Pub Date : 2025-12-24DOI: 10.1016/j.jnca.2025.104425
Wanling Lin , Jou-Ming Chang , Xiao-Yan Li
In data center networks (DCNs), many value-added services involving multiple tenants and distributed sites can be naturally modeled as multi-party communication (MPC) processes, where participants exchange information across infrastructures to support collaborative computation. For MPC, it usually ensures that the private data of the multiple parties involved in the collaborative computation are safe, the computational results maintain acceptable accuracy, and all participants are in the same fair position in a distributed environment. This article considers an unexplored application based on MPC called the distributed adjustable computation scheme (DACS), which allows computation to be invoked when the collected data reaches a specified threshold in the communication. We developed a distributed algorithm using secure multi-protection routing to enable DACS. The proposed algorithm guarantees that each private data can be successfully delivered to the desired recipient even if any faulty component (including server or link) exists in the network. Also, no other than the destination can receive the complete private data. We implement DACS on highly scalable data center networks. Through simulation, experimental results show that DACS is highly reliable and achieves high security efficiency.
{"title":"DACS: Distributed adjustable computation scheme in highly scalable data center networks based on multi-protection routing","authors":"Wanling Lin , Jou-Ming Chang , Xiao-Yan Li","doi":"10.1016/j.jnca.2025.104425","DOIUrl":"10.1016/j.jnca.2025.104425","url":null,"abstract":"<div><div>In data center networks (DCNs), many value-added services involving multiple tenants and distributed sites can be naturally modeled as multi-party communication (MPC) processes, where participants exchange information across infrastructures to support collaborative computation. For MPC, it usually ensures that the private data of the multiple parties involved in the collaborative computation are safe, the computational results maintain acceptable accuracy, and all participants are in the same fair position in a distributed environment. This article considers an unexplored application based on MPC called the distributed adjustable computation scheme (DACS), which allows computation to be invoked when the collected data reaches a specified threshold in the communication. We developed a distributed algorithm using secure multi-protection routing to enable DACS. The proposed algorithm guarantees that each private data can be successfully delivered to the desired recipient even if any faulty component (including server or link) exists in the network. Also, no other than the destination can receive the complete private data. We implement DACS on highly scalable data center networks. Through simulation, experimental results show that DACS is highly reliable and achieves high security efficiency.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"247 ","pages":"Article 104425"},"PeriodicalIF":8.0,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823154","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}
Pub Date : 2025-12-22DOI: 10.1016/j.jnca.2025.104414
Huseyin Ozgur Kamali , Ali Berkay Gorgulu , Murat Karakus , Evrim Guler , Suleyman Uludag
The growing pressures of urbanization, vehicular proliferation, and fragmented parking infrastructure pose significant sustainability and mobility challenges in modern cities. In response, we present SPOSChain (Smart Parking Optimization with SDN and Blockchain), a novel Blockchain-enhanced and Software-Defined Networking (SDN)-based smart parking system that unifies independent parking providers under a decentralized, intelligent coordination framework. SPOSChain introduces a four-layer architecture integrating IoT, data, control, and blockchain layers, to ensure transparency, scalability, and real-time responsiveness. The core parking assignment task is formulated as a fairness-driven optimization problem, which is mathematically equivalent to a parallel job scheduling problem, known to be NP-hard, thereby necessitating the development of efficient heuristic strategies. To this end, we propose, adopt, and evaluate multiple heuristic and hybrid algorithms, including Local Search, Branch-and-Bound, and Genetic Search, culminating in a time-aware Hybrid Search model. Simulation results under diverse vehicle arrival distributions (uniform, normal, and exponential) demonstrate that our approach significantly reduces load imbalance, quantified via the Total of Differences metric, while improving responsiveness and maintaining scalability. SPOSChain not only enables equitable and efficient parking allocation but also supports sustainable urban mobility by reducing driver search time, CO emissions, and network overhead. These results underscore the transformative potential of programmable, decentralized parking systems in future smart city infrastructures.
城市化、车辆激增和零散的停车基础设施带来的日益增长的压力，对现代城市的可持续性和流动性构成了重大挑战。作为回应，我们提出了SPOSChain（基于SDN和区块链的智能停车优化），这是一种新型的基于区块链增强和软件定义网络（SDN）的智能停车系统，它将独立的停车提供商统一在一个分散的智能协调框架下。SPOSChain引入了集成物联网、数据、控制和区块链层的四层架构，以确保透明度、可扩展性和实时响应能力。核心停车分配任务是一个公平驱动的优化问题，它在数学上相当于一个并行作业调度问题，被称为NP-hard，因此需要开发有效的启发式策略。为此，我们提出、采用并评估了多种启发式和混合算法，包括局部搜索、分支定界和遗传搜索，最终形成了具有时间意识的混合搜索模型。不同车辆到达分布（均匀分布、正态分布和指数分布）下的仿真结果表明，我们的方法显著降低了负载不平衡（通过Total of Differences度量进行量化），同时提高了响应能力并保持了可扩展性。SPOSChain不仅能实现公平高效的停车分配，还能通过减少驾驶员搜索时间、二氧化碳排放和网络开销来支持可持续的城市交通。这些结果强调了可编程的、分散的停车系统在未来智慧城市基础设施中的变革潜力。
{"title":"Smart parking optimization with software defined networking and blockchain: SPOSChain","authors":"Huseyin Ozgur Kamali , Ali Berkay Gorgulu , Murat Karakus , Evrim Guler , Suleyman Uludag","doi":"10.1016/j.jnca.2025.104414","DOIUrl":"10.1016/j.jnca.2025.104414","url":null,"abstract":"<div><div>The growing pressures of urbanization, vehicular proliferation, and fragmented parking infrastructure pose significant sustainability and mobility challenges in modern cities. In response, we present <span>SPOSChain</span> (Smart Parking Optimization with SDN and Blockchain), a novel Blockchain-enhanced and Software-Defined Networking (SDN)-based smart parking system that unifies independent parking providers under a decentralized, intelligent coordination framework. <span>SPOSChain</span> introduces a four-layer architecture integrating IoT, data, control, and blockchain layers, to ensure transparency, scalability, and real-time responsiveness. The core parking assignment task is formulated as a fairness-driven optimization problem, which is mathematically equivalent to a parallel job scheduling problem, known to be NP-hard, thereby necessitating the development of efficient heuristic strategies. To this end, we propose, adopt, and evaluate multiple heuristic and hybrid algorithms, including Local Search, Branch-and-Bound, and Genetic Search, culminating in a time-aware Hybrid Search model. Simulation results under diverse vehicle arrival distributions (uniform, normal, and exponential) demonstrate that our approach significantly reduces load imbalance, quantified via the Total of Differences metric, while improving responsiveness and maintaining scalability. <span>SPOSChain</span> not only enables equitable and efficient parking allocation but also supports sustainable urban mobility by reducing driver search time, CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> emissions, and network overhead. These results underscore the transformative potential of programmable, decentralized parking systems in future smart city infrastructures.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"247 ","pages":"Article 104414"},"PeriodicalIF":8.0,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145813948","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}
With the rapid development of science and technology, wireless sensor networks (WSN) are increasingly applied in environmental monitoring, industrial control, and smart cities. However, WSN deployment faces three core challenges that existing algorithms fail to address comprehensively. (1) Insufficient coverage precision. Traditional optimization algorithms (e.g., NSGA-II, MOPSO) often leave local coverage holes due to limited fine-grained search capability. (2) High resource redundancy. Fixed grid or weight-based methods (e.g., MOEA/D) cannot dynamically adjust node distribution according to environmental density, leading to redundant deployment. (3) Unbalanced energy consumption. Single-objective or simplified multi-objective approaches ignore the trade-off between coverage, waste rate, and energy consumption, shortening network lifetime. To tackle these issues, a multi-objective parrot optimizer (MOPO) based on improved Lévy flight and an adaptive elliptical segmentation screening mechanism is proposed for WSN deployment optimization. The randomness of original Lévy flight causes large step-length jumps, making fine-grained searches difficult. Thus, a Sigmoid perturbation mechanism is integrated into Lévy flight to enhance local search accuracy while preserving global exploration. Based on this improvement, an elite non-dominated strategy is combined with an adaptive grid (dynamic adjustment by solution density) and elliptical segmentation selection—this ensures retention of optimal individuals in high-density areas, maintains population diversity, and accelerates exploration of sparse regions. An external archive further preserves a uniform and diverse Pareto solution set. MOPO is tested in obstacle-free/obstacle WSN models with coverage, waste rate, and energy consumption rate as objectives. Comparative experiments with NSGA-II, MOPSO, and MOGWO in different monitoring areas show MOPO ranks first in all Friedman tests. A real-world test (41°10′20″N, 29°04′30″E, 1320 × 610 m2) achieves 94 % target coverage. This proves MOPO effectively solves the three core challenges of WSN deployment, providing a practical and efficient optimization method for large-scale, resource-constrained WSN scenarios.
{"title":"Multi-objective parrot optimizer with improved Lévy flight and adaptive elliptical segmentation - based screening mechanism for layout optimization of wireless sensor networks","authors":"Yun-Hao Zhang, Jie-Sheng Wang, Yu-Xuan Xing, Yu-Feng Sun, Si-Wen Zhang, Xue-Lian Bai","doi":"10.1016/j.jnca.2025.104413","DOIUrl":"10.1016/j.jnca.2025.104413","url":null,"abstract":"<div><div>With the rapid development of science and technology, wireless sensor networks (WSN) are increasingly applied in environmental monitoring, industrial control, and smart cities. However, WSN deployment faces three core challenges that existing algorithms fail to address comprehensively. (1) Insufficient coverage precision. Traditional optimization algorithms (e.g., NSGA-II, MOPSO) often leave local coverage holes due to limited fine-grained search capability. (2) High resource redundancy. Fixed grid or weight-based methods (e.g., MOEA/D) cannot dynamically adjust node distribution according to environmental density, leading to redundant deployment. (3) Unbalanced energy consumption. Single-objective or simplified multi-objective approaches ignore the trade-off between coverage, waste rate, and energy consumption, shortening network lifetime. To tackle these issues, a multi-objective parrot optimizer (MOPO) based on improved Lévy flight and an adaptive elliptical segmentation screening mechanism is proposed for WSN deployment optimization. The randomness of original Lévy flight causes large step-length jumps, making fine-grained searches difficult. Thus, a Sigmoid perturbation mechanism is integrated into Lévy flight to enhance local search accuracy while preserving global exploration. Based on this improvement, an elite non-dominated strategy is combined with an adaptive grid (dynamic adjustment by solution density) and elliptical segmentation selection—this ensures retention of optimal individuals in high-density areas, maintains population diversity, and accelerates exploration of sparse regions. An external archive further preserves a uniform and diverse Pareto solution set. MOPO is tested in obstacle-free/obstacle WSN models with coverage, waste rate, and energy consumption rate as objectives. Comparative experiments with NSGA-II, MOPSO, and MOGWO in different monitoring areas show MOPO ranks first in all Friedman tests. A real-world test (41°10′20″N, 29°04′30″E, 1320 × 610 m<sup>2</sup>) achieves 94 % target coverage. This proves MOPO effectively solves the three core challenges of WSN deployment, providing a practical and efficient optimization method for large-scale, resource-constrained WSN scenarios.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"247 ","pages":"Article 104413"},"PeriodicalIF":8.0,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145813864","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}
Pub Date : 2025-12-19DOI: 10.1016/j.jnca.2025.104416
Zubaida Rehman , Iqbal Gondal , Hai Dong , Mengmeng Ge , Mark A. Gregory , Ikram ul Haq
Eclipse attacks, which isolate victim nodes by monopolizing their peer connections, remain a critical threat to Ethereum’s consensus mechanism. To address this, we present a principled framework for detecting Eclipse attacks in Ethereum peer-to-peer networks, grounded in a formal adversarial model. Existing defenses are either ad-hoc or lack provable guarantees, leaving open questions about their reliability under adaptive adversaries. Our work aims to bridge this gap by formally defining eclipse attack detection as a security property. We specify soundness, completeness, and robustness theorems under bounded adversarial drift, and derive formal guarantees within false positive and false negative bounds, resilience to adversarial manipulation, and multi-node compositional reliability. We then instantiate a lightweight detection framework that maps packet-level traffic features to predictions using ensemble classifiers (Random Forest, XGBoost). The system was validated using a controlled Ethereum testbed and extended with CTGAN-generated synthetic traces to emulate networks of up to 100 nodes. Empirical evaluation shows that our framework achieves up to 96% F1-score with sub-second inference latency, well within Ethereum’s 12-second Proof-of-Stake validator time slots. These findings demonstrate that lightweight statistical features, when coupled with formal analysis, enable accurate, efficient, and scalable detection of network-level partitioning attacks. Our work establishes a deployable and theoretically grounded defense foundation for securing modern blockchain systems against eclipse adversaries.
{"title":"A robust eclipse attack detection framework for Ethereum networks","authors":"Zubaida Rehman , Iqbal Gondal , Hai Dong , Mengmeng Ge , Mark A. Gregory , Ikram ul Haq","doi":"10.1016/j.jnca.2025.104416","DOIUrl":"10.1016/j.jnca.2025.104416","url":null,"abstract":"<div><div>Eclipse attacks, which isolate victim nodes by monopolizing their peer connections, remain a critical threat to Ethereum’s consensus mechanism. To address this, we present a principled framework for detecting Eclipse attacks in Ethereum peer-to-peer networks, grounded in a formal adversarial model. Existing defenses are either ad-hoc or lack provable guarantees, leaving open questions about their reliability under adaptive adversaries. Our work aims to bridge this gap by formally defining eclipse attack detection as a security property. We specify soundness, completeness, and robustness theorems under bounded adversarial drift, and derive formal guarantees within false positive and false negative bounds, resilience to adversarial manipulation, and multi-node compositional reliability. We then instantiate a lightweight detection framework that maps packet-level traffic features to predictions using ensemble classifiers (Random Forest, XGBoost). The system was validated using a controlled Ethereum testbed and extended with CTGAN-generated synthetic traces to emulate networks of up to 100 nodes. Empirical evaluation shows that our framework achieves up to 96% F1-score with sub-second inference latency, well within Ethereum’s 12-second Proof-of-Stake validator time slots. These findings demonstrate that lightweight statistical features, when coupled with formal analysis, enable accurate, efficient, and scalable detection of network-level partitioning attacks. Our work establishes a deployable and theoretically grounded defense foundation for securing modern blockchain systems against eclipse adversaries.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"246 ","pages":"Article 104416"},"PeriodicalIF":8.0,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785055","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}
Pub Date : 2025-12-19DOI: 10.1016/j.jnca.2025.104412
Kaifeng Hua , Shengchao Su , Nannan Zhang
With the growing demand for dynamic resources in the Internet of Vehicles, service migration has become essential for maintaining user service continuity. However, existing techniques often transfer redundant dirty page data during operation state file transfers, leading to high network traffic and significant migration delays, which are unsuitable for the low latency and low traffic requirements of intelligent transportation scenarios. To overcome this issue, this paper proposes a Intelligent Adaptive Container Migration Technique called IACMT, which is based on dynamic filtering of dirty pages with two-stage compression. IACMT features a dirty page filtering mechanism that intelligently identifies active dirty pages by monitoring the frequency of page accesses and modification patterns in real time. This mechanism facilitates the delayed transmission of less critical dirty pages, effectively reducing the data size during the iterative transmission phase. Furthermore, it incorporates a two-stage data compression algorithm that employs run-length encoding (RLE) followed by dynamic Huffman coding. In the initial stage, RLE eliminates redundant byte sequences in the state file. The subsequent output is then adaptively compressed using a dynamic Huffman tree, improving compression efficiency while managing computational overhead. Experimental results show that IACMT reduces data transmission volume by approximately 35 % for typical in-vehicle workloads, while cutting migration time and service interruption duration by around 24 % and 34 %, respectively.
{"title":"Seamless service migration for the Internet of Vehicles in edge computing: A dynamic dirty page filtering and two-stages compression technique","authors":"Kaifeng Hua , Shengchao Su , Nannan Zhang","doi":"10.1016/j.jnca.2025.104412","DOIUrl":"10.1016/j.jnca.2025.104412","url":null,"abstract":"<div><div>With the growing demand for dynamic resources in the Internet of Vehicles, service migration has become essential for maintaining user service continuity. However, existing techniques often transfer redundant dirty page data during operation state file transfers, leading to high network traffic and significant migration delays, which are unsuitable for the low latency and low traffic requirements of intelligent transportation scenarios. To overcome this issue, this paper proposes a <u>I</u>ntelligent <u>A</u>daptive <u>C</u>ontainer <u>M</u>igration <u>T</u>echnique called IACMT, which is based on dynamic filtering of dirty pages with two-stage compression. IACMT features a dirty page filtering mechanism that intelligently identifies active dirty pages by monitoring the frequency of page accesses and modification patterns in real time. This mechanism facilitates the delayed transmission of less critical dirty pages, effectively reducing the data size during the iterative transmission phase. Furthermore, it incorporates a two-stage data compression algorithm that employs run-length encoding (RLE) followed by dynamic Huffman coding. In the initial stage, RLE eliminates redundant byte sequences in the state file. The subsequent output is then adaptively compressed using a dynamic Huffman tree, improving compression efficiency while managing computational overhead. Experimental results show that IACMT reduces data transmission volume by approximately 35 % for typical in-vehicle workloads, while cutting migration time and service interruption duration by around 24 % and 34 %, respectively.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"247 ","pages":"Article 104412"},"PeriodicalIF":8.0,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785053","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}
Pub Date : 2025-12-19DOI: 10.1016/j.jnca.2025.104415
Junfeng Tian , Yujian Wu , Jin Tian , Liuling Qi
Sharding is a widely adopted technique for enhancing blockchain scalability by partitioning the network into multiple sub-shards, thereby distributing the workload. However, traditional dynamic sharding schemes often suffer from delayed adjustments to the shard count, hindering rapid convergence to an optimal workload distribution in practical deployments and limiting overall scalability. Furthermore, the dynamic participation of nodes is frequently overlooked. To address these challenges, this paper proposes PolyembryonyChain (PE-Chain), a novel and efficient hierarchical sharding architecture designed for environments with dynamically participating nodes, with the goal of achieving elastic scalability. Its core innovations are twofold. First, it introduces a dynamic threshold sharding algorithm that adaptively adjusts the number of shards and optimizes the network topology to enable elastic expansion. Second, it incorporates a validator assignment and reconfiguration scheme, specifically tailored for dynamic hierarchical environments, to ensure system security. A comprehensive security analysis and simulation results demonstrate that PE-Chain significantly outperforms state-of-the-art baselines, achieving approximately 28% higher throughput while maintaining low latency, which underscores its superior scalability and practical value.
{"title":"PE-Chain: An efficient hierarchical sharding architecture for dynamically participating node models","authors":"Junfeng Tian , Yujian Wu , Jin Tian , Liuling Qi","doi":"10.1016/j.jnca.2025.104415","DOIUrl":"10.1016/j.jnca.2025.104415","url":null,"abstract":"<div><div>Sharding is a widely adopted technique for enhancing blockchain scalability by partitioning the network into multiple sub-shards, thereby distributing the workload. However, traditional dynamic sharding schemes often suffer from delayed adjustments to the shard count, hindering rapid convergence to an optimal workload distribution in practical deployments and limiting overall scalability. Furthermore, the dynamic participation of nodes is frequently overlooked. To address these challenges, this paper proposes PolyembryonyChain (PE-Chain), a novel and efficient hierarchical sharding architecture designed for environments with dynamically participating nodes, with the goal of achieving elastic scalability. Its core innovations are twofold. First, it introduces a dynamic threshold sharding algorithm that adaptively adjusts the number of shards and optimizes the network topology to enable elastic expansion. Second, it incorporates a validator assignment and reconfiguration scheme, specifically tailored for dynamic hierarchical environments, to ensure system security. A comprehensive security analysis and simulation results demonstrate that PE-Chain significantly outperforms state-of-the-art baselines, achieving approximately 28% higher throughput while maintaining low latency, which underscores its superior scalability and practical value.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"246 ","pages":"Article 104415"},"PeriodicalIF":8.0,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785051","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}
Pub Date : 2025-12-19DOI: 10.1016/j.jnca.2025.104417
Shuang Zeng, Haitao Zhang, Zezhong Yan
Service Level Objective (SLO) assignment involves distributing an application’s end-to-end latency SLO among its microservices, guiding each microservice’s resource allocation based on its assigned sub-SLO. However, existing SLO assignment methods primarily focus on horizontal scaling, overlooking the significant impact of varying resource usage contexts across nodes and container configurations on microservice latency characteristics. Moreover, these methods fail to consider how scaling decisions affect node resource usage. This oversight creates discrepancies between decision-time and runtime latency characteristics, which can lead to SLO violations or resource wastage. This paper proposes CASLO, a joint scaling and deployment method based on context-aware SLO assignment that aims to meet application SLOs with minimal resource usage. It characterizes microservice latency by categorizing influencing factors into node and container contexts, which enables the model to capture dynamic performance under varying conditions. Building on this characterization, CASLO employs Particle Swarm Optimization (PSO) to iteratively estimate each microservice’s tolerance to contextual resource conditions. For each tolerance, it determines the resource usage of each node post-scaling and deployment, addressing discrepancies of latency characteristics between decision-time and runtime and distinguishing latency characteristics across nodes. Based on the determined resource context, CASLO assigns SLOs to each microservice, dynamically configuring container resources to derive scaling and deployment decisions. Resource usage is then calculated to provide feedback to PSO for iterative optimization. Compared to state-of-the-art methods, CASLO achieves 32% reduction in resource usage and decreases the frequency of SLO violations by 61%.
{"title":"CASLO: Joint scaling and deployment for microservices leveraging context-aware SLO assignment","authors":"Shuang Zeng, Haitao Zhang, Zezhong Yan","doi":"10.1016/j.jnca.2025.104417","DOIUrl":"10.1016/j.jnca.2025.104417","url":null,"abstract":"<div><div>Service Level Objective (SLO) assignment involves distributing an application’s end-to-end latency SLO among its microservices, guiding each microservice’s resource allocation based on its assigned sub-SLO. However, existing SLO assignment methods primarily focus on horizontal scaling, overlooking the significant impact of varying resource usage contexts across nodes and container configurations on microservice latency characteristics. Moreover, these methods fail to consider how scaling decisions affect node resource usage. This oversight creates discrepancies between decision-time and runtime latency characteristics, which can lead to SLO violations or resource wastage. This paper proposes CASLO, a joint scaling and deployment method based on context-aware SLO assignment that aims to meet application SLOs with minimal resource usage. It characterizes microservice latency by categorizing influencing factors into node and container contexts, which enables the model to capture dynamic performance under varying conditions. Building on this characterization, CASLO employs Particle Swarm Optimization (PSO) to iteratively estimate each microservice’s tolerance to contextual resource conditions. For each tolerance, it determines the resource usage of each node post-scaling and deployment, addressing discrepancies of latency characteristics between decision-time and runtime and distinguishing latency characteristics across nodes. Based on the determined resource context, CASLO assigns SLOs to each microservice, dynamically configuring container resources to derive scaling and deployment decisions. Resource usage is then calculated to provide feedback to PSO for iterative optimization. Compared to state-of-the-art methods, CASLO achieves 32% reduction in resource usage and decreases the frequency of SLO violations by 61%.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"247 ","pages":"Article 104417"},"PeriodicalIF":8.0,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785050","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}
Pub Date : 2025-12-15DOI: 10.1016/j.jnca.2025.104394
Durgesh Singh , Sasthi C. Ghosh
Millimeter wave (mmWave) D2D communication is highly vulnerable to blockages from dynamic obstacles leading to severe packet loss and deterioration in quality of service. User equipments (UEs) acting as relays are chosen to divert the communication path in case of blockage. However, relays themselves are vulnerable to be blocked due to their and obstacles’ motion. Thus mobility parameters of dynamic obstacles must be captured effectively, but it is challenging as they might not be connected to the network. Additionally, instantaneous change in their orientation in motion is difficult to measure which might vary abruptly compared to their speed which can be sensed with high accuracy. A probabilistic model is developed considering the obstacle’s orientation in motion is unknown. Later the geometrical structure of the problem is analyzed and then exploited to devise a novel technique to derive closed form blockage expression for a given pair of UE. The proposed technique can be applied across exhaustive scenarios of motion of a given UE pair which may get blocked due to dynamic obstacles. Through extensive simulations, we have observed that our proposed approach outperforms both classical received signal strength (RSS) based approach and two recent state of the art approaches. We have also validated our results against an oracle which has complete speed and orientation information regarding UEs and obstacles.
{"title":"Network-assisted relay selection in mmWave D2D communication under presence of dynamic obstacles with unknown orientation","authors":"Durgesh Singh , Sasthi C. Ghosh","doi":"10.1016/j.jnca.2025.104394","DOIUrl":"10.1016/j.jnca.2025.104394","url":null,"abstract":"<div><div>Millimeter wave (mmWave) D2D communication is highly vulnerable to blockages from dynamic obstacles leading to severe packet loss and deterioration in quality of service. User equipments (UEs) acting as relays are chosen to divert the communication path in case of blockage. However, relays themselves are vulnerable to be blocked due to their and obstacles’ motion. Thus mobility parameters of dynamic obstacles must be captured effectively, but it is challenging as they might not be connected to the network. Additionally, instantaneous change in their orientation in motion is difficult to measure which might vary abruptly compared to their speed which can be sensed with high accuracy. A probabilistic model is developed considering the obstacle’s orientation in motion is <em>unknown</em>. Later the geometrical structure of the problem is analyzed and then exploited to devise a novel technique to derive closed form blockage expression for a given pair of UE. The proposed technique can be applied across exhaustive scenarios of motion of a given UE pair which may get blocked due to dynamic obstacles. Through extensive simulations, we have observed that our proposed approach outperforms both classical received signal strength (RSS) based approach and two recent state of the art approaches. We have also validated our results against an oracle which has complete speed and orientation information regarding UEs and obstacles.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"246 ","pages":"Article 104394"},"PeriodicalIF":8.0,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145760433","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}
Pub Date : 2025-12-12DOI: 10.1016/j.jnca.2025.104400
Muhammad Nawaz Khan , Sokjoon Lee , Tariq Hussain , Razaz Waheeb Attar , Mohsin Shah , Amal Hassan Alhazmi
In the digital age of the Internet of Things (IoT), there is a significant shift from traditional computing to an ubiquitous, highly connected, and automated world that provides services to anyone, anywhere, and at any time. An IoT-based system is an unstable network characterized by fluctuating dynamics and fragile connections, resulting in lower performance on congestion, latency, and energy consumption. To effectively manage these functional parameters and implement a dynamic scheduling mechanism, this paper presents a novel scheme, “Dynamic and Adaptive Scheduling of Cognitive Sensors (DASCS) for Collaborative Target Tracking in Energy-Efficient IoT Environments”. In this approach, cognitive sensors dynamically schedule their functions according to their role in the wireless mesh grid and adapt to new states by checking the network traffic conditions. Its dual goals involve reducing network traffic to significantly decrease energy consumption and enhancing network performance by equally distributing energy resources throughout the grid. Furthermore, it works in object detection and monitors the direction of movement within the IoT environment. DASCS improves network performance by increasing packet delivery ratios by 2.31% at the base station and 27.92% at the cluster head, while adding more live sensors, it improves network stability by 38.46%. DASCS also enhances energy efficiency by increasing the average residual energy by 68.8% compared to other benchmark schemes while maintaining a high event detection rate and a low false alarm rate.
{"title":"Dynamic and Adaptive Scheduling of Cognitive Sensors for collaborative target tracking in energy-efficient IOT environments","authors":"Muhammad Nawaz Khan , Sokjoon Lee , Tariq Hussain , Razaz Waheeb Attar , Mohsin Shah , Amal Hassan Alhazmi","doi":"10.1016/j.jnca.2025.104400","DOIUrl":"10.1016/j.jnca.2025.104400","url":null,"abstract":"<div><div>In the digital age of the Internet of Things (IoT), there is a significant shift from traditional computing to an ubiquitous, highly connected, and automated world that provides services to anyone, anywhere, and at any time. An IoT-based system is an unstable network characterized by fluctuating dynamics and fragile connections, resulting in lower performance on congestion, latency, and energy consumption. To effectively manage these functional parameters and implement a dynamic scheduling mechanism, this paper presents a novel scheme, “Dynamic and Adaptive Scheduling of Cognitive Sensors (DASCS) for Collaborative Target Tracking in Energy-Efficient IoT Environments”. In this approach, cognitive sensors dynamically schedule their functions according to their role in the wireless mesh grid and adapt to new states by checking the network traffic conditions. Its dual goals involve reducing network traffic to significantly decrease energy consumption and enhancing network performance by equally distributing energy resources throughout the grid. Furthermore, it works in object detection and monitors the direction of movement within the IoT environment. DASCS improves network performance by increasing packet delivery ratios by 2.31% at the base station and 27.92% at the cluster head, while adding more live sensors, it improves network stability by 38.46%. DASCS also enhances energy efficiency by increasing the average residual energy by 68.8% compared to other benchmark schemes while maintaining a high event detection rate and a low false alarm rate.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"248 ","pages":"Article 104400"},"PeriodicalIF":8.0,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145731203","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}
Pub Date : 2025-12-08DOI: 10.1016/j.jnca.2025.104409
Khaoula Hidawi , Sabrine Ennaji , Elena Ferrari
This paper introduces BlackoutADR, a novel adversarial attack exploiting LoRaWAN’s Adaptive Data Rate (ADR) mechanism in cellular-connected UAV networks, with applicability to other IoT systems as well. By subtly manipulating Received Signal Strength Indicator (RSSI) and Signal-to-Noise Ratio (SNR), BlackoutADR increases UAV transmission power, causing 45% faster battery depletion within 100 s of simulation time and disrupting network operations. Using NS-3 simulations with a 20-UAV FANET, we evaluate its evasion of multiple ML-based IDSs (CNN, LSTM, BiLSTM, FNN, LoRaWAN-specific). Results show BlackoutADR remains undetected due to its subtle manipulations evading even dynamic thresholds, outperforming traditional jamming attacks. To address the identified vulnerability, we outline reactive measures, including dynamic threshold-based IDSs, secure ADR mechanisms, and recommendations for drone manufacturers.
{"title":"BlackoutADR: Exploiting adaptive data rate vulnerabilities in LoRaWAN-based FANETs","authors":"Khaoula Hidawi , Sabrine Ennaji , Elena Ferrari","doi":"10.1016/j.jnca.2025.104409","DOIUrl":"10.1016/j.jnca.2025.104409","url":null,"abstract":"<div><div>This paper introduces <em>BlackoutADR</em>, a novel adversarial attack exploiting LoRaWAN’s Adaptive Data Rate (ADR) mechanism in cellular-connected UAV networks, with applicability to other IoT systems as well. By subtly manipulating Received Signal Strength Indicator (RSSI) and Signal-to-Noise Ratio (SNR), <em>BlackoutADR</em> increases UAV transmission power, causing 45% faster battery depletion within 100 s of simulation time and disrupting network operations. Using NS-3 simulations with a 20-UAV FANET, we evaluate its evasion of multiple ML-based IDSs (CNN, LSTM, BiLSTM, FNN, LoRaWAN-specific). Results show <em>BlackoutADR</em> remains undetected due to its subtle manipulations evading even dynamic thresholds, outperforming traditional jamming attacks. To address the identified vulnerability, we outline reactive measures, including dynamic threshold-based IDSs, secure ADR mechanisms, and recommendations for drone manufacturers.</div></div>","PeriodicalId":54784,"journal":{"name":"Journal of Network and Computer Applications","volume":"246 ","pages":"Article 104409"},"PeriodicalIF":8.0,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705027","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号