Future Generation Computer Systems-The International Journal of Escience最新文献

{"title":"EC5: Edge–cloud collaborative computing framework with compressive communication","authors":"Jingwei Tan ,&nbsp;Fagui Liu ,&nbsp;Bin Wang ,&nbsp;Qingbo Wu ,&nbsp;C.L. Philip Chen","doi":"10.1016/j.future.2025.107715","DOIUrl":"10.1016/j.future.2025.107715","url":null,"abstract":"<div><div>With an increasing number of deep neural network (DNN)-based applications being deployed at the edges, edge–cloud collaborative computing has emerged as a promising solution to alleviate the burden on resource-constrained edges by collaborative inference. However, simply offloading part of DNN to the cloud introduces significant communication overhead during inference. In this paper, we propose EC5, an Edge–Cloud Collaborative Computing framework with Compressive Communication. The compression of the intermediate feature is formulated using information theory and jointly optimized with the DNN through end-to-end multi-task learning. By decomposing DNN parameters into a new space, EC5 enables efficient storage and update of models across various compression levels. An Adaptive Exit scheme is designed to retain high-confidence inputs on the edge for fast inference, reducing reliance on the cloud. Experimental comparisons with baseline methods prove that EC5 significantly conserves network bandwidth and reduces communication instances, with low latency and acceptable accuracy loss, showing flexibility across different communication scenarios.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107715"},"PeriodicalIF":6.2,"publicationDate":"2025-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ENNigma: A framework for Private Neural Networks","authors":"Pedro Barbosa ,&nbsp;Ivone Amorim ,&nbsp;Eva Maia ,&nbsp;Isabel Praça","doi":"10.1016/j.future.2025.107719","DOIUrl":"10.1016/j.future.2025.107719","url":null,"abstract":"<div><div>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.</div><div>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.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107719"},"PeriodicalIF":6.2,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"WFE-Tab: Overcoming limitations of TabPFN in IIoT-MEC environments with a weighted fusion ensemble-TabPFN model for improved IDS performance","authors":"Sergio Ruiz-Villafranca ,&nbsp;José Roldán-Gómez ,&nbsp;Javier Carrillo-Mondéjar ,&nbsp;José Luis Martinez ,&nbsp;Carlos H. Gañán","doi":"10.1016/j.future.2025.107707","DOIUrl":"10.1016/j.future.2025.107707","url":null,"abstract":"<div><div>In recent years we have seen the emergence of new industrial paradigms such as Industry 4.0/5.0 or the Industrial Internet of Things (IIoT). As the use of these new paradigms continues to grow, so do the number of threats and exploits that they face, which makes the IIoT a desirable target for cybercriminals. Furthermore, IIoT devices possess inherent limitations, primarily due to their limited resources. As a result, it is often impossible to detect attacks using solutions designed for other environments. Recently, Intrusion Detection Systems (IDS) based on Machine Learning (ML) have emerged as a solution that takes advantage of the large amount of data generated by IIoT devices to implement their functionality and achieve good performance, and the inclusion of the Multi-Access Edge Computing (MEC) paradigm in these environments provides the necessary computational resources to deploy IDS effectively. Furthermore, TabPFN has been considered as an attractive option for solving classification problems without the need to reprocess the data. However, TabPFN has certain drawbacks when it comes to the number of training samples and the maximum number of different classes that the model is capable of classifying. This makes TabPFN unsuitable for use when the dataset exceeds one of these limitations. In order to overcome such limitations, this paper presents a Weighted Fusion-Ensemble-based TabPFN (WFE-Tab) model to improve IDS performance in IIoT-MEC scenarios. The presented study employs a novel weighted fusion method to preprocess data into multiple subsets, generating different ensemble family TabPFN models. The resulting WFE-Tab model comprises four stages: data collection, data preprocessing, model training, and model evaluation. The performance of the WFE-Tab method is evaluated using key metrics such as Accuracy, Precision, Recall, and F1-Score, and validated using the Edge-IIoTset public dataset. The performance of the method is then compared with baseline and modern methods to evaluate its effectiveness, achieving an F1-Score performance of 99.81%.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107707"},"PeriodicalIF":6.2,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fuzzy energy management strategies for energy harvesting IoT nodes based on a digital twin concept","authors":"Michal Prauzek,&nbsp;Karolina Gaiova,&nbsp;Tereza Kucova,&nbsp;Jaromir Konecny","doi":"10.1016/j.future.2025.107717","DOIUrl":"10.1016/j.future.2025.107717","url":null,"abstract":"<div><div>This study presents a cloud-assisted energy management strategy for energy harvesting Internet-of-Things (IoT) nodes, using a novel digital twin (DT) concept for dynamic optimization of IoT node behavior. The system is built upon a fuzzy-rule-based controller optimized through a differential evolution (DE) algorithm. DE is particularly well-suited for this application, as it is capable of optimizing the controller without requiring gradient information, allowing it to efficiently navigate the complex, nonlinear characteristics of IoT energy management problems. The optimization process tunes nine key fuzzy input coefficients to create an energy-efficient control strategy. The DT concept serves as a virtual replica of the physical IoT node, continuously synchronizing real-time data from sensors and other internal parameters, including energy harvesting rates and component health. Through this real-time feedback loop, the DT enables predictive adjustments to the control system, increasing the longevity and reliability of the IoT devices in harsh and changing environments. Compared to traditional energy management strategies, the proposed method improves energy utilization by 11%, leveraging four years of solar data collected from multiple geographical locations. Moreover, the system achieves a 12% increase in successful transmissions, ensuring greater data availability in the cloud while minimizing device failures and optimizing the use of available energy. The DT concept allows the system to simulate and predict IoT node behavior under various conditions, continuously refining the energy management strategy. This ensures not only optimal energy efficiency but also accounts for component degradation over time, offering long-term adaptability and minimizing the need for manual intervention. Thus, the synergy between the DT concept and DE optimization offers a powerful, scalable solution for managing energy-constrained IoT networks, surpassing conventional expert-designed strategies in both adaptability and performance.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107717"},"PeriodicalIF":6.2,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"QuantuneV2: Compiler-based local metric-driven mixed precision quantization for practical embedded AI applications","authors":"Jeongseok Kim ,&nbsp;Jemin Lee ,&nbsp;Yongin Kwon ,&nbsp;Daeyoung Kim","doi":"10.1016/j.future.2025.107718","DOIUrl":"10.1016/j.future.2025.107718","url":null,"abstract":"<div><div>Mixed-precision quantization methods have been proposed to reduce model size while minimizing accuracy degradation. However, existing studies require retraining and do not consider the computational overhead and intermediate representations (IR) generated during the compilation process, limiting their application at the compiler level. This computational overhead refers to the runtime latency caused by frequent quantization and de-quantization operations during inference. Performing these operations at the individual operator level causes significant runtime delays. To address these issues, we propose <span>QuantuneV2</span>, a compiler-based mixed-precision quantization method designed for practical embedded AI applications. <span>QuantuneV2</span> performs inference only twice – once before quantization and once after quantization – and operates with a computational complexity off <span><math><mrow><mi>O</mi><mrow><mo>(</mo><mi>n</mi><mo>)</mo></mrow></mrow></math></span> that increases linearly with the number of model parameters. We also made the sensitivity analysis more stable by using local metrics like weights, activation values, the Signal-to-Quantization-Noise Ratio (SQNR), and the Mean Squared Error (MSE). We also cut down on computational overhead by choosing the best IR and using operator fusion. Experimental results show that <span>QuantuneV2</span> achieved up to a 10.28% improvement in accuracy and a 12.52% increase in speed compared to existing methods across five models: ResNet18v1, ResNet50v1, SqueezeNetv1, VGGNet, and MobileNetv2. This demonstrates that <span>QuantuneV2</span> enhances model performance while maintaining computational efficiency, making it suitable for deployment in embedded AI environments.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107718"},"PeriodicalIF":6.2,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scalable compute continuum","authors":"Valeria Cardellini ,&nbsp;Patrizio Dazzi ,&nbsp;Gabriele Mencagli ,&nbsp;Matteo Nardelli ,&nbsp;Massimo Torquati","doi":"10.1016/j.future.2024.107697","DOIUrl":"10.1016/j.future.2024.107697","url":null,"abstract":"<div><div>The Compute Continuum paradigm addresses the challenges of heterogeneous and dynamic computing resources, facilitating distributed application execution while enhancing data locality, performance, availability, adaptability, and energy efficiency. By integrating IoT, edge, and cloud resources into a cohesive continuum, applications can operate closer to data sources and end users. This approach supports refined adaptation strategies tailored to specific infrastructure components, enabling reduced latency, optimized bandwidth use, and improved privacy. To fully realize the Compute Continuum’s potential, autonomous and proactive management is essential, leveraging interdisciplinary methods from optimization theory, control theory, machine learning, and artificial intelligence. This special issue highlights advancements in three key areas: resource characterization and scheduling, middleware for application deployment and reconfiguration, and applications in the Compute Continuum. These contributions highlight innovative solutions for resource optimization, dynamic management, and real-world implementations, showcasing the potential of the Compute Continuum to revolutionize distributed computing across diverse domains.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107697"},"PeriodicalIF":6.2,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards efficient privacy-preserving conjunctive keywords search over encrypted cloud data","authors":"Yaru Liu ,&nbsp;Xiaodong Xiao ,&nbsp;Fanyu Kong ,&nbsp;Hanlin Zhang ,&nbsp;Jia Yu","doi":"10.1016/j.future.2025.107716","DOIUrl":"10.1016/j.future.2025.107716","url":null,"abstract":"<div><div>With increasing popularity of cloud computing, more and more users choose to store data on cloud servers. Privacy-preserving keyword search is a critical technology in the field of cloud computing, which can directly search for encrypted data stored on cloud servers. In this paper, we propose a new scheme which can achieve conjunctive keywords search in a privacy-preserving way, and maintain forward security. In order to realize conjunctive keywords search with reduced communication cost and leakage, our scheme constructs a secure index based on the full binary tree data structure. Each leaf node represents a keyword, and the node stores the file identifier containing the keyword. Thus, all files containing searched keywords can be searched at one time without searching one file by one. The search time is only related to the number of search keywords and not related to the number of files. Each non-leaf node stores the keywords of its left and right child nodes, which are mapped to the Indistinguishable Bloom Filter(IBF). To achieve forward security, we choose a random string as the latest state to update trapdoors for each update query. Thus, update trapdoor cannot match with previous search trapdoors to achieve forward security. Finally, detailed experiments and security analysis prove that our scheme is secure and efficient.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107716"},"PeriodicalIF":6.2,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Service-driven dynamic QoS on-demand routing algorithm","authors":"Hao She,&nbsp;Lixing Yan,&nbsp;Chuanfeng Mao,&nbsp;Qihui Bu,&nbsp;Yongan Guo","doi":"10.1016/j.future.2024.107685","DOIUrl":"10.1016/j.future.2024.107685","url":null,"abstract":"<div><div>With the proliferation of Internet of Things (IoT) devices, the scale of networks is growing exponentially. However, dynamically meeting the diverse quality of service (QoS) routing requirements for users and services in large-scale networks remains a critical challenge. To address this issue, this paper proposes a Service-Driven Dynamic QoS On-Demand model and establishes a corresponding QoS optimization objective function. The SHA-256 hash algorithm is utilized to simplify the large-scale network model, effectively reducing the number of Segment Routing (SR) nodes. The proposed Service-Driven Dynamic QoS On-Demand Routing Algorithm (SDDRL) identifies the optimal path, which is then uniformly disseminated by the SDN controller, thereby addressing existing challenges in SDN-IoT networks. Compared to OSPF-based and DDQN-based algorithms, the SDDRL algorithm reduces the average delay by 53.85% and 31.63%, respectively. The proposed algorithm reduces the packet loss rate, improves the average network congestion degree and route calculation time compared to other existing algorithms, and it demonstrates superior performance in handling complex tasks.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107685"},"PeriodicalIF":6.2,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing IoT security: Assessing instantaneous communication trust to detect man-in-the-middle attacks","authors":"Rabeya Basri ,&nbsp;Gour Karmakar ,&nbsp;S.H. Shah Newaz ,&nbsp;Joarder Kamruzzaman ,&nbsp;Linh Nguyen ,&nbsp;Mohammad Mahabub Alam ,&nbsp;Muhammad Usman","doi":"10.1016/j.future.2025.107714","DOIUrl":"10.1016/j.future.2025.107714","url":null,"abstract":"<div><div>Communication trust is regarded as an effective tool to detect various dangerous cyber attacks, including Man-in-the-Middle (MITM) attacks and acts as a complement to zero trust. There exist some approaches in the literature to calculate inter-node communication trust in Wireless Sensor Networks (WSNs) and IoT networks and leverage it to detect attacks. In WSNs, since promiscuous communication mode is used in calculating inter-node communication trust, it is not suitable for IoT networks. For IoT, the packet forwarding behavior of edge nodes is used in calculating inter-node communication trust, which is limited to detect the MITM attacks effectively unless an edge node is compromised and acts as an MITM attacker. Additionally, these trust calculation mechanisms neither leverage communication channel characteristics nor the communication trust between sensor and edge nodes. Protecting IoT networks from various cyber attacks like MITM attacks requires the instantaneous trust calculation using communication channel characteristics. Since active MITM attacks incur delays, consideration of delay in trust calculation appears to be an effective means in identifying attacks. Neither end-to-end (E2E) delay nor delay due to attacks has been used in communication trust calculation in the existing literature. To bridge this research gap and detect active MITM attacks accurately and spontaneously, in this paper, a new conceptual model, named IPCTCM is introduced for instantaneous trust calculation of an IoT communication channel leveraging delay due to active MITM attacks. Two popular time-series data estimation tools, named Kalman filter and LSTM are used to estimate the expected E2E delay to identify delay due to attacks. Our proposed communication trust calculation model is validated using the data, generated by a testbed implementation in our IoT lab at Federation University Australia. Performance evaluation shows our proposed model achieves an attack detection accuracy of 98.9%, which outperforms an existing intrusion detection method with the improvement of 48.1% accuracy. Furthermore, our trust calculation method has broader applicability in other communication domains as well.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107714"},"PeriodicalIF":6.2,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MSCPR: A maintainable vector commitment-based stateless cryptocurrency system with privacy preservation and regulatory compliance","authors":"Xingyu Yang,&nbsp;Lei Xu,&nbsp;Liehuang Zhu","doi":"10.1016/j.future.2025.107713","DOIUrl":"10.1016/j.future.2025.107713","url":null,"abstract":"<div><div>In traditional account-based cryptocurrency systems, maintaining the <em>state</em> of all accounts consumes significant storage space. To reduce storage costs, recently some studies propose to leverage vector commitment schemes to design <em>stateless</em> cryptocurrency systems. In such systems, validators only need to store a commitment to the state vector to validate transactions. However, to prove membership in the state vector, each user is required to locally maintain a <em>position proof</em>. This introduces a burden as users need to update their proofs every time the commitment value changes. Additionally, existing stateless systems often include users’ account balances and transferred values in transactions explicitly, which compromises privacy. To address above issues, we propose a stateless cryptocurrency system based on a maintainable vector commitment scheme. In the proposed system, a bucketing technique is employed to simplify the proof update operations. And we leverage the homomorphic property of vector commitments to preserve the confidentiality of transactions. Furthermore, by constructing an anonymity set, transaction anonymity is ensured. To prevent adversaries from taking advantage of the anonymity, we design a predicate encryption-based regulation scheme. Through a series of simulations, we demonstrate that the proposed system is storage-efficient, with acceptable time overhead for privacy preservation and regulatory compliance.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107713"},"PeriodicalIF":6.2,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142990516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}