The widespread use of the Internet and digital services has significantly increased data collection and processing. Critical domains like healthcare rely on this data, but privacy and security concerns limit its usabi...
详细信息
The widespread use of the Internet and digital services has significantly increased data collection and processing. Critical domains like healthcare rely on this data, but privacy and security concerns limit its usability, constraining the performance of intelligent systems, particularly those leveraging Neural networks (NNs). NNs require high-quality data for optimal performance, but existing privacy-preserving methods, such as Federated Learning and Differential Privacy, often degrade model accuracy. While Homomorphic Encryption (HE) has emerged as a promising alternative, existing HE-based methods face challenges in computational efficiency and scalability, limiting their real-world application. To address these issues, we introduce ENNigma, a novel framework employing state-of-the-art Fully Homomorphic Encryption techniques. This framework introduces optimizations that significantly improve the speed and accuracy of encrypted NN operations. Experiments conducted using the CIC-DDoS2019 dataset - a benchmark for distributed Denial of Service attack detection - demonstrate ENNigma's effectiveness. A classification performance with a maximum relative error of 1.01% was achieved compared to non-private models, while reducing multiplication time by up to 59% compared to existing FHE-based approaches. These results highlight ENNigma's potential for practical, privacy-preserving neural network applications.
Collaborative data sharing underlies applications in systems such as online social networks and cloud storage. A central provider hosts shared data, e.g., a Facebook group page, and provides sharing users with read/wr...
详细信息
Collaborative data sharing underlies applications in systems such as online social networks and cloud storage. A central provider hosts shared data, e.g., a Facebook group page, and provides sharing users with read/write access according to user-defined settings. Historical incidents prove that data storage centralization enables a corrupted provider to censor or diverge users' views of the shared data. Such misbehavior is hardly detectable as users update the shared data unbeknown to each other. Existing solutions suggest server-side data replication, which is storage inefficient, or users' out-of-band communication, which is communication intensive. Addressing these inefficiencies, we propose Integrita, a data-sharing mechanism that protects users' view-consistency needless to out-of-band communication. We present a novel distributed history tree algorithm to partition the shared data among N storage providers, N - 1 of which are Byzantine faulty and colluding. Our data partitioning solution reduces storage overhead by a multiplicative factor of N and allows provable detection of server-side equivocation and identification of corrupted servers. We introduce and achieve anew consistency level, named q-detectable consistency, where users' views inconsistency cannot remain undetected for more than q updates, q being the system parameters' function. Without loss of generality, we use online social networks as a case study to deploy Integrita and supply performance/numerical analysis accordingly.
The Internet of Things (IoT) networks are poised to play a critical role in providing ultra-low latency and high bandwidth communications in various real-world IoT scenarios. Assuring end-to-end secure, energy- aware,...
详细信息
The Internet of Things (IoT) networks are poised to play a critical role in providing ultra-low latency and high bandwidth communications in various real-world IoT scenarios. Assuring end-to-end secure, energy- aware, reliable, real-time IoT communication is hard due to the heterogeneity and transient behavior of IoT networks. Additionally, the lack of integrated approaches to efficiently schedule IoT tasks and holistically offload computing resources, and computational limits in IoT systems to achieve effective resource utilization. This paper makes three contributions to research on overcoming these problems in the context of distributed IoT systems that use the Software Defined Networking (SDN) programmable control plane in symbiosis with blockchain to benefit from the dispersed or decentralized, and efficient environment of distributed IoT transactions over Wide Area networks (WANs). First, it introduces a Blockchain-SDN architectural component to reinforce flexibility and trustworthiness and improve the Quality of Service (QoS) of IoT networks. Second, it describes the design of an IoT-focused smart contract that implements the control logic to manage IoT data, detect and report suspected IoT nodes, and mitigate malicious traffic. Third, we introduce a novel consensus algorithm based on the Proof-of-Authority (PoA) to achieve agreements between blockchain-enabled IoT nodes, improve the reliability of IoT edge devices, and establish absolute trust among all smart IoT systems. Experimental results show that integrating SDN with blockchain outperforms traditional Proof-of-Work (PoW) and Practical Byzantine Fault Tolerance (PBFT) algorithms, delivering up to 68% lower latency, 87% higher transaction throughput, and 45% better energy savings.
This article presents a short (and partial) history of synchronization in systems made up of asynchronous sequential processes (automata). Among other points, it shows that synchronization (which consists in ordering ...
详细信息
ISBN:
(纸本)9783031744976;9783031744983
This article presents a short (and partial) history of synchronization in systems made up of asynchronous sequential processes (automata). Among other points, it shows that synchronization (which consists in ordering operations issued by processes on shared objects) has a different flavor according to the fact that the objects are physical objects (such as a printer or a disk) or logical objects (immaterial objects represented by sequences of bits). It then follows from this physical/logical nature of computing objects that mutual exclusion is to physical objects what consensus is to logical objects. The article also addresses recent results on process synchronization in fully anonymous systems (systems in which processes cannot be distinguished one from the other, and where there is a disagreement on the addresses of the memory registers.
Robots with very weak capabilities placed on the vertices of a graph are required to move toward a common vertex from where they do not move anymore. The task is known as the Gathering problem and it has been extensiv...
详细信息
ISBN:
(纸本)9783031744976;9783031744983
Robots with very weak capabilities placed on the vertices of a graph are required to move toward a common vertex from where they do not move anymore. The task is known as the Gathering problem and it has been extensively studied in the last decade with respect to both general graphs and specific topologies. Most of the challenges faced are due to possible isometries observable from the placement of the robots with respect to the underlying topology. Rings, Grids, Complete graphs are just a few examples of very regular topologies where the placement of the robots and suitable movements are crucial for succeeding in Gathering. Here we are interested in understanding what can be done in Butterfly graphs where really many isometries are present and most importantly unavoidable by any movement. We propose a Gathering algorithm for the so-called leader configurations, i.e., those where the initial placement of the robots admits the detection (and election) of one robot as the leader. We introduce a non-trivial technique to elect the leader which is of its own interest. We also prove that the proposed Gathering algorithm is asymptotically optimal in terms of synchronous rounds required.
Collaborative learning in peer-to-peer networks offers the benefits of distributed learning while mitigating the risks associated with single points of failure inherent in centralized servers. However, adversarial wor...
详细信息
ISBN:
(纸本)9783031820724;9783031820731
Collaborative learning in peer-to-peer networks offers the benefits of distributed learning while mitigating the risks associated with single points of failure inherent in centralized servers. However, adversarial workers pose potential threats by attempting to inject malicious information into the network. Thus, ensuring the resilience of peer-to-peer learning emerges as a pivotal research objective. The challenge is exacerbated in the presence of non-convex loss functions and non-iid data distributions. This paper introduces a resilient aggregation technique tailored for such scenarios, aimed at fostering similarity among peers' learning processes. The aggregation weights are determined through an optimization procedure, and use the loss function computed using the neighbor's models and individual private data, thereby addressing concerns regarding data privacy in distributed machine learning. Theoretical analysis demonstrates convergence of parameters with non-convex loss functions and non-iid data distributions. Empirical evaluations across three distinct machine learning tasks support the claims. The empirical findings, which encompass a range of diverse attack models, also demonstrate improved accuracy when compared to existing methodologies.
Temporal Graph Neural networks (TGNNs) have achieved success in real-world graph-based applications. The increasing scale of dynamic graphs necessitates distributed training. However, deploying TGNNs in a distributed ...
详细信息
ISBN:
(纸本)9789819757787;9789819757794
Temporal Graph Neural networks (TGNNs) have achieved success in real-world graph-based applications. The increasing scale of dynamic graphs necessitates distributed training. However, deploying TGNNs in a distributed setting poses challenges due to the temporal dependencies in dynamic graphs, the need for computation balance during distributed training, and the non-ignorable communication costs across disjointed trainers. In this paper, we propose DisTGL, a distributed temporal graph neural network learning system. Leveraging a temporal-aware partitioning scheme and a series of enhanced communication techniques, DisTGL ensures efficient distributed computation and minimizes communication overhead. Based on that, DisTGL facilitates fast TGNN training and downstream tasks. An evaluation of DisTGL using various TGNN models shows that i) DisTGL achieves acceleration of up to 10x compared to existing TGNN frameworks;and ii) the proposed distributed dynamic graph partitioning reduces cross-machine operations by 25%, while the optimized communication reduce the costs by 1.5-2.5x.
The increasing computational capabilities of Low Earth Orbit (LEO) constellations have significantly augmented their autonomy and operational flexibility. Complex onboard tasks such as observation, sensing, and situat...
详细信息
The increasing computational capabilities of Low Earth Orbit (LEO) constellations have significantly augmented their autonomy and operational flexibility. Complex onboard tasks such as observation, sensing, and situational awareness can be processed and executed directly on the Satellite Edge Computing (SEC) networks. to the time-varying characteristics of inter-satellite links and the uncertainty in the load of edge satellites, efficient offloading of on-board tasks presents significant challenges. We introduce an on-board distributed task offloading method for LEO satellite tasks in emergency to enhance service quality. We initially a dynamic offloading scheme, in which data-source satellites can transmit tasks to edge nodes. Then, formulate the multi-hop satellite network dynamic offloading (MSNDO) problem to minimize system and maximize success ratio of time-sensitive tasks under multiple constraints. Finally, we propose a distributed deep reinforcement learning algorithm that allows individual satellites to design offloading strategies knowing the decision-making patterns of other satellites. Simulation experiments show that the proposed algorithm can utilize the edge satellite processing capabilities more efficiently and significantly improve performance of the SEC system.
Nowadays, energy buildings have a huge impact in society regarding the active role in the management of energy consumption. Hence, building owners are required to avoid energy losses and improve energy efficiency as h...
详细信息
ISBN:
(纸本)9783031820724;9783031820731
Nowadays, energy buildings have a huge impact in society regarding the active role in the management of energy consumption. Hence, building owners are required to avoid energy losses and improve energy efficiency as high as possible. Therefore, it is required to plan an optimization strategy to buy and sell energy in the market ahead of time. To formulate this optimization plan, building owners require the work of specialists responsible for processing, training, forecasting, and evaluation tasks regarding the prediction of energy consumption data from a building for a specific target of time. Therefore, a multiagent-system is needed to allow the cooperation of various agents including the building owner, forecast provider, data structurer and error analysis. Moreover, forecasting algorithms such as artificial neural networks should be taken into consideration in order to process large quantities of energy consumption data during the training and forecasting phases.
We address the problem of enforcing global invariants, i.e., system-level properties, in Collective Adaptive systems, such as distributed and decentralized Internet of Things (IoT) solutions. In particular, we propose...
详细信息
ISBN:
(纸本)9783031751066;9783031751073
We address the problem of enforcing global invariants, i.e., system-level properties, in Collective Adaptive systems, such as distributed and decentralized Internet of Things (IoT) solutions. In particular, we propose a novel approach adopting Attribute-based memory Updates (AbU), a calculus modeling declarative, event-driven systems with attribute-based communication. Our methodology leverages a combination of precise node-level scheduling and local reasoning, with local invariants derived from the system-level property to guarantee. This distributed and decentralized approach promotes efficient enforcing while ensuring desired system-wide behavior, without the need for a central controlling authority.
暂无评论