Multi-subject video generation aims to synthesize videos from textual prompts and multiple reference images, ensuring that each subject preserves natural scale and visual fidelity. However, current methods face two challenges: scale inconsistency, where variations in subject size lead to unnatural generation, and permutation sensitivity, where the order of reference inputs causes subject distortion. In this paper, we propose MoFu, a unified framework that tackles both challenges. For scale inconsistency, we introduce Scale-Aware Modulation (SMO), an LLM-guided module that extracts implicit scale cues from the prompt and modulates features to ensure consistent subject sizes. To address permutation sensitivity, we present a simple yet effective Fourier Fusion strategy that processes the frequency information of reference features via the Fast Fourier Transform to produce a unified representation. Besides, we design a Scale-Permutation Stability Loss to jointly encourage scale-consistent and permutation-invariant generation. To further evaluate these challenges, we establish a dedicated benchmark with controlled variations in subject scale and reference permutation. Extensive experiments demonstrate that MoFu significantly outperforms existing methods in preserving natural scale, subject fidelity, and overall visual quality.

Personalized image generation is crucial for improving the user experience, as it renders reference images into preferred ones according to user visual preferences. Although effective, existing methods face two main issues. First, existing methods treat all items in the user's historical sequence equally when extracting user preferences, overlooking the varying semantic similarities between historical items and the reference item. Disproportionately high weights for low-similarity items distort user visual preferences for the reference item. Second, existing methods heavily rely on consistency between generated and reference images to optimize generation, which leads to underfitting user preferences and hinders personalization. To address these issues, we propose Retrieval Augmented Personalized Image GenerAtion guided by Recommendation (RAGAR). Our approach uses a retrieval mechanism to assign different weights to historical items according to their similarities to the reference item, thereby extracting more refined users' visual preferences for the reference item. Then we introduce a novel rank task based on the multi-modal ranking model to optimize the personalization of the generated images instead of forcing depend on consistency. Extensive experiments and human evaluations on three real-world datasets demonstrate that RAGAR achieves significant improvements in both personalization and semantic metrics compared to five baselines.

Multi-graph multi-label learning (MGML) represents each object as a bag-of-graphs with multiple labels, but demands large-scale labeled data whose acquisition is often difficult and costly. Self-supervised contrastive learning (SCL) mitigates label dependence by leveraging data augmentation to construct discriminative pretext tasks, proving effective for multi-instance learning. However, when applied to MGML, SCL faces two key challenges: (1) it distinguishes individual instances by their differences, whereas MGML requires modeling label correlations; (2) it assumes semantic invariance under augmentation, but structural perturbations in MGML alter label semantics. To tackle these challenges, we propose a self-suPervised contrastive rE-learning framework for mulTi-grAph multi-labeL classification (PETAL). Specifically, to model label correlations, we first define a unified label space to learn label prototypes and align features with them, yielding prototype-aligned representations. We then design a multi-granularity contrastive loss over these representations, which captures label dependencies by contrasting at the bag level, graph level, and bag-graph level. Moreover, to ensure semantic invariance, we develop a contrastive re-learning strategy based on prototype-aligned representations to generate augmentation-free positive samples. This guarantees consistent multi-label distributions without structural perturbations. Experiments on six datasets demonstrate that PETAL achieves an average improvement of 4.12% over state-of-the-art self-supervised and supervised baselines.

Although camouflaged object segmentation has advanced rapidly in recent years, existing methods are still confined to visual mask prediction under fixed task assumptions. They cannot interactively respond to user requests, nor can they proactively understand and reason about the user’s intent. Our work tackles this issue by proposing a novel task, Language-Guided Reasoning Camouflaged Object Segmentation (LRCOS). Given a camouflaged image and an implicit query text instruction that requires reasoning, LRCOS aims to output intent-consistent segmentation mask. To establish a benchmark for this task, we build CamoQuery, comprising 12,437 image–mask samples and 25971 implicit query text instructions. To better reflect real-world camouflaged scenarios, we additionally collect MCD, a multi-instance camouflage dataset where multiple camouflaged targets co-exist within the same scene, increasing the need for reasoning. Building on CamoQuery, we further propose COSA, a vision–language segmentation assistant that segments the intended camouflaged object from implicit queries and produces a reasoning explanation. Experiments on CamoQuery demonstrate that COSA has strong reasoning segmentation capability in camouflaged scenes and exhibits zero-shot capability.

Lifelong sequential recommendation aims to model users' long-term interests by leveraging their entire interaction history, but the high computational overhead caused by ultra-long sequences poses a core challenge. To address this, existing methods generally adopt the subsequence learning strategies to shorten the input sequence, which can be divided into two categories: (1) Sequence compression methods compress long sequences into multiple subsequence representations through strategies such as uniform segmentation and clustering; (2) Top-k retrieval methods filter the subsequence related to the target item from long sequences via target attention mechanisms or retrieval mechanisms and learn its representation. However, these two types of methods still face challenges in subsequence representation learning: (1) Sequence compression methods struggle to simultaneously balance the high similarity of items within subsequences and smooth temporal continuity (i.e., small temporal intervals between adjacent items), resulting in incorrect learning of subsequence representations; (2) Top-k retrieval methods lose a large amount of effective context information when the length of the retrieved subsequence is much smaller than the original sequence, resulting in incomplete sequence representations. To overcome these challenges, we propose a novel lifelong sequential recommendation method with adaptive subsequence compression and contextual fusion. Specifically, an adaptive subsequence compression module is first designed: it utilizes gradient policy sampling to achieve adaptive segmentation of subsequences, thereby retaining their temporal continuity, and introduces a reward function to enhance the similarity of items within subsequences. Second, a subsequence context fusion module is constructed: it leverages causal cross-attention to fuse recent interactions with their subsequence context and to capture correlations among recent items, thereby learning more complete and accurate sequence representations. We conduct extensive experiments on three public long-sequence recommendation datasets. Experimental results demonstrate that our proposed method consistently outperforms a variety of strong baselines in both predictive accuracy and computational efficiency.

Multi-Graph Learning (MGL) is a fundamental machine learning paradigm that represents objects as bags-of-graphs, each encoding a distinct structural property, and has broad applications in bioinformatics, chemistry, computing power networks, and software defect detection. However, the inherent scarcity of labeled data poses a significant bottleneck for supervised MGL approaches. While self-supervised contrastive learning offers a compelling solution, its direct application to MGL faces three key challenges: (1) existing graph neural networks, primarily for single-graph modeling, struggle to yield discriminative bag-level representations from bags-of-graphs; (2) conventional contrastive objectives are limited to single-level settings, failing to capture cross-hierarchical dependencies; and (3) standard data augmentation often disrupts intrinsic graph and bag structures, undermining semantic consistency. To address these issues, we propose the Hierarchical Graph-Bag-Network (HGBN), a self-supervised MGL framework that constructs hierarchical representations in the form of a graph-bag-network. HGBN employs an asymmetric hierarchical graph neural network to learn discriminative graph-level and bag-level representations, introduces cross-hierarchical contrastive objectives to align graph-level and bag-level semantics, and leverages the asymmetric network outputs to form positive and negative pairs, preserving intrinsic structural and semantic consistency. Experiments on eight benchmark multi-graph datasets demonstrate that HGBN consistently outperforms both supervised and self-supervised state-of-the-art baselines, achieving average improvements of 4.82% in accuracy and F1 score.

The proliferation of Unmanned Aerial Vehicles (UAVs) in 6G-enabled edge computing presents a dual challenge of ensuring secure data transmission and performing accurate object detection on resource-constrained platforms. This paper proposes a Synergistic Encryption and Detection framework (SED-UAV) that integrates a lightweight chaotic image encryption algorithm, LB-ICA, and an enhanced small object detector, YOLO-LFP. The LB-ICA algorithm leverages an improved chaotic map and a plaintext-aware key mechanism using SHA-256. It achieves a key space of over 2100 and an average encryption speed of 20.3 ms per block, providing robust security against differential and chosen-plaintext attacks with high efficiency suitable for UAVs. For detection, the YOLO-LFP model enhances the YOLOv8 baseline with a hybrid attention module, an adaptive feature fusion strategy, and a dedicated small object head. Experimental results on the VisDrone2019 benchmark show YOLO-LFP achieves a state-of-the-art performance of 44.7% [email protected], significantly outperforming existing models. This work provides a comprehensive solution for deploying secure, real-time computer vision applications on UAV platforms.

Data augmentation is an effective technique for tackling data sparsity in sequential recommendation (SR). Existing methods generate new data during the model training to improve the performance. However, deploying them on a backbone model requires retraining, architecture modification, or introducing additional modules and learnable parameters. These processes are time-consuming and costly for well-trained models, especially when the model and data scales become large. In this work, we explore the test-time augmentation (TTA) for SR, which augments the input sequences during the inference phase and then fuses the model's predictions to improve final accuracy. It avoids the significant overhead associated with training-time augmentation. We first experimentally examine the potential of existing augmentation operators for TTA and find that the Substitute and Mask consistently achieve better performance. Further analysis reveals that these two operators retain the original sequential pattern while adding appropriate perturbations. Moreover, the random selection of augmentation positions creates suitable augmented samples from both semantic and temporal perspectives. Meanwhile, we find that the fixed operation ratio limits the diversity of augmented data, and the TTA may impair the model's performance on long sequences. In addition, the two operators still face time-consuming similarity-based item selection or interference from mask tokens. Based on the analysis and limitations, we present TNoise and TMask. The former injects uniform noise into the representation, avoiding the computational overhead of item selection. The latter blocks mask tokens from participating in model calculations (TMask-B) or directly removes interactions that should have been replaced with mask tokens (TMask-R). Further, we sample the augmentation ratio from a uniform distribution to improve the data diversity. For short sequences, we introduce a sequence smoothing and lengthening method based on inter-item interpolation. For long sequences, we set a threshold to avoid the negative effects of TTA. Comprehensive experiments demonstrate the effectiveness, efficiency, and generalizability of our method.

Nowadays, the development of Collaborative Edge Computing (CEC) has greatly accelerated deep neural network (DNN) inference by enabling edge service providers (ESPs) in the JointCloud federation to deploy computational resources and well-trained DNN models at the edge of the network for collaborative inference with mobile devices (MDs), thereby promoting the rapid advancement of intelligent applications. This raises the need for an effective DNN inference offloading mechanism between MDs and ESPs. However, existing schemes often lack market efficiency and fail to ensure desirable economic properties. To this end, we propose a truthful double auction mechanism for delay-aware DNN inference offloading (TDAD), which integrates a dynamic programming approach with an adaptive resource allocation and pricing strategy to maximize social welfare while ensuring truthfulness, budget balance, and individual rationality. Specifically, TDAD first employs a binary search-based delay-aware partitioning and offloading method to determine the minimum feasible resource profile for each MD, and then applies a dynamic programming-based double auction to match MDs' inference demands with ESPs' combinatorial resources, and compute the corresponding payments and rewards. Theoretical analysis proves that TDAD satisfies the desired economic properties, while experiments in realistic CEC environments validate its effectiveness and efficiency.

The core function of quantum networks is to establish high-fidelity quantum entanglement for long-distance communication. However, the main challenge is to efficiently allocate resources under limited conditions, maximize throughput, satisfy end-to-end (E2E) fidelity requirements, and prevent quantum decoherence caused by inefficient routing algorithms. Current research focuses on optimizing either throughput or fidelity, with a lack of approaches that optimize both simultaneously; furthermore, existing algorithms suffer from high computational complexity. To tackle these challenges, this study proposes a Satisfying Fidelity Threshold Routing and Adaptive Purification Strategy (SFTRAP). SFTRAP maximizes throughput for each request by selecting multiple paths and dynamically choosing links for entanglement purification based on the current state of link resources, thus minimizing throughput loss while satisfying fidelity threshold. The strategy also adaptively adjusts the number of purification rounds according to the fidelity threshold, thereby optimizing the time required for deep purification and enhancing algorithmic efficiency. For multi-request scenarios, SFTRAP employs a priority sorting mechanism that takes into account both path cost and path freedom, which refines request scheduling and path selection to create more efficient request combinations, thus further boosting the overall network throughput. Simulation results indicate that SFTRAP surpasses state-of-the-art methods in terms of both throughput and algorithmic efficiency, highlighting its potential for optimizing resources in quantum networks.

Hot rolling production scheduling (HRPS) is an essential process in modern steel manufacturing. Its effectiveness is influenced by three primary challenges: selecting suitable slabs to maximize efficiency and maintain product quality, adhering to increasingly stringent carbon tax regulations, and managing uncertainties in processing times stemming from fluctuations in rolling speed. This article presents a novel robust HRPS problem under carbon tax regulation (RHRPSP-CTR) that considers slab selection, carbon tax regulation, and uncertain processing times simultaneously for the first time. Based on a budgeted uncertainty set, we develop a robust counterpart model that utilizes a classical dualization scheme and dynamic programming recursive equations to address the challenges associated with evaluating the worst case cost of carbon emissions (CEs) and determining the worst case completion time for each slab caused by processing time uncertainty, respectively. Recognizing the characteristics of slab selection and the high computational complexity in the large-scale RHRPSP-CTR, we propose an adaptive large neighborhood search algorithm incorporating two enhancement strategies: a slab selection rule and a max-min weight update mechanism. Extensive computational experiments demonstrate that the proposed method yields optimal solutions for small-scale problem instances and high-quality, robust solutions for large-scale instances within a relatively short computation time. Moreover, the results indicate that compared with deterministic HRPS schemes, robust HRPS schemes lead to only a slight increase in cost. Notably, a higher carbon tax does not necessarily lead to lower CEs, and a larger uncertainty budget coefficient or uncertainty range does not always result in higher CEs.

The efficiency of data retrieval is crucial for the lifecycle traceability of large-scale manufacturing products. This article proposes a novel on-chain and off-chain parallel retrieval method based on blockchain and Merkle Patricia Trie (MPT), which is called parallel retrieve index MPT (PRIMPT). In the proposed method, index MPT (InMPT) is designed to store the index information and the off-chain address of transactions based on MPT. In addition, the transaction division and the node division are proposed to realize the parallelization. Among them, the transaction division adds a new stage branch node under the root node of InMPT and divides transactions according to their stages, while the node division groups the nodes in the blockchain according to the derived most efficient group number and assigns retrieval requests to these node groups. Experiment results show that the PRIMPT method improves the efficiency by 44% compared with the state-of-the-art on-chain and off-chain method.

Sequential recommendation (SR) has received much attention and made promising progress in the past few years due to its high alignment with real recommendation scenarios. It models users' preferences and behavior patterns from their historical behavior sequences and provides personalized recommendations. However, the widespread problem of data sparsity limits the performance of sequential recommendation models. To tackle this, data augmentation (DA) provides a feasible solution by improving the quantity, quality, or diversity of the training samples without the need for additional data collection. In this survey, we present a systematic and timely review of research efforts on data augmentation for sequential recommendation. We start by providing a clear formulation of the problem and task. Then, we develop a unified taxonomy that categorizes existing augmentation methodologies regarding their augmentation objects and principles. Next, we conduct a comparative discussion on the advantages and disadvantages of different categories, supplemented with quantitative performance evaluations, time-complexity analyses, and visual case studies of representative methods, aiming to provide actionable guidance for the selection and development of augmentation methods in real-world scenarios. Finally, we present the future research directions and summarize this survey.

In mobile edge computing (MEC), computation offloading serves as an effective solution to bridge the gap between the stringent latency requirements of computational tasks and the limited processing capabilities of terminal devices (TDs). However, complex inter-task dependencies, dynamic network conditions, and the decentralized architecture of MEC systems pose significant challenges to efficient and adaptive task offloading. To address these challenges, this paper investigates dependency-aware task offloading and resource optimization in MEC environments. First, we propose a task feature extraction method based on dependency-aware graph neural networks (FEDG), which captures the hierarchical structure and varying importance of subtask dependencies by adaptively learning the aggregation weights of predecessor nodes and edges. Then, to address the joint dependency-aware task offloading and resource allocation problem under partial observability in MEC networks, we design a Dependency-aware Graph-based Multi-Agent deep reinforcement learning (DGMA) algorithm. DGMA integrates adaptive prioritized experience replay and correlation-based selective parameter sharing to improve learning efficiency and accelerate convergence in multi-agent environments. Extensive simulations demonstrate that DGMA achieves superior performance in terms of delay, energy consumption, offloading utility, and deadline violation rate.

We consider the energy-constrained task allocation problem in large-scale Aerial Edge Computing (AEC) systems, which encompasses a series of tightly coupled decision-making processes, includingwhichtasks need to be processed by unmanned aerial vehicles (UAVs),howto allocate these tasks and balance energy across UAVs for delay-sensitive requirements. However, little attention has been devoted to exploring the above coupled decision-making problem in AEC with various resource and energy constraints, which is further complicated by energy dynamics (UAV battery states), task-specific consumption, and allocation-feedback balance. In this paper, we formulate a multi-dimensional joint optimization problem, simultaneously optimizing task allocation and energy rewarding to maximize long-term system rewards while balancing service quality and energy efficiency. To this end, we propose a green aerial edge computing framework where partial UAVs are equipped with energy harvesting modules to collect ambient energy. To circumvent the intractable computational complexity arising from the coupled energy states of massive UAVs, we design a distributed solution method based on the mean field game, which decouples the dense multi-agent interactions into a game between an individual UAV and the aggregate population state, thereby transforming the complex global optimization problem into a set of equivalent scalable subproblems. We develop an optimal energy valuation scheme to guide UAV behavior. Numerical results show that our mechanism can effectively ensure sustainable system operation while maintaining high quality of service for metaverse users, outperforming existing methods in both system sustainability and service responsiveness.

We consider the efficient resource allocation task in the blockchain-based mobile edge computing (MEC) system that requires decentralized transaction management to validate transactions between edge servers (ESs) and mobile devices (MDs). In such task allocation process (where MDs' resources are limited and privacy-sensitive), it is a significant challenge to guarantee individual rationality with satisfactory system stability while enabling flexible task offloading under various locality constraints (e.g., communication distance, bandwidth and delay). In this paper, we formulate the target problem as a blockchain-assisted task-resource matching model, and then propose a near optimal locality-aware resource allocation mechanism over smart contract to enable automatic and efficient transactions in MEC system. More specifically, for the service agents selection, we design the preference-based selection strategy to get highest estimated profit. For the flexible task offloading, we develop the minimum delay task graph partitioning algorithm to determine the optimal task offloading solution for MD under different resource bundles. For the task-resource matching, we propose a task-resource matching game (based on potential game) with the second lowest cost strategy to determine the matching of task-resource and decide the price of resource bundle. For the transaction verification and block allocation, we propose a social welfare-driven consensus mechanism to enable verified transaction and fair block allocation in a reward-free way. Strict theoretical analysis and extensive simulations demonstrate that our mechanism guarantees individual rationality, Nash Equilibrium, and stable near optimal solution.

As mobile applications become increasingly computation-intensive, mobile devices (MDs) face growing limitations due to their constrained computational capabilities and battery life. Collaborative Edge Computing (CEC) has emerged as a promising solution to address these challenges by enabling multiple edge service providers (ESPs) to offer computation offloading services to MDs. As such, a CEC resource trading market is essential for efficient interactions between MDs and ESPs. However, jointly determining the offloading ratios, allocating combinatorial computation and communication resources, and designing appropriate pricing strategies in a dynamic market remains a significant challenge. To this end, we propose a truthful online double auction based resource allocation mechanism for partial computation offloading (TRAPO) that explicitly accounts for the stochastic nature of both MDs and ESPs. TRAPO first leverages spatial diversity to construct a set of bids for each MD by mapping their task requirements into resource demands through considering MDs' preferences and partial offloading. Next, we match resource-demanding MDs with resource-supplying ESPs based on adaptive valid price thresholds to maximize social welfare, and calculate the payments of MDs and the rewards of ESPs. Theoretical analyses demonstrate that TRAPO satisfies truthfulness, budget balance, individual rationality, and computational tractability. Simulation experiments further verify the effectiveness and efficiency of TRAPO.

Network Function Virtualization (NFV) facilitates on-demand and flexible service provisioning to meet the escalating demands of Internet of Things (IoT) applications, enabled by Service Function Chain (SFC) technique. The widespread deployment of 5G has connected a massive number of devices and users to IoT networks, accelerating the expansion of IoT scales. IoT users’ service requirements exhibit heightened diversity and dynamism. Consequently, the SFC placement problem in Next Generation Multi-domain IoT (NGMIoT) networks has garnered significant attention. How to efficiently place SFCs under uncertain resource demands to adapt to evolving service request dynamics poses substantial challenges. Therefore, this paper investigates the Robust SFC Placement (RSFCP) problem in NGMIoT networks under resource demand uncertainty. Specifically, we formulate the RSFCP problem as an integer linear programming model to minimize overall SFC placement cost while ensuring service quality. We further prove the RSFCP problem is NP-hard and propose a greedy strategy based heuristic SFC placement algorithm to solve it. Finally, extensive simulation experiments are conducted to evaluate performance, demonstrating that the proposed algorithm outperforms benchmark mechanisms in terms of service acceptance rate and placement cost.

With the rapid development of Artificial Intelligence Generated Content (AIGC), single data centers are increasingly unable to meet the growing demands for data and computational resources in distributed large-scale language model (LLM) training. In this context, distributed training across heterogeneous data centers has become a necessary choice to enhance computational power and flexibility. However, the networks in heterogeneous data centers are polymorphic, with diverse communication protocols and network architectures. This heterogeneity renders traditional routing devices ineffective in recognizing and processing gradient data. Moreover, frequent copying and excessive parsing of gradient data by routing devices across heterogeneous data centers significantly increase model training time. To address these challenges, we propose XAForward, a method for accelerating distributed LLM in heterogeneous data centers using eXpress Data Path (XDP). Specifically, XAForward introduces a polymorphic-compatible protocol that reconstructs the header of gradient data packets to enable efficient data forwarding across different communication protocols in heterogeneous data centers. Additionally, to accelerate distributed LLM computing and reduce gradient data copying and excessive parsing during training, XAForward leverages kernel-bypass techniques based on XDP for packet processing and kernel-level data forwarding using network index identifiers. Experimental results show that, compared to state-of-the-art methods, XAForward reduces the distributed LLM training time by approximately 35% to 40%.

Computing Power Networks (CPNs) integrate heterogeneous resources across the cloud–edge–end continuum to support wide-area distributed computational services, but the geographical separation of computation and data makes cross-domain data access a major bottleneck. Intelligent cross-domain data orchestration in CPNs is difficult because replica selection and end-to-end path planning must be jointly optimized under Service Level Agreement (SLA) and resource constraints, while each domain observes only partial congestion and resource information. This paper presents AE-MAAC, an attention-enhanced multi-agent reinforcement learning framework that formulates cross-domain data orchestration as a Multi-Agent Markov Decision Process (MMDP) with a hierarchical composite action space and constraint-aware masking under a centralized-training–decentralized-execution paradigm. An attention-based state representation captures heterogeneous cross-domain topology and resource information, an attention-enhanced centralized critic strengthens inter-domain credit assignment in large-scale settings, and parallel dual-policy actors together with a parallel experience ensemble and prioritized sampling improve training stability in large constrained action spaces. Extensive simulations across three CPN scales show that AE-MAAC achieves the highest average episode reward. In the representative 5×10 network, it reaches an average episode reward of 431.7 with a 94.2% request success rate and a 259.8 ms average end-to-end delay, while yielding a lower multi-objective cost than state-of-the-art RL baselines.

Rumor blocking approaches in social networks aim to identify a small set of counter-rumor seed nodes and compete with rumor cascades to quickly stop the propagation of rumors. However, current rumor blocking methods assume complete knowledge of rumor node positions, which is often unattainable in real-world scenarios. In this paper, we introduce the concept of Uncertainty Rumor Blocking, where we address the uncertainty surrounding rumor node locations by considering a set of suspicious nodes, each associated with a probability indicating the likelihood of rumor propagation. As traditional node selection algorithms become inadequate under uncertain conditions, we propose a Graph Neural Network-based Inverse Reinforcement Learning (G-IRL) approach to effectively select counter-rumor seed nodes. Through comprehensive experimentation on three datasets, we demonstrate the consistent superiority of our G-IRL over state-of-the-art baseline methods for node selection in the context of uncertainty rumor containment.

Network measurement sketches enable efficient traffic monitoring but require careful parameter configuration to balance accuracy and memory efficiency. We presentRA-Sketch, a unified framework for generating memory-optimal sketch configurations that satisfy user-defined error constraints across diverse network measurement tasks. Unlike existing approaches that rely on computationally intensive experimental testing, RA-Sketch introduces: 1)Poisson-distributed collision modelingto construct error predictors for both frequency-independent tasks (membership query, heavy-hitter detection, and super-spreader detection) and frequency-dependent tasks (flow size distribution, frequency estimation, and cardinality estimation), eliminating the need for empirical validation; 2) Ahierarchical search strategycombining power-of-two scaling and binary search, reducing iterations through optimized parameter initialization. RA-Sketch supports 10+ sketch architectures including Bloom Filter, Elastic Sketch, HeavyKeeper, MEC Sketch, MRAC, CM Sketch, CO Sketch, gSkt, rSkt1 among others. Evaluations on real-world network traces demonstrate: 1) up to 6–7 orders-of-magnitude faster configuration than benchmark-based methods; 2) Prediction errors are within 10% for heavy-hitter detection and super-spreader detection in most evaluated settings, while prediction errors for membership query, flow size distribution, frequency estimation, and cardinality estimation are close to zero; 3) Memory utilization approaches theoretical minima. The framework’sgenerality and efficiency enable real-time reconfiguration of sketches under dynamic network conditions.

The rise of edge intelligence is driving distributed machine learning toward a new paradigm of edge-collaborative computing. To overcome the severe communication bottleneck in this paradigm, In-Network Aggregation is a critical enabling technology. However, its effectiveness is fundamentally undermined by the profound resource heterogeneity of edge networks. Specifically, edge devices, adapting to hardware constraints, operate at varying numerical precisions, leading to significant data inflation as gradients are aggregated. Compounding this, unevenly distributed network resources and traditional, precision-oblivious routing strategies often misallocate critical, high-precision gradients to low-quality paths. This mismatch creates severe network congestion, crippling the efficiency of distributed training. To address this, we propose the Precision-Aware Hierarchical In-Network Aggregation (PAHInA) framework, the first, to our knowledge, to perform routing optimization for in-network aggregation that explicitly considers precision heterogeneity. The core of PAHInA is an intelligent control-plane scheduler that co-optimizes for gradient priority and path cost, dynamically planning the most cost-effective aggregation strategy for each flow. This fine-grained scheduling guarantees that high-priority gradients are routed through premium, low-latency paths, minimizing global communication overhead. On the data plane, we leverage the eXpress Data Path (XDP) for high-performance packet processing to reduce aggregation-induced overhead. Extensive simulations show that, compared to state-of-the-art baselines, PAHInA significantly mitigates network congestion, reducing end-to-end communication time by up to 33% and boosting overall training throughput by approximately 30%.

In cloud-edge-end (CEE) collaboration, the resource optimization based on deep reinforcement learning have achieved significant performance improvements in time-slot systems. However, some studies only focus on computing delay and energy consumption in each time slot, ignoring the impact of task backlog queues on system performance. In addition, the delay-oriented optimization tends to offload a large number of tasks to servers, failing to fully utilize the computing resource of mobile devices. To address these issues, we propose the multi-dimensional preference-driven resource optimization framework (MPROF). This study includes several key points: 1) constructing a three-layer heterogeneous architecture that applying the collaboration among edges for CEE; 2) proposing the task backlog estimation mechanism, which mitigates the impact of previous unfinished tasks on the current time slot; 3) proposing the group relative direct-preference policy optimization (GRDPO) that incorporates the preference information for efficient task offloading, and combines it with mathematical programming for the system resource optimization. The simulation experiments are conducted across multiple typical scenarios. The results show that, the proposed framework outperforms existing mainstream methods in task offloading, system delay, task backlog, and energy consumption control, demonstrating certain practical application prospects.

Multi-access edge computing (MEC) enhances users’ computing capabilities. However, it still faces the queue instability challenge due to the conflict between limited communication-computation resources and the massive access of users. In this paper, we invoke a multiple one-to-one queuing system in a non-orthogonal multiple access (NOMA)-MEC network, where NOMA enables massive access, and study its network stability and queuing delay control mechanisms. Meanwhile, we jointly consider the many-to-one queuing system. We formulate two network power minimization stochastic problems, each subject to the stability of the multiple one-to-one/many-to-one queuing system. We propose Lyapunov-based algorithms to solve these problems analytically and achieve a computational complexity of approximatelyO(N)withNusers. We further investigate the queuing-delay control mechanism in the multiple one-to-one queuing system-based NOMA-MEC network. Numerical results show that the proposed algorithms converge efficiently, effectively stabilize queuing systems, satisfy the queuing-delay requirement, and further demonstrate that: 1) the proposed multiple one-to-one queuing system-based NOMA-MEC exhibits a better balanced load distribution between users and the MEC server compared to orthogonal multiple access (OMA)-MEC; 2) we obtain a more efficient tradeoff between network power consumption and queue backlog compared to the many-to-one queuing system; and 3) we obtain superior queue stabilization capacity under bursty data and network power suppression conditions. This also validates NOMA’s high spectral efficiency and fine-grained remote execution control at the MEC server.

We consider the collaborative inference acceleration task via cloud-edge-end collaboration, which involves a series of tightly coupled decision-making steps, includingwhichDNN model to be selected,how muchto compress model,howto partition model, andwhereto offload partitioned submodels. In practical deployments, these decisions jointly affect both fine-tuning and inference performance, and must jointly account for such aspects as the model being used, the computational resources and local datasets available at each device, as well as network latencies, which significantly increases the complexity of optimizing the problem. Yet, no existing studies focus on such joint optimization problem for these tightly coupled decisions. In this paper, we model this problem as a multi-dimensional optimization problem, jointly optimizing collaborative inference and fine-tuning by selecting the DNN model, compression level, partition strategy, and computational resource allocation, with the objective of minimizing the overall energy consumption of the learning-inference process, subject to accuracy and latency constraints. To this end, we propose an algorithmic framework called JQODI combining a time-energy tree diagram to represent the learning process, a dynamic programming solution strategy, and a data-driven theoretical approach to predict the expected total number of training epochs that meet the accuracy requirements. We prove that JQODI approximates the optimal solution with polynomial complexity. Numerical results demonstrate that JQODI surpasses state-of-the-art methods in both energy efficiency and latency.

To comply with privacy regulations and self-protection from cloud outages, increasingly more users are leveraging multiple cloud providers for service deployment. However, due to the isolation of highly Non-Identically and Independently Distributed (Non-IID) workload datasets and deficiencies of existing methods as reflected in our experimental studies, no cloud providers can single-handedly capture the workload patterns of such users for accurate workload prediction. Accordingly, we proposeP3Forecast, aPersonalizedPrivacy-Preserving Cloud workload prediction framework based on Federated Generative Adversarial Networks (GANs), which allows cloud providers with Non-IID workload data to collaboratively train workload prediction models as preferred while protecting privacy. We first design a data synthesis quality assessment method based on Dynamic Time Warping (calledpattern-aware DTW), which is insusceptible to time series length and reliable for the comparison of temporal patterns. By usingpattern-aware DTWas the model aggregation weights, we adopt the Federated Learning (FL) of a GAN model for the augmentation of IID workload training datasets per cloud provider. Then, we further design a post-training method of local workload prediction models, which consists of a query mechanism based on comprehensive evaluation of data synthesis informativeness and an adaptive learning rate adjustment strategy for stable convergence. Extensive experiments driven by real-world workloads demonstrate that compared with the state-of-the-art,P3Forecastimproves workload prediction accuracy by 23.7%-64.5% on average over all cloud providers, while ensuring the fastest convergence in Federated GAN training.

Mobile Edge Computing (MEC) decentralizes the network's computing and storage capabilities from centralized infrastructure to edge nodes located closer to end-users, enabling context-aware service deployment, low-latency service response, and efficient computation for mobile users. However, achieving energy-efficient service deployment while maintaining service delivery quality remains a significant challenge due to the wide geographic distribution of edge nodes, the dynamic variation of service workloads, and the differences between service deployment and delivery cycles. To address these challenges, we first design a feature encoding strategy and a self-attention-based encoder to extract contextual features, which are fused to support adaptive decision timescale regulation driven by service semantics and system load dynamics. Then, we propose a novel Dual-Timescale Energy-Efficient Service Deployment and Delivery (DT-EESD) framework integrated with hierarchical learning. The upper layer leverages an enhanced decision-making mechanism to optimize proactive service deployment and base station switching on a larger timescale, aiming to reduce long-term network costs. The lower layer employs a fine-grained real-time optimization approach to dynamically handle service delivery and resource allocation on a smaller timescale, effectively responding to dynamic service requests. By incorporating an expected reward-based learning mechanism, the framework efficiently handles the temporal coupling between deployment and delivery cycles. Extensive experiments demonstrate that DT-EESD outperforms baseline algorithms, achieving at least a 10.89% reduction in average system cost while also improving model convergence, reducing delay, and enhancing resource utilization.

Federated edge learning (FEEL) greatly facilitates the development of ubiquitous intelligence by combining federated learning and edge computing. However, traditional FEEL implementations assume fixed-sized local datasets, neglecting the potential of edge devices to acquire sensory information actively. Such a simplistic scenario leads to overestimating data availability and underestimating resource utilization in networks with varying resource capacity. Moreover, the existing FEEL-oriented systems with integrated sensing, communication, and computation (ISCC) have separate-based designs, leading to an inefficient use of wireless resources. To alleviate these issues, we propose a novel FEEL-oriented ISCC framework in edge collaborative networks, by leveraging the integrated sensing and communication (ISAC) technique to achieve the dual purpose of data sensing and parameter transmission. Then, over the designed framework, we present FEEL convergence analysis under non-independent and identically distributed (non-iid) and iid data. Correspondingly, we formulate a joint beamforming and flexible time duration optimization problem to maximize the convergence speed of FEEL, subject to limited resources on the devices and requirements for data sensing and communication. To address the problem efficiently, we propose an alternative optimization framework, in which the successive convex approximation (SCA) method is adopted to solve the nonconvex beamforming design subproblem, and a low-complexity method is derived for optimal time allocation. Extensive results reveal that the proposed framework can achieve excellent performance in model training accuracy by efficiently utilizing limited resources in edge collaborative networks, under iid and non-iid data.

In cloud-edge collaborative networking, intelligent devices often need to offload tasks that they cannot handle to edge or cloud servers for processing. So the key problem is how to deploy various types of services and offload tasks to the appropriate servers. Deep Reinforcement Learning (DRL) algorithms have been widely used to address these issues. However, existing DRL solutions typically use centralized training methods, which can not address the challenge of obtaining global network states in practical scenarios due to the extremely large network scale and privacy concerns. In this paper, we designed a Knowledge-Defined Cloud-Edge Collaborative Networking (KDCECN) architecture for managing network information and proposed a Partially Observed Lightweight exchange Hierarchical DRL algorithm (PO-LeHDRL). This algorithm fully considers factors such as the tolerable delay of tasks, the size of tasks, and different service deployment strategies, which solves the joint service placement and task offloading problems in the cloud-edge collaborative network in a distributed training and decision-making manner. The innovation of this scheme lies in achieving data privacy protection. In this scheme, edge nodes do not need to share their respective network measurement information and apply Laplace noise to the reward value using the differential privacy mechanism, effectively reducing additional communication overhead and protecting the privacy of network data on edge devices. The experimental results show that compared with the baselines, our scheme can improve the system's task completion rate and reduce the task completion delay, and it shows scalability across different network environments.

Traditional computation offloading and resource allocation strategies encounter several issues that lead to poor service experience and resource wastage. The resource allocation scheme lacks the flexibility to adapt to the time-varying offloading demands of User Equipment (UEs). Furthermore, there is an imbalance between UEs seeking better service and Service Providers (SPs) aiming to minimize cost expenditures. In this paper, we propose a knowledge-defined networking-based Multi-Layer Computation Offloading and Resource Allocation strategy optimization (ML-CORA) architecture. Based on the ML-CORA, we design a Multi-Layer Local-Global Collaborative computation offloading and resource allocation strategy optimization (ML2GC) algorithm. The basic level of the ML2GC algorithm expresses and optimizes computation offloading demands from the perspective of UE (local), while the meta level optimizes the resource allocation strategy on demand from the perspective of the SP (global), achieving a collaborative multi-objective optimization for a win-win system between UEs and SPs. The two-layer structure of the ML2GC algorithm outputs continuous and discrete actions respectively, which improves the flexibility and efficiency of the algorithm while effectively balancing the interests of all parties and promoting efficient resource utilization. Simulation results based on the real-world dataset of Shanghai Telecom indicate that the ML2GC algorithm significantly improves both social welfare and resource utilization compared to baseline algorithms.

Sharding, as a Layer-1 scaling technique, is widely recognized as a promising solution to address the scalability limitations faced by blockchains. However, distributing accounts across shards generates numerous cross-shard transactions (CSTs) and inter-shard workload imbalances, potentially compromising scalability. Existing state-of-the-art methods typically optimize transaction distribution for subsequent epochs by periodically reallocating accounts using graph partitioning or community detection to balance loads and minimize CSTs. Nevertheless, these methods often involve computationally expensive account migration and overlook real-world transaction skewness, exacerbating workload imbalances. To address the above problems, we propose a two-phase account group migration service with dynamic load awareness in this paper. This service can monitor and analyze system state to determine the necessity of account migration. Once triggered, it employs a two-phase algorithm that combines account grouping with global group-level re-partitioning to balance workloads and minimize CSTs. Additionally, group-level migration further helps reduce computational overhead. We evaluate the designed service by replaying large-scale real Ethereum transactions. Experimental results demonstrate that, compared with the baselines, our method can not only improves system throughput and reduces transaction confirmation latency, but also enhances overall scalability.

Offloading a plethora of end workflows to edge servers in mobile edge computing (MEC) systems involves a series of coupled decision-making steps, includinghow muchedge resources will be allocated for each workflow,whichsubtasks will be offloaded, andhowto determine edge-end transaction prices to ensure system stability. Particularly, these decisions must jointly account for workflow characteristics, the resources available at each edge server, as well as local constraints (e.g., communication distance, task latency, and bandwidth conditions), which again increases the difficulty of optimizing the problem. However, no existing study addresses such joint optimization problems for these tightly coupled decisions. To fill this gap, a minimum-delay workflow partitioning algorithm is first designed to determine the optimal task offloading solution under various resource conditions. Based on this algorithm, two locality-based social welfare maximization models (basic and dynamic) are constructed. Specifically, for basic model, a multi-stage task matching game with the second lowest cost strategy is developed to determine the resource selection and pricing. For the dynamic model with uncertain requests, an online learning algorithm is introduced to track the dynamic valuations of mobile devices and to ensure that the resulting task allocation solution achieves an upper-bounded regret. Strict theoretical analysis demonstrates that our mechanism guarantees individual rationality, Nash Equilibrium, and stable approximation ratio. Simulation results verify the effectiveness and efficiency of our mechanism, and show that the proposed mechanisms obtain at most 18% higher social welfare than existing studies.

The robust network topology of the Internet of Things (IoT) system facilitates uninterrupted service provisioning when encountering device failures. Traditional topology optimization strategies use link-level algorithms to design robust network topologies for IoT device deployment, ensuring network resilience against failures. These algorithms struggle to provide a robust topology for large-scale networks due to the high complexity and computational cost of optimizing each link individually. To overcome this limitation, we introduceLEGO-Motif, a motif-based IoT topology generation algorithm inspired by preferential attachment (PA) and evolutionary theory. By sequentially integrating network motifs, similar to assembling LEGO bricks, the algorithm efficiently enhances topology robustness while reducing computational overhead. Specifically, we propose a novel metric based on motif density to measure topology robustness; then, guided by this metric, we design a topology generation algorithm that ensures optimal topology with high robustness against cyberattacks throughout its growth, inspired by an evolutionary neural network framework. The LEGO-Motif algorithm introduces novel recombination, PA-based mutation, and pruning operators to enhance optimization performance and reduce running-time costs. Comprehensive case studies and evaluations show that LEGO-Motif outperforms current topology optimization algorithms, achieving more robust network topologies with reduced running time, which offers a promising optimal solution for deploying the IoT topology.

By generating new yet effective data, data augmentation has become a promising method to mitigate the data sparsity problem in sequential recommendation. Existing works focus on augmenting the original data but rarely explore the issue of imbalanced relevance and diversity for augmented data, leading to semantic drift problems or limited performance improvements. In this paper, we propose a novel Balanced data Augmentation Plugin for Sequential Recommendation (BASRec) to generate data that balance relevance and diversity. BASRec consists of two modules: Single-sequence Augmentation and Cross-sequence Augmentation. The former leverages the randomness of the heuristic operators to generate diverse sequences for a single user, after which the diverse and the original sequences are fused at the representation level to obtain relevance. Further, we devise a reweighting strategy to enable the model to learn the preferences based on the two properties adaptively. The Cross-sequence Augmentation performs nonlinear mixing between different sequence representations from two directions. It produces virtual sequence representations that are diverse enough but retain the vital semantics of the original sequences. These two modules enhance the model to discover fine-grained preferences knowledge from single-user and cross-user perspectives. Extensive experiments verify the effectiveness of BASRec. The average improvement is up to 72.0% on GRU4Rec, 33.8% on SASRec, and 68.5% on FMLP-Rec. We demonstrate that BASRec generates data with a better balance between relevance and diversity than existing methods.

Multi-behavior recommendation exploits auxiliary behaviors (e.g., view, cart) to help predict users' potential target behavior (e.g., purchase) on a given item. However, existing works suffer from two issues: (1) They generally consider only a single chain from auxiliary behaviors to the target behavior, referred to as a purchase chain (e.g., view -> cart -> purchase), ignoring other valuable purchase chains (e.g., view ->purchase) that are beneficial for recommendation performance. (2) Most studies presume that interacted items in auxiliary behaviors are good for recommendations, and pay little attention to the negative transfer problem. That is, some auxiliary behaviors may negatively transfer the influence to the modeling of target ones (e.g., items viewed but not purchased). To alleviate these issues, we propose a novel Multiple Purchase Chains (MPC) model with negative transfer elimination for multi-behavior recommendation. Specifically, we construct multiple purchase chains from auxiliary to target behaviors according to users' historical interactions, while the representations of a previous behavior will be fed to initialize the next behavior on the chain. Then, we construct a negative graph for the latter behavior and learn the negative representations of users and items which will be filtered out to eliminate negative transfer. Experimental results on two real datasets outperform the best baseline by 40.97% and 47.26% on average in terms of Recall@10 and NDCG@10 respectively, demonstrating the effectiveness of our method.

Prompt tuning is a promising method to fine-tune a pre-trained language model without retraining its large-scale parameters. Instead, it attaches a soft prompt to the input text, whereby downstream tasks can be well adapted by merely learning the embeddings of prompt tokens. Nevertheless, existing methods still suffer from two challenges: (i) they are hard to balance accuracy and efficiency. A longer (shorter) soft prompt generally leads to a better (worse) accuracy but at the cost of more (less) training time. (ii) The performance may not be consistent when adapting to different downstream tasks. We attribute it to the same embedding space but responsible for different requirements of downstream tasks. To address these issues, we propose an Efficient Prompt Tuning method (EPT) by multi-space projection and prompt fusion. Specifically, it decomposes a given soft prompt into a shorter prompt and two low-rank matrices, significantly reducing the training time. Accuracy is also enhanced by leveraging low-rank matrices and the short prompt as additional knowledge sources to enrich the semantics of the original short prompt. In addition, we project the soft prompt into multiple subspaces to improve the performance consistency, and then adaptively learn the combination weights of different spaces through a gating network. Experiments on 13 natural language processing downstream tasks show that our method significantly and consistently outperforms 11 comparison methods with the relative percentage of improvements up to 12.9%, and training time decreased by 14%.

Involving collaborative information in Large Language Models (LLMs) is a promising technique for adapting LLMs for recommendation. Existing methods achieve this by concatenating collaborative features with text tokens into a unified sequence input and then fine-tuning to align these features with LLM's input space. Although effective, in this work, we identify two limitations when adapting LLMs to recommendation tasks, which hinder the integration of general knowledge and collaborative information, resulting in sub-optimal recommendation performance. (1) Fine-tuning LLM with recommendation data can undermine its inherent world knowledge and fundamental competencies, which are crucial for interpreting and inferring recommendation text. (2) Incorporating collaborative features into textual prompts disrupts the semantics of the original prompts, preventing LLM from generating appropriate outputs. In this paper, we propose a new paradigm, Collaborative LoRA (CoRA), with a collaborative query generator. Rather than input space alignment, this method aligns collaborative information with LLM's parameter space, representing them as incremental weights to update LLM's output. This way, LLM perceives collaborative information without altering its general knowledge and text inference capabilities. Specifically, we employ a collaborative filtering model to extract user and item embeddings and inject them into a set number of learnable queries. We then convert collaborative queries into collaborative weights with low-rank properties and merge the collaborative weights into LLM's weights, enabling LLM to perceive the collaborative signals and generate personalized recommendations without fine-tuning or extra collaborative tokens in prompts. Extensive experiments confirm that CoRA effectively integrates collaborative information into LLM, enhancing recommendation performance.

Haoyu Xu, Pengxiang Lan, Enneng Yang, Guibing Guo, Jianzhe Zhao, Linying Jiang, Xingwei Wang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Graph stream summarization refers to the technique used to process graph streams-unbounded sequences of edges-by constructing compressed representations that support approximate queries on both graph topology and temporal information in computing power networks. However, existing methods struggle to achieve accurate and efficient temporal queries due to two key limitations: (1) inefficient integration of temporal information, leading to high latency in both edge processing and query execution; and (2) redundant multilayer structures that accumulate errors, significantly reducing query accuracy. In this paper, we propose GRIT, an accurate and efficient Graph stReam summarIzation for Temporal query. GRIT introduces a new structure FlatIndex, which organizes temporal information in a flattened form, playing a critical role in minimizing error accumulation and ensuring accurate temporal queries. To further enhance edge processing efficiency, we introduce a lazy update strategy, which updates only a single element in the FlatIndex upon edge insertion, significantly reducing insertion latency. Moreover, our greedy-based decomposition (GBD) algorithm decomposes the target query range into the minimal number of intervals corresponding to the FlatIndex, enabling efficient execution of temporal queries over arbitrary time ranges. Extensive experiments on five real-world datasets demonstrate that GRIT improves query accuracy by 2-3 orders of magnitude, while reducing query latency by 1-2 orders of magnitude and increasing throughput by 7-13 times compared to state-of-the-art methods.

Network resilience refers to the ability of a network to maintain its functionality despite perturbations, where resilience/robustness is shown when a substantial proportion of its nodes remain engaged even under changes. Such phenomenon is common in real-world networks, such as computing power networks. Previous works demonstrate that the coreness of a user/node effectively captures the dynamics of user engagement. However, most existing works only consider changes in a single coreness value and thus fail to measure the overall network resilience. Subsequent works are either inefficient or do not consider coreness-decreased scenario. In this paper, we propose and study the collapsed follower maximization problem, aiming to maximize the number of coreness-decreased vertices by finding a group collapser (collapsing a set of vertices) with a given budget. We prove that the problem is NP-hard and W[2]-hard parameterized by the budget b. To address the problem, we first present a Greedy algorithm that iteratively finds the best collapser in each of the budget b iterations. To further optimize the Greedy algorithm, we propose GreedyOpt, which leverages the shell component structure to accelerate the computation of follower for one collapser and prune the search space. Extensive experimental results on 8 real-world datasets show that the effectiveness and efficiency of our algorithms.

Graph Neural Networks (GNNs) are the rising standard for graph tasks, yet their distributed training in servers or computing power network remains challenging. Cross-machine sampling and data loading often create bottlenecks, leading to inefficient resource utilization. In this paper, we present Hybrid2, a distributed GNN training system that combines full-graph and mini-batch training through a novel hybrid-batch training method. It also adopts hybrid feature extraction, leveraging both local caching and remote access to improve feature retrieval efficiency. The integration of these methods in Hybrid² results in a dual hybrid-gain effect. First, it reduces sampling and loading overhead by pre-aggregating neighbors for each target vertex, minimizing the layers to sample and load. Second, it accelerates data loading by dynamically identifying and locally caching the most frequently accessed vertices during training, maximizing memory efficiency. Experimental results demonstrate that Hybrid² brings substantial performance improvements across key components of distributed GNN training. Network communication overhead is reduced by up to tens of times, while both sampling and loading achieve at least several-fold speedups. These gains contribute to an overall training acceleration exceeding 20× compared to DistDGL, all with comparable GPU memory usage and no loss in accuracy. Compared to the state-of-the-art system, it achieves nearly 3× speedup while using fewer resources.

Multimodal recommendation aims to model user and item representations comprehensively with the involvement of multimedia content for effective recommendations. Existing research has shown that it is beneficial for recommendation performance to combine (user- and item-) ID embeddings with multimodal salient features, indicating the value of IDs. However, there is a lack of a thorough analysis of the ID embeddings in terms of semantics in the literature. In this paper, we revisit the value of ID embeddings for multimodal recommendation and conduct a thorough study regarding its semantics, which we recognize as subtle features of content and structure. Based on our findings, we propose a novel recommendation model by incorporating ID embeddings to enhance the salient features of both content and structure. Extensive experiments on three real-world datasets (Baby, Sports, and Clothing) demonstrate the superiority of our method over state-of-the-art multimodal recommendation methods and the effectiveness of fine-grained ID embeddings.

Mobile devices’ multiple network interfaces create the foundation for multipath optimization, yet widespread implementation remains challenging. Existing multipath solutions primarily rely on simulation environments or require hardware/software adaptations, limiting real-world adoption. Meanwhile, 89.8% of mobile users prefer WiFi over cellular networks due to data cost concerns, despite cellular networks often providing better stability. We propose WiCell, an application-transparent multipath enhancement system for mobile networks that addresses these challenges through: (1) a deployment-friendly approach requiring no hardware or software modifications; (2) a tunneling-based architecture with edge proxy servers that dynamically adjusts transmission based on real-time link condition feedback; and (3) two operational modes tailored to different user requirements - WiFi-Primary Adaptive Redundancy Mode for balanced performance and data conservation, and Dual Transmission Mode for maximum connection stability. Our comprehensive evaluation demonstrates WiCell’s effectiveness across 30 common applications. In WiFi-Primary Adaptive Redundancy Mode, WiCell achieves a 74.6% reduction in video stalling rate while consuming only 22.12% of the cellular data compared to full redundancy approaches. In Dual Transmission Mode, WiCell reduces maximum latency by 77.4% compared to WiFi-only transmission, with 61.36% lower average latency than WiFi and 65.84% lower than cellular-only connections. These results validate WiCell as a practical solution for enhancing wireless network performance while maintaining application transparency and addressing user concerns about cellular data consumption.

As network size continues to increase and data flows between LANs become more frequent, in-band network telemetry methods across LANs require more balanced path lengths and the privacy of telemetry data. Existing schemes have large variance in telemetry path lengths and do not focus on the riskiness of cross-LAN telemetry. In this paper, we design SRv6-ALINT to direct telemetry paths through SRv6 and upload telemetry data at boundary nodes. In the small-scale network case, SRv6-ALINT solve the path generation strategy to get the theoretical optimal value of the path length variance through a solver. In the case of larger scale networks, SRv6-ALINT propose DFS-stitch, an algorithm that guarantees that the path length variance is as small as possible when the number of paths generated is minimized with no duplicate edges and covering the entire network. For segment list compression in SRv6, SRv6-ALINT propose sequential and binary compression algorithms to compress the obtained paths, reduce the segment list depth and decrease the bandwidth. The evaluation shows that the path length variance generated by DFS-stitch is smaller compared to existing schemes. The compressed segment list length of both segment list compression algorithms is reduced to approximately 50% of the uncompressed length.

Node equivalence is common in graphs, such as computing networks, encompassing automorphic equivalence (preserving adjacency under node permutations) and attribute equivalence (nodes with identical attributes). Despite their importance for learning node representations, these equivalences are largely ignored by existing graph models. To bridge this gap, we propose a GrAph self-supervised Learning framework with Equivalence (GALE) and analyze its connections to existing techniques. Specifically, we: 1) unify automorphic and attribute equivalence into a single equivalence class; 2) enforce the equivalence principle to make representations within the same class more similar while separating those across classes; 3) introduce approximate equivalence classes with linear time complexity to address the NP-hardness of exact automorphism detection and handle node-feature variation; 4) analyze existing graph encoders, noting limitations in message passing neural networks and graph transformers regarding equivalence constraints; 5) show that graph contrastive learning are a degenerate form of equivalence constraint; and 6) demonstrate that GALE achieves superior performance over baselines.

Graphs, fundamental in modeling various research subjects such as computing networks, consist of nodes linked by edges. However, they typically function as components within larger structures in real-world scenarios, such as in protein-protein interactions where each protein is a graph in a larger network. This study delves into the Graph-of-Net (GON), a structure that extends the concept of traditional graphs by representing each node as a graph itself. It provides a multi-level perspective on the relationships between objects, encapsulating both the detailed structure of individual nodes and the broader network of dependencies. To learn node representations within the GON, we propose a position-aware neural network for Graph-of-Net which processes both intra-graph and inter-graph connections and incorporates additional data like node labels. Our model employs dual encoders and graph constructors to build and refine a constraint network, where nodes are adaptively arranged based on their positions, as determined by the network’s constraint system. Our model demonstrates significant improvements over baselines in empirical evaluations on various datasets.

Network measurement sketches enable efficient traffic monitoring but require careful parameter configuration to balance accuracy and memory efficiency. We present RA-Sketch, a framework for generating memory-optimal sketch configurations that satisfy user-defined error constraints across diverse network measurement tasks. Unlike existing approaches that rely on computationally intensive experimental testing, RA-Sketch introduces: 1) Poisson-distributed collision modeling to construct error predictors for both frequency-independent tasks (membership query, heavy-hitter detection) and frequency-dependent tasks (frequency/cardinality estimation), eliminating the need for empirical validation; 2) A hierarchical search strategy combining power-of-two scaling and binary search, reducing iterations through optimized parameter initialization. RA-Sketch supports 10+ sketch architectures including Bloom Filter, Elastic Sketch, HeavyGuardian, HeavyKeeper, CM/CO Sketch, gSkt, rSkt1 and so on. Evaluations on real-world network traces demonstrate: 1) 6–7 orders of magnitude faster configuration than benchmark-based methods; 2) Prediction errors ≤10% for heavy-hitter detection, while prediction errors for membership query, and frequency/cardinality estimation are close to zero; 3) Memory utilization approaches theoretical minima. The framework’s generality and efficiency enable real-time reconfiguration of sketches under dynamic network conditions.

Per-Flow cardinality measurement in high-speed networks is essential for network security and traffic analysis applications. Flow cardinality refers to the number of distinct elements within a flow, such as the number of unique destination IPs associated with a given source IP. While extensive research has been conducted on single-flow cardinality estimation, achieving accurate per-flow cardinality measurement with real-time performance and low memory overhead remains challenging in large-scale network environments, particularly given the highly skewed distribution of flow cardinalities where mouse flows with smaller cardinalities dominate, and elephant flows with larger cardinalities are fewer. This paper introduces MEC-Sketch, a memory-efficient cardinality estimation data structure that leverages the inherently skewed distribution of flow cardinalities in network traffic. MEC-Sketch employs a dual-component architecture: a heavy part utilizing a majority vote algorithm for precise super-spreader detection, and a light part implementing compact cardinality estimators for memory-efficient measurement of mouse flows. We address two fundamental technical challenges: (1) adapting the majority vote algorithms to operate with cardinality estimators that lack native support for real-time queries, and (2) implementing an effective mapping strategy between large estimators in the heavy part and small estimators in the light part during elephant-mouse flow separation. Comprehensive evaluations on real-world network traces demonstrate that MEC-Sketch significantly outperforms state-of-the-art solutions in terms of estimation accuracy, memory efficiency, and computational performance for both cardinality estimation and super-spreader detection tasks.

State-of-the-art deep learning models rely on large GPU clusters and various parallelism strategies, which in turn depend on collective communication (CC) operators to synchronize data. While vendor libraries (e.g., NCCL, RCCL) provide standard CC algorithms, they often suffer from bandwidth bottlenecks in imbalanced topologies. Recent synthesis-based methods improve performance but face three key limitations: poor scalability due to the combinatorial explosion of scheduling space, lack of support for multistage execution, and suboptimal communication throughput. We propose Canvas, a scalable and near-optimal CC scheduling framework that addresses these challenges. Canvas introduces: (1) Hierarchical synthesis to decompose the global scheduling problem into tractable subproblems for scalability. (2) Collective decomposition to enable structured, multi-stage algorithm generation. (3) Cross-micro-batch pipeline scheduling to parallelize communication across micro-batches and maximize link utilization. Evaluations show that Canvas achieves up to 1.98× bandwidth speedup over TACCL and 3.56× over TE-CCL, and synthesizes algorithms for 512-GPU topologies within 1.77 hours, whereas TACCL fails to produce results within 24 hours.

Modern distributed training systems face escalating communication bottlenecks as GPU clusters scale to accommodate large models. While collective communication libraries and automated synthesizers address algorithmic efficiency, they suffer from three critical limitations including labor-intensive manual intervention requirement, overreliance on predefined input optimization, and suboptimal isolated configuration optimization. To solve these problems, we present TuCCL, a systematic framework that co-optimizes communication algorithms and runtime parameters through three innovations: Topology-Aware Sketch Generation that automatically produces high-performance primitives, Hierarchical Configuration Optimization modeling nonlinear parameter-performance relationships, and Multi-phase Resource-Aware Configuration Optimization enabling joint configuration tuning with adaptive search space pruning. Evaluations demonstrate TuCCL’s superiority over state-of-the-art systems with 1.75x–11.49x bandwidth improvements for AllGather/AllReduce on NVIDIA V100/A100 clusters, 90.2% faster configuration search than grid methods, and 1.22x-2.52x end-to-end training speedups across diverse model scales.

Large Language Models (LLMs) have advanced rapidly, but their growing parameter scales and memory demands pose critical challenges for distributed training. Although GPU memory capacity improves steadily, model sizes expand much faster, causing frequent Out-of-Memory (OOM) errors and rising training costs. Existing memory optimization approaches, such as ZeRO-3 and offloading, alleviate per-GPU memory pressure but introduce excessive collective communication, limited computation–communication overlap, and degraded scalability. We present MAZ3, a distributed training framework that mitigates these limitations through three key techniques: (1) Collaborative CPU–GPU memory management, storing full parameters in CPU memory and broadcasting them within nodes to reduce global synchronization; (2) Fine-grained communication–computation overlap, aligning collective operations with model computation to hide latency; (3) Hierarchical aggregation operators, leveraging intra-node NVLink and inter-node NIC channels concurrently to minimize communication overhead. We implement MAZ3 on a multi-GPU cluster and evaluate it with large-scale models. Results show that MAZ3 reduces inter-node model communication (gradients and parameters) by 33%, improves training efficiency by 40.3%, and increases throughput by 67.9% compared with ZeRO-3. Moreover, MAZ3 retains the memory efficiency of ZeRO-3 while approaching the training efficiency and throughput of ZeRO-2 Offload, achieving a balanced trade-off between memory optimization and performance.

Accurate encrypted malicious traffic detection is crucial for improving network service quality. Existing methods leverage the widespread application of machine learning (ML) to distinguish encrypted malicious traffic from normal traffic by learning the statistical characteristics of traffic. However, the scarcity of high-quality annotated encrypted malicious traffic data, especially malicious traffic samples, limits the performance of these supervised learning methods. Additionally, annotating network traffic is challenging as it requires domain-specific expert knowledge. Therefore, this paper proposes an encrypted malicious traffic detection framework based on the active learning method. This framework achieves high recognition rates using a limited number of samples. It employs a hybrid weighted uncertainty sampling strategy that utilizes the independence coefficient method to weight uncertainty measurements across multiple scales. This improves the reliability during the automatic instance selection process. In the experimental section, we achieved a detection accuracy exceeding 93 % using a data subset comprising only 1 % of the original dataset. Furthermore, we validated the robustness of the proposed framework through calibration rate measurements in the experiments.

Latency-sensitive services across the end-edge-cloud continuum require not only low mean latency but explicit control of tail latency and data freshness. We propose Control-Loop Freshness-aware Service Function Chain orchestration (CLFSFC), a freshness-aware orchestration framework for Service Function Chains (SFCs) that jointly selects model variants and function placements. We define a Control-Loop Freshness (CLF) objective that combines 95th/99th-percentile (P95/P99) end-toend latency with an Age of Information (AoI) proxy. To make this objective operational under uncertainty, we allocate per-stage risk budgets via the union bound and convert mean/variance profiles into percentile constraints using Cantelli's inequality, yielding a two-stage greedy solver with interpretable quotas. We implement CLF-SFC with offline profiling of YOLOv5 n/s/m variants across end/edge/cloud devices, and evaluate it with profiling-driven measurements on COCO 2017 under synthesized network regimes. Across bandwidth and round-trip time settings, CLF-SFC reduces P95/P99 latency and Service Level Objective (SLO) violations relative to Edge-only, Cloud-only, and a riskagnostic heuristic; at high bandwidth it remains comparable to the Shortest-Latency-Path (SLP) baseline. The proposed CLF-SFC framework naturally fits embodied-AI pipelines where perception, fusion, and policy modules operate in a closed loop. By explicitly incorporating the Age-of-Information (AoI), our orchestration ties data freshness to control stability, complementing tail-latency minimization.

To address the challenge of data acquisition in unknown disaster environments — such as earthquake zones, flood-affected regions, or industrial accident sites — this paper proposes a Hybrid Centralized - Distributed Collaborative Coverage (HCC) strategy for UAV swarms. The HCC framework integrates centralized trajectory planning with distributed obstacle avoidance to achieve rapid, efficient, and safe coverage of Points of Interest (PoIs). Specifically, a Centralized Collaborative Optimization (CCO) strategy is designed to compute secure cooperative coverage schemes on an edge server, while a Distributed Obstacle-avoidance Coverage Strategy (DOCS) enables each UAV to perform safety-aware navigation based on local sensing. Unlike sequential exploration, the proposed method supports parallel and synchronized coverage execution, ensuring that data acquisition across the entire region occurs concurrently. Simulation results demonstrate that the proposed method outperforms benchmark algorithms in terms of coverage rate, safety, energy efficiency, and network lifetime.

Blockchain is increasingly integrated in Multiaccess Edge Computing (MEC) to coordinate secure and lowlatency resource provisioning and service orchestration among resource-constrained embodied AI devices. However, conventional blockchains perform costly transaction verification during propagation, which can be exploited by spam transaction attacks and overload resource-limited edge devices. To mitigate the substantial overhead of verification, we propose a two-stage BLS multi-signature backed transaction propagation mechanism for blockchain-enabled MEC: a small-scope random-walk phase with deep verification and signing, followed by a large-scope propagation phase with probabilistic verification. In the first stage, nodes conduct deep verification and sign valid transactions using the BLS multi-signature, then forward the signed transaction to a small and randomly sampled subset of neighbors to rapidly accumulate valid signatures. In the second stage, transactions whose aggregated signature count exceeds a threshold will be broadcast throughout the entire blockchain network and undergo deep verification with a specific probability, relieving edge nodes' verification burden. Moreover, verifiers record signers associated with failed deep verifications. Signers whose failures exceed a system threshold are quarantined to restrain the spread of spam transactions. Experimental results demonstrate that the proposed mechanism reduces energy consumption by at least 60% and 18.6% compared with the original and benchmark mechanisms respectively, while maintaining nearly identical transmission performance and ensuring that the proportion of invalid transactions propagated to honest nodes does not exceed 14%.

The rise of Embodied AI, including autonomous robots, cooperative autonomous driving, and augmented reality agents, is driving a deep integration of intelligent systems with the physical world, imposing stringent demands on the underlying network for real-time, low-latency interaction. However, these Embodied AI systems typically operate in complex polymorphic network environments, simultaneously handling heterogeneous identifiers such as content, IP, and geographic location. This causes traditional routing mechanisms to suffer from significant latency overhead and compatibility bottlenecks due to protocol conversion and adaptation, severely limiting the performance and responsiveness of Embodied AI applications. To address this challenge, we propose the Polymorphic Compatible Segment Routing (PCSR) protocol. PCSR adopts an innovative paradigm of decoupling the protocol from the infrastructure. It dynamically maps native protocol semantics to lightweight 8 -byte identifiers and utilizes compatibility logic encapsulated at the packet tail to achieve smooth compatibility with traditional networks without requiring large-scale modification of existing equipment. Furthermore, we built a zero-copy forwarding engine using the kernel eXpress Data Path (XDP) technology to fundamentally optimize data transmission efficiency. Experimental validation on a 10-node heterogeneous testbed shows that PCSR reduces end-toend latency by 32.7% and maintains a high throughput rate in hybrid network environments. This work demonstrates that PCSR provides an efficient and deployable routing solution for latency-critical, cross-domain collaborative services required by Embodied AI.

Using ambient magnetic for indoor localization has been a research focus in recent years. However, intricate magnetic feature patterns in complicated indoor ambiance, with particular emphasis on the multi-scale dynamics arising from diverse user motion states, further hinder localization accuracy and universality. To address these challenges, this paper originally proposes a novel mixed-frequency magnetic data representation learning-based framework (MixLoc) for accurate and universal localization. Our core idea is to systematically model the multi-scale dynamics-affected pedestrian indoor localization as a location-semantic learning problem based on mixed-frequency magnetic data. First, we propose a data augmentation method to automatically construct a mixed-frequency magnetic dataset. Then, we propose a novel encoder to learn temporal and spatial representations from these data and extract subtle differences among multi-scale sequences. Finally, a novel localization model is proposed to accurately infer locations by capturing significant global and local features from both temporal and spatial representations. The evaluation results based on comprehensive experiments show that the accuracy of MixLoc increases about 42% compared to other state-of-the-art approaches.

Due to the destruction of ground communication facilities, emergency communication networks are crucial for emergency rescue missions. UAV-Assisted Communication Network (UACN) is becoming a promising approach to establish emergency communication networks. However, the complexity and dynamics of UACN in disaster areas impose strict requirements on routing algorithms. To this end, this paper proposes an efficient routing scheme based on a double-layer UAV-assisted communication architecture to optimize the network performance of UACN and help emergency rescue missions to be completed efficiently. Specifically, we propose a double-layer UAV-assisted communication architecture based on the high coverage requirements of emergency scenarios and the delay-sensitive requirements of emergency rescue. Then, we propose an Attention mechanism-based Message passing Deep Reinforcement Learning (AMDRL) to optimize the routing problem in view of the fact that existing schemes rarely make full use of the information of the network environment. AMDRL uses the characteristics of graph neural network (GNN) to interact with the network topology environment and extract available information through the information passing process between topological links. Finally, we conduct simulation experiments in Knowledge-Defined Networking (KDN). The results show that the network performance of AMDRL is better than that of the baseline algorithm.

In-band Network Telemetry(INT) enables fine-grained telemetry by editing the packet header with the capability of programmable data plane to carry network status. However, INT could cause significant telemetry overhead without an effective system design. Existing measurement systems attempt to reduce this overhead by employing fixed-frequency INT sampling. Nonetheless, these methods lead to frequent measurements of network ports with large flows while neglecting ports with small flows for extended periods. In this paper, we introduce a lightweight passive telemetry system based on INT, called INTSelection. The core idea is to use a flow selection algorithm at the centralized controller so as to measure all active ports. Compared with the current method, INT-Selection reduces the bandwidth overhead by 58.2% and 2.7%.

Real-time network traffic classification plays a crucial role in ensuring Quality of Service and network security, and machine learning (ML) based methods achieve high classification accuracy but induce significant computational overhead. While SmartNIC solutions can offload classification tasks thus reducing CPU burdens, they still suffer from various limitations, manifested in limited computing capabilities, insufficiency in dynamic load handling and high latency from heterogeneous computing architectures. To solve these problems, we propose SmartTC, with three key designs: (1) SmartTC employs hardware-software co-design to optimize SmartNIC processing power, (2) SmartTC adopts a trafficaware dynamic batch submission strategy that adjusts submission policies based on real-time network load, and (3) SmartTC proposes parallel pipeline scheduling that ensures efficient task execution while minimizing communication overhead. Finally, we implement SmartTC on the BlueField-3 DPU and conduct extensive experiments for evaluations, and comparison results demonstrate that SmartTC significantly outperforms existing solutions. For example, it reduces average traffic classification time by up to$\mathbf{1 6. 8 \%}$under low loads and$\mathbf{9 0. 9 \%}$under high loads. Besides, SmartTC does not affect Bluefield-3 network services, and saves host CPU usage by at least two cores.

While the P4-based programmable data plane supports flexible configuration regardless of specific hardware or protocols, potential software and hardware failures threaten its performance. However, existing tools fail to efficiently test and automatically repair faults, relying on time-consuming and error-prone manual manipulation. To address these challenges, we propose REP4, a system that automatically detects, localizes, and repairs bugs in programmable data planes. First, it specifies high-level intents to express comprehensive data plane functions and eliminate invalid paths in the control flow graph, thereby accelerating testing. Second, faulty P4 code and table rules are automatically located by employing taint tracking. Finally, REP4 updates the erroneous fragments through constraint-based repair and performs regression testing for validation. We deploy REP4 on real switches, and extensive experiments on open-source P4_16 programs show that REP4 saves nearly half of the test packets than existing approaches with 100 % function coverage and localizes different types of root causes in 55s in the switch.p4 without false positives, and automatically updates$\mathbf{P 4}$code and table rules in 1.5s, with all updates successfully passing regression testing.

The encapsulated route images provided by vendors hide detailed implementation codes of routing protocol, making it challenging to implement the same routing protocol on heterogeneous devices, as well as to locate and repair the configuration errors. While some works attempt to address the above issues, they fail in large-scale deployment, manifested in insufficiency in complex routing behaviors, lack of clear mapping between routing behavior and protocol standards, and difficulty in obtaining an open-source license. To solve all the problems, we propose RTB, a novel tool to mine the causality between network configurations and routing behaviors, in order to assist operations personnel in better pinpointing configuration errors. First, RTB generates test cases based on configuration features using a coverage-guided combinatorial testing algorithm. Then, RTB employs a differential network model to obtain differential datasets storing the correspondence between configurations and routing behaviors. Finally, RTB utilizes the differential datasets to create multilayer causal graphs, so as to reason about the implementation details and differences in routing protocols, which assists network operation and maintenance personnel in locating configuration errors. Experimental results show that RTB only needs 35.3 % test cases, and improves code coverage by 10 % over Metha. Moreover, RTB can mine the causality with 98 % accuracy, which is 5 % more efficient than existing tools.

Network device misconfigurations can lead to security vulnerabilities and operational failures. Existing validation methods primarily rely on program analysis or machine learning, which are often labor-intensive and data-sensitive. While Large Language Models (LLMs) excel in text analysis, their direct application to configuration validation faces challenges such as limited domain knowledge, hallucinations, and non-interpretability. To address these issues, we propose ConfigChecker, an automated network configuration validation system that integrates LLMassisted structured knowledge extraction with knowledge graphbased reasoning. ConfigChecker involves two key phases: (1) Knowledge Graph Construction - We construct three instructiontuning datasets (EV, PG, and RE) to fine-tune LLMs for structured command information extraction, including view transitions, parameter constraints, and inter-command dependencies. The extracted knowledge is stored in a knowledge graph, providing a foundation for configuration validation. (2) Configuration Validation - A graph-based command validator ensures syntax and semantic correctness, while knowledge graph reasoning and indexing mechanism are used to detect inter-command dependency violations. We evaluate six instruction-tuned LLMs for structured knowledge extraction, achieving up to 90% accuracy on the EV and RE datasets. Additionally, our system shows a 9%-20% improvement in validation accuracy compared to three mainstream large models. Furthermore, the average response time per configuration line remains below 15 milliseconds. Results demonstrate that ConfigChecker significantly enhances validation accuracy and efficiency, providing a novel integration of LLMs and knowledge graphs for automated configuration validation.

Network traffic measurement is critical for effective network management. Sketch has been proven to be a promising network traffic measurement solution. Considering the skewed distribution of network traffic, where low-frequency mouse flows dominate and high-frequency elephant flows are fewer, recent sketch-based solutions employ hierarchical designs to enhance memory efficiency and accuracy. However, these solutions inevitably introduce additional challenges, including increased memory access overhead, severe hash collisions between elephant and mouse flows, and limited adaptability to dynamic network environments. In this paper, we propose LA-Sketch, an adaptive level-aware data structure. First, LA-Sketch employs a level-aware classifier to intelligently map each flow to its corresponding level, thereby reducing memory access overhead caused by hierarchical designs and mitigating hash collisions between elephant and mouse flows. Second, we introduce an adaptive counter configuration method that dynamically adjusts the number of counters at each level according to diverse network traffic distributions, which theoretically minimizes overall hash collisions. Finally, to adapt to the continuously changing network traffic characteristics, we propose an adaptive online training method that enables LA-Sketch's classifier to maintain high performance using only sketch query values for training, avoiding the significant overhead of massive traffic data collection. Extensive evaluations on two real-world network traces across five measurement tasks demonstrate that LA-Sketch outperforms state-of-the-art hierarchical sketches.

Heterogeneous networks utilize devices from multiple vendors to enhance failure prevention, but also pose challenges in network device management. Network configuration translation, as a critical aspect of management, facilitates the upgrading of pairwise nodes that back up each other and device replacements across different vendors. Manually translating configuration requires a significant amount of time and demands that administrators have ample experience. To address this, we introduce the Network Configuration Translation MultiAgent System (NetMAS), designed to manage heterogeneous manuals and streamline configuration translation. By efficiently handling diverse manuals, NetMAS simplifies the translation of network configurations across different devices and vendors. NetMAS consists of three core components: a hierarchical manual retriever, a detachable network configuration translator, and a memory module. Leveraging the advanced function-calling capabilities of the Large Language Models, NetMAS eliminates the need for external training, significantly enhancing overall efficiency. Our experimental results demonstrate that NetMAS can effectively retrieves configuration manuals. The retrieval accuracy reaches 94.11% when considering the top-5 results. Furthermore, it supports manuals from various vendors and devices. When translating from Cisco to Huawei, H3C, and Juniper, NetMAS improves accuracy by an average of$25.48 \times$and$12.38 \times$, respectively, compared to directly using GPT-4o and Deepseek.

In communication scenarios where information is highly valuable and privacy protection is essential, such as internal networks of governments and companies, network security plays a crucial role. The emergence of programmable networks offers a new approach to achieving network security. This paper proposes an endogenous security method based on programmable networks and, for the first time, uses the programmable data plane programming language P4 to implement dynamic address hopping and key field obfuscation. This method conceals the real IP, MAC, and Port information of the data packet during the communication process, thereby increasing the difficulty of identifying and tracking the location and identity of the attacker. At the same time, the implementation of dynamic address hopping and key field obfuscation does not require communication with the control plane; instead, it is implemented within the data plane, thereby saving on communication delay and memory usage. Through simulation experiments, the additional communication delay and CPU utilization introduced by the PNEST were evaluated. The results showed that the communication delay increased to approximately 8 ms and the CPU utilization to approximately 5% during the data packet transmission process, demonstrating the feasibility and practicality of the PNEST.

To be still yet still moving. - Do Hyun Choe

Pengxiang Lan, Haoyu Xu, Enneng Yang, Yuliang Liang, Guibing Guo, Jianzhe Zhao, Xingwei Wang. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Graph self-supervised learning aims to learn the intrinsic graph representations from unlabeled data, with broad applicability in areas such as computing networks. Although graph contrastive learning (GCL) has achieved remarkable progress by generating perturbed views via data augmentation and optimizing sample similarity, it performs poorly in heterophilic graph scenarios (where connected nodes are likely to belong to different classes or exhibit dissimilar features). In heterophilic graphs, existing methods typically rely on random or carefully designed augmentation strategies (e.g., edge dropping) for contrastive views. However, such graph structures exhibit intricate edge relationships, where topological perturbations may completely alter the semantics of neighborhoods. Moreover, most methods focus solely on local contrastive signals while neglecting global structural constraints. To address these limitations, inspired by graph coloring, we propose a novel Coloring learning for heterophilic graph Representation framework, CoRep, which: 1) Pioneers a coloring classifier to generate coloring labels, explicitly minimizing the discrepancy between homophilic nodes while maximizing that of heterophilic nodes. A global positive sample set is constructed using multi-hop same-color nodes to capture global semantic consistency. 2) Introduces a learnable edge evaluator to guide the coloring learning dynamically and utilizes the edges' triplet relations to enhance its robustness. 3) Leverages Gumbel-Softmax to differentially discretize color distributions, suppressing noise via a redundancy constraint and enhancing intra-class compactness. Experimental results on 14 benchmark datasets demonstrate that CoRep significantly outperforms current state-of-the-art methods.

Model merging integrates multiple expert models with diverse capabilities into a unified framework, facilitating collaborative learning. However, most existing methods assume simultaneous access to all models, which is often impractical in real-world scenarios where models are received sequentially. While some studies have investigated continual model merging (CMM)--which involves sequentially merging multiple models--the challenge of balancing prior knowledge (stability) and incorporating new tasks (plasticity) remains unresolved. This paper, for the first time, formally defines the stability and plasticity of CMM from the perspective of orthogonal projection. Subsequently, we analyze the relationships among the spaces spanned by task data, historical gradients, and accumulated gradients. Building on this, we propose a data-free \textbf{D}ual \textbf{O}rthogonal \textbf{P}rojection (DOP) method, which eliminates data dependence and mitigates interference between the merged model and models for old and new tasks by projecting their parameter differences onto their respective approximate data spaces. Finally, to solve potential conflicts between stability and plasticity, we reformulate DOP as a multi-objective optimization problem and employ a multi-gradient descent algorithm to obtain a Pareto-optimal solution. Extensive experiments across multiple architectures and task configurations validate that our approach significantly outperforms state-of-the-art CMM methods.

As distributed deep learning training (DLT) systems scale, collective communication has become a significant performance bottleneck. While current approaches optimize bandwidth utilization and task completion time, existing communication libraries (CCLs) backends fail to efficiently manage GPU resources during algorithm execution, limiting the performance of advanced algorithms. This paper proposes ResCCL, a novel CCL backend designed for Resource-Efficient Scheduling to address key limitations in current systems. ResCCL enhances execution efficiency by optimizing scheduling at the primitive level (e.g., send and recvReduceCopy), enabling flexible thread block (TB) allocation, and generating lightweight communication kernels to minimize runtime overhead. Our approach tackles the global scheduling problem, reduces idle TB resources, and enhances communication bandwidth. Evaluation results demonstrate that ResCCL achieves up to 2.5× improvement in bandwidth performance compared to both NCCL and MSCCL. It reduces SM resource overhead by 77.8% and increases TB utilization by 41.6% while running the same algorithms. In end-to-end DLT, ResCCL boosts Megatron's throughput by up to 39%.

Data augmentation has become a promising method of mitigating data sparsity in sequential recommendation.Existing methods generate new yet effective data during model training to improve performance.However, deploying them requires retraining, architecture modification, or introducing additional learnable parameters.These steps are time-consuming and costly for well-trained models, especially when the model scale becomes large.In this work, we explore the test-time augmentation (TTA) for sequential recommendation, which augments the inputs during the model inference and then aggregates the model's predictions for augmented data to improve final accuracy.It avoids significant time and cost overhead from the previously mentioned steps.We first experimentally disclose the potential of existing augmentation operators for TTA and find that the Mask and Substitute consistently achieve better performance.Further analysis reveals that these two operators are effective because they retain the original sequential pattern while adding appropriate perturbations.Meanwhile, we argue that these two operators still face time-consuming item selection or interference information from mask tokens.Based on the analysis and limitations, we present TNoise and TMask.The former injects uniform noise into the original representation, avoiding the computational overhead of item selection.The latter blocks

Logical reasoning-based recommendation methods employ logical rules to mitigate the adverse effects of noise items in short interaction sequences on recommendation accuracy.However, there are two problems with existing methods: 1) As the length of the interaction sequence increases, introducing more noise items exacerbates the negative impact on logical reasoning, thereby reducing the accuracy of these methods.2) They are often dominated by the user's single primary interest, which prevents simultaneous consideration of users' multiple-aspect interests in long sequences.To address these issues, we propose a novel dEnoising Multi-Interestaware Logical rEasoning (EMILE) method for long-sequence recommendation.Specifically, we design a logical rule-based interest extractor that enhances the importance of preferred items in constructing user interests while minimizing the negative impact of disliked items.This extractor effectively mitigates the adverse effects of noise items in long interaction sequences.Furthermore, we propose a novel multi-interest learning strategy that optimizes two new objective functions-interest probability distribution contrastive loss and interest logical reasoning contrastive loss-to ensure the model simultaneously considers multiple-aspect interests.These two objective functions require that the target item is more * Corresponding authors.

Graph Neural Networks (GNNs)-based recommendation algorithms typically assume that training and testing data are drawn from independent and identically distributed (IID) spaces. However, this assumption often fails in the presence of out-of-distribution (OOD) data, resulting in significant performance degradation. In this study, we construct a Structural Causal Model (SCM) to analyze interaction data, revealing that environmental confounders (e.g., the COVID-19 pandemic) lead to unstable correlations in GNN-based models, thus impairing their generalization to OOD data. To address this issue, we propose a novel approach, graph representation learning via causal diffusion (CausalDiffRec) for OOD recommendation. This method enhances the model's generalization on OOD data by eliminating environmental confounding factors and learning invariant graph representations. Specifically, we use backdoor adjustment and variational inference to infer the real environmental distribution, thereby eliminating the impact of environmental confounders. This inferred distribution is then used as prior knowledge to guide the representation learning in the reverse phase of the diffusion process to learn the invariant representation. In addition,we provide a theoretical derivation that proves optimizing the objective function of CausalDiffRec can encourage the model to learn environment-invariant graph representations, thereby achieving excellent generalization performance in recommendations under distribution shifts. Our extensive experiments validate the effectiveness of CausalDiffRec in improving the generalization of OOD data, and the average improvement is up to 10.69% on Food, 18.83% on KuaiRec, 22.41% on Yelp2018, and 11.65% on Douban datasets.

The distributionally robust optimization (DRO)-based graph neural network methods improve recommendation systems' out-of-distribution (OOD) generalization by optimizing the model's worst-case performance. However, these studies fail to consider the impact of noisy samples in the training data, which results in diminished generalization capabilities and lower accuracy. Through experimental and theoretical analysis, this paper reveals that current DRO-based graph recommendation methods assign greater weight to noise distribution, leading to model parameter learning being dominated by it. When the model overly focuses on fitting noise samples in the training data, it may learn irrelevant or meaningless features that cannot be generalized to OOD data. To address this challenge, we design a Distributionally Robust Graph model for OOD recommendation (DRGO). Specifically, our method first employs a simple and effective diffusion paradigm to alleviate the noisy effect in the latent space. Additionally, an entropy regularization term is introduced in the DRO objective function to avoid extreme sample weights in the worst-case distribution. Finally, we provide a theoretical proof of the generalization error bound of DRGO as well as a theoretical analysis of how our approach mitigates noisy sample effects, which helps to better understand the proposed framework from a theoretical perspective. We conduct extensive experiments on four datasets to evaluate the effectiveness of our framework against three typical distribution shifts, and the results demonstrate its superiority in both independently and identically distributed distributions (IID) and OOD.

Unmanned aerial vehicle (UAV) assisted non-orthogonal multiple access (NOMA) communications show great potential in providing simultaneous service to various users. This paper proposes a novel reconfigurable intelligent surface (RIS) aided UAV-swarm NOMA system, where the Coordinated Multi-Point (CoMP) technique is employed to manage inter-cell interference for all users. The primary objective is to maximize the overall security rate by jointly optimizing the UAV swarm trajectories, power distribution among the UAVs, and the reflection coefficients of the RIS. To address the non-convex nature and interdependence of this optimization problem while accommodating long-term dynamic requirements, we introduce an optimization framework that combines alternating optimization (AO) with the weighted-QMIX algorithm. Specifically, successive convex optimization (SCA) is utilized within the AO method to determine the optimal power allocation and RIS coefficients. Meanwhile, the weighted-QMIX algorithm conducts online trajectory optimization for all UAVs through a multi-agent deep reinforcement learning process. Numerical results demonstrate that: 1) the proposed method achieves a higher communication security rate compared to systems lacking RIS, NOMA, or stationary UAVs; and 2) the proposed algorithm outperforms a pure deep reinforcement learning (DRL) algorithm in terms of both computational complexity and security performance.

This paper investigates the physical layer security (PLS) issue of unmanned aerial vehicle (UAV) aided multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) networks with randomly distributed passive eavesdroppers (Eves). Considering Nakagami-m fading, we propose a novel secure communication protocol that integrates transmit antenna selection (TAS) and maximum ratio combining (MRC) diversity technology. Specifically, to tackle the challenges of low spectrum efficiency and PLS performance, we propose two TAS solutions: TAS-max UN and TAS-max UF, the first one aims to enhance the performance of UN and the other one focuses on UF. To mitigate the impact of passive eavesdropping, a secure protected zone is established around the UAV to limit the Eve’s ability. Accordingly, we derive the closed-form expressions for the ergodic secrecy rate to evaluate the impact of spatial randomness. Then, the accuracy of the derived expressions is verified through Monte-Carlo simulations.To further support the theoretical analysis, asymptotic expressions under the high-SNR regime are derived, which offer valuable insights into secrecy rate trends and model convergence. Moreover, we propose a three-dimensional UAV deployment optimization framework that adopts a hybrid approach combining grid-based evaluation and Genetic Algorithm refinement, which improves ESR performance while significantly reducing computational complexity. In addition, a Simulated Annealing based power allocation scheme is introduced to optimize the power coefficient aF, achieving enhanced secrecy rate with improved search efficiency and adaptability. Extensive simulation results confirm that the proposed TAS/MRC framework, together with the secure protected zone, consistently outperforms conventional OMA and MRT schemes in terms of secrecy rate and robustness. The impact of key system parameters, including power allocation, UAV altitude, antenna configuration, and Eve density, is also thoroughly analyzed.

In this work, we investigate the physical layer security (PLS) of ambient backscatter communication non-orthogonal multiple access (AmBC-NOMA) networks where non-colluding eavesdroppers (Eves) are randomly distributed. In the proposed system, a base station (BS) transmits a superimposed signal to a typical NOMA user pair, while a backscatter device (BD) simultaneously transmits its unique signal by reflecting and modulating the BS’s signal. Meanwhile, Eves passively attempt to wiretap the ongoing transmissions. Notably, the number and locations of Eves are unknown, posing a substantial security threat to the system. To address this challenge, the BS injects artificial noise (AN) to mislead the Eves, and a protected zone is employed to create an Eve-exclusion area around the BS. Theoretical expressions for outage probability (OP) and intercept probability (IP) are provided to evaluate the system’s reliability-security trade-off. Asymptotic behavior at high signal-to-noise ratio (SNR) is further explored, including the derivation of diversity orders for the OP. Numerical results validate the analytical findings through extensive simulations, demonstrating that both the AN injection and protected zone can effectively enhance PLS. Furthermore, analysis and insights of different key parameters, including transmit SNR, reflection efficiency at the BD, power allocation coefficient, power fraction allocated to desired signal, Eve-exclusion area radius, Eve distribution density, and backscattered AN cancellation efficiency, on OP and IP are also provided.

Time-Sensitive Networking (TSN), renowned for its deterministic properties, has become a pivotal technology under-pinning real-time industrial control in Cyber-Physical Systems. Existing research emphasizes enhancing the transmission services of TSN networks for control applications by improving flow schedulability and minimizing end-to-end delay. However, these studies abstract the performance requirements of control applications into rigid, impractical constraints for flow scheduling, disrupting the connection between control optimization and transmission enhancement, and eventually undermining genuine progress in industrial control. Within a co-optimization framework of communication and control, this paper proposes AFS-RT, an Adaptive TSN Flow Scheduling method for Robotic arm Teleoperation, a representative industrial control application. Specifically, through a comprehensive analysis of the teleoperation case, we first integrate slot allocation-based flow scheduling with remote control to formulate a control-driven co-optimization model. To tackle the complexities arising from the implicit mapping between communication and control, we augment the Deep Reinforcement Learning agent responsible for slot allocation with slot-correlation-guided feature extraction, improving feature comprehension by leveraging inherent correlations between slots and thereby boosting the agent’s decision-making capabilities. Extensive testbed and simulation experiments demonstrate that AFS-RT significantly improves teleoperation performance under diverse network conditions compared to SOTA algorithms.

Segment Routing over IPv6 (SRv6) is an innovative and adaptable source routing technique that enhances interconnection services. It plays a pivotal role in next-generation networking technologies, providing crucial support for network telemetry, computing power networks, and related technologies. The end-to-end connectivity capability of SRv6 is highly regarded by ISPs, driving its widespread deployment in networks. However, configuring an SRv6 network can be challenging and prone to errors due to the complexity of low-level configuration languages and numerous protocol parameters. To address this issue, we present AutoSRv6, a system designed to synthesize SRv6 configurations for large, evolving networks using high-level abstractions of network topology and policies. AutoSRv6 leverages formal constraint-solving techniques and SMT solvers to compute protocol parameters and generate configuration files that align with network policies. Furthermore, AutoSRv6 incorporates a mechanism to overcome the constraints imposed by hardware, mapping the end-to-end path to a SID (Segment Identifier) sequence. We have developed a prototype of AutoSRv6 and conducted experiments on diverse network topologies, evaluating its performance with various network policies. The results show that autoSRv6 can generate the network configuration satisfying the policy, the time cost of IGP synthesis is better than the existing method, and the length of the segment list is optimized by more than 2 times.

The goal of continual learning (CL) is to learn from a series of continuously arriving new tasks without forgetting previously learned old tasks. To avoid catastrophic forgetting of old tasks, orthogonal gradient projection (OGP) based CL methods constrain the gradients of new tasks to be orthogonal to the space spanned by old tasks. This strict gradient constraint will limit the learning ability of new tasks, resulting in lower performance on new tasks. In this paper, we first establish a unified framework for OGP-based CL methods. We then revisit OGP-based CL methods from a new perspective on the loss landscape, where we find that when relaxing projection constraints to improve performance on new tasks, the unflatness of the loss landscape can lead to catastrophic forgetting of old tasks. Based on our findings, we propose a new Dual Flatness-aware OGD framework that optimizes the flatness of the loss landscape from both data and weight levels. Our framework consists of three modules: data and weight perturbation, flatness-aware optimization, and gradient projection. Specifically, we first perform perturbations on the task's data and current model weights to make the task's loss reach the worst-case. Next, we optimize the loss and loss landscape on the original data and the worst-case perturbed data to obtain a flatness-aware gradient. Finally, the flatness-aware gradient will update the network in directions orthogonal to the space spanned by the old tasks. Extensive experiments on four benchmark datasets show that the framework improves the flatness of the loss landscape and performance on new tasks, and achieves state-of-the-art (SOTA) performance on average accuracy across all tasks.

Machine learning (ML) typically relies on the assumption that training and testing distributions are identical and that data are centrally stored for training and testing. However, in real-world scenarios, distributions may differ significantly, and data are often distributed across different devices, organizations, or edge nodes. Consequently, it is to develop models capable of effectively generalizing across unseen distributions in data spanning various domains. In response to this challenge, there has been a surge of interest in federated domain generalization (FDG) in recent years. FDG synergizes federated learning (FL) and domain generalization (DG) techniques, facilitating collaborative model development across diverse source domains for effective generalization to unseen domains, all while maintaining data privacy. However, generalizing the federated model under domain shifts remains a complex, underexplored issue. This article provides a comprehensive survey of the latest advancements in this field. Initially, we discuss the development process from traditional ML to domain adaptation (DA) and DG, leading to FDG, as well as provide the corresponding formal definition. Subsequently, we classify recent methodologies into four distinct categories: federated domain alignment (FDAL), data manipulation (DM), learning strategies (LSs), and aggregation optimization (AO), detailing appropriate algorithms for each. We then overview commonly utilized datasets, applications, evaluations, and benchmarks. Conclusively, this survey outlines potential future research directions.

Distributed machine learning over geo-distributed clouds enables joint training of data located in different regions, alleviating the burden of transferring large volumes of training datasets, which greatly saves bandwidth. However, the limited capacity of WAN links slows down the inter-cloud communications, which significantly decelerates the synchronization of distributed machine learning over geo-distributed clouds. Besides, the multi-tenancy in clouds results in multiple training tasks running simultaneously, whose synchronizations consistently compete for the limited WAN bandwidth with each other, which further aggravates the training performance of each task. While existing works optimize synchronizations through techniques like gradient compression, multi-resource interleaving and so on, none of them targets at the synchronization congestion especially due to multi-tenant learning, which results in inferior training performance.To solve these problems, we propose a simple but effective scheme, SCC, for fast and efficient multi-tenant learning via synchronization congestion control. SCC monitors the cross-cloud network conditions and evaluates the synchronization congestion level based on the round-trip transmission time for each synchronization. Then SCC alleviates synchronization congestion via controlling the synchronization frequency according to the synchronization congestion level in a probabilistic way. Extensive experiments are conducted within our testbeds consisted of 16 NVIDIA V100 GPUs to evaluate the performance of SCC, and comparison results show that SCC can reduce the average training completion time and makespan by up to 28.6% and 43.2% over SAP-SGD [1]. Targeted experiments are conducted to demonstrate the effectiveness and robustness of SCC.

Network measurement is critical for various network applications, but scaling measurement techniques to the network-wide level is challenging for existing sketch-based solutions. In software switches, centralized deployment provides low resource usage but suffers from poor load balancing. In contrast, collaborative measurement achieves load balancing through flow distribution across software switches but requires high resource usage. This paper presents a novel distributed deployment framework that overcomes the limitations above. First, our framework is lightweight such that it splits sketches into segments and allocates them across forwarding paths to minimize resource usage and achieve load balancing. This also enables per-packet load balancing by distributing computations across software switches. Second, through a novel collaborative strategy, our framework achieves finer-grained flow distribution and further optimizes load balancing. Third, we further optimize load balancing by eliminating the mutual influence among forwarding paths. We evaluate the proposed framework on various network topologies and different sketches. Results indicate our solution matches the load balancing of collaborative measurement while approaching the low resource usage of centralized deployment. Moreover, it achieves superior performance in per-packet load balancing, which is not considered in previous deployment solutions.

Advanced router configuration synthesizers aim to prevent network outages by automatically synthesizing configurations that implement routing protocols. However, the lack of interpretability makes operators uncertain about how low-level configurations are synthesized and whether the automatically generated configurations correctly align with routing intents. This limitation restricts the practical deployment of synthesizers.In this paper, we present NetKG, an interpretable configuration synthesis tool.(i) NetKG leverages a knowledge graph as the intermediate representation for configurations, reformulating the configuration synthesis problem as a configuration knowledge completion task; (ii) NetKG regards network intents as query tasks that need to be satisfied in the current configuration space, achieving this through knowledge reasoning and completion; (iii) NetKG explains the synthesis process and the consistency between configuration and intent through the configuration knowledge involved in reasoning and completion.We show that NetKG can scale to realistic networks and automatically synthesize intent-compliant configurations for static routes, OSPF, and BGP. It can explain the consistency between configuration and intent at different granularities through a visual interface. Experimental results indicate that NetKG synthesizes configurations in 2 minutes for a network with up to 197 routers, which is 7.37x faster than the SMT-based synthesizer.

With the recent advancement in network technologies, the vehicle based medical networks extend medical services to mobile vehicles, thereby offering flexible and efficient healthcare services for vehicle users in need. The integration of vehicle based medical network and edge computing enables computation intensive medical service tasks to be offloaded on edge servers, to provide fast service response for vehicle users. An efficient task offloading and resource allocation strategy is critical for Vehicle based Medical Edge Computing System (VMECS) to satisfy real-time and reliability requirements while ensuring service performance. To this end, in this paper, we investigate the problem of task computation allocation in VMECS networks. By introducing deep reinforcement learning, we first present a novel VMECS architecture to automatically achieve the optimal task offloading and resource allocation through the multi-agent collaboration, thereby improving service performance. Then, we formulate the problem of task offloading and resource allocation in VMECS networks as an optimization model with the aim of maximizing task success rate by jointly considering communication interferences, resource allocation and delay requirements. To solve it, we further devise a Distributed distributional deterministic policy gradients based Task offloading and Resource allocation (DTR) algorithm. Final simulation results demonstrate that compared with benchmark algorithms, DTR algorithm can obtain higher task success rate, smaller service time, and less task processing time.

With the rapid growth in data center network bandwidth far outpacing improvements in CPU performance, traditional software middleboxes running on servers have become inefficient. The emerging data processing units aim to address this by offloading network functions from the CPU. However, as DPUs are still a new technology, there lacks comprehensive evaluation of their capabilities for accelerating middleboxes. This paper benchmarks and analyzes the performance of offloading middleboxes onto the NVIDIA BlueField-2 DPU. Three key DPU capabilities are explored: flow tables offloading, ARM subsystem packet processing, and connection tracking hardware offload. By applying these to implement representative middleboxes for firewall, packet scheduling, and load balancing, their performance is characterized and compared to conventional CPU-based versions. Results reveal the high throughput of flow tables offloading for stateless firewalls, but limitations as pipeline depth increases. Packet scheduling using ARM cores is shown to currently reduce performance versus CPU-based scheduling. Finally, while connection tracking hardware offload boosts load balancer bandwidth, it also weakens connection creation abilities. Key lessons on efficient middleboxes offloading strategies with DPUs are provided to guide further research and development. Overall, this paper offers useful benchmarking and analysis of emerging DPUs for accelerating middleboxes in modern data centers.

In modern data centers and enterprise networks, software switches have become critical components for achieving flexible and efficient network management. Due to resource constraints in software switches, sketches have emerged as a promising approach for network traffic measurement. However, their accuracy is often impacted by hash collisions. Existing hash functions treat all collisions equally, failing to account for the differing impacts of collisions involving elephant flows versus mouse flows. We propose FSA-Hash, a novel flow-size-aware hashing scheme that separates elephant flows from each other and from mouse flows, minimizing the most detrimental collisions. FSA-Hash is designed based on two insights: separating elephant flows from mouse flows avoids overestimating mouse flows, while separating elephant flows from each other enables accurate heavy-hitter detection. We implement FSA-Hash using machine learning models trained on network traffic data (LFSA-Hash), and also design a lightweight online variant (OLFSA-Hash) that learns the hash model solely from sketch queries on the software switch, obviating traffic collection overheads. Evaluations across four sketches and two tasks demonstrate FSA-Hash’s superior accuracy over standard hash functions. Moreover, OLFSA-Hash closely matches LFSA-Hash’s performance, making it an attractive option for adaptively refining the hash model without monitoring traffic.

In post-disaster emergency medical rescue operations, rapidly establishing an adaptive and flexible edge computing (EC) network, balancing data offloading with energy consumption, and ensuring the stable operation of the network have become urgent priorities. To address these challenges, we proposed an unmanned aerial vehicle (UAV)-assisted microservice mobile edge computing (MEC) architecture. The architecture can be rapidly deployed to provide temporary network coverage and EC services in disaster-stricken areas. A transformer-based resource management (TBRM) approach is utilized to optimize data offloading efficiency and reduce energy consumption, thereby maximizing the service time of the architecture. To enhance the security and reliability of the architecture, four microservices are designed to manage the full UAV lifecycle, and UAV identity authentication is implemented through dual digital signature certificates. Large-scale simulation experiments have demonstrated the effectiveness of the architecture in complex rescue scenarios, providing strong technical support for postdisaster medical rescue efforts.

Blockchain enhances trust and collaboration among entities through its inherent features of transparency, immutability, and traceability, leading to its extensive integration into Multi-access Edge Computing (MEC). However, existing transaction propagation mechanisms require MEC devices to consume significant computing resources for complex transaction verification, increasing their vulnerability to malicious attacks. Adversaries can exploit this by flooding the blockchain network with spam transactions, aiming to deplete device energy and disrupt system performance. To cope with these issues, this paper proposes a reputation-based energy-efficient transaction propagation mechanism that alleviates spam transaction attacks while reducing computing resources and energy consumption. Firstly, we design a subjective logic-based reputation scheme that assesses node trust by integrating local and recommended opinions and incorporates opinion acceptance to counteract false evidence. Then, we optimize the transaction verification method by adjusting transaction discard and verification probabilities based on the proposed reputation scheme to curb the propagation of spam transactions and reduce verification consumption. Finally, we enhance the transaction transmission strategy by prioritizing nodes with higher reputations, enhancing both resilience to spam transactions and transmission reliability. A series of simulations demonstrate the effectiveness of the proposed mechanism.

With the scale of the Internet of Things (IoT) system growing constantly, node failures frequently occur due to device malfunctions or cyberattacks. Existing robust network generation methods utilize heuristic algorithms or neural network approaches to optimize the initial topology. These methods do not explore the core of topology robustness, namely how edges are allocated to each node in the topology. As a result, these methods use massive iterative processes to optimize the initial topology, leading to substantial time overhead when the scale of the topology is large. We examine various robust networks and observe that uniform degree distribution is the core of topology robustness. Consequently, we propose a novel UNIformity driven robusT topologY generation scheme (UNITY) for IoT systems to prevent the node degree from becoming excessively high or low, thereby balancing node degrees. Comprehensive experimental results demonstrate that networks generated with UNITY have an “olive-like” topology consisting of a substantial number of medium-degree nodes and possess strong robustness against both random node failures and targeted attacks. This promising result indicates that the UNITY makes a significant advancement in designing robust IoT systems.

In large-scale dynamic network environments, optimizing the computation offloading and resource allocation strategy is key to improving resource utilization and meeting the diverse demands of User Equipment (UE). However, traditional strategies for providing personalized computing services face several challenges: dynamic changes in the environment and UE demands, along with the inefficiency and high costs of real-time data collection; the unpredictability of resource status leads to an inability to ensure long-term UE satisfaction. To address these challenges, we propose a Knowledge-Defined Networking (KDN)-based Adaptive Edge Resource Allocation Optimization (KARO) architecture, facilitating real-time data collection and analysis of environmental conditions. Additionally, we implement an environmental resource change perception module in the KARO to assess current and future resource utilization trends. Based on the real-time state and resource urgency, we develop a deep reinforcement learning-based Adaptive Long-term Computation Offloading and Resource Allocation (AL-CORA) strategy optimization algorithm. This algorithm adapts to the environmental resource urgency, autonomously balancing UE satisfaction and task execution cost. Experimental results indicate that AL-CORA effectively improves long-term UE satisfaction and task execution success rates, under the limited computation resource constraints.

The advent of Multi-access Edge Computing (MEC) has empowered Internet of Things (IoT) devices and edge servers to deploy sophisticated Deep Neural Network (DNN) applications, enabling real-time inference. Many concurrent inference requests and intricate DNN models demand efficient multi-DNN inference in MEC networks. However, the resource-limited IoT device/edge server and expanding model size force models to be dynamically deployed, resulting in significant undesired energy consumption. In addition, parallel multi-DNN inference on the same device complicates the inference process due to the resource competition among models, increasing the inference latency. In this paper, we propose a Resource-aware and Dynamic DNN Deployment (R3D) scheme with the collaboration of end-edge-cloud. To mitigate resource competition and waste during multi-DNN parallel inference, we develop a Resource Adaptive Management (RAM) algorithm based on the Roofline model, which dynamically allocates resources by accounting for the impact of device-specific performance bottlenecks on inference latency. Additionally, we design a Deep Reinforcement Learning (DRL)-based online optimization algorithm that dynamically adjusts DNN deployment strategies to achieve fast and energy-efficient inference across heterogeneous devices. Experiment results demonstrate that R3D is applicable in MEC environments and performs well in terms of inference latency, resource utilization, and energy consumption.

With the advent of network function virtualization (NFV), many users resort to network service provisioning through virtual network function instances (VNFIs) run on the standard physical server in clouds. Following this trend, NFV markets are emerging, which allow a user to procure VNFIs from cloud service providers (CSPs). In such procurement process, it is a significant challenge to ensure differential privacy and truthfulness while explicitly considering dynamic resource provisioning, location sensitiveness and budget of each VNFI. As such, we design a differentially private and truthful reverse auction with dynamic resource provisioning (PTRA-DRP) to resolve the VNFI procurement (VNFIP) problem. To allow dynamic resource provisioning, PTRA-DRP enables CSPs to submit a set of bids and accept as many as possible, and decides the provisioning VNFIs based on the auction outcomes. To be specific, we first devise a greedy heuristic approach to select the set of the winning bids in a differentially privacy-preserving manner. Next, we design a pricing strategy to compute the charges of CSPs, aiming to guarantee truthfulness. Strict theoretical analysis proves that PTRA-DRP can ensure differential privacy, truthfulness, individual rationality, computational efficiency and approximate social cost minimization. Extensive simulations also demonstrate the effectiveness and efficiency of PTRA-DRP.

A reconfigurable intelligent surface (RIS) assisted semi-grant-free (SGF) non-orthogonal multiple access (NOMA) system is investigated. Unlike existing works that only focus on short-term resource allocation, we study a long-term power-saving optimization problem under queue stability constraints and utilize Lyapunov stability theory to deal with delay-aware resource allocation. We first transform the long-term problem into a series of per-time-slot problems by exploiting the Lyapunov theory. Then, the objective function is minimized by alternatingly optimizing the power allocation, channel assignment, and RIS reflection coefficients. In particular, the channel assignment subproblem is solved by invoking a many-to-one matching algorithm. The power allocation sub-problem is addressed by the developed fractional programming algorithm. The reflection coefficients design sub-problem is solved by a penalty-based method, which tackles the rank one constraint and optimizes reflection coefficients. The numerical results validate the effectiveness and show that it can achieve queue stability by setting the Lyapunov parameters. It also shows that the proposed RIS-assisted SGF NOMA system outperforms without RIS and random RIS phase-shift baselines.

With the continuous development of Internet technology, social networks have provided convenient conditions for information dissemination. The rapid dissemination of information has provided us with great convenience, but some criminals extensively spread rumors based on such convenience, adversely affecting social stability. In this context, two key challenges arise: the survivability of rumors, which refers to their persistence and long-term impact, and the dynamic viewpoint changes of individuals, which influence how rumors spread and diminish over time. This article proposes a dynamic-susceptible-exposed-infected-recovered (DSEIR) rumor propagation model based on human social behavior to solve the spread of rumors problem. This model considers the characteristics of rumors spread in social networks with the Markov chain and makes the simulation more authentic. To suppress rumor propagation, we introduce the concept of rumor survivability and propose a dynamic truth movement blocking strategy, which adapts to people’s evolving viewpoints to curb the influence of rumors effectively. Finally, we analyze the proposed model and blocking strategy in four real networks. The experimental results show that the proposed propagation model can authentically simulate the propagation of rumors in social networks and the proposed blocking strategy efficiently suppresses the rumor propagation. We have released our code here:https://github.com/gf9264/DSEIR.

With the widespread application of 5G and artificial intelligence (AI) technology, the Internet of Things (IoT) has been expanding and integrated into various aspects of our daily lives. However, this also poses challenges such as the ubiquitous demand for communication and computing resources, and data privacy issues. Considering its flexible deployment, high security, and ease of scalability, blockchain-enabled edge computing IoT network (BECIN) has become a promising solution to provide secure and fast communication and computing services. However, existing research on computation offloading in edge computing largely overlooks the stochastic arrival of computational tasks and the potential variability in the number, locations, and resource provisions of edge computing service providers. Therefore, we propose a dynamic, self-adjusting BECIN framework aimed at providing long-term stable, efficient, and secure edge computing data offloading services for ground users in a specific region. This framework supports the dynamic entry and exit of edge computing service providers. Additionally, we introduce a novel dynamic Dueling DDQN approach to update the offloading and resource management policies based on changes in resource provisioning. Experimental results demonstrate the feasibility and superior performance of our framework on system cost and system latency.

Recently, Edge-Cloud Collaborative (ECC) has emerged as an efficient and promising technique to empower various computation-intensive applications in Digital Twin Network (DTN). The integration of ECC and DTN serves to bridge the gap between data analysis and physical states. In ECC, a reliable and optimal task offloading scheme is required to maximize resource utilization and provide satisfying services to End Users (EU). However, existing offloading schemes still face significant challenges, such as the instability and complexity of network topologies, the intricacies of massive data, and the lack of trust among EU. In this paper, we propose anenhancinGedge-clOud collaboraTion wiTh blockchain-assistEd digital twin intelligence offloadiNgscheme (GOTTEN) which transmits large-scale tasks generated by DTs to Edge Station (ES) or Cloud Station (CS) in dynamic DTN scenarios. We first formulate this resource allocation and task offloading problem and provide an appropriate initial solution which guarantees that tasks generated by DTs can be accurately mapped to physical entities, while optimizing block allocation and reducing the decision space of task offloading. Then, we employ the Lagrange Multiplier based Distributed Island model-enhanced Genetic Algorithm (LM-DIGA) to transform our formulated problem into a convex form and achieve an optimal resource allocation under a specific scheme. Additionally, our proposed architecture also leverages blockchain verification mechanisms to enhance system stability, strengthening privacy protection for DT data as well. Finally, extensive simulation results demonstrate that, compared with seven baselines, our proposed scheme achieves a 10 percent the total system delay and privacy overhead with regard to other schemes in ECC.

Mobile edge computation (MEC) is envisioned as a prospective approach for processing the computation-intensive and delay-sensitive tasks of smart mobile devices (SMDs) through offloading them to base stations (BSs) nearby. In fact, efficient task offloading mechanisms are crucial to accomplish an MEC system. The key challenge is to make on-spot decisions upon the arrival of each task and at the same time achieve truthfulness of each SMD. The challenge further escalates, when the unique characteristics of an MEC system, such as locality constraint, delay constraint, etc., are explicitly considered. To solve the challenge, we present a truthful online combinatorial auction-based mechanism (TOCA) for task offloading in an MEC system. Specifically, we first devise the candidate offloading scheme determination algorithm, aiming to determine the candidate offloading schemes of an SMD upon the arrival of its task. Next, we devise the winning offloading scheme selection and pricing algorithm based on the online primal-dual optimization framework, to decide the winning scheme among the SMD's candidate offloading schemes and calculate its payment. By solid theoretical analysis, we verify that TOCA achieves truthfulness, individual rationality and computational efficiency and a smaller competitive ratio. Trace-driven simulation studies validate the effectiveness and efficacy of TOCA.

A UAV team shows tremendous potential for various mobile scenarios. However, some evidences reveal their vulnerability to False Data Injection (FDI) attacks, which can significantly jeopardize the flight security or even lead to catastrophic incidents. Existing studies primarily focus on detecting or defending against FDI attacks at the trajectory control of individual UAVs, leaving a gap in a comprehensive secure mechanism that can simultaneously detect, localize, and compensate for such attacks across an entire UAV team. The complexity of developing such a solution is magnified by the multiple design goals, the inherent sophistication of UAV team, and practical attack assumptions. In this paper, we propose a joint secure framework based on multi-task deep learning to simultaneously detect FDI attacks, localize the compromised components, and compensate control signals to mitigate the impact of FDI attacks on promising UAV teams. Specifically, we design an all-in-one deep learning model framework with a temporal-spatial information extraction module and a hierarchical multi-task module to perform three tasks simultaneously. Moreover, we introduce an iterative learning method with experience replay to counteract knowledge decay during model training. Extensive experiments and real flight demonstrations are presented to validate the improved performance and the benefits of our proposed secure method.

Software-Defined Data Center Networks (SDDCNs) utilizes Software Defined Networking (SDN) as a network architecture to achieve highly flexible, programmable, and automated management of Data Center Networks (DCNs). The high energy consumption of DCNs remains a persistent and significant challenge. Thus, the energy saving is crucial and imperative for DCNs. Current energy-efficient solutions primarily rely on flow consolidation and dynamic device sleeping techniques to reduce energy consumption. However, these approaches often yield long Flow Completion Time (FCT), potentially resulting in violations of service-level agreements, particularly for delay-sensitive applications. In this paper, we formulate the problem of minimizing the power consumption in SDDCNs as key objective while ensuring timely FCT for delay-sensitive applications. To solve this problem, we first introduce the Active Network Generation (ANG) approach, which generates a minimal active subnet with the least number of active devices while meeting the current traffic demand. Subsequently, we propose two algorithms based on the type of applications: the Delay-Tolerant Flow Route (DTFR) algorithm for delay-tolerant applications and the Delay-Sensitive Flow Route (DSFR) algorithm for delay-sensitive applications. Simulation results demonstrate that our propose solution achieves an energy-saving rate of up to 67.77% and significantly reduces FCT compared to benchmark solutions.

Vehicle repositioning is widely used in Mobility on-Demand (MoD) systems to address supply-demand imbalances and improve order completion rates. Existing methods typically offer repositioning recommendations focused on enhancing vehicle coordination toward supply-demand re-balance. However, these methods often overlook the possibility that drivers may not follow these recommendations due to their personal preferences, leading to recommendation-decision inconsistency and further disrupting the supply-demand balance. To address this issue, we propose a preference-aware vehicle repositioning recommendation strategy for MoD systems, named FREE, which is based on a Coulomb Force directed approach. The core idea is to strike a balance between vehicle coordination and consistency between recommendations and driver decisions. First, we introduce a Coulomb force-based representation (CFR) to model coordination among vehicles. In this model, the interactions between vehicles and orders are represented as forces that drive the repositioning of vehicles. Next, we develop a driver preference learning model that accurately captures drivers’ preferences using triplet and consistency loss. We then integrate these preferences with the CFR into a multi-agent deep reinforcement learning (MADRL) based repositioning algorithm to generate optimal recommendations. Finally, we validate the effectiveness of FREE through simulations using real-world data, demonstrating its superiority over existing benchmarks.

Existing High Efficiency Video Coding (HEVC) selective encryption algorithms only consider the encoding characteristics of syntax elements to keep format compliance, but ignore the semantic features of video content, which may lead to unnecessary computational and bit rate costs. To tackle this problem, we present a content-aware tunable selective encryption (CATSE) scheme for HEVC. First, a deep hashing network is adopted to retrieve groups of pictures (GOPs) containing sensitive objects. Then, the retrieved sensitive GOPs and the remaining insensitive ones are encrypted with different encryption strengths. For the former, multiple syntax elements are encrypted to ensure security, whereas for the latter, only a few bypass-coded syntax elements are encrypted to improve the encryption efficiency and reduce the bit rate overhead. The keystream sequence used is extracted from the time series of a new improved logistic map with complex dynamic behavior, which is generated by our proposed sine-modular chaotification model. Finally, a reversible steganography is applied to embed the flag bits of the GOP type into the encrypted bitstream, so that the decoder can distinguish the encrypted syntax elements that need to be decrypted in different GOPs. Experimental results indicate that the proposed HEVC CATSE scheme not only provides high encryption speed and low bit rate overhead, but also has superior encryption strength than other state-of-the-art HEVC selective encryption algorithms.

Competitive opinion maximization (COM) aims to determine some individuals (i.e., seed nodes) from social networks, propagating the desired opinions toward a target entity to their neighbors through social relationships when facing with its competitors (components) and maximize the opinion spread after the specific time. Current studies on COM are still in its infancy, while the only work merely considers the scenario that the strategy of competitors is known but ignores the unknown scenario. In addition, previous studies on COM cannot easily address the situation where some users might dynamically change their opinions. To address the COM issue, we investigate the multistage COM and propose a brand-new Q-learning-based opinion maximization framework (QOMF). Our QOMF consists of two components: dynamic opinion propagation and seeding process. We formulate the COM problem by maximizing relative effective opinions. To produce a dynamic opinion series more realistically, we design an opinion propagation model by joining the activation process and a dynamic opinion process. Moreover, we also verify that the opinion propagation model can reach convergence within finite iterations. To acquire the seed nodes, we design a multistage Q-learning seeding scheme by considering known and unknown competitor strategies, respectively. Experimental results on three real datasets demonstrate that the proposed method outperforms the benchmarks on reaching relatively effective opinions.

Recommendation unlearning is an emerging task to erase the influences of user-specified data from a trained recommendation model. Most existing research follows the paradigm of partitioning the original dataset into multi-fold and then retraining corresponding sub-models while those influences are totally removed. Despite the effectiveness, two key problems remain unexplored: (i) Existing work becomes inefficient and computationally expensive to retrain all sub-models, especially when facing large amounts of unlearning data. (ii) User preferences are dynamically changing. If users express negative opinions on some interacted items they used to prefer, how can we adaptively erase the outdated preferences behind such transformation from the trained model? Although these unlearning data contain outdated information, there is still a lot of helpful knowledge worth preserving. Existing methods ignore this preservation during unlearning and may remove all the knowledge in the interactions, compromising the final performance. In light of these limitations, we propose a novel unlearning framework called GFEraser, which transforms the unlearning into an efficient guided filtering process to avoid time-consuming retraining and retain beneficial knowledge. Specifically, we develop an intra-user negative sampling strategy to learn the outdated preferences that need to be erased. Under the guidance of differential maximization agreement and attention-based fusion module, the original representations are adaptively filtered and aggregated based on the learned preferences. Besides, we leverage contrastive learning to preserve the invariant user preferences, maintaining the final performance. Finally, we devise a new metric called Ranking Decrease Rate to evaluate the unlearning effect. Experimental results demonstrate that GFEraser can maintain reliable recommendation performance while achieving efficient outdated preferences unlearning, up to 37 \(\times\) acceleration.

Preference logical reasoning utilizes user-item interactions (e.g., ratings and reviews) to infer user preferences and discover user decision paths from the knowledge graph to enhance the explainability of item recommendations. However, existing algorithms assume that the ratings and reviews of any item are always consistent, ignoring situations where items with high ratings have negative reviews or items with low ratings but positive reviews. This leads to inaccurate learning of user preferences. In fact, through experimental analysis of two real datasets, we found that on average, about 10% of the interactive data exhibited this inconsistency, that is, items with high ratings but negative reviews appear in the recommendation list. To address this issue, we propose a general preference logical reasoning method based on preference operators. Specifically, we capture the semantic information of users toward the item (its corresponding attributes) in reviews and define two preference operators ( like and dislike ) for the item to correct ambiguous neutral ratings or false ratings that do not reflect true preferences. In the process of preference path reasoning, the like preference operator increases the occurrence probability of liked items, while the dislike preference operator reduces the occurrence probability of disliked items. By fusing the preference operators in the preference path, we obtain consistent user preferences and enhance the explainability of item recommendations. The experimental results on four real datasets demonstrate that our method can effectively improve the performance of all comparison baselines in terms of recommendation accuracy and user decision explainability.

Graph Contrastive Learning (GCL) leverages data augmentation techniques to produce contrasting views, enhancing the accuracy of recommendation systems through learning the consistency between contrastive views. However, existing augmentation methods, such as directly perturbing interaction graph (e.g., node/edge dropout), may interfere with the original connections and generate poor contrasting views, resulting in sub-optimal performance. In this article, we define the views that share only a small amount of information with the original graph due to poor data augmentation as noisy views (i.e., the last 20% of the views with a cosine similarity value less than 0.1 to the original view). We demonstrate through detailed experiments that noisy views will significantly degrade recommendation performance. Further, we propose a model-agnostic Symmetric Graph Contrastive Learning (SGCL) method with theoretical guarantees to address this issue. Specifically, we introduce symmetry theory into graph contrastive learning, based on which we propose a symmetric form and contrast loss resistant to noisy interference. We provide theoretical proof that our proposed SGCL method has a high tolerance to noisy views. Further demonstration is given by conducting extensive experiments on three real-world datasets. The experimental results demonstrate that our approach substantially increases recommendation accuracy, with relative improvements reaching as high as 12.25% over nine other competing models. These results highlight the efficacy of our method. The code is available at https://github.com/user683/SGCL .

Federated learning stands out as a promising approach within the domain of edge computing, providing a framework for collaborative training on distributed datasets without necessitating data sharing. However, federated learning involves the frequent transmission of machine learning model updates between the server and clients, resulting in high communication costs. Additionally, heterogeneous clients can further complicate the Federated Learning process and deteriorate performance. To address these challenges, we propose Adaptive Self-Knowledge Distillation-based Quality- and Reputation-Aware Cross-Device Federated Learning (ASDQR) - an efficient communication and inference framework designed for heterogeneous clients. ASDQR initiates the process by selecting high-reputation and high-quality clients to be involved in federated learning, significantly impacting communication efficiency and inference effectiveness. ASDQR also introduces a model of adaptive local self-knowledge distillation that incorporates multiple local personalized historical knowledge for more accurate inference, allowing the historical level to be dynamically adjusted across time. Finally, we present an inference-effective aggregation scheme that assigns higher weights to important and reliable local model updates based on clients’ contribution degrees when performing global model aggregation. ASDQR consistently outperforms baseline methods across all datasets and communication rounds, achieving 9.0% higher accuracy than FedAvg, 6.59% higher than MOON, 0.29% higher than FedProx, 0.2% higher than PFedSD, and 0.08% higher than FedMD on the MNIST dataset at 100 communication rounds. Similar improvements are observed on CIFAR, HAR, and WISDM datasets, demonstrating the robustness and efficiency of ASDQR in federated learning with non-IID data.

Existing selective encryption schemes for High Efficiency Video Coding (HEVC) only focus on the encoding characteristics of syntax elements in entropy coding and lack an understanding of the video content. Consequently, a large amount of unnecessary encryption operations are utilized to protect insensitive video frames, resulting in low encryption efficiency. In this article, we propose a content-aware selective encryption scheme for H.265/HEVC, which encrypts only the groups of pictures (GOPs) containing sensitive content and thus offers high efficiency. In our scheme, a deep hashing network is first adopted to retrieve video frames to determine the content-sensitive GOPs. Then, multiple prediction and residual syntax elements in sensitive GOPs are encrypted using a keystream sequence generated by the hyper-chaotic Lorenz system. In addition, the direct current coefficient of each \(4\times 4\) transform block is exchanged with a pseudo-randomly selected non-zero alternating current coefficient to further offer stronger visual distortion. Finally, the sign bits used for marking each GOP-type are reversibly embedded into the encrypted syntax elements to facilitate the decoder to distinguish the GOPs that need to be decrypted. Experimental results indicate that the proposed content-aware selective encryption scheme can efficiently protect sensitive content and is robust against all common attacks. Furthermore, it outperforms other state-of-the-art HEVC selective encryption algorithms in terms of security performance.

Configurations deployed in routers govern the routing and addressing of IP-based networks. Automated tools for synthesizing and verifying configurations have been deployed to replace manual configurations, preventing network outages from misconfigurations. However, these tools offer limited assistance in real production networks as manual configuration remains the primary and preferred approach for daily operations. To address this issue, we present NetGenius, a configuration recommendation tool that assists operators in manually editing network configurations like a code editor. NetGenius employs a graph neural network to estimate the importance of neighboring commands to the center command, providing recommendations based on importance scores when the center command is used as input. First, NetGenius leverages a generic graph model named Knowledge Graph to model existing configurations and derives configuration features for each center command by composing it with its neighbors. Subsequently, NetGenius conducts a general approach based on the graph neural network to estimate and calculate corresponding feature importance scores. Finally, NetGenius uses a recommendation mechanism to recommend configurations based on importance scores while adhering to specific constraints. We extensively evaluate the performance of NetGenius using real-world configurations from different vendors. The experimental results demonstrate that NetGenius achieves the highest recommendation accuracy of 98.59% across various datasets. Moreover, when applied to a large-scale network comprising$152,475$configuration commands, NetGenius completes its training within 6 seconds and generates recommendations in 5 milliseconds.

In-band Network Telemetry (INT) technology enables fine-grained network monitoring by encapsulating intra-switch network status into INT probes, which is essential in data center networks for ensuring Quality of Service (QoS). Existing INT-based telemetry systems leverage centralized controllers to compute non-overlapping probe paths, thereby facilitating lightweight and network-wide measurements. However, these systems fail to adapt effectively to network topology changes caused by link or device failures, primarily due to inflexible path planning under dynamic conditions. To address this problem, we propose INT-Source, a unified policy-based network-wide telemetry system for probing and forwarding. First, we design a data plane forwarding mechanism for INT probes to ensure telemetry coverage during topology changes and reduce telemetry overhead. Second, we design a probe packet structure and introduce a switch-based probe verification and discard mechanism to prevent redundant link probing. Third, we introduce two algorithms for INT-Source: a Single-Source algorithm to facilitate deployment and a Multi-Source algorithm to enable lightweight and scalable telemetry. Our evaluation shows that INT-Source reduces bandwidth overhead to 12.5% compared to existing methods across three network topologies. Even with a 10% link failure rate, INT-Source is able to monitor 93.1% of network ports, demonstrating strong robustness.

With advances in the edge computing (EC) and federated learning (FL) technologies in jointcloud, the edge FL service market has emerged recently and it requires trading edge resources between model requesters and data owners to complete FL tasks, which needs to incentivize sufficient data owners to participate in model training tasks. However, the limitations of resource trading and incentive design for edge FL service market have not been well addressed. In this paper, we propose a two-level blockchain-aided resource trading mechanism for encouraging appropriate edge servers to compete for dynamic FL tasks from the market while incentivizing data owners to participate in the FL tasks. At the upper level, we apply the deep learning-based reverse auction to model the dynamics of the task server selection process, with the aim of maximizing the total social welfare of the edge FL service market, where the edge server, as a seller, considers not only the data contribution of edge devices but also the cost of using blockchain when bidding. At the lower level, the edge servers offer rewards in exchange for the data owners’ participation, while the parameter aggregation is completed through the blockchain in a decentralized manner, which improves the FL’s robustness. Then, we utilize the Stackelberg game to model the dynamic process that the data owners compete for the servers’ revenue. We conduct extensive simulation experiments and the experimental results show that the proposed mechanism is able to get maximized social welfare and provide effective insights and strategies for the resource trading in the edge FL market to complete the federated training.

More and more application providers (APs) start to deploy their geo-distributed services in multiple cloud environments, such as JointCloud, federated clouds and InterCloud. Thus, massive cross-cloud traffic is generated from the services of APs, who need to pay Internet service providers (ISPs) for using their bandwidth. As such, an effective cross-cloud link bandwidth allocation and pricing mechanism is needed between APs and ISPs. Existing fixed-price scheme lacks market efficiency. Thus, we propose a truthful padding-based auction mechanism (TPAM) for cross-cloud bandwidth, which introduces the padding method and well-designed pricing strategy to ensure desirable properties. This mechanism is flexible enough to allow each AP to win the whole request, or win the specified proportional request, or lose and get nothing. Specifically, we first devise a linear-program-based method to calculate the padding vector for each candidate AP. Next, we design a padding-based method to determine the winning APs and match them with ISPs who offer the cheapest bandwidth. Finally, we design a critical-value-based pricing strategy and a marginal-cost-based pricing strategy for APs and ISPs to achieve truthfulness and budget balance. Theoretical analyses prove that TPAM achieves truthfulness, budget balance, individual rationality, asymptotic efficiency and computational tractability. Trace-driven simulation results also validate the effectiveness and efficiency of TPAM.

With the expansion of production networks, new challenges arise in scaling telemetry systems to accommodate the massive number of network devices. In-band Network Telemetry (INT) is widely adopted for its fine-grained and accurate measurements. However, system robustness and performance scalability remain key challenges for INT-based network-wide telemetry systems. Existing INT-based measurements manage the network as a whole and rely on a centralized controller for path planning and telemetry data collection. As networks scale and the probability of failures increases, frequent re-planning leads to prolonged telemetry interruptions. In this work, we propose INT-Partition, a hierarchical and fault-tolerant in-band network telemetry system with a divide-and-conquer paradigm. INT-Partition conducts telemetry in two stages: network partitioning and telemetry within each partition, enabling scalability for mega-scale networks. Our evaluations, including mega-scale simulations using BMv2 software switches and small-scale validations on Tofino hardware switches, demonstrate the effectiveness of our approach. With 1% of network equipment out of order, INT-Partition covers over 89.67% of the area and accurately locates faults. As the network scales, telemetry planning and deployment time is reduced by 75.62% to 96.02%, and hot reloading enables seamless switching of telemetry deployment.

Fuzzy-rough sets-based feature selection (FRFS), as an effective data pre-processing technique, has drawn significant attention with the growing prevalence of large-scale datasets. However, centralized FRFS approaches suffer from the following shortcomings: 1) low computational efficiency, 2) bottlenecks in memory and computational resources, and 3) strict limitation of collaborative implementation using nonshared datasets owned by different data providers. These limitations highlight the growing necessity of integrating FRFS into a distributed FS framework. Nevertheless, most existing distributed FS schemes are reliant on a designated central server to collect and merge the local results from all slave nodes, which may result in several challenges including single point of failure risk, lack of trust and reliability, and lack of transparency and traceability. To relieve the above issues, this paper proposes a blockchain assisted distributed FS framework, successfully implementing a distributed solution for FRFS (BDFRFS). Firstly, this framework introduces blockchain to merge, reach consensus and publish the global results generated during each iteration of FRFS, including the currently selected feature subset with its corresponding similarity matrix and dependency degree. This not only eliminates the reliance of central server and alleviates the burden on the central server, but also enhances the credibility and traceability of the results. Additionally, the implementation of FRFS is designed within this framework, utilizing three strategies to improve the efficiency of centralized FRFS: 1) eliminating the irrelevant and redundant features prior to the executing FRFS; 2) removing redundant and unnecessary computations involved in generating the similarity matrices; and 3) enabling parallel computation of dependency degrees. Finally, the experimental results conducted on eight large-scale datasets demonstrate that the proposed framework can significantly reduce the runtime cost and improve the classification accuracy compared to centralized FRFS and several distributed FS approaches.

Mobile edge computing (MEC) pushes cloud computing capabilities to the network edge, which provides real-time processing and caching flexibility for service-based applications. Conventionally, the individual node solution is insufficient to tackle the increasing computation workload and provide diverse services, especially for unpredictable spatiotemporal service request patterns. To address this, we first propose a hierarchical collaborative computing (HCC) framework to serve users’ demands by reaping sufficient computing capability in Cloud, ubiquitous service area in edge layer, and idle resources in device layer. To better unleash the benefits of HCC and pursue long-term performance, we investigate heterogeneity-aware resource management by collaborative service placement, task scheduling, and resource allocation both in-node and cross-node. We then propose an online optimization framework that first decouples the decisions across different slots. For each instant mixed integer non-linear programming problem, we introduce the surrogate Lagrangian relaxation method to reduce complexity and design hybrid numerical techniques to solve the subproblems. Theoretical analysis and extensive simulation results demonstrate the efficiency of the HCC framework in decreasing system cost on devices, and our proposed algorithms can effectively utilize the resources in the collaborative space to achieve the trade-off between system cost minimization and service placement cost stability.

With the widespread adoption of the Internet of Things (IoT) and various smart medical devices, the volume of medical data has dramatically increased, making the processing of medical Internet of Things (IoMT) data increasingly challenging. Due to the integration of edge computing and cloud computing, IoMT can allocate increased computing and storage resources in proximity to the terminal, addressing the low-latency requirements of computationally intensive tasks. While existing initiatives have shifted services to edge servers, they have not taken into account the joint impact of task priorities and mobile computing services on Mobile Edge Computing (MEC) networks. Fortunately, the rapidly advancing field of Artificial Intelligence (AI) has proven effective in some resource allocation applications in recent years. In this article, we propose a mobile edge computing-based intelligent healthcare multitasking processing system aimed at addressing the issue of service prioritization in medical scenarios. Considering energy consumption and latency, we present a multi-objective task-aware service offloading algorithm under the framework of end-edge-cloud collaborative IoMT systems, employing deep deterministic policy gradients (DDPG). Adaptability to the diversity of different services is achieved through dynamic adjustments based on various business types and system requirements. Finally, the effectiveness of DDPG for IoMT is validated using real-world data.

Edge intelligence (EI) entails deploying artificial intelligence algorithms at the network’s edge. Utilizing edge computing infrastructure enables local data processing and decision-making, resulting in decreased latency, bandwidth consumption, and improved privacy security. With increasing the number of Internet of Things (IoT) devices and the development of 6G communication technologies, there is a growing demand for fast, efficient, and low-latency data processing, which has led to the rise of EI. This survey comprehensively reviews and analyzes the current state of research on EI from the perspective of technological development, with a particular focus on the following key aspects: (1) We review the basic concepts of EI and its distinctions from traditional cloud computing and edge computing; (2) We explore the technological framework of EI in detail, including key technologies such as edge computing and federated learning, and analyze how these technologies integrate with modern communication technologies like IoT devices and 6G networks; (3) We discuss the challenges faced when implementing EI technologies, such as data privacy and security issues, device resource constraints, and propose corresponding solutions or research directions. The survey also outlines the main research directions and technical challenges driving the future development of EI, providing valuable insights and guidance for researchers and practitioners in the field.

Edge-Cloud-End Collaborative (ECEC) computing emerges as a promising paradigm to support computationally intensive applications within Digital Twin (DT) networks, providing flexibility and real-time performance for heterogeneous task scheduling and resource allocation. However, effectively balancing key Service Level Agreement (SLA) parameters, such as response delay, availability, and throughput in dynamic network environments remains challenging, especially for complex, large-scale applications. Existing solutions typically lack SLA-oriented proactive resource reservation schemes. To address these limitations, we propose an SLA-driven hierarchical bidirectional closed-loop DT Network Architecture (SLA-DTNA), comprising four layers: real network, local DT, edge DT, and cloud DT. The proposed architecture systematically decomposes SLA requirements into measurable system parameters, aiming to minimize overall system cost. Specifically, a differentiated task management mechanism is designed at the local DT layer to ensure Quality of Service (QoS) for critical tasks. At the edge DT layer, we propose a heuristic-based lightweight scheduling algorithm leveraging DT capabilities for efficient task resource mapping and reduced scheduling complexity. At the cloud DT layer, we apply the Deep Deterministic Policy Gradient (DDPG) algorithm for adaptive resource reservation, dynamically adapting schemes based on task preferences and historical behavior patterns. Simulation and experimental results validate that the proposed SLA-DTNA enables fine-grained and intelligent resource allocation, enhances overall network performance, and effectively satisfies dynamic SLA requirements.

Traditional single-cloud storage suffers from single points of failure, leading to low data availability. As a result, it fails to meet users' demands for reliable cloud storage services. Therefore, the current cloud storage paradigm has shifted to distributed-cloud storage (e.g., multi-cloud storage, JointCloud storage), where users store multiple replicas of data across multiple Cloud Service Providers (CSPs). However, this imposes significant storage pressure on CSPs. To reduce costs and maximize profits, some malicious CSPs may delete user data, undermining trust in cloud services and hindering the growth of the cloud computing industry. To address this issue, we propose a novel distributed-cloud storage based on permissioned blockchain, which effectively reduces storage costs while ensuring data availability. Firstly, we integrate Byzantine Fault Tolerance in permissioned blockchain with erasure coding (EC) to replace the traditional multi-cloud multi-replica storage approach. This integration significantly reduces storage costs while providing an efficient means for data recovery. Based on blockchain, we further propose a data integrity auditing approach that eliminates reliance on semi-trusted third-party auditors and enables decentralized data integrity verification. Combined with this auditing approach, our EC-based data recovery approach ensures data availability while enhancing users' trust in distributed-cloud storage. Theoretical analysis indicates that our scheme reduces storage overhead from$O(n)$to$O(1)$with$n$CSPs while ensuring data availability. Meanwhile, experimental results demonstrate that computational overhead is reduced by approximately 78% compared to traditional multi-cloud multi-replica storage, achieving the cost-effective and highly reliable distributed-cloud storage.

The simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) can provide a fullcoverage agile radio environment. A unique STAR-RIS-assisted coordinated multi-point (CoMP) framework is investigated in this paper, where cell-center and cell-edge users are embellished by the reflection and transmission features of the STAR-RIS. Unlike previous works controlling the transmission and reflection phase-shift independently, we consider a more practical coupled phase-shift model. We formulate an online active and passive beamforming problem to maximize long-term energy efficiency (EE) with time-varying locations and channels. Moreover, we propose a hybrid learning framework combining model-free and model-based optimization techniques. For the model-free method, we invoke a risk-sensitive multi-agent deep reinforcement learning algorithm to accelerate the online optimization of the passive beamforming of all STAR-RISs. For the model-based method, we invoke fractional programming (FP) to optimize the coordinated zero-forcing beamformer among all base stations and to realize an exact reward evaluation for each action in the DRL algorithm. Compared to DRL algorithms that optimize passive and active beamforming together, we dramatically shrink the action and state spaces. Comprehensive numerical results explain that the STAR-RIS enhanced CoMP system accomplishes a more excellent EE than the benchmark STAR-RIS cases and that without STAR-RIS.

This paper investigates the design of jointly supporting physical layer security (PLS) and covert communications (CCs) in an active simultaneously transmitting and reflecting reconfigurable intelligent surface (a-STAR-RIS) assisted integrated sensing and communication (ISAC) system. Due to the unified waveform design of ISAC signals, we consider a challenging scenario with two targets being suspicious attackers, where one warden target potentially detects the confidential transmission behavior of covert users and another eavesdropper target attempts to intercept the broadcasted confidential information of security users. We investigate the joint beamforming design at the base station (BS) and the a-STAR-RIS to achieve a high-quality sensing beampattern while meeting covertness and security communication requirements. (1) For the ideal scenario with perfect channel state information (CSI) and precise target locations, we propose an alternative optimization (AO) method to address the optimization problem involving highly coupled variables. Specifically, the optimal beamforming design at the BS is handled using the semi-definite relaxation (SDR) technique, while the beamforming design at the a-STAR-RIS is addressed through a penalty-based iterative algorithm. (2) A more practical case with uncertain target locations and imperfect CSI is considered to achieve a robust beamforming design, where the non-deterministic outage probability constraints are effectively transformed by employing the Bernstein-type inequality. Numerical results demonstrate the superiority of the a-STAR-RIS over the baseline cases and certify that the proposed algorithms can effectively balance the tradeoff among the sensing quality, covert and secure communication requirements. Besides, results also show that the proposed robust beamforming scheme can construct adequate sensing beampattern, even with imperfect CSI and uncertain target locations.

Limited-supervised multi-label learning (LML) leverages weak or noisy supervision for multi-label classification model training over data with label noise, which contain missing labels and/or redundant labels. Existing studies usually solve LML problems by assuming that label noise is independent of the input features and class labels, while ignoring the fact that noisy labels may depend on the input features (instance-dependent) and the classes (label-dependent) in many real-world applications. In this paper, we propose limited-supervised Multi-label Learning with Dependency Noise (MLDN) to simultaneously identify the instance-dependent and label-dependent label noise by factorizing the noise matrix as the outputs of a mapping from the feature and label representations. Meanwhile, we regularize the problem with the manifold constraint on noise matrix to preserve local relationships and uncover the manifold structure. Theoretically, we bound noise recover error for the resulting problem. We solve the problem by using a first-order scheme based on proximal operator, and the convergence rate of it is at least sub-linear. Extensive experiments conducted on various datasets demonstrate the superiority of our proposed method.

Multi-task learning (MTL) aims to empower a model to tackle multiple tasks simultaneously. A recent development known as task arithmetic has revealed that several models, each fine-tuned for distinct tasks, can be directly merged into a single model to execute MTL without necessitating a retraining process using the initial training data. Nevertheless, this direct addition of models often leads to a significant deterioration in the overall performance of the merged model. This decline occurs due to potential conflicts and intricate correlations among the multiple tasks. Consequently, the challenge emerges of how to merge pre-trained models more effectively without using their original training data. This paper introduces an innovative technique called Adaptive Model Merging (AdaMerging). This approach aims to autonomously learn the coefficients for model merging, either in a task-wise or layer-wise manner, without relying on the original training data. Specifically, our AdaMerging method operates as an automatic, unsupervised task arithmetic scheme. It leverages entropy minimization on unlabeled test samples from the multi-task setup as a surrogate objective function to iteratively refine the merging coefficients of the multiple models. Our experimental findings across eight tasks demonstrate the efficacy of the AdaMerging scheme we put forth. Compared to the current state-of-the-art (SOTA) task arithmetic merging scheme, AdaMerging showcases a remarkable 11\% improvement in performance. Notably, AdaMerging also exhibits superior generalization capabilities when applied to unseen downstream tasks. Furthermore, it displays a significantly enhanced robustness to data distribution shifts that may occur during the testing phase.

Multi-task learning (MTL) compresses the information from multiple tasks into a unified backbone to improve computational efficiency and generalization. Recent work directly merges multiple independently trained models to perform MTL instead of collecting their raw data for joint training, greatly expanding the application scenarios of MTL. However, by visualizing the representation distribution of existing model merging schemes, we find that the merged model often suffers from the dilemma of representation bias. That is, there is a significant discrepancy in the representation distribution between the merged and individual models, resulting in poor performance of merged MTL. In this paper, we propose a representation surgery solution called ``Surgery" to reduce representation bias in the merged model. Specifically, Surgery is a lightweight task-specific plugin that takes the representation of the merged model as input and attempts to output the biases contained in the representation from the merged model. We then designed an unsupervised optimization objective that updates the Surgery plugin by minimizing the distance between the merged model's representation and the individual model's representation. Extensive experiments demonstrate significant MTL performance improvements when our Surgery plugin is applied to state-of-the-art (SOTA) model merging schemes.

Damage or relocation of network devices necessitates the replacement of devices. Additionally, manufacturers of devices may also change due to factors such as price, policies, and functional limitations. This brings great needs of converting existing configurations to fix new devices and ensuring consistent network policy behaviors, which is configuration translation. Manually translating requires network administrators to possess significant configuration expertise and to spend extreme time translating thousands of commands for a core device. Therefore, we propose ConfigTrans, a novel configuration translation method for commands with and without parameters based on constraint solving and large language models (LLMs), respectively. For commands with parameters, we explore constraints about parameters, commands, and format limitations, and design a heuristic algorithm to solve them. For commands without parameters, we utilize an LLM to find the translation. The most likely several commands are selected based on semantic similarity, and the final result and keywords in it are chosen by the LLM. To meet the requirements of the view, the result commands are arranged while supplementing missing higherview commands. Experimental results show that our method achieves an accuracy rate of 82.47 % for translating commands with parameters and 75.5 % for commands without parameters.

Network measurement is critical for various network applications, but scaling measurement techniques to the network-wide level is challenging for existing sketch-based solutions. Centralized sketch deployment provides low resource usage but suffers from poor load balancing. In contrast, collaborative measurement achieves load balancing through flow distribution across switches but requires high resource usage. This paper presents a novel lightweight distributed deployment framework that overcomes the limitations above. First, our framework is lightweight such that it splits sketches into segments and allocates them across forwarding paths to minimize resource usage and achieve load balancing. This also enables per-packet load balancing by distributing computations across switches. Second, our framework is also optimized for load balancing by coordinating between flows and enabling finer-grained flow distribution. We evaluate the proposed framework on various network topologies and different sketch deployments. Results indicate our solution matches the load balancing of collaborative measurement while approaching the low resource usage of centralized deployment. Moreover, it achieves superior performance in per-packet load balancing, which is not considered in previous deployment policies. Our work provides efficient distributed sketch deployment to strike a balance between load balancing and resource usage enabling effective network-wide measurement.

Network measurement plays a critical role in numerous network applications that rely on fundamental flow processing tasks such as frequency estimation, heavy hitter detection, and distribution estimation. Sketch has emerged as an efficient approach for network measurement due to its low overhead. However, most sketch-based solutions target static windows while enabling sliding window-based measurement remains an open challenge. This paper introduces two novel general frameworks applicable to diverse sketch models for sliding window-based network measurement: a traditional sliding window framework and a fine-grained flow-level framework. The traditional framework divides the window into parts and uses centralized flushing to remove expired parts. The flow-level framework tracks timestamps to maintain exact flow characteristics over one period, preventing truncation. To optimize memory usage, a bit-wise adaptive allocation algorithm allows dynamic borrowing of unused counter bits. The frameworks are evaluated on sketches for different flow processing tasks. Results show the frameworks are widely generalizable, reduce error substantially compared to existing approaches, and provide more efficient memory usage.

Blockchain, a promising distributed ledger for decentralization, plays a crucial role in JointCloud computing, offering secure data storage and sharing as well as reliable transaction tracking and resource allocation. While promising, it faces storage limitations due to its full-replication strategy. This issue has been addressed by scholars through storage partitioning mechanisms like BFT-Store and PartitionChain. These mechanisms leverage Erasure Coding with the Byzantine Fault Tolerant consensus protocol to overcome storage constraints. However, challenges persist: i) high computational complexity in encoding and decoding leading to prolonged computation time; ii) extensive use of verified signatures causing increased network message transmission and communication burden; iii) during system re-initialization, under-performing scalability and substantial time consumption due to the involving of all nodes in data decoding and re-encoding. To overcome these challenges, we propose two novel storage partitioning mechanisms (i.e., GPartition-Store and ParallelC-Store) for the permissioned blockchain that reduce the computational complexity of the coding processes by dividing nodes into multiple groups called storage units (SUs), avoid additional communication of generating verification proofs by employing the Bloom Filter, and improve the stability and scalability of the system by implementing the re-initialization process exclusively within a specific SU. Particularly, the computational complexity of encoding/decoding can be further degreased by about g2/g3and g/g2compared to PartitionChain, by GPartition-Store and ParallelC-Store respectively (g is the number of SUs). Compared with the full-replication strategy, BFT-Store and PartitionChain, the experimental results demonstrate that the proposed mechanisms improve the Quality of Service (QoS) of the blockchain system including performance (i.e., efficiency and throughput), scalability and stability, while guaranteeing the availability.

The Fifth-Generation mobile networks (5G) and Beyond 5G (B5G) have been proposed to support a variety of application scenarios, such as enhanced Mobile Broadband (eMBB), ultra-Reliable Low-Latency Communications (uRLLC), and massive Machine Type Communications (mMTC). On the other hand, Mobile Edge Computing (MEC) and Network Functions Virtualization (NFV) technologies have been widely advocated by service providers to meet diverse service demands and reduce operational costs. To alleviate the pressure on the edge network, resource consumption can be minimized by considering the reuse of Virtual Network Function (VNF) instances. However, implementing VNF chain deployment with latency guarantees and resource efficiency in a distributed network architecture remains an urgent issue to be addressed. In this paper, we explore the Service Function Chains (SFCs) orchestration problem with distributed edge network resources, aiming to design efficient service flow routing and resource allocation schemes to significantly respond to local user requests. We propose a low-complexity Distributed SFCs Orchestration algorithm with VNF Reuse (DSOR), which initially uses local information at the edge to explore the VNFs orchestration scheme and executes the distributed service orchestration. Subsequently, service chains are deployed based on asynchronous consensus to enhance network utility and reduce resource costs. Finally, the performance of DSOR is evaluated through extensive simulation experiments. The experimental results indicate that DSOR can improve the utilization of network resources, as well as the response rate to edge service requests.

Sequential recommendation aims to provide users with personalized suggestions based on their historical interactions. When training sequential models, padding is a widely adopted technique for two main reasons: 1) The vast majority of models can only handle fixed-length sequences; 2) Batch-based training needs to ensure that the sequences in each batch have the same length. The special value 0 is usually used as the padding content, which does not contain the actual information and is ignored in the model calculations. This common-sense padding strategy leads us to a problem that has never been explored in the recommendation field: Can we utilize this idle input space by padding other content to improve model performance and training efficiency further?

With the development of intelligent logistics technology, some companies began to use robots instead of humans to load cargo. This paper studies a novel container loading problem based on robotic loader system (CLP-RLS). Different from the existing robot-packable pattern in the literature, the robotic loader system in this paper consists of a depalletizing robot, an automatic telescopic roller line and a loading robot, in which the loading robot will enter the carriage along with the automatic telescopic roller line. In CLP-RLS, it is necessary to consider not only many practical constraints already in the literature, including load balancing, orientation, stability, and multi-drop but also two new constraints related to robotic loader system: pallet continuity constraint (the loading sequence of cargo on the same pallet is continuous) and robot position constraint (the robot can only load cargo incrementally from the front to the back of the truck). Due to the difficulty in modeling CLP-RLS and the large scale of the real-case instances, we present a tree search approach based on wall-building to solve CLP-RLS, intending to find a feasible loading scheme for the RLS to minimize the length required to load all cargo. The effectiveness of the proposed approach is verified through both real-case instances and numerical instances.

Network configuration plays a vital role in quality assurance of network services, requiring considerable effort and time. Automatic network configuration approaches are promising due to their capacity to automatically generate and verify configurations. However, these methods suffer from drawbacks, such as generated configuration content being largely unknown to network operators and inefficient for large-scale networks. Manual configuration is thus still the primary way of managing networks. To facilitate editing processes of manual configuration, a network-wide tool for recommending custom keywords is in urgent need. In this article, we propose a keyword recommendation tool that recommends custom keywords across various network devices. We observe that network devices of the same type and role tend to have a unified template and similar configurations, which enables recommending custom configurations between them. However, the vision entails the following three challenges. First, configurations need to be modeled accurately. Second, a wide variety of network protocols need to be supported. Third, relationships between custom keywords might be implicit and difficult to find. To address the challenges, we first built a configuration knowledge graph that could accurately model configurations, extract latent relationships between keywords, and generate explainable recommendations. Then we applied a recommendation framework to the graph for appropriate keyword recommendations. Lastly, to validate the performance, we conduct recommendations on real configurations over 26 000 times. Experimental results indicate that the overall coverage rate for matching expected configurations reaches 79.396%, and the redundancy rate is less than 20%.

Cellular networks are vital for emerging applications like the Metaverse, which impose demanding quality and quantity requirements. This necessitates frequent reconfiguration of both new and existing base stations to balance network service quality (e.g., ultra-low latency and high bandwidth) and resource consumption. Existing data-driven configuration methods learn from historical data, but have two key limitations. First, they yield only approximate solutions, lacking precision. Second, poor bootstrapping for new base stations with previously unobserved attributes. In this paper, we pioneer intent-driven configuration synthesis by designing an intent language and utilizing satisfiability modulo theory (SMT) for cellular networks to enable exact and precise solutions. We formulate synthesis as an SMT problem, permitting verification of precision. First, we cast configuration generation as a program synthesis problem via novel modeling to bridge the intent-configuration gap. Second, we extend SMT synthesis to scale to large networks. However, vanilla SMT approaches have poor scalability. Hence, we propose an optimization using sampling for constraint verification instead of exhaustive forward solving. We also design a domain-specific optimization to prune the sample space and improve efficiency. Experiments on various network scales demonstrate the effectiveness of our proposed SMT-based cellular network configuration synthesis.

The dynamical system perspective has been used to build efficient image classification networks and semantic segmentation networks. Furthermore, the Runge–Kutta (RK) methods are powerful tools for building networks from the dynamical systems perspective. Hence, the Runge–Kutta segmentation network (RKSeg) for medical image segmentation was born. Skip connections and multiple scaling are often used in common models but lack mathematical explanations. RKSeg interprets and uses skip connections based on the RK methods. Therefore, RKSeg greatly improves segmentation efficiency. However, it does not explain and use multiple scales from a dynamical system perspective but only inherits the multi-scale scheme of existing models. We compensate for this shortcoming by interpreting and using multiple scales based on the RK methods. In addition, the network structure also limits the excellent image classification networks as the backbones of RKSegs. Therefore, we modify the network structure to support more image classification networks as backbones. As a result, we propose a novel network structure RKSeg+. Our proposed RKSeg+ achieves better segmentation results with fewer parameters than RKSeg. Furthermore, RKSeg+, well configured with few parameters, outperforms state-of-the-art models on six of the ten organ datasets in the Medical Segmentation Decathlon.

Multi-graph Multi-label learning (Mgml) aims to classify a set of objects of interest, such as text or images, using a bag-of-graphs representation. PreviousMgmlworks have limitations as they only learn labels at the bag level, lose structural information in learning by transferring graphs into instances, and cannot handle noisy labels. This paper presents a robustcoarseandfine-grainedNoise Multi-graph Multi-label (cfMGNML) learning framework that builds the learning model over the graphs and empowers label prediction at both thecoarse(bag) andfine-grained(graph in each bag) levels with noisy labels. To identify label noise, a label probability matrix is defined to act on the scoring function of each label, with a higher probability value indicating that the label is more likely to be the corresponding graph or bag label. The problem is regularized with the manifold constraint on the label probability matrix to preserve local relationships within the data and uncover its essential manifold structure. Meanwhile, a thresholding rank-loss objective is proposed to rank the labels for the graphs and bags and minimize the hamming loss at one step simultaneously. To tackle the non-convex optimization problem, an effective sub- gradient descent algorithm is developed. Experiments over various datasets demonstrate the proposed method achieves superior performance than the state-of-the-art algorithms.

Continual learning (CL) aims to learn new tasks without forgetting previous tasks. However, existing CL methods require a large amount of raw data, which is often unavailable due to copyright considerations and privacy risks. Instead, stakeholders usually release pre-trained machine learning models as a service (MLaaS), which users can access via APIs. This paper considers two practical-yet-novel CL settings: data-efficient CL (DECL-APIs) and data-free CL (DFCL-APIs), which achieve CL from a stream of APIs with partial or no raw data. Performing CL under these two new settings faces several challenges: unavailable full raw data, unknown model parameters, heterogeneous models of arbitrary architecture and scale, and catastrophic forgetting of previous APIs. To overcome these issues, we propose a novel data-free cooperative continual distillation learning framework that distills knowledge from a stream of APIs into a CL model by generating pseudo data, just by querying APIs. Specifically, our framework includes two cooperative generators and one CL model, forming their training as an adversarial game. We first use the CL model and the current API as fixed discriminators to train generators via a derivative-free method. Generators adversarially generate hard and diverse synthetic data to maximize the response gap between the CL model and the API. Next, we train the CL model by minimizing the gap between the responses of the CL model and the black-box API on synthetic data, to transfer the API's knowledge to the CL model. Furthermore, we propose a new regularization term based on network similarity to prevent catastrophic forgetting of previous APIs. Our method performs comparably to classic CL with full raw data on the MNIST and SVHN datasets in the DFCL-APIs setting. In the DECL-APIs setting, our method achieves 0.97×, 0.75× and 0.69× performance of classic CL on the more challenging CIFAR10, CIFAR100, and MiniImageNet, respectively.

Programmable switches allow data plane to program how packets are processed, which enables flexibility for network management tasks, e.g., packet scheduling and flow measurement. Existing studies focus on program deployment at a single switch, while deployment across the whole data plane is still a challenging issue, especially manifested in the difficulty in joint correct implementation of P4 programs, resource load balancing of network devices, and optimization of network performance. In this paper, we present RED, a Resource-Efficient and Distributed program deployment solution for programmable switches. First of all, we analyze data plane programs to estimate the resource utilization and divide them into two categories for further processing. Then, the proposed merging and splitting algorithms are selectively applied to merge or split the pending programs. Finally, we consolidate the scarce resources of the whole data plane for distributed program deployment. Extensive experiments with both testbed and large-scale simulations are conducted and comparison results show that 1) RED achieves network-wide resource balancing in a distributed way and the latency of processing packets within the switch was reduced by 16.7%. 2) RED improves the speedup by two orders of magnitude compared to P4Visor in merging program and merges more 18% tables than SPEED; 3) If the resources required to run a P4 program exceed the resource limit of the switch, it cannot be deployed on the switch. RED makes the overwhelmed programs to be deployed on switches and switch throughput increased by 10.7%.

Unmanned aerial vehicle (UAV) assisted communications have been regarded as an effective solution to provide instant services. This paper proposes a novel reconfigurable intelligent surface (RIS) assisted multi-UAV system, where ground users are formed as multicast groups and served by multiple UAVs with coordinated multi-point technique. The goal is to maximize the sum of the minimum rates for all groups by jointly optimizing the trajectories, the cooperative beamforming of the clustered UAVs, and the passive beamforming of the RIS. A hybrid learning scheme is proposed, integrating a multi-agent deep reinforcement learning algorithm, RES-QMIX, and a majorization-minimization (MM)-based alternating optimization. First, the RES-QMIX algorithm is proposed to optimize the trajectories of all UAVs. Then, the alternating optimization is invoked to decouple the joint beamforming into two sub-problems, and each is transformed into a convex quadratic cone programming problem with the MM algorithm. Moreover, the alternating optimization is employed to estimate the reward of the action in RES-QMIX algorithm, thus reducing the action space and achieving the joint optimization. Numerical results show that: 1) The proposed hybrid learning framework achieves fast convergence and outperforms heuristic algorithms; 2) The proposed system obtains a more significant communication rate than CoMP and RIS-only systems.

Social networks currently serve as one of the primary sources from which people obtain news, with the spread of rumors emerging as a major concern. The goal of rumor suppression is to minimize the number of individuals affected by rumors through various methods, such as blocking and disseminating the truth. Although this problem has evolved into a popular research topic, existing solutions often overlook the temporal impact of rumor-refuting information and the influence of user opinions on rumor spreading. In the study, we first investigate the two-stage rumor minimization problem. The problem primarily considers two situations about only the propagation of rumors and the simultaneous propagation of rumor and rumor-refuting information, aiming to minimize the impact of rumors. We propose the two-stage user opinion rumor propagation model (TSUORP), which fully incorporates the timing of official releases of rumor-refuting information and their influence on the generation of rumors propagation. Based on this, we propose an approach using the knowledge graph convolutional network (KGCN) algorithm to rapidly and effectively select rumor-refuting information seed nodes based on user opinions. To assess the validity of our proposed approach, we perform experiments on three authentic datasets, showcasing its notable advantages.

Existing H.265/HEVC selective encryption (SE) schemes do not take into account the semantic features of input videos, nor do they adjust the encryption syntax elements according to the sensitivity of video content, which greatly limits their applicability. In this paper, we propose a chaos-based tunable H.265/HEVC SE scheme with semantic understanding. First, a deep hashing network is employed to identify content-sensitive videos by analyzing the semantic features of video sequences. Then, the non-sensitive videos and the retrieved sensitive ones are encrypted with different encryption strengths, respectively. Specifically, for non-sensitive videos, seven syntax elements with bypass-coded bins are selected for encryption at a constant bit rate. Hence, the encrypted bitstream keeps exactly the same compression ratio. To provide heavier visual distortion for content-sensitive videos, the regular-coded bins of four syntax elements and the intra prediction mode (IPM) are encrypted based on their corresponding encoding characteristics as well. Additionally, the selected syntax elements are all masked using a keystream generated by a chaotic system to ensure real-time constraints. Experimental results demonstrate that our suggested scheme offers format compatibility and is secure against all common attacks. Meanwhile, it outperforms state-of-the-art SE schemes in terms of security strength. Furthermore, the proposed scheme can be flexibly used in a wide range of applications according to the user’s requirements for encryption strength and bit rate.

In multi-objective evolutionary neural architecture search (NAS), existing predictor-based methods commonly suffer from the rank disorder issue that a candidate high-performance architecture may have a poor ranking compared with the worse architecture in terms of the trained predictor.To alleviate the above issue, we aim to train a Pareto-wise end-to-end ranking classifier to simplify the architecture search process by transforming the complex multi-objective NAS task into a simple classification task. To this end, a classifier-based Pareto evolution approach is proposed, where an online classifier is trained to directly predict the dominance relationship between the candidate and reference architectures. Besides, an adaptive clustering method is designed to select reference architectures for the classifier, and an α-domination assisted approach is developed to address the imbalance issue of positive and negative samples. The proposed approach is compared with a number of state-of-the-art NAS methods on widely-used test datasets, and computation results show that the proposed approach is able to alleviate the rank disorder issue and outperforms other methods. Especially, the proposed method is able to find a set of promising network architectures with different model sizes ranging from 2M to 5M under diverse objectives and constraints.

Fuzzy-rough sets (FRS) provide an outstanding theoretical tool for feature selection (FS). Whilst promising, the FRS model is sensitive to noisy information and ineffectively applicable to the data with large class density difference, with existing FRS-based FS methods only tackling one of these challenges. Therefore, to overcome both of these issues, this article presents a robust FS algorithm using linear reconstruction measure for the first time. First, a pseudo FRS model is proposed, where the distribution-aware linear reconstruction relation serving as the fuzzy similarity relation is constructed by considering the insight of meaningful information (i.e., distribution information of samples and density information of classes) to enhance the robustness and the pseudofuzzy rough approximations are further redefined based on$k$-Nearest Neighbor ($k$NN) granules determined by the linear reconstruction coefficients to empower the antinoise ability. Then, the pseudo FRS model is employed to guide the robust FS algorithm from the perspective ofredundant filter,strongly relevant priority, anddiscriminative selectionto determine the final feature subset. The experimental results on 31 datasets and practical applications (i.e., cancer diagnosis and face recognition) demonstrate that the reduct gained by the proposed approach generally outperforms those attained by alternative implementations of FRS-based FS and state-of-the-art FS techniques.

Simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs), as a revolutionary technique, can boost transmission security by controlling unfavorable environments for signal eavesdropping and reducing interference. Integrating unmanned aerial vehicles (UAVs) with STAR-RISs has generated considerable interest due to its enhanced deployment flexibility. However, developing secure communication capabilities using flying STAR-RIS remains an open issue. Therefore, this work investigates the secrecy energy efficiency (SEE) maximization problem for the uplink non-orthogonal multiple access (NOMA) systems, where the UAV-mounted STAR-RIS is employed against the eavesdroppers. Specifically, we consider the joint optimization of the power control, the transmission/reflection coefficients, and the UAV/STAR-RIS’s placement for static and mobile scenarios. The problems are also subject to the minimum data rate requirements and the safety flight region. To tackle the intractable problems, we first adopt the iterative-based method to solve the problem under the static scenario. After that, we invoke the fractional programming and successive convex approximation methods to get the power control scheme, the semidefinite relaxation method to get the transmission/reflection (T/R) coefficients design, and the search-based method to obtain the UAV/STAR-RIS position. Extending to the mobile scenario, we adopt the double deep Q-network (DDQN) algorithm to learn the online UAV trajectory design policy from a long-term perspective. Numerical results unveil that: 1) the proposed iterative-based joint optimization algorithm for static scenarios achieves a near-optimal solution; 2) the NOMA communications aided by the UAV-mounted STAR-RIS achieve significant SEE gain over the conventional reflection-only RIS and the fixed STAR-RIS cases; 3) the DDQN-based algorithm for mobile scenario achieves a near-optimal solution and obtains a valuable performance gain over the short-sighted greedy algorithm.

Integrated sensing and communication (ISAC) has been regarded as an emerging technique to satisfy the sensing requirements for future 6G networks. However, the confidential communication information embedded in the probing waveform could be eavesdropped by the radar targets, which leads to insecurity issues for ISAC systems. To this end, we propose a coordinated multi-point transmission (CoMP) empowered secure ISAC system. Unlike existing work focusing on a single base station (BS), multiple BSs are coordinated to improve sensing performance and communication security. Specifically, non-orthogonal multiple access (NOMA) is employed to improve spectrum efficiency and facilitate spectrum sharing between sensing and communication functions. By importing the artificial noise (AN) to disrupt eavesdropper reception, a joint radar association and beamforming design optimization problem is formulated to maximize the minimum beampattern gain, subject to the maximum power constraint and secure communication requirements. The mixed-integer non-convex optimization problem is first transformed into more tractable forms. Then, a near-optimal solution is obtained by applying an accelerated stochastic coordinate descent algorithm for radar association and the penalty-based iterative algorithm for beamforming design. Moreover, the optimization problem is further extended to more practical cases with uncertain target directions. Our numerical results show: i) the proposed AN-aided COMP-NOMA empowered ISAC system can support much higher high-quality radar sensing, while simultaneously guaranteeing secure communication; ii) the proposed scheme significantly outperforms the relevant benchmark schemes in terms of the beampattern gain; iii) the proposed joint optimization algorithm can achieve high beampattern gain, even with uncertain target directions.

Recommender systems are influenced by many confounding factors (i.e., confounders) which result in various biases (e.g., popularity biases) and inaccurate user preference. Existing approaches try to eliminate these biases by inference with causal graphs. However, they assume all confounding factors can be observed and no hidden confounders exist. We argue that many confounding factors (e.g., season) may not be observable from user–item interaction data, resulting inaccurate user preference. In this article, we propose a deconfounded recommender considering unobservable confounders. Specifically, we propose a new causal graph with explicit and implicit feedback, which can better model user preference. Then, we realize a deconfounded estimator by the front-door adjustment, which is able to eliminate the effect of unobserved confounders. Finally, we conduct a series of experiments on two real-world datasets, and the results show that our approach performs better than other counterparts in terms of recommendation accuracy.

Multi-scenario and multi-task recommendation can use various feedback behaviors of users in different scenarios to learn users’ preferences and then make recommendations, which has attracted attention. However, the existing work ignores feature interactions and the fact that a pair of feature interactions will have differing levels of importance under different scenario-task pairs, leading to sub-optimal user preference learning. In this article, we propose a M ulti-scenario and M ulti-task aware F eature I nteraction model, dubbed MMFI , to explicitly model feature interactions and learn the importance of feature interaction pairs in different scenarios and tasks. Specifically, MMFI first incorporates a pairwise feature interaction unit and a scenario-task interaction unit to effectively capture the interaction of feature pairs and scenario-task pairs. Then MMFI designs a scenario-task aware attention layer for learning the importance of feature interactions from coarse-grained to fine-grained, improving the model’s performance on various scenario-task pairs. More specifically, this attention layer consists of three modules: a fully shared bottom module, a partially shared middle module, and a specific output module. Finally, MMFI adapts two sparsity-aware functions to remove some useless feature interactions. Extensive experiments on two public datasets demonstrate the superiority of the proposed method over the existing multi-task recommendation, multi-scenario recommendation, and multi-scenario & multi-task recommendation models.

Sequential recommendation has now been more widely studied, characterized by its well-consistency with real-world recommendation situations. Most existing works model user preference as the transition pattern from the previous item to the next, ignoring the time interval between these two items. However, we find that the time intervals in different sequences may vary significantly and thus result in the ineffectiveness of user modeling due to the issue ofpreference drift. Thus we propose an assumption that a sequence with uniformly distributed time intervals (denoted as uniform sequence) is more beneficial for preference learning than that with greatly varying time intervals. We then conduct an empirical study on four real datasets and the results support this assumption. Therefore, we advocate to augment sequence data from the perspective of time intervals, which is not studied in the literature. Specifically, we design five operators (Ti-Crop, Ti-CateReorder, Ti-Mask, Ti-Substitute, Ti-Insert) to transform the original non-uniform sequence to uniform sequence with the consideration of time intervals. Then, we devise a control strategy to execute data augmentation on item sequences in different lengths and a looseness range to ensure the generalization (or diversity) of generated data. Finally, we implement these improvements on a state-of-the-art model CoSeRec and proposeTimeInterval AwareCoSeRec(TiCoSeRec). Experimental results on four datasets demonstrate that TiCoSeRec achieves significantly better performance than other 11 counterparts recommendation techniques.

Traditional routing algorithms cannot dynamically change network environments due to the limited information for routing decisions. Meanwhile, they are prone to performance bottlenecks in the face of increasingly complex business requirements. Some approaches, such as deep reinforcement learning (DRL) have been proposed to address the routing problems. However, they hardly utilize the information about the network environment fully. The Knowledge Defined Networking (KDN) architecture inspires us to develop new learning mechanisms adapted to the dynamic characteristics of the network topology. In this paper, we propose an effective scheme to solve the routing optimization problem by adding a graph neural network (GNN) structure to DRL, called Message Passing Deep Reinforcement Learning (MPDRL). MPDRL uses the characteristics of GNN to interact with the network topology environment and extracts exploitable knowledge through the message passing process of information between links in the topology. The goal is to achieve the load balance of network traffic and improve network performance. We have conducted experiments on three Internet Service Provider (ISP) network topologies. The evaluation results show that MPDRL obtains better network performance than the baseline algorithms.

Recently, computation offloading methods have greatly improved the Quality of Experience (QoE) in Multi-access Edge Computing (MEC) by offloading tasks to the edge servers. Since well-coordinated actions of Terminal Devices (TDs) are critical to improving the performance of the entire individual system, many practical MEC-based applications, i.e., firefighting robots and unmanned aerial vehicles, require great teamwork among TDs. However, real-world scenarios are usually bound by resource conditions. For instance, network connectivity may weaken or experience interruptions during emergency situations. In cases where the communication medium is utilized by multiple TDs, achieving effective coordination poses a significant challenge. In this paper, we propose a computation offloading scheme based on Scheduled Multi-agent Deep Reinforcement Learning (SMDRL) to make the most efficient decision in a resource-constrained scenario. First, we design a virtual energy queue based on the MEC system and maximize the QoE (related to service delay and energy consumption) in a real-time manner. Subsequently, we propose a scheduled multi-agent deep reinforcement learning algorithm to support each TD in learning how to encode messages, select actions, and schedule itself based on the received messages. Furthermore, a TopK mechanism is introduced. This mechanism chooses the most crucial TDs to broadcast their messages, and then the computation offloading problem in a communication-constrained MEC environment can be solved in a low-communication manner. Also, we prove that even under limited communication conditions, our proposed methods can still lead to the close-to-optimal performance. The final performance analysis shows that the developed scheme has significant advantages over other representative schemes.

Due to the outstanding merits such as mobility, high maneuverability, and flexibility, Unmanned Aerial Vehicles (UAVs) are viable mobile power transmitters that can be rapidly deployed in geographically constrained regions. They are good candidates for supplying power to energy-limited Sensor Nodes (SNs) with Wireless Power Transfer (WPT) technology. In this paper, we investigate a UAV-enabled WPT system that transmits power to a set of SNs at unknown positions. A key challenge is how to efficiently gather the locations of SNs and design a power transfer scheme. We formulate a multi-objective optimization problem to jointly optimize these objectives: maximization of UAV's search efficiency, maximization of total harvested energy, minimization of UAV's flight energy consumption and maximization of UAV's energy utilization efficiency. To tackle these issues, we present a two-stage strategy that includes a UAV Motion Control (UMC) algorithm for obtaining the coordinates of SNs and a Dynamic Genetic Clustering (DGC) algorithm for power transfer via grouping SNs into clusters. First, the UMC algorithm enables the UAV to autonomously control its own motion and conduct target search missions. The objective is to make the energy-restricted UAV find as many SNs as feasible without any apriori knowledge of their information. Second, the DGC algorithm is used to optimize the energy consumption of the UAV by combining a genetic clustering algorithm with a dynamic clustering strategy to maximize the amount of energy harvested by SNs and the energy utilization efficiency of the UAV. Finally, experimental results show that the proposed algorithms outperform their counterparts.

Mobile edge computation (MEC) has recently emerged as a promising computing paradigm for supporting latency-sensitive mobile applications. Due to the limited resources of the edge servers (ESs), efficient resource allocation mechanisms are key to realize the MEC paradigm. In such a resource allocation process, it is a significant challenge to guarantee truthfulness while enabling flexible task offloading and satisfying the locality constraint. To address such a challenge, we propose a truthful auction-based resource allocation mechanism with flexible task offloading (TARFO) in an MEC system. Specifically, we first design the minimum delay task graph partitioning algorithm, aiming at calculating the minimum completion time and the task offloading solutions under different resource profiles. Based on this algorithm, for each smart mobile device (SMD), we further determine the set of feasible non-dominated resource profiles and the corresponding task offloading solutions. We next propose an efficient primal-dual approximation winning bid selection algorithm to determine the set of the winning bids and a critical value based pricing algorithm to calculate the payments of the winning bids. Strict theoretical analysis demonstrates TARFO can ensure truthfulness, individual rationality, computational efficiency and a smaller approximation ratio. Simulation results verify the effectiveness and efficiency of TARFO.

As a distributed learning paradigm, federated learning (FL) can be applied in mobile edge computing (MEC) to support real-time artificial intelligence by leveraging edge computation resources while preserving data privacy in the end devices. However, the unpredictable wireless connections between end devices and edge servers in MEC (e.g., frequent handovers and unstable wireless channels) may result in the loss of important model parameters, which slows down the FL training process and degrades the quality of the global model. In this paper, we propose an adaptive collaborative federated learning (ACFL) scheme to accelerate the convergence and improve model reliability by mitigating communication-based parameter loss under a three-layer MEC architecture. First, a dynamic epoch adjustment method is proposed to reduce communication rounds by dynamically adjusting the training epochs in end devices. In addition, to accelerate the FL convergence, we present an edge server collaborative training scheme by leveraging a multi-layer computing architecture, where edge servers utilize their maintained data to collaboratively train models with end devices. Finally, extensive simulations are conducted and show that ACFL can efficiently improve model reliability and accelerate the convergence of the FL process in MEC.

The rapid adoption of IPv6 has increased network access scale while also escalating the threat of Distributed Denial of Service (DDoS) attacks. By the time a DDoS attack is recognized, the overwhelming volume of attack traffic has already made mitigation extremely difficult. Therefore, continuous network monitoring is essential for early warning and defense preparation against DDoS attacks, requiring both sensitive perception of network changes when DDoS occurs and reducing monitoring overhead to adapt to network resource constraints. In this paper, we propose a novel DDoS incident monitoring mechanism that uses macro-level network traffic behavior as a monitoring anchor to detect subtle malicious behavior indicative of the existence of DDoS traffic in the network. This behavior feature can be abstracted from our designed traffic matrix sample by aggregating continuous IPv6 traffic. Compared to IPv4, the fixed-length header of IPv6 allows more efficient packet parsing in preprocessing. As the decision core of monitoring, we construct a lightweight Binary Convolution DDoS Monitoring (BCDM) model, compressed by binarized convolutional filters and hierarchical pooling strategies, which can detect the malicious behavior abstracted from input traffic matrix if DDoS traffic is involved, thereby signaling an ongoing DDoS attack. Experiment on IPv6 replayed CIC-DDoS2019 shows that BCDM, being lightweight in terms of parameter quantity and computational complexity, achieves monitoring accuracies of 90.9%, 96.4%, and 100% when DDoS incident intensities are as low as 6%, 10%, and 15%, respectively, significantly outperforming comparison methods.

With the widespread popularity of the Internet of Things and various intelligent medical devices, the amount of medical data is rising sharply, and thus medical data processing has become increasingly challenging. Mobile edge computing technology allows computing power to be allocated at the edge closer to users, which enables efficient data offloading for healthcare systems. However, existing studies on medical data offloading seldom guarantee effective data privacy and security. Moreover, the research equipping data offloading architectures with Blockchain neglect the delay and energy consumption costs incurred in using Blockchain technology for medical data offloading. Therefore, in this paper, we propose a data offloading scheme for healthcare based on Blockchain technology, which achieves optimal medical resource allocation and simultaneously minimizes the cost of offloading tasks. Specifically, we design a smart contract to ensure secure data offloading. And, we formulate the cost problem as a Markov Decision Process, solved by a policy search-based deep reinforcement learning (Asynchronous Advantage Actor-Critic) scheme, where we jointly consider offloading decisions, allocation of computing resources and radio transmission bandwidth, and Blockchain data security audits. The security of our smart-contract-based mechanism is theoretically and empirically proved, while extensive experimental results also show that our solution can obtain superior performance gains with lower cost than other baselines.

Network verification has recently made strides, focusing on the satisfiability of configurations and policies or the performance and versatility of their methods. However, they generally ignore explainability, which is the ability to explain why a network violates or satisfies a certain forwarding policy. In this paper, we propose an explainable network verification framework XNV, which uses a novel interpretable fault analysis method to construct an effective explainable network verifier using knowledge graph (KG). XNV provides appropriate explanations to help operators understand the verification results, improving the transparency and trustworthiness of the verification system. First, XNV uses the KG as an intermediate representation of the configuration semantic level, storing the configuration semantics and routing protocol states. Then, XNV constructs human-logical fault trees for policies and implements root-cause analysis of policy violations based on KG queries and minimum cut set matching. Experiments and case evaluations show that our system provides good interpretability while balancing performance, accelerated understanding, and handling of misconfigurations.

With the development of network measurement technologies, a hybrid measurement architecture can effectively optimize the sketch structure in switches, making it more adaptable to the current complex and volatile network environment. However, current optimization technologies based on hybrid measurement architectures generally suffer from insufficient automation, difficulty of learning effective numerical features, and lack of generality, resulting in poor scalability in real deployment. To solve these problems, we propose theTalentSketchframework, based on which we further developDeepSketchfor effective sketch optimization. First, we useSeq2Seqto automatically identify target flows instead of relying on manual thresholds. Second, we propose a new training strategy that extracts low-precision flows for models with weak learning capabilities. Last, we develop a new sketch optimization framework that can optimize different kinds of sketches only by changing the training data for generality. A large number of experimental results show thatDeepSketchexhibits superior performance. For example: (1) the accuracy of optimized sketches has increased by 20% to 73%, (2) Without replacing the model structure, the accuracy of the optimized sketches can generally reach over 80%. (3) The impact of low sampling rates on accuracy is less than 1% on various sketches.

In-band network telemetry (INT) allows for fine-grained network monitoring, without requiring communication with the controller at each hop. Existing INT-based network-wide telemetry systems achieve low-overhead monitoring with non-overlapping path planning algorithms. However, these systems do not constrain the length of the generated probing paths, which will lead to packet loss when the size of the packet with collected telemetry data exceeds the MTU limit. To address this issue, we propose MTU-adaptive path segmentation algorithms step by step in this paper. Initially, we present two single-path planning algorithms: the INT-optimize algorithm, which produces a single path that covers the entire network with the lowest southbound communication overhead, and the INT-low-cost algorithm, which further accelerates the INT-optimize. Next, to consider the MTU limit, we propose the single-MTU adaptive INT-Segment algorithm to divide the single long path generated in the previous step into multiple path segments. In addition, we generalize the MTU-adaptive network telemetry problem and propose a multi-MTU adaptive INT-Segment solution to achieve high-performance network telemetry in networks with multiple MTU settings. Extensive evaluations demonstrate that our proposed MTU-adaptive solutions can achieve sub-second network-wide telemetry for large-scale networks, with less than 2.9ms to calculate the probing paths for an 18-pod FatTree. Furthermore, our multi-MTU adaptive INT-Segment solution significantly reduces the number of INT Sinks and INT Sources by 13.25%-42.39% when deployed in multi-MTU networks while maintaining stable telemetry data collection time. Compared with the state-of-the-art INT-path, our solution adapts the probing path to the network MTU limit, producing a telemetry data collection efficiency improvement of 10%-94%.

The current global economy is undergoing a transformative phase, emphasizing collaboration among multiple competing entities rather than monopolization. Economic globalization is accelerating the adoption of globalized cloud services, and in line with this trend, cloud 2.0 introduces the concept of “cloud cooperation”. JointCloud, as a novel computing model for Cloud 2.0, advocates for the establishment of an evolving cloud ecosystem. However, a critical challenge arises due to the lack of direct incentives for a cloud to join the JointCloud ecosystem, leading to uncertainty regarding the rationale for the existence of the JointCloud ecosystem. To address this ambiguity, we draw inspiration from supply chain competition and formulate the market dynamics of resources within the JointCloud ecosystem. Our focus is particularly on the analysis of data resource trade within the JointCloud market. To comprehensively analyze the JointCloud market, we propose a market game that examines the competition among clouds within the ecosystem. We theoretically prove that a Nash Equilibrium always exists under the JointCloud market. Subsequently, we conduct an in-depth analysis of the profits of cloud resource manufacturers and cloud resource retailers as the number of clouds varies within the JointCloud ecosystem. Based on our analysis, we further explore the incentives for a cloud to participate in the JointCloud ecosystem. We then evaluate the performance of the proposed market game through extensive experiments, illustrating how process variables and profits change with the market size. The experiments demonstrate that the trends of various variables are aligned with our analysis obtained from the market game. Compared with the Cournot model, our proposed model captures the market power of both manufacturers and retailers, resulting in a model that closely mirrors real market dynamics. Our findings provide valuable insights into the cloud market within Cloud 2.0, offering guidance for stakeholders navigating the evolving landscape of cloud cooperation and competition.

Enforcing scheduling policies with software schedulers at end-hosts leads to high CPU consumption, low throughput, and inaccuracies. To address these issues, offloading packet schedulers to the network interface card presents a promising research direction. However, existing attempts suffer from inflexible on-NIC scheduling that cannot execute complex hierarchies of network policies. In this article, we propose FlowValve+, a general framework for multi-queue packet scheduling on SoC-based SmartNICs. The key insight behind FlowValve+is to abstract the inherent queueing system as a single FIFO queue and perform specialized tail drop to mix the FIFO queue with expected flow proportions. FlowValve+leverages hardware accelerations to produce high throughput while substantially reducing CPU and memory usage on end-hosts. We prototype FlowValve+on Netronome Agilio 40GbE and NVIDIA Bluefield-2 100GbE SmartNICs to demonstrate its ability to accurately enforce network policies while driving TCP traffic at 40 Gbps and 80 Gbps on both platforms, respectively. Moreover, FlowValve+can save two CPU cores compared to DPDK packet schedulers.

Network Function Virtualization (NFV) provides a flexible way to provision new services by decoupling network functions from hardware and implementing them as Virtual Network Functions (VNFs). However, the rapid development of technologies greatly promotes the explosion of diverse services, which directly results in the exponential increase of heterogeneous traffic. In addition, such a tremendous amount of heterogeneous traffic will generate bursts in a more dynamic and unexpected manner, so it becomes extremely hard to satisfy the customer demands. Aiming at addressing these challenges, this work proposes a positive and elastic VNF deployment mechanism for service provisioning, which introduces three novelties:1) a Gated Recurrent Unit (GRU) based traffic prediction model is established to predict the unexpected and dynamically changing traffic behaviors in advance with the accuracy over 98%; 2) a closed-loop system is formed, in which the prediction model can learn and evolve continuously to respond to more complex scenarios; 3) different states of VNF are introduced and dynamically switched to deal with the current demands with reduced cost by avoiding frequent VNF initialization and destroy.The experimental results indicate that the proposed mechanism outperforms the state-of-the-art methods, which include achieving over 98% prediction accuracy, improving the service acceptance rate by more than 18%, and reducing the overall cost by more than 20%.

With the proliferation of devices connected to the Internet of Things (IoT), the complexity of network management has increased. To intelligently manage large-scale networks, we propose a Knowledge-Defined Edge Computing Networks (KDECN) architecture. Edge Nodes (ENs) deployed in the KDECN architecture are responsible for collecting and preprocessing the relevant information uploaded by User Devices (UDs), and provide computation resources for UDs. Futhermore, since multiple UDs share system computation resources, one computing decision will affect the subsequent decision-making of other UDs. Thus, accurately predicting the demands for UD task requests is a key challenge to maximize long-term execution utility. To this end, we deploy the LSTM-based Task Request Demand Prediction (TRDP) method on the management plane of KDECN architecture to predict the task request quantity of UDs in each future time slot. In order to maximize long-term execution utility of the system, we propose a Deep Reinforcement Learning (DRL)-based Long-term Computation Offloading and computation Resource Allocation (L-CORA) algorithm. Specifically, the proposed L-CORA algorithm makes computing decisions based on the prediction of the offloading task quantity and the personalized demands of UDs to ensure the long-term quality of computing service. Extensive experiments with Shanghai real-world datasets to prove that the KDECN-based L-CORA algorithm effectively improves the average utility of the system.

Sequential recommendation is an important task to predict the next-item to access based on a sequence of interacted items. Most existing works learn user preference as the transition pattern from the previous item to the next one, ignoring the time interval between these two items. However, we observe that the time interval in a sequence may vary significantly different, and thus result in the ineffectiveness of user modeling due to the issue of preference drift. In fact, we conducted an empirical study to validate this observation, and found that a sequence with uniformly distributed time interval (denoted as uniform sequence) is more beneficial for performance improvement than that with greatly varying time interval. Therefore, we propose to augment sequence data from the perspective of time interval, which is not studied in the literature. Specifically, we design five operators (Ti-Crop, Ti-Reorder, Ti-Mask, Ti-Substitute, Ti-Insert) to transform the original non-uniform sequence to uniform sequence with the consideration of variance of time intervals. Then, we devise a control strategy to execute data augmentation on item sequences in different lengths. Finally, we implement these improvements on a state-of-the-art model CoSeRec and validate our approach on four real datasets. The experimental results show that our approach reaches significantly better performance than the other 9 competing methods. Our implementation is available: https://github.com/KingGugu/TiCoSeRec.

The goal of continual learning (CL) is to continuously learn new tasks without forgetting previously learned old tasks. To alleviate catastrophic forgetting, gradient projection based CL methods require that the gradient updates of new tasks are orthogonal to the subspace spanned by old tasks. This limits the learning process and leads to poor performance on the new task due to the projection constraint being too strong. In this paper, we first revisit the gradient projection method from the perspective of flatness of loss surface, and find that unflatness of the loss surface leads to catastrophic forgetting of the old tasks when the projection constraint is reduced to improve the performance of new tasks. Based on our findings, we propose a Data Augmented Flatness-aware Gradient Projection (DFGP) method to solve the problem, which consists of three modules: data and weight perturbation, flatness-aware optimization, and gradient projection. Specifically, we first perform a flatness-aware perturbation on the task data and current weights to find the case that makes the task loss worst. Next, flatness-aware optimization optimizes both the loss and the flatness of the loss surface on raw and worst-case perturbed data to obtain a flatness-aware gradient. Finally, gradient projection updates the network with the flatness-aware gradient along directions orthogonal to the subspace of the old tasks. Extensive experiments on four datasets show that our method improves the flatness of loss surface and the performance of new tasks, and achieves state-of-the-art (SOTA) performance in the average accuracy of all tasks.

Basket representation plays an important role in the task of next-basket recommendation. However, existing methods generally adopts pooling operations to learn a basket's representation, from which two critical issues can be identified. First, they treat a basket as a set of items independent and identically distributed. We find that items occurring in the same basket have much higher correlations than those randomly selected by conducting data analysis on a real dataset. Second, although some works have recognized the importance of items repeatedly purchased in multiple baskets, they ignore the correlations among the repeated items in a same basket, whose importance is shown by our data analysis. In this paper, we propose a novel Basket Representation Learning (BRL) model by leveraging the correlations among intra-basket items. Specifically, we first connect all the items (in a basket) as a hyperedge, where the correlations among different items can be well exploited by hypergraph convolution operations. Meanwhile, we also connect all the repeated items in the same basket as a hyperedge, whereby their correlations can be further strengthened. We generate a negative (positive) view of the basket by data augmentation on repeated (non-repeated) items, and apply contrastive learning to force more agreements on repeated items. Finally, experimental results on three real datasets show that our approach performs better than eight baselines in ranking accuracy.

In recent years, probabilistic network analysis has received significant attention as a crucial aspect of network management. With the increasing use of distributed routing protocols in networks, computing the probabilities of multiple properties becomes a challenging task, as different property violations derive from different failure scenarios. While some previous work has attempted to prune the space of failure scenarios by identifying a set of links whose failure does not change whether a property holds, this approach often requires multiple calls to the system, leading to redundant computation when verifying multiple properties. To solve this problem, we propose NetDiceSyn, a scalable multi-property probabilistic network configuration analyzer. As verifying each property requires exploring a state tree, NetDiceSyn proposes to explore all state trees synchronously and compute a shared data plane for multiple states in different state trees, thus reducing the redundant computation between different state trees and verifying multiple properties simultaneously. Besides, we design a state sharing method to reduce the redundant storage of states during synchronous exploration. Additionally, we propose abstract topology to merge equivalent states within a state tree to speed up probabilistic verification in series link networks. Extensive experiments are conducted with real-world network topologies, and the results show that NetDiceSyn outperforms state-of-the-art methods, providing up to an order of magnitude speedup when verifying hundreds of properties for networks with hundreds of links.

In recent years, the number of network applications is constantly increasing, and network congestion often occurs. To ensure the network Quality of Service (QoS), different types of traffic are classified according to their requirements, and similar traffic is transmitted to the same queue for scheduling. The switch generally uses fair queuing and its extension schemes to schedule traffic. These schemes achieve different bandwidth allocation by configuring different queue weights, so as to obtain a lower packet loss rate. However, the switch can provide us with very few statistical parameters, so using a large number of statistical parameters for adaptive weight adjustment is challenging in implementation. At the same time, the weight range supported by the switch is large, but the action space supported by reinforcement learning is limited, which cannot represent the entire queue weight space. Although deep reinforcement learning can solve the problem with large space, the existing switches can not well support the calculation of neural network model. In this paper, we propose a scheduling optimization mechanism combining Q-learning and genetic algorithm, called QGSO, which is used to schedule traffic in real switches. Firstly, we model the scheduling optimization problem as a Markov decision process (MDP) and use Q-learning to solve it in order to select the optimal queue weights according to the state of the environment. Secondly, we use genetic algorithm to filter out a group of optimal queue weights from the weight space, achieving compression of the solution space. Finally, we use a hardware testbed to test and verify the effectiveness of the algorithm. The experimental results show that our algorithm can effectively schedule traffic and achieve a lower packet loss rate.

As a core component of the Internet of Vehicles, reasoning about the trajectory of pedestrians or vehicles in complex road conditions plays a critical role in autonomous driving and socially aware robotic navigation. Most existing methods do not adequately consider the effects of heterogeneous traffic agents. Toward this end, we propose the traffic trajectory prediction algorithm based on the convolutional attention network (TraGCAN) to predict the trajectories of heterogeneous traffic agents in dense traffic. The algorithm of the proposed method examines the behavior of different traffic agents in terms of both time and space dimensions to identify their movement patterns and interactions. We construct the spatial relationship of traffic agents as a graph structure and introduce a graph convolutional network to extract spatial interactions. In addition, we design a spatial attention mechanism to adaptively calculate weights for all spatial interactions to capture different influences from neighboring agents. To improve the accuracy of trajectory prediction, the algorithm considers the influence of the heterogeneous characteristics of traffic agents on their motion behaviors. We evaluated the performance of the proposed TraGCAN on heterogeneous traffic data sets, and the results demonstrate that the error of TraGCAN is reduced by 15% compared to existing methods.

Multiaccess edge computing (MEC) provides task offloading services to facilitate the integration of idle resources with the network and bring cloud services closer to the end user. By selecting suitable servers and properly managing resources, task offloading can reduce task completion latency while maintaining the Quality of Service (QoS). Prior research, however, has primarily focused on tasks with strict time constraints, ignoring the possibility that tasks with soft constraints may exceed the bound limits and failing to analyze this complex task constraint issue. Furthermore, considering additional constraint features makes convergent optimization algorithms challenging when dealing with such complex and high-dimensional situations. In this article, we propose a new computational offloading decision framework by minimizing the long-term payment of computational tasks with mixed bound constraints. In addition, redundant experiences are gotten rid of before the training of the algorithm. The most advantageous transitions in the experience pool are used for training in order to improve the learning efficiency and convergence speed of the algorithm as well as increase the accuracy of offloading decisions. The findings of our experiments indicate that the method we have presented is capable of achieving fast convergence rates while also reducing sample redundancy.

Cross-silo federated learning (FL) is a privacy-preserving distributed machine learning where organizations acting as clients cooperatively train a global model without uploading their raw local data. Recently, the cross-silo FL in multiaccess edge computing (MEC) is used in increasing industrial applications. Most existing research on cross-silo FL pays attention to the performance aspect, ignoring the incentive mechanism for high-quality client selection and long participation in model training for efficient and stable FL, which has prevented the widespread adoption of cross-silo FL in MEC. In this article, we propose an incentive mechanism with quality-Aware and reputation-Aware based on the infinitely repeated game for cross-silo FL named VARF. VARF selects high-quality and high-reputation edge nodes (ENs) as candidates for model training in the cross-silo FL by a heuristic algorithm and then motivates the selected ENs to actively contribute their resources. VARF also models the long-term behavior of ENs in cross-silo FL as an infinitely repeated game and derives a stable and long-term cooperative strategy for clients while maximizing the amount of local data for model learning in cross-silo FL. Extensive simulations with real-world data sets demonstrate that the performance of VARF is more beneficial than other benchmarks. Meanwhile, experimental results show that cloud platforms (CPs) and ENs eventually form a long and stable cooperative relationship under the trigger strategy.

Applying the Multiple Drones System (MDS) to perform the repetitive and dangerous tasks for human in many complex environments has become a trend all around the world, due to the increasing capacity of mobile communication and the increasing intelligence of drone robots. However, to fulfill the target with less cost as much as possible, drones need to collaborate deeply with each other to make the optimal decision, which is now important and challenging. In this work, we focus on addressing the efficient task allocation among large-scale drones with the object of minimizing the resource waste and cost, which is proved to be NP-hard. Specifically, we first introduce the Digital Twin (DT) technology to dynamically construct the spatial structure for drones, in which a density clustering based algorithm is proposed to decompose the large-scale task allocation problem among all drones into smaller sub-problems among partial drones. Then, for each sub-problem, we propose an improved auction algorithm to allocate the sub-tasks to local drones according to the task difficulty and drone ability. The experimental results indicate that the proposed method outperforms the state-of-the-art methods in terms of the moving distance, resource utilization and task completion time, etc.

With the rapid development of network function virtualization (NFV), more and more enterprises and operators are seeking network service provisioning via service chains of virtual network functions (VNFs), instead of depending on proprietary hardware appliances. Following this trend, an NFV market is emerging, where users can procure different VNF instances (VNFIs) and their combinations from multiple network service providers (NSPs) in a pay-as-you-go way. In such a procurement process, how to guarantee truthfulness while enabling flexible resource provisioning in form of VNFIs is a significant challenge. In this paper, we propose a truthful reverse combinatorial auction-based mechanism to solve the combinatorial VNFI procurement problem. To support flexible resource provisioning, this mechanism allows NSPs to be multi-minded, and determine the provisioning VNFIs according to the auction results. Specifically, we design a heuristic algorithm to determine the winning bids in polynomial time. Furthermore, we devise a critical-payment-based pricing algorithm to induce NSPs to disclose their real costs, aiming to achieve truthfulness. Rigorous theoretical analysis shows the proposed mechanism can guarantee truthfulness, individual rationality and computational efficiency. Simulation results also verify the effectiveness and efficiency of the proposed mechanism.

The rapid growth of mobile device (e.g., smart phone and bracelet) has spawned a lot of new applications, during which the requirements of applications are increasing, while the capacities of some mobile devices are still limited. Such contradiction drives the emergency of computation migration among mobile edge devices, which is a lack of research currently. In this article, we focus on addressing the computation migration oriented resource allocation problem among mobile edge devices. Specifically, we first construct a framework for Mobile Social Cloud(MSC), in which the mobile devices with rich resources are abstracted as resource suppliers and those resource-lacking devices are abstracted as resource demanders. Then, a mathematical model is formulated and an evolutionary algorithm is proposed to effectively solve this model based on decomposition, dominance, and genetic operations. Moreover, the parallel computing is introduced to further improve the efficiency of the proposed algorithm. The experimental results indicate that the proposed algorithm outperforms the other state-of-the-art methods and it improves the calculation efficiency by about 178 percent (2 cores) and 262 percent (3 cores) by introducing parallel computing.

Misinformation and rumors can spread rapidly and widely through online social networks, seriously endangering social stability. Therefore, rumor blocking on social networks has become a hot research topic. In the existing research, when users receive two opposing opinions, they tend to believe the one arrives first. In this article, we argue that users will dialectically trust the information based on their own opinions rather than the rule of first-come-first-listen. We propose a confidence-based opinion adoption (CBOA) model, which considers the opinion and confidence according to the traditional linear threshold (LT) model. Based on this model, we propose the directed graph convolutional network (DGCN) method to select the$k$most influential positive cascade nodes to suppress the propagation of rumors. Finally, we verify our method on four real network datasets. The experimental results show that our method can sufficiently suppress the propagation of rumors and obtains smaller number of rumor nodes than the baseline algorithms.

Opinion Maximization (OM) aims at determining a small set of influential individuals, spreading the expected opinions of an object (e.g., product or individual) to their neighbors through the social relationships and eventually producing the largest opinion spread. In previous studies, once the corresponding nodes are activated, their opinions usually keep unchanged, which fails to capture the real scenarios where the opinion of each node on the object can dynamically change over time. In this view, we propose a Dynamic Opinion Maximization Framework (DOMF) to settle the OM problem, which consists of two parts: dynamic opinion formation and adaptive seeding process. Specifically, we formulate the OM problem by maximizing rational opinions, and prove that: 1) the OM problem within a constant ratio is NP-hard, and 2) the objective function does not satisfy the monotonicity and submodularity properties anymore. To model the dynamic opinion issue, we propose adaptive cooperation model based on Q-learning theory, which is proved to be capable of eventually reaching convergence. Moreover, to dynamically generate the initial seed nodes, we design the Multi-stage Heuristic Algorithm (MHA). Experimental results demonstrate that each component of our model is effective, and the proposed approach improves the rational opinion spread.

Recently, most of the Internet of things (IoT) infrastructures are highly centralized with single points of failure, which results in serious security and privacy issues of IoT data. Fortunately, blockchain technique can provide a decentralized and secure IoT framework to deal with security issues based on the characteristics of decentralization, non-tampering, openness, transparency, and traceability. However, the blockchain consensus protocol guarantees the safety and reliability of data, but it also brings problems such as scalability limitations and poor storage extensibility, resulting in the inability to directly integrate blockchain and the IoT in existing conditions. In this article, a private three-tier local blockchain, Three-tier architecture Blockchain (TBchain), is proposed to solve the problem by splitting part of the transactions in the public blockchain and locking them in a higher-level blockchain TBchain. Additionally, the private blockchain TBchain is connected to the public blockchain to build a hierarchical blockchain network to provide privacy protection for the IoT data stored on the blockchain. Finally, we implement an IoT framework based on TBchain and the InterPlanetary File System (IPFS) to realize the decentralized IoT, which guarantees the user’s access control right to personal data. Experimental results show that the IoT framework based on TBchain and IPFS realizes the user’s access control right to personal data by verifying in advance to ensure the confidentiality and security of shared data, and improves the security and privacy of IoT data and transactions. Moreover, we prove that the scalability and storage extensibility of the blockchain is positively correlated with the number of data blocks in TBchain.

Monocular 3D reconstruction is to reconstruct the shape of object and its other information from a single RGB image. In 3D reconstruction, polygon mesh, with detailed surface information and low computational cost, is the most prevalent expression form obtained from deep learning models. However, the state-of-the-art schemes fail to directly generate well-structured meshes, and we identify that most meshes have severe Vertices Clustering (VC) and Illegal Twist (IT) problems. By analyzing the mesh deformation process, we pinpoint that the inappropriate usage of Chamfer Distance (CD) loss is a root cause of VC and IT problems in deep learning model. In this article, we initially demonstrate these two problems induced by CD loss with visual examples and quantitative analyses. Then, we propose a fine-grained reconstruction method CD 2 by employing Chamfer distance twice to perform a plausible and adaptive deformation. Extensive experiments on two 3D datasets and comparisons with five latest schemes demonstrate that our CD 2 directly generates a well-structured mesh and outperforms others in terms of several quantitative metrics.

This article investigates how to enhance the Multi-access Edge Computing (MEC) systems performance with the aid of device-to-device (D2D) communication computation offloading. By adequately exploiting a novel computation offloading mechanism based on D2D collaboration, users can efficiently share computational resources with each other. However, it is challenging to distinguish valuable information that truly promotes a collaborative decision, as worthless information can hinder collaboration among users. In addition, the transmission of large volumes of information requires high bandwidth and incurs significant latency and computational complexity, resulting in unacceptable costs. In this article, we propose an efficient D2D-assisted MEC computation offloading framework based on Attention Communication Deep Reinforcement Learning (ACDRL), which simulates the interactions between related entities, including device-to-device collaboration in the horizontal and device-to-edge offloading in the vertical. Second, we developed a distributed cooperative reinforcement learning algorithm that includes an attention mechanism that skews computational resources towards active users to avoid unnecessary resource wastage in large-scale MEC systems. Finally, to improve the effectiveness and rationality of cooperation among users, we introduce a communication channel to integrate information from all users in a communication group, thus facilitating cooperative decision-making. The proposed framework is benchmarked, and the experimental results show that the proposed framework can effectively reduce latency and provide valuable insights for practical design compared to other baseline approaches.

A novel simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) assisted collabo-rative multi-cell network is proposed. Cell-center and cell-edge users are enhanced by the STAR-RIS's reflection and transmission functions, respectively. We propose an online joint active and passive beamforming framework to maximize this system's long-term energy efficiency (EE) under time-varying channels and users' requirements. We first invoke fractional programming (FP) to optimize the coordinated zero-forcing beamforming among all base stations and to construct an interference-free transmission during each time slot. Then, a parallel deep reinforcement learning (DRL) algorithm is proposed to facilitate the online optimization of the passive beamforming of all STAR-RISs. Finally, the beamforming calculated by the FP algorithm is utilized as a part of the reward function of the DRL. As a result, the size of the action and state space is reduced, and the dynamic joint optimization is realized. Extensive numerical results reveal that: 1) the proposed algorithm induces a low computation complexity compared with conventional DRL algorithms, and 2) the STAR-RIS enhanced system can achieve higher EE than systems without RIS or with conventional reflection/transmission-only RISs.

Enforcing scheduling policies at end-hosts with software schedulers suffers from high CPU consumption, low throughput, and inaccuracy. Offloading scheduling functions to the network interface card (NIC) provides a promising direction to address these problems. However, existing efforts in scheduling offloading suffer from inflexible on-NIC packet schedulers, which cannot execute complex hierarchies of network policies. In this paper, we present FlowValve, the first parallel packet scheduler for Network Processor (NP)-based SmartNICs that offloads critical functions of Linux traffic control, including packet classifying and scheduling. The key insight behind FlowValve is to abstract inherent queues attached to the NIC interface (wire side) as a single FIFO queue and perform specialized tail drop to mix the FIFO queue with expected flow proportions. FlowValve takes advantage of on-chip multi-core parallelism and hardware accelerations to produce high throughput. Meanwhile, it substantially reduces CPU and memory burdens on end-hosts. We prototype FlowValve on a Netronome Agilio SmartNIC and demonstrate its effectiveness against non-offloaded kernel schedulers and DPDK QoS Scheduler. We find that FlowValve outperforms both in accurately enforcing network policies while driving line rate performance (i.e., 40Gbps), which contributes to saving at least two CPU cores.

With the massive and rapid growth of network traffic, sketching techniques are widely used to estimate a variety of network flow-level metrics, e.g., flow size, top-k flows, and the number of flows. However, how to cope with the constantly changing network environment and achieve high accuracy with limited switch resources is still a challenging problem. Existing studies do not well address the issues of identifying and correcting the inaccurately estimated flows. Therefore, we propose TalentSketch, an adaptive and high-precision hybrid measurement framework based on LSTM. TalentSketch uses LSTM to learn the sketch features with the captured traffic information. It could identify the inaccurately estimated flows with an error-prone flow model and correct them with a well-trained regression model. We conduct extensive experiments to verify the performance of TalentSketch. Experimental results show that, without increasing switch resource overhead, TalentSketch improves the measurement accuracy of different kinds of sketches by 12%∼23%. More importantly, it can track network fluctuations and provide feedback on the overall accuracy of a sketch in real-time.

In-band network telemetry (INT) enables hop-by-hop fine-grained network monitoring without interacting with the controller at every hop. Existing INT-based network-wide telemetry systems achieve low-overhead monitoring with the non-overlapped path planning algorithms. However, they do not bound the length of the generated probing paths, which may lead to packet loss when the collected telemetry data exceeds the MTU limit. In this paper, we propose an MTU-adaptive path segmentation algorithm to solve this problem. First, we provide two single-path planning algorithms: the INT-optimize algorithm produces a single path that covers the entire network with the lowest southbound communication overhead, and the INT-low-cost algorithm further improves the path planning efficiency of the INT-optimize. By taking the MTU limit into account, we further propose INT-Segment, a novel path segmentation algorithm to split the single long path produced from the previous step into multiple path segments. Extensive evaluations show that the proposed INT-Segment can realize sub-second network-wide telemetry for large-scale networks. It takes less than 2.9ms to calculate the probing paths for an 18-pod FatTree. Compared with the state-of-the-art INT-path, our solution makes the probing path well adapted to the network MTU limit and improves the telemetry efficiency by 10%-94%.

In-band network telemetry (INT) delivers high-precision network monitoring by collecting device-internal states entirely on the data plane. For rapid congestion awareness and network troubleshooting, it is necessary to conduct network-wide telemetry by generating multiple monitoring paths covering the entire network graph. Solving the optimal path planning problem used the eulerian trail initially at a time complexity of$O(k(3E+V-15k/2))$. For mega-scale data center networks, such a high complexity is unacceptable because the algorithm cannot adapt well to occasional topology changes. In this work, we propose improved INT-path and INT-react, two refined path planning algorithms with a much reduced time complexity of only$O(E)$. Furthermore, INT-react also considers balanced path generation to reduce the longest path length for synchronized collection of telemetry data from each monitoring path. The evaluation shows that on average it costs 2.10s for the improved INT-path to solve the optimal path planning for a network of 9500 switches, while the computation is completed within only 0.283s on average for INT-react. In addition, INT-react reduces the longest path length. INT-react's path planning is so fast that it promptly reacts to topology changes and is ready to be deployed in mega-scale production networks.

In the data center network, mixed flows which have contradictory service requirements are transmitted simultaneously. Switches usually aggregate similar flows to the same queue after flow classification and schedule them using fair queuing and its extension schemes capable of flow isolation. These schemes implement diverse bandwidth allocation by assigning different weights to queues. Existing solutions rely on rich statistics such as packet arrival rate and delay to realize dynamic bandwidth allocation. However, many statistics are difficult to accurately measure or even obtain in real switches due to resource limitations. Providing differentiated services for mixed flows under such restrictions is still a challenge. To solve this issue, this paper proposed a reinforcement learning-based scheduling optimization (RLSO) mechanism. First, mixed flows scheduling is modeled as the Markov decision process (MDP) and Q-learning is used to find the approximate optimal solution with a few statistics. Second, the solution space is compressed to reduce the complexity of the algorithm and adapt to the limited performance of switches. Finally, the performance of the proposed mechanism is evaluated on a hardware testbed with workloads that include coarsegrained and fine-grained flows. The results show that RLSO can effectively schedule mixed flows.

Multiple types of flows with contradictory service requirements, namely mixed flows, coexist in the data center network. Similar flows will be aggregated into the same queue after flow classification and are scheduled in switches by using fair queueing and its extension schemes which are capable of flow isolation. These schemes allocate different bandwidths by adjusting weights to realize differentiated services. However, existing solutions only focus on the requirements of some flows, which leads to the failure to satisfy the requirements of other flows. Therefore, it is necessary to make a trade-off between the service requirements of different flows when allocating bandwidth. In this paper, a max-min fairness based scheduling optimization (MMFSO) mechanism is proposed to schedule mixed flows. First, the bandwidth requirements of each queue are calculated by statistics of the switch. To reduce the influence of sampled statistics while forecasting bandwidth requirements, we introduce the exponentially weighted moving average for bandwidth requirements computation. Second, the bandwidth is allocated to each queue according to the max-mix fairness. The queue weight is determined by the allocated bandwidth of the queue. Finally, the performance of the proposed mechanism is evaluated on the hardware testbed in which workloads include coarse-grained flows and fine-grained flows. The results show that MMFSO can effectively schedule mixed flows.

In order to support the 5GB5G network with huge connections, large traffic, and low latency, the evolution of the IP network to the IPv6 network is an inevitable trend. Although IPv6 network has significantly improved network performance, there are still a series of network failures. After the failure occurs, it is necessary for the routing algorithm to create a backup path to ensure the continuous provision of network services. So, the selection of the backup path directly determines the performance of the network service. SRv6 takes advantage of the programmable capability of the IPv6 extension header to implement differentiated reliable routing, satisfies the QoS requirements of different network services, and improves user experience. First, we design the network model for single link failure, define the network service, and design the two-dimensional flow table. Then, the backup path traversal algorithm is designed to obtain the universal set of backup paths in the network, and the optimal backup path for a specific network service is selected through comprehensive evaluation method. Experimental results show that the differentiated reliable routing mechanism proposed in this paper exhibits better performance in terms of bandwidth, delay, jitter and packet loss rate compared with the benchmark mechanism.

Software defined networking (SDN) have the advan-tages of centralized control, global visibility, and programmabil-ity, but these features also bring new security issues, such as Topological Poisoning Attack (TPA), where attackers can attack topology discovery services by stealing host locations or forging link information. Considering the three levels of identity, data package and path, this paper designs a chain authentication defense scheme. The scheme includes authentication mechanism, transaction information storage mechanism, source IP authenti-cation mechanism and smart contract notification mechanism. The received packets are authenticated by digital signature algorithm, and the trusted identity and location information are stored securely. At the same time, an improved block storage structure is designed to avoid data redundancy, and malicious information is processed by smart contract notification and stream rule installation. The experimental results show that the defense scheme designed in this paper can effectively defend against TPA attacks. Compared with the benchmark mechanism, the deployment of this scheme has less impact on controller performance and less impact on the delay of topology discovery in SDN.

In this paper, the joint task offloading and resource allocation are investigated for the semi-grant-free (SGF) non-orthogonal multiple access (NOMA) assisted mobile edge computing (MEC) system. Moreover, simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) are deployed to improve the quality of wireless communications under the mode switching protocol. Each MU can partially or fully offload its task to the base station (BS) based on its differentiated channel conditions and computing capacity in the proposed MEC system. We formulate the joint task offloading, channel assignment, power allocation, and the RIS coefficients design problem to save energy consumption. The formulated problem is modeled from a long-term optimization perspective as a multi-agent Markov game (MG). Then, a multi-agent deep reinforcement learning (MADRL) based joint task offloading and resource allocation (JTORA) algorithm is proposed to solve the problem. The simulation results confirm that the applied SGF-NOMA scheme can significantly reduce energy consumption under a stringent latency constraint. Moreover, the effectiveness of the STAR-RIS and the proposed algorithm are confirmed.

The problem of resource allocation in a high mobility network is always meaningful while challenging. Due to the mobility characteristic, the main difficulty lies in solving different optimization problems in a limited time, so traditional time-consuming optimization solvers are no longer applicable. This paper considers NGMA-enabled heterogeneous networks (HetNets) and uses end-to-end multi-agent reinforcement learning (MARL) to optimize the resource allocation problem. Even though MARL can use many cooperative agents to divide action space, it may lead to a significant increase in agent number, which is harmful to credit assignment. To tackle this problem, we employ a unique design that can fix the number of agents in any case. Agent credit assignment is then considered to guide agents to work cooperatively and ensure each efficient action gets a suitable reward. Also, a novel learning process named Learn to Improve is utilized to make our method more general. Numerical results and comparison experiments show the effectiveness and robustness of our methods.

Abstract 5G cellular IoT has several advantages compared to other access technologies, enabling operators to serve a wider area and more IoT devices. However, in the urban transportation system, a massive number of vehicles exhaust the available resources in the cell, resulting in excessive load in the 5G cellular network. This article proposes a novel reinforcement learning based V2V routing (RLbR) framework, which offloads non‐realtime traffic into the V2V network and significantly relieves the load of 5G cellular network. Meanwhile, we propose a V2V routing algorithm. Specifically, the Q ‐values of neighbouring vehicles are firstly calculated according to the cache factor CF and energy factor EF and evaluate the quality of neighbouring vehicles. Then, the position factor PF is calculated, based on which, the vehicle forwards the data packet. In addition, an environment model is designed to accelerate the convergence of Q ‐table. The results show that the RLbR framework brings the highest offload rate compared to the other three frameworks, and simultaneously, the proposed algorithm improves the lifetime of the V2V network and performs well in terms of delivery ratio and average delay.

Abstract Best‐effort service model of traditional routing is gradually hard to meet the personalized demands under the rapid development of network technologies (e.g. 5G and IPv6). Therefore, service customization should be considered. In this work, a service customized routing mechanism based on deep learning in IPv6 network is proposed, which includes deep learning‐based service customization module, reliability evaluation module, and routing calculation module. The first module uses neural network to learn the complex service customization function, which can quickly output win‐win customized service strategies based on user demands. The second module can quantify the reliability of service routing paths, where not only the link status of IPv6 Neighbor Unreachable Detection (NUD) is considered, but also propose link performance weights to ensure the reliability of differentiated service performance. The third module uses the gray wolf optimization algorithm to calculate an optimal routing path to forward services with the customized strategies as the constraints and the maximum reliability and minimum cost as the goal. Finally, the mechanism is tested on the IPv6 Source Address Validation Improvement (SAVI) platform, which can reduce the execution time by 12.25% and improve the average routing reliability, user and ISP satisfaction by 9.0%, 40.45% and 7.4%, respectively.

A future Internet of Things (IoT) will feature a service-oriented architecture consisting of lightweight computing platforms offering individual, loosely coupled microservices. Often, an end-user will request a bespoke service that will require a composition of two or more microservices offered by different service providers. However, the underlying complexities of soft compositions and the increased security risks are inherent in such a massively decentralized and distributed architecture. The use of trust management to secure the IoT is well studied in the literature. However, there are limitations to its use in service compositions in the IoT. Specifically, transparent (agnostic) trust composition and decomposition remain key problems for this area. A novel model for trustworthy service compositions in the IoT, SC-TRUST, is therefore proposed to deal with these challenges. In this study, the trust properties of service compositions and the effect of service workflows on transparent trust composition and decomposition are investigated. Based on the findings, relevant trust evaluation functions are derived to guide the compositions. SC-TRUST was implemented in a suitable application and its performance, in terms of the utility derived and the trust accuracy, convergence, and resiliency, was evaluated. The results show that SC-TRUST improves the quality of service compositions and adequately mitigates trust-related attacks, thus increasing both efficiency and security.

Computation offloading decisions play a crucial role in implementing mobile-edge computing (MEC) technology in the Internet of Things (IoT) services. Mobile network operators (MNOs) can employ computation offloading techniques to reduce task completion delay and improve the Quality of Service (QoS) for users by optimizing the system’s processing delay and energy consumption. However, different IoT applications (e.g., entertainment and autonomous driving) generate different delay tolerances and benefits for computational tasks from the MNO perspective. Therefore, simply minimizing the delay of all tasks does not satisfy the QoS of each user. The system architecture design should consider the significance of users and the heterogeneity of tasks. Unfortunately, rare work has been done to discuss this practical issue. In this article, from the perspective of MNO, we investigate the computation offloading optimization problem of multiuser delay-sensitive tasks. First, we propose a new optimization model, which designs different optimization objectives for the cost and revenue of tasks. Then, we transform the problem into a Markov decision processes problem, which leads to designing a multiagent iterative optimization framework. For the strategic optimization of each agent, we further propose a cooperative multiagent deep reinforcement learning (CMDRL) algorithm to optimize two different objectives at the same time. Two agents are integrated into the CMDRL framework to enable agents to collaborate and converge to the global optimum in a distributed manner. At the same time, the priority experience replay method is introduced to improve the utilization rate of effective samples and the learning efficiency of the algorithm. The experimental results show that our proposed method can effectively achieve a significantly higher profit than the alternative state-of-the-art method and exhibit a more favorable computational performance than benchmark deep reinforcement learning methods.

Non-orthogonal multiple access (NOMA) assisted semi-grant-free (SGF) transmission has been viewed as one of the promising technologies to meet massive connectivity requirements of the next-generation networks. A novel intelligent reconfigurable surface (IRS) assisted SGF NOMA transmission system is proposed, where the IRS is employed to satisfy the channel gain requirements for grant-based users (GBUs) and grant-free users (GFUs). The dynamic optimization on the sub-carrier assignment and power allocation for roaming GFUs, and the amplitude control and phase shift design for reflecting elements of the IRS, is formulated. Aiming at maximizing the long-term data rate of all GFUs, the optimization problem is first modeled as a multi-agent Markov decision problem. Then, three multi-agent deep reinforcement learning based frameworks are proposed to solve the problem under three different IRS cases, including the ideal IRS, non-ideal IRS with continuous phase shifts, and non-ideal IRS with discrete phase shifts. Specifically, for each GFU agent, a sub-carrier assignment deep Q-network (DQN) and a power allocation deep deterministic policy gradient (DDPG) are integrated to dynamically assign network resources for each GFU. For the only IRS agent, two DDPGs are integrated to dynamically assign phase shift and amplitude for each reflecting element of ideal IRS. The single DDPG for dynamically assigning continuous phase shifts, and parallel DQNs for dynamically assigning discrete phase shifts for non-ideal IRS with fixed amplitude are also proposed. Simulation results demonstrate that: 1) The network sum rates of all GFUs can achieve a significant improvement with the aid of IRS, comparing with the system without IRS. 2) The network sum rates of the NOMA assisted SGF transmissions are superior to that of OMA assisted GF transmissions.

Competitive opinion maximization (COM) in signed social networks targets at selecting a subset of influential individuals (i.e., seed nodes), spreading the desired opinions of the product to their neighbors against its opponents, and eventually achieving the maximum opinion propagation. Current studies mainly focus on competitive influence maximization and opinion maximization. However, COM in signed social networks has not been studied in depth. In this article, we study the COM in signed social networks and propose a novel reinforcement-learning-based opinion maximization framework (RLOM) to solve the COM problem. The proposed RLOM is composed of two phases: the activated dynamic opinion model and the reinforcement-learning-based seeding process. We theoretically prove the COM problem to be NP-hard. To model the opinion propagation process, we propose the activated dynamic opinion model based on a stateless Q-learning approach. Moreover, we propose the reinforcement-learning-based seeding scheme, which is leveraged in an unknown opponent strategy. Experiment results verify the effectiveness of our method in terms of effective opinions on three signed datasets.

This article proposes an adaptive localized decision variable analysis approach under the decomposition-based framework to solve the large-scale multiobjective and many-objective optimization problems (MaOPs). Its main idea is to incorporate the guidance of reference vectors into the control variable analysis and optimize the decision variables using an adaptive strategy. Especially, in the control variable analysis, for each search direction, the convergence relevance degree of each decision variable is measured by a projection-based detection method. In the decision variable optimization, the grouped decision variables are optimized with an adaptive scalarization strategy, which is able to adaptively balance the convergence and diversity of the solutions in the objective space. The proposed algorithm is evaluated with a suite of test problems with 2-10 objectives and 200-1000 variables. Experimental results validate the effectiveness and efficiency of the proposed algorithm on the large-scale multiobjective and MaOPs.

The performance of decomposition-based algorithms is sensitive to the Pareto front shapes since their reference vectors preset in advance are not always adaptable to various problem characteristics with no a priori knowledge. For this issue, this article proposes an adaptive reference vector reinforcement learning (RVRL) approach to decomposition-based algorithms for industrial copper burdening optimization. The proposed approach involves two main operations, that is: 1) a reinforcement learning (RL) operation and 2) a reference point sampling operation. Given the fact that the states of reference vectors interact with the landscape environment (quite often), the RL operation treats the reference vector adaption process as an RL task, where each reference vector learns from the environmental feedback and selects optimal actions for gradually fitting the problem characteristics. Accordingly, the reference point sampling operation uses estimation-of-distribution learning models to sample new reference points. Finally, the resultant algorithm is applied to handle the proposed industrial copper burdening problem. For this problem, an adaptive penalty function and a soft constraint-based relaxing approach are used to handle complex constraints. Experimental results on both benchmark problems and real-world instances verify the competitiveness and effectiveness of the proposed algorithm.

Mobile edge computing (MEC) emerges as an appealing paradigm to provide time-sensitive computing services for industrial Internet of Things (IIoT) applications. How to guarantee truthfulness and budget-balance under locality constraints is an important issue to the allocation and pricing design of the MEC system. In this paper, we propose a truthful combinatorial double auction mechanism, which integrates the padding concept and the efficient pricing strategy to guarantee desirable properties in constrained MEC environments. This mechanism takes into account the locality characteristics of the MEC systems, where mobile devices (MDs) only offload tasks to edge servers (ESs) in the proximity with various requirements, and ESs only serve their neighboring MDs with limited resources. To be specific, for allocation, a linear programming (LP)-based padding method is used to obtain the near-optimal solution in the polynomial time. For pricing, a critical-value-based pricing strategy and a VCG-based pricing strategy are designed for MDs and ESs to achieve truthfulness and budget-balance. Our theoretical analysis confirms that TCDA is able to hold a set of desirable economic properties, including truthfulness, individual rationality, and budget-balance. Furthermore, simulation results validate the theoretical analysis, and verify the effectiveness and efficiency of TCDA.

Network service provisioning becomes flexible and programmable with the help of Network Function Virfitualization (NFV), since NFV abstracts various service functions into software components called Virtual Network Function (VNF) and VNFs can be flexibly and quickly composed to form new services. It is commonly known that sharing the same VNF among different services can improve the resource utilization. However, we should be aware that such sharing also leads to serious resource preemption. In addition, VNF sharing aggravates the generation of the performance bottleneck, which then causes the rate mismatch problem between the upstream and downstream VNFs belonging to the same service chain. In this article, we propose a dynamic and flexible algorithm to jointly address the VNF sharing resource allocation and the rate coordination between the upstream and downstream VNFs. Specifically, 1) the VNFs are shared among different service chains with a fairness factor considered for the purpose of reducing the resource preemption probability and improving the resource utilization; 2) the backpressure indicator of each VNF is defined to judge its pressure condition, based on which we can dynamically adjust the processing rates between it and its downstream or upstream VNFs by maximizing the idle resource utilization. The experimental results indicate that the proposed algorithm outperforms the other methods in terms of the average delay, the flow completion time, the throughput and the backlog, etc. Meanwhile, the proposed algorithm achieves more stable performance than the other methods.

A novel intelligent reconfigurable surface (IRS) assisted coordinated multi-point (CoMP) system is proposed. Our objective is to maximize the energy efficiency (EE) of this system by jointly optimizing base station (BS) clustering, user association, sub-carrier assignment, power allocation, and optimal design of the IRS, while satisfying the users’ quality of service requirements. Considering the amplitude and phase shift characteristics, both ideal and non-ideal IRS are investigated. The formulated problem is proved to be NP-hard. By analyzing its structure, we decouple it into the power allocation sub-problem, the BS clustering, UE association, and sub-carrier assignment sub-problem, and the reflection coefficients design sub-problem. For the power allocation sub-problem, we invoke the fractional programming to find the optimal solution. For the reflection coefficients design sub-problem of ideal IRS, the optimal solution is derived with the Lagrangian dual method. Whereas quantization-based method is employed to find the discrete phase shifts for non-ideal IRS. We finally propose a genetic algorithm (GA) to represent the potential solutions of sub-carrier assignment, and combine the other two optimization algorithms as fitness estimator in GA. Numerical results validate the feasibility, fast convergence, and the flexibility of the proposed algorithm. It shows that the proposed scheme outperform the system without IRS and that with a random initialized IRS.

The rapid development and application of 5G/B5G generate tremendous amount of traffic which in turn cause great burden for the corresponding transmission network. One typical way to address such challenge is to sink the content (e.g., 4K and 8K videos) from the remote cloud to the edge servers. In this case, how to efficiently visiting and getting these contents becomes a new problem, in which the cooperation between cloud and edge should be taken into consideration. In this regard, this work builds an edge and cloud cooperative routing and caching system which consists of three main modules of content popularity prediction, cooperative caching and cooperative routing. Specifically, the content prediction is designed by jointly leveraging the technologies of Long Short-Term Memory (LSTM) and Temporal Pattern Attention (TPA) to dig the traffic features and predict the future content popularity. Based on the prediction results and the technology of reinforce learning, the cooperative caching module designs both a reactive content replacement and an active content caching strategies. After that, the cooperative routing is carried out to help customers visiting and obtaining these content efficiently with the objective of minimizing the overhead. The experimental results indicate that the proposed methods outperform the state-of-the-art benchmarks in terms of the caching hit rate, the average throughput, the successful content delivery rate and the average routing overhead.

Smartphone-based indoor localization has attracted considerable attentions in both research and industrial areas. However, the localization accuracy and robustness are still challenging problems due to low-cost noisy devices, especially in those complicated localization environments. Considering that pedestrian dead-reckoning (PDR) devices are widely equipped in recent smartphones, we propose a novel indoor localization fusing algorithm that integrates both wireless fidelity (WiFi) features and PDR features. By formulating the fusing indoor localization as a recursive function approximation problem, a sliding-window-based displacement scheme is designed to generate a time-series-based feature data set. We further apply the long short-term memory (LSTM) network for data fusion and localization on this data set by taking advantage of its benefits in time-series prediction and characterization. To evaluate the performance of the proposed algorithm, we compare it with state-of-the-art filter-based localization algorithms in three typical movements and three postures of holding smartphones. Extensive experiment results demonstrate the accuracy and robustness of the proposed algorithm in indoor localization, even in some extreme environments.

The energy consumed by data centers has been growing rapidly in recent years. Among all the major contributors to the power consumption of entire data centers, data center network (DCN) can account for up to 20% of the total power consumption. In this article, we first devise a power-efficient software-defined DCN (PESD-DCN) framework, which can achieve desirable power efficiency, avoid potential link congestion, and reduce frequent device state transition. Then, we formulate the optimization problem of maximizing the radio full-utilized devices to all devices. To solve it, we propose correlation-aware flow routing (CFR) algorithm, which leverages correlation-aware flow consolidation (CFC) technique to improve energy efficiency, avoid the potential link congestion, and reduce frequent device state transition. Moreover, to further improve the DCN energy efficiency, we propose flow rerouting, link rate adaptation, and device sleeping (FLD) algorithm. Finally, simulation results demonstrate that PESD-DCN can achieve a good performance. More specifically, in comparison to the other baseline algorithms, PESD-DCN can achieve up to 79.19% energy efficiency, 67.1% decrease in switch state transition (SST), and 55.4% decrease in link state transition (LST).

Industry 4.0 focuses on continuous interconnection services, allowing for the continuous and uninterrupted exchange of signals or information between related parties. The application of messaging protocols for transferring data to remote locations must meet specific specifications such as asynchronous communication, compact messaging, operating in conditions of unstable connection of the transmission line of data, limited network bandwidth operation, support multilevel Quality of Service (QoS), and easy integration of new devices. The Message Queue Telemetry Transport (MQTT) protocol is used in software applications that require asynchronous communication. It is a light and simplified protocol based on publish-subscribe messaging and is placed functionally over the TCP/IP protocol. It is designed to minimize the required communication bandwidth and system requirements increasing reliability and probability of successful message transmission, making it ideal for use in Machine-to-Machine (M2M) communication or networks where bandwidth is limited, delays are long, coverage is not reliable, and energy consumption should be as low as possible. Despite the fact that the advantage that MQTT offers its way of operating does not provide a serious level of security in how to achieve its interconnection, as it does not require protocol dependence on one intermediate third entity, the interface is dependent on each application. This paper presents an innovative real-time anomaly detection system to detect MQTT-based attacks in cyber-physical systems. This is an online-semisupervised learning neural system based on a small number of sampled patterns that identify crowd anomalies in the MQTT protocol related to specialized attacks to undermine cyber-physical systems.

Software-Defined-Networking (SDN) is progressively dominating the dynamic management for timely network trouble shooting and fine grained traffic scheduling in data center networks. One critical issue in SDN is to reduce the communication overhead between the switches and the controller. Such overhead is mainly caused by handling miss-match packets, because for each miss-match packet, a switch will send a request to the controller asking for forwarding rule. Existing approaches to address this problem generally need to deploy intermediate proxy or authority switches to hold rule copies, so as to reduce the number of requests sent to the controller. In this paper, we argue that using the intrinsic buffer in a SDN switch can also greatly reduce the communication overhead without using additional devices. If a switch buffers each miss-match packet, only a few header fields instead of the entire packet are required to be sent to the controller. Experiment results show that this can reduce 78.7 percent control traffic and 37 percent controller overhead at the cost of increasing only 5.6 percent switch overhead on average. If the proposed flow-granularity buffer mechanism is adopted, only one request message needs to be sent to the controller for a new flow with many arrival packets. Thus the control traffic and controller overhead can be further reduced by 64 percent and 35.7 percent respectively on average without increasing the switch overhead.

The safety driving-related content demands of vehicle users increase rapidly, especially with the development of autonomous driving. It is significantly necessary to obtain the safety-related transportation information of an area when vehicles are drove there, whether or not they are controlled by human being. However, vehicular content caching can bring issues in distributed-fashion, such as high response delay and low content response ratio because of the poor traffic condition and the obstructions of buildings. As a consequence, we adopt UAVs (Unmanned Aerial Vehicles) to assist the driving safety-related content caching for vehicles. Besides, since the power energy and the caching storage of UAVs are limited, it is needed to design an optimal caching scheme to guarantee the driving safety-related content demands of vehicle users as well as reduce the energy consumption of UAVs. In this article, we propose a novel deep Q-learning based air-assisted vehicular caching scheme to respond to the driving safety-related content requests of vehicle users. First, a three-layered content response architecture is introduced, where an airship is leveraged to take charge of the scheduling of UAVs to improve the content response. Then, a multi-objective mathematical model is built to describe the specific problem of the proposed scheme. Finally, deep Q-learning is applied to solve the multi-objective problem by learning from the history content requests of vehicle users. Extensive experiments have been conducted which show the proposed scheme outperforms its counterparts in terms of content hit ratio, response delay, being scheduling probability and packet buffering time.

The rapid development of blockchain technology and its various applications have attracted huge attention in the last five years. The consensus mechanism and incentive mechanism are the backbone of a blockchain network. The consensus mechanism plays a crucial role in sustaining the network security, integrity, and efficiency. The incentive mechanism motivates the distributed nodes to “mine” so as to participate the consensus mechanism. The existing mechanisms bear the fairness and justice issues. In this paper, from the perspective of mechanism design, we propose a consensus-incentive mechanism through applying continuous double auction theory, which is abbreviated as ABC mechanism. Our mechanism consists of four stages, including initiation stage, auction stage, completion stage, and confirmation stage. The auction model in use is the continuous double auction to ensure the transactions are stored in a real-time manner. Through extensive experimental evaluations, our mechanism is proven to improve the fairness and justice of the blockchain network.

With the explosive growth of mobile devices at the edge of mobile social networks (MSNs), the amount of the content that needs to be transmitted is exploded. Traditional content delivery mechanisms leverage only local information to make routing decisions, which results in both high latency and low delivery rate. Software-defined networking (SDN) is a novel network paradigm, the design philosophy of which could be applied to MSN for improving the content delivery performance. In this article, the centralized control thought of SDN is introduced into MSN to efficiently process social information. The classical routing algorithm of BubbleRap is improved from the perspective of network density, which is the basis of designing the sparse and dense routing mechanisms for MSNs. In addition, flexibly switching between these two routing mechanisms is implemented by a discriminating scheme, achieving efficient yet adaptive routing. The experimental results show that the delivery ratio of sparse routing is up to 83%, and the dense routing could reach up to 93%.

Wireless communication demands increase sharply with the explosive growth of mobile devices. The communications mainly depend on the infrastructure-based networks, e.g., WLANs and cellular networks. However, such wireless connections may be unavailable in crowded areas (e.g., concert and conference hall) or interrupted by infrastructure failures caused by earthquake or tsunami. These promote the evolution of local communication systems over device-to-device communication, such as Bluetooth and Wi-Fi Direct (WFD). However, none of the existing studies construct a full-featured local communication system, and they do not consider how to support the user mobility either. In this article, we implement and evaluate the performance of a WFD-based local communication system. First, we improve the intragroup communication by the native implementation of WFD on the Android platform, and propose an application-layer forwarding solution for the intergroup communication, which can be applied to three or more connected groups. Then, we put forward a self-adaptive handover mechanism taking user mobility and node failures into account. To deal with the uncertainty in the handover decision procedure, a fuzzy-logic-based normalized quantitative decision algorithm (FNQD) with the weights derived from the fuzzy analytic hierarchy process (FAHP) is utilized. Finally, we evaluate the performance of the system through both simulation and experiment analysis. Results show that we can get a maximum throughput of 31.7 Mb/s for the intragroup communication and a maximum goodput of 4.76 Mb/s for the intergroup communication. What is more, mobile devices could perform various types of handover according to their roles and status, which could improve the robustness of the local communication system.

With the drastic increase in the number and capabilities of mobile devices, user demands for bandwidth-hungry applications such as video streaming and multimedia file sharing are pushing the limits of current cellular systems. Offloading resources sharing services from the cellular infrastructure to device-to-device (D2D) networks can improve spectrum efficiency and expand system capacity. This paper designs and implements new D2D networks formation mechanism based on commercial off-the-shelf smartphones, and data transmission mechanism among these devices using Wi-Fi Direct(WFD). In this system, we realize the single-hop and multi-hop real-time video streaming transmissions, and compare the performance between the multicast and the unicast transmission mechanisms. To demonstrate and evaluate the system, we implement a prototype system using various brand handsets with Andorid OS and conduct experiments. The results show that the system can transmit real-time videos at the bitrate of 2200kbps and frame rate (fps) of 25 for 4-5 receivers simultaneously within 75 meters in case of single-hop transmission. As for performing multi-hop transmission, the real-time videos can be transmitted to 1 recevier at the bitrate of 5000kbps and fps of 25.

Device-to-Device (D2D) communication is considered as an efficient approach to improve current wireless networks. However, for inter-group communication, since devices are using the same IP address segment, IP conflicts can be induced into D2D networks. To address this problem, we propose an Information-Centric resource sharing system for D2D communication network. In the system, both P2P and WLAN network adaptor interfaces are adopted simultaneously for establishing a D2D network. A socket reusing mechanism is proposed for inter-group communication. In addition, to improve content distribution and retrieval, we base our system on Information-Centric networking (ICN) paradigm. We design ICN flow tables on network nodes to realize request forwarding and data backtracking. To demonstrate the feasibility and effectiveness, we have implemented the system on commercial Android devices using Wi-Fi Direct technology. Experimental results show that the proposed Information-Centric resource sharing system is feasible for D2D network in terms of data transmission and energy consumption. The average latency for transmitting 1MB data is 257.23ms; i.e., the throughput is up to 31.1 Mbps. The average energy consumption for transmitting 100MB data is only about 0.66 mAh.

Traffic optimization in Information Centric Networks (ICN) inevitably involves making full use of in-network storages. Although ICN provides efficient content delivery and traffic offloading by exploiting nearby cached content and reducing redundant transmissions for same content requests, it inherently follows an opportunistic fashion to utilize in-network caches. In this paper, we improve network traffic distribution for ICN by leveraging multicasting. Specifically, introducing Software-Defined Networking (SDN), a virtual point-based algorithm is proposed to achieve many-to-many multicasting with global optimization. Furthermore, we propose a tunnel-based bandwidth allocation mechanism to improve the scalability of the system. Extensive simulation results show that multicast tree can be constructed optimally and meanwhile our proposal significantly improves network performance in terms of load balancing and network utilizazation.

Emerging technological innovations introduce the possibility to reconfigure the data center topology at runtime. The development of reconfigurable architectures can adapt their topology to account for changing demands, e.g., using Flyways-like augmented links. However, there is no common notion established in this area of how to evaluate the topology, the underlying theoretical analysis is not yet well studied. In this paper, we present the upper bound of the network diameter using spectral graph theory, to the best of our knowledge, which is a first theoretical attempt on understanding the nature of the reconfigurable data center networks. We further prove the algebraic connectivity is insensitive to the weight changes. Finally, based the algebra-connectivity λ2, we give a comprehensive link-augmentation validity, which can be implemented in current DCNs potentially.

Moment invariants of images are important features for pattern recognition and image processing. Many methods have been proposed to derive moment invariants of images under different group actions. However, the completeness and independence of a set of moment invariants are two open problems. In this paper, we use the moving frame method to derive affine moment invariants of color images. The moving frame for the normalized color moment space under the action of the affine group is presented. Using this moving frame, we obtain a complete and independent set of affine moment invariants of color images. This system of affine moment invariants is also invariant under diagonal photometric changes. Experimental results are provided to validate the correctness of the derivation.

Security through trust presents a viable solution for threat management in the Internet of Things (IoT). Currently, a well-defined trust management framework for collaborative applications on the IoT platform does not exist. In order to estimate reliably the trust values of nodes within a system, the trust should be measured by suitable parameters that are based on the nodes’ functional properties in the application context. Existing models do not clearly outline the parametrization of trust. Also, trust decay is inadequately modeled in most current models. In addition, trust recommendations are usually inaccurately weighted with respect to previous trust, thereby increasing the effect of bad recommendations. A new model, CTRUST, is proposed to resolve these shortcomings. In CTRUST, trust is accurately parametrized while recommendations are evaluated through belief functions. The effects of trust decay and maturity on the trust evaluation process were studied. Each trust component is neatly modeled by appropriate mathematical functions. CTRUST was implemented in a collaborative download application and its performance was evaluated based on the utility derived and its trust accuracy, convergence, and resiliency. The results indicate that IoT collaborative applications based on CTRUST gain a significant improvement in performance, in terms of efficiency and security.

To provide satisfactory Quality of Service (QoS) on the move, efficient mobility management is indispensable to provide mobile users with seamless and ubiquitous wireless connectivity. However, both the conventional centralized mobility architecture and the upcoming distributed mobility management face fundamental challenges such as sub-optimal routing, scalability, and so on. The emerging software-defined networking (SDN) architecture can efficiently manage network operations, and accordingly provides a new direction to address the challenges in mobility management. In this paper, we propose an SDN-based Mobility Management (SDN-MM) scheme to support seamless Intro/Inter domain handover with route optimization. SDN-MM decouples mobility management and packet forwarding functions by installing route optimizing and mobility control logics in an SDN controller, but exempting it from traffic redirecting. In SDN-MM, a comprehensive set of signaling operations are designed in order to provide transparent and efficient mobility support for ongoing sessions in each handover scenario, which prevents packet loss and tunneling overhead, and accordingly provide improved QoS to mobile users. For data communications, an SDN controller in SDN-MM pre-calculates the optimal end-to-end route before a handover, and decides whether to migrate traffic to the route by balancing the performance gain and the signaling overhead, which greatly improves bandwidth resource utilization. Finally, we develop a novel analytical model to evaluate the performance of SDN-MM, including signaling overhead, handover latency, and packet delivery cost. The simulation results have been provided to demonstrate that the proposed SDN-MM can greatly improve handover performance and maintain high resource utilization efficiency as well.

The ubiquity of ever-more-capable mobile devices, especially smartphones, brings forth participatory sensing to collect and interpret information. It can achieve unprecedented quantity of data. However, it is arduous to guarantee quality of data because everyone can contribute data without scrutinization. It is an important issue in quality-oriented participatory sensing. Our idea to address this issue is motivating participants to contribute accurate data for improving data quality directly. In this article, we propose a reputation-based incentive mechanism, RIM, to realize the idea. More specifically, we identify the participants who collect the accurate data and regard them as the reputable ones. Then, the reputable participants are granted a higher chance to obtain rewards so that other people will try to follow such users and become reputable as well. Namely, RIM can encourage and steer users to collect accurate data in the long term. We analyze our incentive mechanism by formalization and premise implications. For a feasibility study of participatory sensing and verification of the implications, we implement and deploy a participatory sensing application focusing on monitoring environmental noise in a specific location as a case study and conduct a simulation based on the case study to further evaluate the proposed incentive mechanism. The results from the case study and the simulation present that RIM can remarkably increase the quality of collected data in participatory sensing while corroborating our theoretical implications.

Radio frequency identification (RFID) networks planning (RNP) is a challenging task on how to deploy RFID readers under certain constraints. Existing RNP models are usually derived from the flat and centralized-processing framework identified by vertical integration within a set of objectives which couple different types of control variables. This paper proposes a two-level RNP model based on the hierarchical decoupling principle to reduce computational complexity, in which the costefficient planning at the top levels is modeled with a set of discrete control variables (i.e., switch states of readers), and the quality of service objectives at the bottom level are modeled with a set of continuous control variables (i.e., physical coordinate and radiate power). The model of the objectives at the two levels is essentially a multiobjective problem. In order to optimize this model, this paper proposes a specific multiobjective artificial bee colony optimizer called H-MOABC, which is based on performance indicators with reinforcement learning and orthogonal Latin squares approach. The proposed algorithm proves to be competitive in dealing with two-objective and three-objective optimization problems in comparison with state-of-the-art algorithms. In the experiments, H-MOABC is employed to solve the two scalable real-world RNP instances in the hierarchical decoupling manner. Computational results shows that the proposed H-MOABC is very effective and efficient in RFID networks optimization.

Jointing visual-semantic embeddings (VSE) have become a research hotpot for the task of image annotation, which suffers from the issue of semantic gap, i.e., the gap between images' visual features (low-level) and labels' semantic features (high-level). This issue will be even more challenging if visual features cannot be retrieved from images, that is, when images are only denoted by numerical IDs as given in some real datasets. The typical way of existing VSE methods is to perform a uniform sampling method for negative examples that violate the ranking order against positive examples, which requires a time-consuming search in the whole label space. In this paper, we propose a fast adaptive negative sampler that can work well in the settings of no figure pixels available. Our sampling strategy is to choose the negative examples that are most likely to meet the requirements of violation according to the latent factors of images. In this way, our approach can linearly scale up to large datasets. The experiments demonstrate that our approach converges 5.02x faster than the state-of-the-art approaches on OpenImages, 2.5x on IAPR-TCI2 and 2.06x on NUS-WIDE datasets, as well as better ranking accuracy across datasets.

The quality of experience (QoE) of users has become an important factor concerned by service providers to keep users for their services. The measurable quality of service (QoS) refers to technical performance, not the satisfaction of users, but it is closely related to user's QoE. Most existing researches mainly focus on studying the QoE-QoS relationship for services in terrestrial network with rare attention on satellite networks. Multimedia services delivery over satellite networks is a promising service in emerging future Internet, thus it is appealing to analyze how service QoE depend on QoS in satellite networks. In this paper, we build a simulated satellite network based on OPENT software to measure the QoS parameters and obtain the distorted videos/voices. Then, based on the original and distorted video/voice sequences, we perform a subjective test to obtain the subjective opinion scores representing user's QoE. Finally, on the basis of the collected data, the influence of single QoS parameter on QoE is analyzed and the QoS parameter thresholds to different QoE levels are provided.

CBOW (Continuous Bag-Of-Words) is one of the most commonly used techniques to generate word embeddings in various NLP tasks. However, it fails to reach the optimal performance due to uniform involvements of positive words and a simple sampling distribution of negative words. To resolve these issues, we propose OptRank to optimize word ranking and approximate negative sampling for bettering word embedding. Specifically, we first formalize word embedding as a ranking problem. Then, we weigh the positive words by their ranks such that highly ranked words have more importance, and adopt a dynamic sampling strategy to select informative negative words. In addition, an approximation method is designed to efficiently compute word ranks. Empirical experiments show that OptRank consistently outperforms its counterparts on a benchmark dataset with different sampling scales, especially when the sampled subset is small. The code and datasets can be obtained from https://github.com/ouououououou/OptRank.

The infrastructure-based networks will be unavailable in the case of infrastructure failures (e.g., earthquake and tsunami) or in crowded areas (e.g., concert and conference hall). This promotes the evolution of location communication systems, which also benefit offloading computing, mobile edge computing and mobile crowdsourcing. In this paper, we utilize Wi-Fi Direct (WFD) to develop a local communication system. We not only present an intra-group communication solution by the native characteristics of WFD, but also propose a general solution for the bidirectional inter-group communication, which is beyond the scope of WFD specifications. Experimental results show that the maximum throughput of intra-group communication could reach up to 31.7 Mbps, and the maximum throughput of inter-group communication is 8.26 Mbps.

With the severe spectrum shortage in conventional cellular networks, millimeter-wave (mmWave) technology has been proposed and is expected to be used in small cells to meet the increase spectrum efficiency, especially for those small cells are deployed with high density and forming as a multi-hop backhaul structure different from conventional schemes with priori given routes, a cross-layer optimization model aiming at maximizing the energy efficiency while taking into account both route selection and resource allocation. This paper decomposes the original problem into two sub-problems: resource allocation sub-problem at the link-physical layer and route selection sub-problem at the network layer. To reflect the interplay property between these two problems, we further propose two joint optimization strategies. The first is using genetic algorithm (GA) for resource allocation with given routes and using linear programming for routes selection with given resource allocation. The second is using the linear programming to evaluate the fitness of each individual in the GA. The simulation results show our solution can efficiently scale with the energy and spectrum efficiency and properly route the traffic without overloading any one of the base stations.

with the strong impact of OTT (Over The Top) business in the mobile Internet era, operators urgently need to discover the user value information from massive data to help them provide personalized accurate services and expand business customer services. The construction of social user groups based on mobile communication data can help operators to accurately analyze customer social structures, thus promoting quality service and improving marketing quality. In this paper, we design a set of social group construction algorithm based on user behavior characteristics excavated from massive user data in mobile communication network. Due to the huge volume of mobile communication data sets, a parallel design based on MapReduce is exploited. The experimental results show that the ADBLINKw algorithm performs well on the efficiency and community detection quality.

With the integration of industrialization and informatization, there are a large number of embedded industrial wireless physical devices that are susceptible to malicious intrusion by malicious nodes within the industrial wireless networks, which leads to perform unique internal attacks, such as DOS attacks, selfish attacks, Sinkhole attacks and black hole attacks. In order to detect and resist external attacks, the SDN technology is introduced into the existing industrial wireless network and the network equipment is managed by a centralized controller. At the same time, the controller shields the differences between network devices in the network bottom layer. Open control enables network users to define their own network routing and forwarding strategies, making the network more flexible and intelligent. Moreover, SDN controller can make global routing strategy according to the global state of the network and the underutilized resources in the network we introduce SDN paradigm into the industrial wireless network and propose a secure routing mechanism for industrial wireless networks based on SDN. The internal malicious nodes are detected by computing the node comprehensive trust value of node. We design a biologic heuristic algorithm based on the foraging principle of physarum as a network security routing algorithm to calculate the network security transmission path. The performance of packet delivery rate and average end to end delay are compared and analyzed. The result shows that the routing mechanism depicted in the thesis has better performance, and the security of data transmission is greatly improved.

MSN (Mobile Social Network) enables nodes (mobile devices) to realize packet delivery by leveraging social relationships of mobile users. However, MSN has to adapt with the daily increasing content (e.g., video) requirement requested by mobile users. Based on the fact that ICN (Information-Centric Networking) supports mobility naturally, we propose an MSN content-centric routing scheme. It is inspired by the in-network caching and named-content properties of Named Data Networking. By the analysis of the historical requested content names of MSN users, a novel interest distance metric is proposed, based on which, the forwarding node is selected for an interest packet. Meanwhile, by the analysis of the historical encounters of MSN nodes, a novel encounter regularity metric is proposed, based on which, the forwarding node is selected for a data packet. Furthermore, in order to respond the forth-coming requests rapidly, nodes preferentially cache the content which with high requested probability. By comparing with the existed schemes, simulation experiments represent that our scheme is effective and feasible.

Software Defined Networking (SDN) decouples the control plane from the data plane, which increases network scalability and flexibility. But malicious applications on SDN controller can cause the entire network to crash. So, we design a reputation and incentive mechanism on SDN to reduce application's malicious access. In the proposed module, first of all, the application behavior is analyzed and the malicious accesses are identified, which are used to build the reputation and incentive mechanism. Second, the analysis results of the application behavior are combined through beta probability density to obtain the reputation rating. The reward or punishment will be given based on the behavior and reputation of the application under the selected social strategy. Simulation results show that the system can accurately identify malicious behavior and reduce malicious requests, with an acceptable runtime overhead about 300 microseconds.

Software-defined satellite networks (SDSN) move the control logic off the forwarding devices and into the external controller, which achieves network flexibility, programmability, evolvability, and visibility. The controller placement in SDSN, as the foundation of SDSN implementation, has been seldom studied. In this paper, we proposed a three-layer hierarchical controller architecture for software-defined geostationary earth orbit/low earth orbit (GEO/LEO) satellite networks. Specifically, network operations control center (NOCC) is deployed as super controller, GEO satellites are domain controllers, and a part of LEO satellites are slave controllers. Based on this architecture, we further proposed a slave controller selection strategy to facilitate cost reduction and stability enhancement. The feasibility of this controller architecture is validated in terms of the maximum switch to controller and controller to controller delays. Besides, the influences of service request distribution on these two control delays are analyzed.

Network function virtualisation and software‐defined networking are two emerging technologies which together enable a new service provisioning paradigm called service function chain (SFC). However, the SFC should be scalable enough to accommodate one or more service functions joining or leaving it during its lifecycle. In this work, the authors first formulate this problem as an integer linear programming (ILP) model, and then address this ILP model to obtain the optimal solution. Due to the extremely high execution time for solving the ILP model, they next proposed a dynamic heuristic to solve this problem. In particular, the proposed heuristic leverages two kinds of strategies to handle the scale‐in (i.e. removing existing service functions from SFCs) and scale‐out (i.e. adding new service functions to SFCs) requests. Since each SFC has a corresponding service function path (SFP) constructed, the first kind of strategy proposes to serve the arriving scale‐in and scale‐out requests based on the in‐use SFP, while the second kind of strategy intends to optimise the in‐use SFP to achieve low cost and packet loss probability proactively. Finally, the experimental results indicate that the proposed heuristic can achieve better performance than the existing algorithm.

The Virtual Network Function (VNF) embedding problem is important for service provision in the context of Network Function Virtualisation (NFV). However, this problem is proved to be NP‐hard and challenging, and requires to be explored further. In this study, the authors first formulate it as an Integer Linear Programming (ILP) model for optimal solutions. Then, to compensate for the high running time of solving the ILP model, they propose a heuristic approach which fulfils the embedding process by jointly taking the global network connectivity and the local substrate node capacity into consideration. The simulation on real‐world network topologies demonstrates that the proposed approach can provide solutions within 1.7 times of the optimal solution offered by ILP. In addition, the experiments also suggest that the proposed approach can provide up to 2.75 times reduction in the overall cost than the other benchmarks.

In the software defined network (SDN) environment, network function virtualisation enables the virtual machine migration. Owing to the fact that transferring large amount of data will impede competing workflows, virtual network function (VNF) migration has brought a new perspective. Many optimised algorithms focusing on limiting migration time and migration cost have been proposed. In this study, the authors address the problem from a different perspective. They view the network topology from a global perspective and focus on the network effect of the whole network caused by VNF migration in the context of SDN. They introduce a parameter delay to formulate the network effect and an effect model is proposed to evaluate the migration effect of the network. In addition, a heuristic algorithm is proposed to minimise network effect while balancing network load and improving the service considering the migration cost and resources limit at the same time. The practicability and efficiency of the proposed model and algorithm are validated by simulation evaluation. By comparing their proposed algorithm with traditional benchmarks and closely related benchmarks, the experimental results show that their proposed algorithm largely reduces the network effect, while at the same time limiting the run time.

Dynamic Host Configuration Protocol (DHCP) provides dynamic use of IP addresses, but it presents challenges to meet smart terminals with great mobility and transient network access patterns. Existing studies have tried to solve this problem through adjusting DHCP lease, which controls how long a host owns an address. However, few studies clearly express the relations among the lease, address utilization and DHCP overhead. In this paper, we uncover how the leases affect address utilization and DHCP overhead with two methods, based on which, we can set the leases for the smart terminals flexibly and judiciously. First of all, we present an emulation technique to evaluate address utilization and DHCP overhead under different leases. It provides an experimental basis for setting the lease for the whole WLAN. Evaluation results show that if the lease is set to 120 min instead of 60 min by default, it can reduce 41.78% DHCP overhead on average and still reserve at least 9.2% address space for the possibly emerging terminals. Then, we model the relationship between the lease and address utilization, as well as the relationship between the lease and DHCP overhead. According to these models, we propose a load-aware DHCP lease time optimization algorithm, which helps to set different leases for each area of the WLAN based on theoretical analysis. Evaluation results show that compared with the default lease for the whole WLAN, a lease combination of {15, 120, 120} for different areas can reduce 36.85% DHCP overhead on average and guarantee there is always 10% available address space.

The name-centric mechanism relies on reverse paths to achieve content delivery. In vehicular environments, reverse paths may frequently break due to high mobility, and consequently, content communications might frequently fail. Taking into account the advantages of the name-centric mechanism and address-centric mechanism, we are motivated to integrate the address-centric mechanism with the name-centric mechanism and aim to overcome the above limitations. Based on this motivation, we propose a location-related content communication method with mobility support in vehicular environments. In this proposal, a user can acquire contents from the nearest provider in a unicast way, so the content communication failures caused by reverse-path disruption are avoided and the communication cost and latency are lowered. Moreover, this method achieves the provider and receiver handovers to guarantee that the content communications can be successfully performed. Finally, this proposal is analyzed and evaluated and the data show that it effectively improves the content communication success rate and reduces the cost and latency.

Due to the centralized control, network-wide monitoring and flow-level scheduling of Software-Defined-Networking (SDN), it can be utilized to achieve Quality of Service (QoS) for cloud applications and services, such as voice over IP, video conference and online games, etc. However, most existing approaches stay at the QoS framework design and test level, while few works focus on studying the basic QoS techniques supported by SDN. In this paper, we enable SDN with QoS guaranteed abilities, which could provide end-to-end QoS routing for each cloud user service. First of all, we implement an application identification technique on SDN controller to determine required QoS levels for each application type. Then, we implement a queue scheduling technique on SDN switch. It queues the application flows into different queues and schedules the flows out of the queues with different priorities. At last, we evaluate the effectiveness of the proposed SDN-based QoS technique through an experimental analysis. Results show that when the output interface has sufficiently available bandwidth, the delay can be reduced by 28% on average. In addition, for the application flow with the highest priority, our methods can reduce 99.99% delay and increase 90.17% throughput on average when the output interface utilization approaches to the maximum bandwidth limitation.

In-network caching is an important feature of ICN (Information Centric Networking). There are prior arts focusing on designing a highly efficient caching strategy by exploiting either node data or content data respectively. However, simply exploiting these data itself is not enough to reduce the cost of network operation and increase the quality of user experience, as there is no consideration on supplementary action of these data. In this paper, a multidimensional data learning based strategy (MDDL) is proposed to cache the selected content in a few suitable nodes. To understand the current state of node and content, a multidimensional state attribution data model including network, node and content data is proposed. Based on the model, the mapping relationship between the attribution data and the matching relationship value is mined. Utilizing this mapping function, a matching algorithm to predict the matching relationship between the node and the content in the next time period is proposed. In order to improve the accuracy of the matching algorithm, a safe semi supervised support vector machine (S4VM) algorithm is introduced. Simulation results show that MDDL outperforms CEE, BETW and LCD when looking at cost reduction of network operation and enhancement in quality of user experience.

Traditional cellular networks are moving towards heterogenous cellular networks (HetNets) to satisfy the stringent demand for data rates and capacity. To enable the new applications in 5G, such as haptic communications, we face new challenges of achieving high availability with low latency in HetNets. In this paper, we introduce coordinated multi-point (CoMP) and carrier aggregation (CA) techniques in HetNets to guarantee the availability of all UEs, where CoMP improves the single-path availability, and CA enhances availability via multi carrier gain combining. To characterize the availability, we first derive an exact closed-form expression for the availability of a random UE in a CoMP&CA-enabled HetNets. To achieve the maximum UE availability, we formulate a max-min optimization problem. To solve it, we then propose a joint two-step optimization algorithm (JTOA), and our results showcase the effective of our proposed JTOA, and the effective of CoMP in availability improvement in HetNets.

One critical issue in SDN is to reduce the communication overhead between the switches and the controller. Such overhead is mainly caused by handling miss-match packets, because for each miss-match packet, a switch will send a request to the controller asking for forwarding rule. Existing approaches to address this problem generally need to deploy intermediate proxy or authority switches to hold rule copies, so as to reduce the number of requests sent to the controller. In this paper, we argue that using the intrinsic buffer in a SDN switch can also greatly reduce the communication overhead without using additional devices. If a switch buffers each miss-match packet, only a few header fields instead of the entire packet are required to be sent to the controller. Experiment results show that this can reduce 78.7% control traffic and 37% controller overhead at the cost of increasing only 5.6% switch overhead on average. If the proposed flow-granularity buffer mechanism is adopted, only one request message needs to be sent to the controller for a new flow with many arrival packets. Thus the control traffic and controller overhead can be further reduced by 64% and 35.7% respectively on average without increasing the switch overhead.

In this paper, we propose a decentralized identity management system based on Blockchain. The function of the system mainly includes identity authentication and reputation management. The technical advantages of the Blockchain makes the data in the system safe and credible. In addition, we use smart contracts to write system rules to ensure the reliability of user information. We bind the user's entity information with the public key address and determine the true identity of a virtual user on the Blockchain. We use the token to represent the reputation which is shown to be an effective reputation model, making the participants in the system prefer to maintain and manage their personal reputation. Our system makes it possible for users to securely manage their identity and reputation on the Internet.

In electronic commerce, buyers and sellers conduct transactions without physical interactions. In reputation systems, the trustworthiness of sellers is achieved by aggregating the ratings shared by other buyers with whom the sellers have ever conducted transactions. However, the ratings provided by buyers for evaluating the same seller could be diverse due to their different judgment criteria, which is referred as the subjectivity problem of reputation systems. It indicates that the ratings shared by some buyers may mislead other buyers with different personalities, making it challenging to aggregate the ratings properly in reputation systems. In this paper, in order to cope with the subjectivity problem, a hybrid architecture of reputation systems is proposed, which is based on coalition formation game theory. In the proposed module, buyers with the same subjectivity will automatically form a club, and share their ratings so as to build seller reputation within their club. The utility of a club is the profit created by the reputation system, which is further divided among the buyers of the club. Two utility allocation algorithms have been investigated, i.e., the proportional and Shapley allocations, respectively. Theoretical analysis and experimental results have shown that buyers with the same personality have the incentive to form a separate pure club if specific conditions are satisfied.

This paper proposes a novel optimization scheme by hybridizing an artificial bee colony optimizer (HABC) with a bee life-cycle mechanism, for both stationary and dynamic optimization problems. The main innovation of the proposed HABC is to develop a cooperative and population-varying scheme, in which individuals can dynamically shift their states of birth, foraging, death, and reproduction throughout the artificial bee colony life cycle. That is, the bee colony size can be adjusted dynamically according to the local fitness landscape during algorithm execution. This new characteristic of HABC helps to avoid redundant search and maintain diversity of population in complex environments. A comprehensive experimental analysis is implemented that the proposed algorithm is benchmarked against several state-of-the-art bio-inspired algorithms on both stationary and dynamic benchmarks. Then the proposed HABC is applied to the real-world applications including data clustering and image segmentation problems. Statistical analysis of all these tests highlights the significant performance improvement due to the life-cycle mechanism and shows that the proposed HABC outperforms the reference algorithms.

Recently, code reuse attacks (CRAs) have emerged as a new class of ingenious security threatens. Attackers can utilize CRAs to hijack the control flow of programs to perform malicious actions without injecting any codes. Existing defenses against CRAs often incur high memory and performance overheads or require extending the existing processors' instruction set architectures (ISAs). To tackle these issues, we propose a hardware-based control flow integrity (CFI) that employs physical unclonable functions (PUF)-based linear encryption architecture (LEA) to protect against CRAs with negligible hardware extending and run time overheads. The proposed method can protect ret and indirect jmp instructions from return oriented programming (ROP) and jump oriented programming (JOP) without any additional software manipulations and extending ISAs. The pre-process will be conducted on codes once the executable binary is loaded into memory, and the real-time control flow verification based on LEA can be done while ret and jmp instructions are executed. Performance evaluations on benchmarks show that the proposed method only introduces 0.61% run-time overhead and 0.63% memory overhead on average.

In this paper, we design an LAP based routing algorithm in General Networks (GN) to solve the problem of information consistency of the full network under OSPF with the following operations: (i) decomposing GN into one or more Single-link Networks (SNs) with the approach of depth-first walk, (ii) re-composting the SNs to a network with regular topology structure by adding links, (iii) searching the undirected complete graph of three nodes round by round until it converges to a simple network topology based on region binding, and (iv) processing different converged network topologies with different LAP based routing algorithms. The proposed algorithm is compared with Dijkstra algorithm over some random network topologies. Simulation results show that the proposed algorithm can solve the problem of information consistency of the full network under OSPF and has better performance than Dijkstra algorithm.