, and app stores have instituted various protections, privacy in mobile apps remains a growing concern for parents and wider society. In this paper, we explore the potential privacy issues and threats that exist in these apps. We investigate 20195 mobile apps from the Google Play store that are designed particularly for children (Family apps) or include children in their target user groups (Normal apps). Using both static and dynamic analysis, we find that 4.47% of Family apps request location permissions, even though collecting location information from children is forbidden by the Play store, and 81.25% of Family apps use trackers (which are not allowed in children’s apps). Even major developers with 40+ kids apps on the Play store use ad trackers. Furthermore, we find that most permission request notifications are not well designed for children, and 19.25% apps have inconsistent content age ratings across the different protection authorities. Our findings suggest that, despite significant attention to children’s privacy, a large gap between regulatory provisions, app store policies, and actual development practices exist. Our research sheds light for government policymakers, app stores, and developers.

The amount of cybersecurity-related information is extraordinarily increasing, given the fast-growing number of cybersecurity attacks and the significant influence brought by them. How to efficiently obtain and precisely understand the relevant knowledge in the sea of information on cybersecurity becomes a challenge. In this article, we propose an innovative cybersecurity retrieval scheme that supports automatic indexing and searching of cybersecurity information based on semantic contents and hidden metadata. The proposed scheme leverages a customized neural model that incorporates new linguistic features and word embedding by identifying the entities related to cybersecurity incidents from the text. We implement a novel cybersecurity search engine to demonstrate effective, understandable and pragmatic cybersecurity information retrieval based on the proposed schema. Comprehensive performance evaluation over real-world datasets has been conducted to validate the new algorithms and techniques developed for cybersecurity information retrieval. The new engine makes it possible to conduct augmented search, cybersecurity analytics, and visualization, with the ultimate goal of providing direct and efficient results to help people obtain and truly understand cybersecurity information.

Federated learning (FL) and split learning (SL) are two popular distributed machine learning approaches. Both follow a model-to-data scenario; clients train and test machine learning models without sharing raw data. SL provides better model privacy than FL due to the machine learning model architecture split between clients and the server. Moreover, the split model makes SL a better option for resource-constrained environments. However, SL performs slower than FL due to the relay-based training across multiple clients. In this regard, this paper presents a novel approach, named splitfed learning (SFL), that amalgamates the two approaches eliminating their inherent drawbacks, along with a refined architectural configuration incorporating differential privacy and PixelDP to enhance data privacy and model robustness. Our analysis and empirical results demonstrate that (pure) SFL provides similar test accuracy and communication efficiency as SL while significantly decreasing its computation time per global epoch than in SL for multiple clients. Furthermore, as in SL, its communication efficiency over FL improves with the number of clients. Besides, the performance of SFL with privacy and robustness measures is further evaluated under extended experimental settings.

The large transformer-based language models demonstrate excellent performance in natural language processing. By considering the transferability of the knowledge gained by these models in one domain to other related domains, and the closeness of natural languages to high-level programming languages, such as C/C++, this work studies how to leverage (large) transformer-based language models in detecting software vulnerabilities and how good are these models for vulnerability detection tasks. In this regard, firstly, we present a systematic (cohesive) framework that details source code translation, model preparation, and inference. Then, we perform an empirical analysis of software vulnerability datasets of C/C++ source codes having multiple vulnerabilities corresponding to the library function call, pointer usage, array usage, and arithmetic expression. Our empirical results demonstrate the good performance of the language models in vulnerability detection. Moreover, these language models have better performance metrics, such as F1-score, than the contemporary models, namely bidirectional long short term memory and bidirectional gated recurrent unit. Experimenting with the language models is always challenging due to the requirement of computing resources, platforms, libraries, and dependencies. Thus, this paper also analyses the popular platforms to efficiently fine-tune these models and present recommendations while choosing the platforms for our framework.

Training machine learning algorithms on sensitive, illegal to possess, and psychologically harmful data is challenging because researchers have to do training without handling the data. Moreover, the nature of the data imposes strict control, monitoring, and examination of all the activities involved, including communication, execution, and release of algorithms, datasets, outputs, and results. In this regard, this work proposes a new multi-zoned framework called MaLFraDA. MaLFraDA has soft air gaps between its zones to isolate and control communication in and out of the framework. Besides, it includes (i) a vetter to investigate and approve incoming model/algorithm, and outgoing information, (ii) encrypted data vaults, and (iii) airlock instances for secure execution/computation. MaLFraDA, with an extension, runs popular distributed machine learning algorithms such as federated and split learning using multiple data custodians.

Graph Neural Networks (GNNs) have achieved tremendous success in many graph mining tasks benefitting from the message passing strategy that fuses the local structure and node features for better graph representation learning. Despite the success of GNNs, and similar to other types of deep neural networks, GNNs are found to be vulnerable to unnoticeable perturbations on both graph structure and node features. Many adversarial attacks have been proposed to disclose the fragility of GNNs under different perturbation strategies to create adversarial examples. However, vulnerability of GNNs to successful backdoor attacks was only shown recently.

The Android system manages access to sensitive APIs by permission enforcement. An application (app) must declare proper permissions before invoking specific Android APIs. However, there is no official documentation providing the complete list of permission-protected APIs and the corresponding permissions to date. Researchers have spent significant efforts extracting such API protection mapping from the Android API framework, which leverages static code analysis to determine if specific permissions are required before accessing an API. Nevertheless, none of them has attempted to analyze the protection mapping in the native library (i.e., code written in C and C++), an essential component of the Android framework that handles communication with the lower-level hardware, such as cameras and sensors. While the protection mapping can be utilized to detect various security vulnerabilities in Android apps, such as permission over-privilege, imprecise mapping will lead to false results in detecting such security vulnerabilities. To fill this gap, we thereby propose to construct the protection mapping involved in the native libraries of the Android framework to present a complete and accurate specification of Android API protection. We develop a prototype system, named NatiDroid, to facilitate the cross-language static analysis and compare its performance with two state-of-the-practice tools, termed Axplorer and Arcade. We evaluate NatiDroid on more than 11,000 Android apps, including system apps from custom Android ROMs and third-party apps from the Google Play. Our NatiDroid can identify up to 464 new API-permission mappings, in contrast to the worst-case results derived from both Axplorer and Arcade, where approximately 71% apps have at least one false positive in permission over-privilege. We have disclosed all the potential vulnerabilities detected to the stakeholders.

This work applies Game Theory to developing secure software. With the perspective of Game Theory, one can see secure software development as a game between software developers and software security engineers, who play this game repeatedly in processes such as agile Software Development Life Cycle (SDLC). The problem we observe is that there can be conflicts between these two players regarding who should find and fix certain software vulnerabilities. To solve this problem, our approach uses Mechanism Design in Game Theory to design games that enforce cooperation between these two players. In doing so, we identify the source of the conflicts between them by looking at the components of the software. These components may be the methods or functions in the software, or individual modules, or similar building blocks. The novelty of our work is that our mechanism constructs a game which allocates software components between these two players such that they work cooperatively while trying to maximize their own payoffs.

Federated learning (FL) and split learning (SL) are state-of-the-art distributed machine learning techniques to enable machine learning training without accessing raw data on clients or end devices. However, their comparative training performance under real-world resource-restricted Internet of Things (IoT) device settings remains barely studied. This work provides empirical comparisons of FL and SL in real-world IoT settings regarding (i) learning performance with heterogeneous data distributions and (ii) on-device execution overhead. Our analyses in this work demonstrate that the learning performance of SL is better than FL under an imbalanced data distribution but worse than FL under an extreme non-IID data distribution. Recently, FL and SL are combined to form splitfed learning (SFL) to leverage each of their benefits (e.g., parallel training of FL and lightweight on-device computation requirement of SL). Our work considers FL, SL, and SFL, and mounts them on Raspberry Pi devices to evaluate their performance, including training time, communication overhead, power consumption, and memory usage with resource-restricted IoT devices. Besides evaluations, we apply two optimizations. First, we generalize SFL by carefully examining the possibility of a hybrid type of model training at the server-side. The generalized SFL merges sequential (dependent) and parallel (independent) processes of model training and thus is beneficial to a system with a large scale of IoT devices, specifically at the server-side operations. Second, we propose pragmatic techniques to substantially reduce the communication overhead by up to four times for the SL and (generalized) SFL.

Machine learning (ML) has been widely used for malware detection on different operating systems, including Android. To keep up with malware's evolution, the detection models usually need to be retrained periodically (e.g., every month) based on the data collected in the wild. However, this leads to poisoning attacks, specifically backdoor attacks, which subvert the learning process and create evasion ‘tunnels’ for manipulated malware samples. To date, we have not found any prior research that explored this critical problem in Android malware detectors. Although there are already some similar works in the image classification field, most of those similar ideas cannot be borrowed to solve this problem, because the assumption that the attacker has full control of the training data collection or labelling process is not realistic in real-world malware detection scenarios. In this article, we are motivated to study the backdoor attack against Android malware detectors. The backdoor is created and injected into the model stealthily without access to the training data and activated when an app with the trigger is presented. We demonstrate the proposed attack on four typical malware detectors that have been widely discussed in academia. Our evaluation shows that the proposed backdoor attack achieves up to 99 percent evasion rate over 750 malware samples. Moreover, the above successful attack is realised by a small size of triggers (only four features) and a very low data poisoning rate (0.3 percent).

Coverage-guided fuzzing is one of the most effective solutions for vulnerability discovery. Among coverage-guided fuzzing, full-speed fuzzing, such as UnTracer, traces test cases only when they discover new coverage. Due to the high expense of tracing test cases, full-speed fuzzers improve the efficiency of fuzzing by tracing only coverage-increasing test cases. However, the existing full-speed fuzzer (i.e., UnTracer) is based on basic block coverage, suffering a severe problem called edge collision. Moreover, such fuzzers neglect the path frequency, which affects fuzzing effectiveness. In this article, we propose CSI-Fuzz, a fuzzer utilizing coverage sensitive instrumentation to address the problems of existing full-speed fuzzing. CSI-Fuzz directly instruments at edges, which solves the problem of edge collision. Meanwhile, CSI-Fuzz sets path identifiers to count the frequency of covered paths. Our CSI-Fuzz can be recognized as an add-on and seamlessly applied to existing coverage-guided fuzzers. We accordingly implement CSI-Fuzz based on two widely-adopted fuzzers, AFL and AFLFast, to evaluate its performance. The experiments demonstrate that CSI-Fuzz discovers more edges than AFL, AFLFast, and UnTracer. Additionally, CSI-Fuzz exposes more bugs than the other fuzzers.

The amount of encrypted Internet traffic almost doubles every year thanks to the wide adoption of end-to-end traffic encryption solutions such as IPSec, TLS and SSH. Despite all the benefits of user privacy the end-to-end encryption provides, the encrypted internet traffic blinds intrusion detection system (IDS) and makes detecting malicious traffic hugely difficult. The resulting conflict between the user's privacy and security has demanded solutions for deep packet inspection (DPI) over encrypted traffic. The approach of those solutions proposed to date is still restricted in that they require intensive computations during connection setup or detection. For example, BlindBox, introduced by Sherry et al. (SIGCOMM 2015) enables inspection over the TLS-encrypted traffic without compromising users' privacy, but its usage is limited due to a significant delay on establishing an inspected channel. PrivDPI, proposed more recently by Ning et al. (ACM CCS 2019), improves the overall efficiency of BlindBox and makes the inspection scenario more viable. Despite the improvement, we show in this paper that the user privacy of Ning et al.'s PrivDPI can be compromised entirely by the rule generator without involving any other parties, including the middlebox. Having observed the difficulties of realizing efficiency and security in the previous work, we propose a new DPI system for encrypted traffic, named "Practical and Privacy-Preserving Deep Packet Inspection (P2DPI)''. P2DPI enjoys the same level of security and privacy that BlindBox provides. At the same time, P2DPI offers fast setup and encryption and outperforms PrivDPI. Our results are supported by formal security analysis. We implemented our P2DPI and comparable PrivDPI and performed extensive experimentation for performance analysis and comparison.

The rapid spread of COVID-19 has made manual contact tracing difficult. Thus, various public health authorities have experimented with automatic contact tracing using mobile applications (or "apps"). These apps, however, have raised security and privacy concerns. In this paper, we propose an automated security and privacy assessment tool - COVIDGUARDIAN - which combines identification and analysis of Personal Identification Information (PII), static program analysis and data flow analysis, to determine security and privacy weaknesses. Furthermore, in light of our findings, we undertake a user study to investigate concerns regarding contact tracing apps. We hope that COVIDGUARDIAN, and the issues raised through responsible disclosure to vendors, can contribute to the safe deployment of mobile contact tracing. As part of this, we offer concrete guidelines, and highlight gaps between user requirements and app performance.

Named Data Networking (NDN) is an emerging network architecture, which is built by keeping data as its pivotal point. The in-network cache, one of the important characteristics, makes data packets to be available from multiple locations on the Internet. Hence data access control and their enforcement mechanisms become even more critical in the NDNs. In this paper, we propose a novel encryption-based data access control scheme using Role-Based Encryption (RBE). The inheritance property of our scheme provides a natural way to achieve efficient data access control over hierarchical content. This in turn makes our scheme suitable for large scale real world content-centric applications and services such as Netflix. Further, the proposed scheme introduces an anonymous signature-based authentication mechanism to reject bogus data requests nearer to the source, thereby preventing them from entering the network. This in turn helps to mitigate better denial of service attacks. In addition, the signature mechanism supports unlinkability, which is essential to prevent leakages of individual user's access patterns. Another major feature of the proposed scheme is that it provides accountability of the Internet Service Providers (ISPs) using batch signature verification. Moreover, we have developed a transparent and secure dispute resolution and payment mechanism using smart-contract and blockchain technologies. We present a formal security analysis of our scheme to show it is provably secure against Chosen Plaintext Attacks. We also demonstrate that our scheme supports more functionalities than the existing schemes and its performance is better in terms of computation, communication and storage.

Internet of Things (IoT) is an emergent and evolving technology, interconnecting the cyber and physical worlds. IoT technology finds applications in a broad spectrum of areas such as homes, health, water and sanitation, transportation, and environmental monitoring. However, the endless opportunities and benefits of IoT come with many security challenges due to the reduced computation, communication, storage, and energy capabilities of the IoT smart devices. Several computationally lightweight cryptographic protocols exist for these resource-constrained IoT smart devices. However, lightweight solutions render the resource-rich ends of the IoT systems (e.g., edge, fog, or cloud modes) vulnerable as nodes at those ends have the capacity for computationally heavier cryptographic protocols, and they operate in relatively more malicious environments. This asymmetric computational nature of IoT systems requires security protocols that can adapt to the resource availability at the node that they operate. This survey describes the IoT structure, computational capabilities of the devices at the end, edge, fog, and cloud platforms, and classifies existing lightweight cryptographic protocols. The comparative analysis of the existing lightweight cryptographic solutions along with their advantages, drawbacks, and vulnerabilities highlights the need for elastic cryptographic protocols which are capable of adapting to the asymmetric capabilities of the different nodes in IoT systems.

Compression is widely used in Internet applications to save communication time, bandwidth and storage. Recently invented by Jarek Duda asymmetric numeral system (ANS) offers an improved efficiency and a close to optimal compression. The ANS algorithm has been deployed by major IT companies such as Facebook, Google and Apple. Compression by itself does not provide any security (such as confidentiality or authentication of transmitted data). An obvious solution to this problem is an encryption of compressed bitstream. However, it requires two algorithms: one for compression and the other for encryption. In this work, we investigate natural properties of ANS that allow to incorporate authenticated encryption using as little cryptography as possible. We target low-level security communication and storage such as transmission of data from IoT devices/sensors. In particular, we propose three solutions for joint compression and encryption (compcrypt). The solutions offer different tradeoffs between security and efficiency assuming a slight compression deterioration. All of them use a pseudorandom bit generator (PRBG) based on lightweight stream ciphers. The first solution is close to original ANS and applies state jumps controlled by PRBG. The second one employs two copies of ANS, where compression is switched between the copies. The switch is controlled by a PRBG bit. The third compcrypt modifies the encoding function of ANS depending on PRBG bits. Security and efficiency of the proposed compcrypt algorithms are evaluated. The first compcrypt is the most efficient with a slight loss of compression quality. The second one consumes more storage but the loss of compression quality is negligible. The last compcrypt offers the best security but is the least efficient.

A new collaborative learning, called split learning, was recently introduced, aiming to protect user data privacy without revealing raw input data to a server. It collaboratively runs a deep neural network model where the model is split into two parts, one for the client and the other for the server. Therefore, the server has no direct access to raw data processed at the client. Until now, the split learning is believed to be a promising approach to protect the client's raw data; for example, the client's data was protected in healthcare image applications using 2D convolutional neural network (CNN) models. However, it is still unclear whether the split learning can be applied to other deep learning models, in particular, 1D CNN.

Federated learning (FL) and split neural networks (SplitNN) are state-of-art distributed machine learning techniques to enable machine learning without directly accessing raw data on clients or end devices. In theory, such distributed machine learning techniques have great potential in distributed applications, in which data are typically generated and collected at the client-side while the collected data should be processed by the application deployed at the server-side. However, there is still a significant gap in evaluating the performance of those techniques concerning their practicality in the Internet of Things (IoT)-enabled distributed systems constituted by resource-constrained devices. This work is the first attempt to provide empirical comparisons of FL and SplitNN in real-world IoT settings in terms of learning performance and device implementation overhead. We consider a variety of datasets, different model architectures, multiple clients, and various performance metrics. For the learning performance (i.e., model accuracy and convergence time), we empirically evaluate both FL and SplitNN under different types of data distributions such as imbalanced and non-independent and identically distributed (non-IID) data. We show that the learning performance of SplitNN is better than FL under an imbalanced data distribution but worse than FL under an extreme non-IID data distribution. For implementation overhead, we mount both FL and SplitNN on Raspberry Pi devices and comprehensively evaluate their overhead, including training time, communication overhead, power consumption, and memory usage. Our key observations are that under the IoT scenario where the communication traffic is the primary concern, FL appears to perform better over SplitNN because FL has a significantly lower communication overhead compared with SplitNN. However, our experimental results also demonstrate that neither FL or SplitNN can be applied to a heavy model, e.g., with several million parameters, on resource-constrained IoT devices because its training cost would be too expensive for such devices. Source code is released and available: https://github.com/Minki-Kim95/Federated-Learning-and-Split-Learning-with-raspberry-pi.

The Internet of Things (IoT) is transforming major industries, including but not limited to healthcare, agriculture, finance, energy, and transportation. IoT platforms are continually improving with innovations, such as the amalgamation of software-defined networks (SDNs) and network function virtualization (NFV) in the edge-cloud interplay. Deep learning (DL) is becoming popular due to its remarkable accuracy when trained with a massive amount of data such as generated by IoT. However, DL algorithms tend to leak privacy when trained on highly sensitive crowd-sourced data such as medical data. The existing privacy-preserving DL algorithms rely on the traditional server-centric approaches requiring high processing powers. We propose a new local differentially private (LDP) algorithm named LATENT that redesigns the training process. LATENT enables a data owner to add a randomization layer before data leave the data owners' devices and reach a potentially untrusted machine learning service. This feature is achieved by splitting the architecture of a convolutional neural network (CNN) into three layers: 1) convolutional module (CNM); 2) randomization module; and 3) fully connected module. Hence, the randomization module can operate as an NFV privacy preservation service in an SDN-controlled NFV, making LATENT more practical for IoT-driven cloud-based environments compared to existing approaches. The randomization module employs a newly proposed LDP protocol named utility enhancing randomization, which allows LATENT to maintain high utility compared to existing LDP protocols. Our experimental evaluation of LATENT on convolutional deep neural networks demonstrates excellent accuracy (e.g., 91%-96%) with high model quality even under low privacy budgets (e.g., ε = 0.5).

Communication in modern cars is managed by a controller area network (CAN) bus protocol and its extensions for electronic control units (ECUs). The CAN bus is a preferred method for reliable real-time broadcast communication. However, unprotected CAN communications make the vehicles vulnerable to a variety of practical malicious wired/wireless attacks. In this work, we analyze the existing frame-level authentication protocol and identify weaknesses and limitations. To address this, we provide a protocol suite for entity authentication, key management, a secure message flow for remote transmission request frames and session key update to be applied for vehicle connection with external devices. We prove the security of our protocol in the random oracle model and assess its resistance against known attacks. We formally verify the security of our protocol using the Tamarin tool. Our simulation results indicate that our protocol improves efficiency.

Industrial Internet of Things (IIoT) is revolutionizing many leading industries such as energy, agriculture, mining, transportation, and healthcare. IIoT is a major driving force for Industry 4.0, which heavily utilizes machine learning (ML) to capitalize on the massive interconnection and large volumes of IIoT data. However, ML models that are trained on sensitive data tend to leak privacy to adversarial attacks, limiting its full potential in Industry 4.0. This article introduces a framework named PriModChain that enforces privacy and trustworthiness on IIoT data by amalgamating differential privacy, federated ML, Ethereum blockchain, and smart contracts. The feasibility of PriModChain in terms of privacy, security, reliability, safety, and resilience is evaluated using simulations developed in Python with socket programming on a general-purpose computer. We used Ganache_v2.0.1 local test network for the local experiments and Kovan test network for the public blockchain testing. We verify the proposed security protocol using Scyther_v1.1.3 protocol verifier.

Each layer of nodes and communication networks in edge computing, from cloud to the end device (i.e, often considered as resource-constrained IoT devices), exhibits a different level of trust for each stakeholder - e.g., edge nodes may not be fully trusted by IoT devices and the cloud. Moreover, asymmetric nature of resources between layers makes it hard to establish a balance between security and performance - e.g., lightweight cryptography may degrade security level against untrusted nodes while heavyweight ones may not be feasible for the light-weight end devices. An advanced encryption scheme such as the Identity-Based Broadcast Encryption (IBBE) is a popular technique to reduce storage and communication overhead. However, IBBE requires heavy computation to the end devices and still does not fully satisfy the security requirements that exist in the layers of edge computing. This paper presents a new IBBE with outsourced partial decryption for hybrid security models that each layer in edge computing requires. It balances the computational overhead based on asymmetric nature that nodes in each layer have. Particularly, with new schemes, the ciphertext can be transformed from its initial format. The cloud encrypts their data for multiple end devices and store them in the edge nodes, but those interim nodes can blindly transform the ciphertext from the cloud into a form which (i) is decryptable by only an authorized end device, and (ii) imposes smaller decryption and data transmission burden to end devices, regardless of the number of recipients. Our security analysis shows that new schemes are selectively and adaptively secure. We implement our solution and show that new schemes reduce the communication overhead from an edge node to end devices and the computation overhead on the end devices, compared to the original IBBE schemes.

Fuzzing is a promising technique for detecting security vulnerabilities. Newly developed fuzzers are typically evaluated in terms of the number of bugs found on vulnerable programs/binaries. However, existing corpora usually do not capture the features that prevent fuzzers from finding bugs, leading to ambiguous conclusions on the pros and cons of the fuzzers evaluated. In this paper, we propose to address the above problem by generating corpora based on search-hampering features. As a proof-of-concept, we designed FEData, a prototype corpus that currently focuses on three search-hampering features to generate vulnerable programs for fuzz testing. Unlike existing corpora that can only answer "how", FEData can also further answer "why" by exposing (or understanding) the reasons for the identified weaknesses in a fuzzer. The "why" information serves as the key to the improvement of fuzzers. Based on the "why" information, our FEData programs enabled us to identify the weakness of AFLFast, called cycle explosion, behind. We further developed an improved version of AFLFast, called AFLFast+, which has overcome the cycle explosion problem. AFLFast+ retains the efficiency of AFLFast in path search while maintaining or even surpassing the bug-finding capability of AFL for the corpus evaluated.

Current anomaly detection systems (ADSs) apply statistical and machine learning algorithms to discover zero-day attacks, but such algorithms are vulnerable to advanced persistent threat actors. In this paper, we propose an adversarial statistical learning mechanism for anomaly detection, outlier Dirichlet mixture-based ADS (ODM-ADS), which has three new capabilities. First, it can self-adapt against data poisoning attacks that inject malicious instances in the training phase for disrupting the learning process. Second, it establishes a statistical legitimate profile and considers variations from the baseline of the profile as anomalies using a proposed outlier function. Third, to deal with dynamic and large-scale networks such as Internet of Things and cloud and fog computing, we suggest a framework for deploying the mechanism as Software as a Service in the fog nodes. The fog enables the proposed mechanism to concurrently process streaming data at the edge of the network. The ODM-ADS mechanism is evaluated using both NSL-KDD and UNSW-NB15 datasets, whose findings indicate that ODM-ADS outperforms seven other peer algorithms in terms of accuracy, detection rates, false positive rates, and computational time.

This paper proposes a key reconciliation mechanism using Reed Solomon code to improve the effectiveness of secret key generation based on Received Signal Strength (RSS) variations in a wireless channel between two communicating parties. We use a two threshold quantization algorithm which can operate as a lossless system improving the secret bit extraction rate. We present our Reed Solomon (RS) code based algorithm for reconciling the independently derived secret keys through sharing of only the syndrome bits. We evaluate our reconciliation algorithm using simulated channel measurements and real environment data gathered between an unmanned aerial vehicle (UAV) and a controller in a semi mobile environment. We show that by selecting appropriate parameters of the RS code, it is possible to generate matching keys at the transmitter and the receiver which can be used to secure the communication at wireless PHY or upper layers.

Reliable data transmissions are challenging in industrial wireless sensor networks as channel conditions change over time. Rapid changes in channel conditions require accurate estimation of the routing path performance and timely update of the routing information. However, this is not well fulfilled in existing routing approaches. Addressing this problem, this paper presents combined global and local update processes for efficient route update and maintenance, and incorporates them with a hierarchical proactive routing framework. While the global process updates the routing path with a relatively long period, the local process with a shorter period checks potential routing path problems. A theoretical modeling is developed to describe the processes. Through simulations, the presented approach is shown to reduce end-to-end delay up to 30 times for large networks, while improving packet reception ratio (PRR) in comparison with hierarchical and proactive routing protocols ROL/NDC, DSDV, and DSDV with IPv6 Routing Protocol for Low-Power and Lossy Networks' Trickle algorithm. Compared with reactive routing protocols AODV and Ad Hoc On-demand Multipath Distance Vector, it provides similar PRR while reducing end-to-end delay over 15 times.

Mobile applications are being increasingly deployed on a massive scale in various mobile sensor grid database systems. With limited resources from the mobile devices, how to process the huge number of queries from mobile users with distributed sensor grid databases becomes a critical problem for such mobile systems. While the fundamental semantic cache technique has been investigated for query optimization in sensor grid database systems, the problem is still difficult due to the fact that more realistic multidimensional constraints have not been considered in existing methods. To solve the problem, a new semantic cache scheme is presented in this paper for location-dependent data queries in distributed sensor grid database systems. It considers multidimensional constraints or factors in a unified cost model architecture, determines the parameters of the cost model in the scheme by using the concept of Nash equilibrium from game theory, and makes semantic cache decisions from the established cost model. The scenarios of three factors of semantic, time, and locations are investigated as special cases, which improve existing methods. Experiments are conducted to demonstrate the semantic cache scheme presented in this paper for distributed sensor grid database systems.

The Distributed Network Protocol version 3 (DNP3) provides Secure Authentication (DNP3-SA) as the mechanism to authenticate unicast messages from a master station to its outstations in supervisory control and data acquisition systems. In large-scale systems, it may be necessary to broadcast a critical request from a master station to multiple outstations at once. The DNP3 protocol standard describes the use of broadcast communication; however, it does not specify its security. This paper is the first to present DNP3 Secure Authentication for Broadcast (DNP3-SAB), a new lightweight security scheme for broadcast mode communication. This scheme is based on hash chain and only makes use of the existing cryptographic primitives specified in DNP3-SA. The scheme integrates itself into the DNP3-SA key update process. The proposed scheme is modeled, validated, and verified using colored Petri Nets against the most common protocol attacks such as modification, injection, and replay. Performance analysis on our scheme and the existing DNP3-SA modes (NACR and AGM) shows that DNP3-SAB reduces the communication overhead significantly at the cost of an increase with a constant term in processing and storage overhead. This benefit is maintained even when DNP3-SAB is under attack.

As one of the most widely used wireless network technologies, IEEE 802.11 wireless local area networks (WLANs) have found a dramatically increasing number of applications in soft real-time networked control systems (NCSs). To fulfill the real-time requirements in such NCSs, most of the bandwidth of the wireless networks need to be allocated to high-priority data for periodic measurements and control with deadline requirements. However, existing quality of service (QoS)-enabled 802.11 medium access control (MAC) protocols do not consider the deadline requirements explicitly, leading to unpredictable deadline performance of NCS networks. Consequentially, the soft real-time requirements of the periodic traffic may not be satisfied, particularly under congested network conditions. This paper makes two main contributions to address this problem in wireless NCSs. A deadline-constrained MAC protocol with QoS differentiation is presented for IEEE 802.11 soft real-time NCSs. It handles periodic traffic by developing two specific mechanisms, a contention-sensitive backoff mechanism and an intra-traffic-class QoS differentiation mechanism. A theoretical model is established to describe the deadline-constrained MAC protocol and evaluate its performance of throughput, delay, and packet-loss ratio in wireless NCSs. Numerical studies are conducted to validate the accuracy of the theoretical model and to demonstrate the effectiveness of the new MAC protocol.

The Body Area Network (BAN) is an emerging technology that focuses on monitoring physiological data in, on and around the human body. BAN technology permits wearable and implanted sensors to collect vital data about the human body and transmit it to other nodes via low-energy communication. In this paper, we investigate interactions in terms of data flows between parties involved in BANs under four different scenarios targeting outdoor and indoor medical environments: hospital, home, emergency and open areas. Based on these scenarios, we identify data flow requirements between BAN elements such as sensors and control units (CUs) and parties involved in BANs such as the patient, doctors, nurses and relatives. Identified requirements are used to generate BAN data flow models. Petri Nets (PNs) are used as the formal modelling language. We check the validity of the models and compare them with the existing related work. Finally, using the models, we identify communication and security requirements based on the most common active and passive attack scenarios.

Distributed Network Protocol Version 3 (DNP3) is the de-facto communication protocol for power grids. Standard-based interoperability among devices has made the protocol useful to other infrastructures such as water, sewage, oil and gas. DNP3 is designed to facilitate interaction between master stations and outstations. In this paper, we apply a formal modelling methodology called Coloured Petri Nets (CPN) to create an executable model representation of DNP3 protocol. The model facilitates the analysis of the protocol to ensure that the protocol will behave as expected. Also, we illustrate how to verify and validate the behaviour of the protocol, using the CPN model and the corresponding state space tool to determine if there are insecure states. With this approach, we were able to identify a Denial of Service (DoS) attack against the DNP3 protocol.

Internet services are important part of daily activities for most of us. These services come with sophisticated authentication requirements which may not be handled by average Internet users. The management of secure passwords for example creates an extra overhead which is often neglected due to usability reasons. Furthermore, password-based approaches are applicable only for initial logins and do not protect against unlocked workstation attacks. In this paper, we provide a non-intrusive identity verification scheme based on behavior biometrics where keystroke dynamics based-on free-text is used continuously for verifying the identity of a user in real-time. We improved existing keystroke dynamics based verification schemes in four aspects. First, we improve the scalability where we use a constant number of users instead of whole user space to verify the identity of target user. Second, we provide an adaptive user model which enables our solution to take the change of user behavior into consideration in verification decision. Next, we identify a new distance measure which enables us to verify identity of a user with shorter text. Fourth, we decrease the number of false results. Our solution is evaluated on a data set which we have collected from users while they were interacting with their mail-boxes during their daily activities.

Increasing use of computerized systems in our daily lives creates new adversarial opportunities for which complex mechanisms are exploited to mend the rapid development of new attacks. Behavioral Biometrics appear as one of the promising response to these attacks. But it is a relatively new research area, specific frameworks for evaluation and development of behavioral biometrics solutions could not be found yet. In this paper we present a conception of a generic framework and runtime environment which will enable researchers to develop, evaluate and compare their behavioral biometrics solutions with repeatable experiments under the same conditions with the same data.

Smartphones are getting increasingly popular and several malwares appeared targeting these devices. General countermeasures to smartphone malwares are currently limited to signature-based antivirus scanners which efficiently detect known malwares, but they have serious shortcomings with new and unknown malwares creating a window of opportunity for attackers. As smartphones become host for sensitive data and applications, extended malware detection mechanisms are necessary complying with the corresponding resource constraints. The contribution of this paper is twofold. First, we perform static analysis on the executables to extract their function calls in Android environment using the command readelf. Function call lists are compared with malware executables for classifying them with PART, Prism and Nearest Neighbor Algorithms. Second, we present a collaborative malware detection approach to extend these results. Corresponding simulation results are presented.

The increase of online services, such as eBanks, WebMails, in which users are verified by a username and password, is increasingly exploited by identity theft procedures. Identity Theft is a fraud, in which someone pretends to be someone else is order to steal money or get other benefits. To overcome the problem of identity Theft an additional security layer is required. Within the last decades the option of verifying users based on their keystroke dynamics was proposed during login verification. Thus, the imposter has to be able to type in a similar way to the real user in addition to having the username and password. However, verifying users upon login is not enough, since a logged station/mobile is vulnerable for imposters when the user leaves her machine. Thus, verifying users continuously based on their activities is required. Within the last decade there is a growing interest and use of biometrics tools, however, these are often costly and require additional hardware. Behavioral biometrics, in which users are verified, based on their keyboard and mouse activities, present potentially a good solution. In this paper we discuss the problem of identity theft and propose behavioral biometrics as a solution. We survey existing studies and list the challenges and propose solutions.

In this paper we demonstrate how to monitor a smartphone running Symbian operating system and Windows Mobile in order to extract features for anomaly detection. These features are sent to a remote server because running a complex intrusion detection system on this kind of mobile device still is not feasible due to capability and hardware limitations. We give examples on how to compute relevant features and introduce the top ten applications used by mobile phone users based on a study in 2005. The usage of these applications is recorded by a monitoring client and visualized. Additionally, monitoring results of public and self-written malwares are shown. For improving monitoring client performance, Principal Component Analysis was applied which lead to a decrease of about 80% of the amount of monitored features.

We introduce a lightweight biometric solution for user authentication over networks using online handwritten signatures. The algorithm proposed is based on a modified Hausdorff distance and has favorable characteristics such as low computational cost and minimal training requirements. Furthermore, we investigate an information theoretic model for capacity and performance analysis for biometric authentication which brings additional theoretical insights to the problem. A fully functional proof-of-concept prototype that relies on commonly available off-the-shelf hardware is developed as a client-server system that supports Web services. Initial experimental results show that the algorithm performs well despite its low computational requirements and is resilient against over-the-shoulder attacks.

Secure communications between large number of sensor nodes that are randomly scattered over a hostile territory, necessitate efficient key distribution schemes. However, due to limited resources at sensor nodes such schemes cannot be based on post deployment computations. Instead, pairwise (symmetric) keys are required to be pre-distributed by assigning a list of keys, (a.k.a. key-chain), to each sensor node. If a pair of nodes does not have a common key after deployment then they must find a key-path with secured links. The objective is to minimize the keychain size while (i) maximizing pairwise key sharing probability and resilience, and (ii) minimizing average key-path length. This paper presents a deterministic key distribution scheme based on Expander Graphs. It shows how to map the parameters (e.g., degree, expansion, and diameter) of a Ramanujan Expander Graph to the desired properties of a key distribution scheme for a physical network topology.