Ensuring global connectivity, wide coverage, and access to high-performance computation and communication resources remains a significant challenge for vehicular users, especially in remote regions and congested urban environments. While terrestrial networks offer high data rates, reliable connectivity, and low latency, they are limited in terms of global coverage, urban–rural communication, and high computational capacities. Space-Air-Ground Integrated Networks have emerged as a promising solution to overcome these limitations; however, efficiently managing the unpredictable task demands of vehicular users and identifying suitable computational resources for offloading in such heterogeneous networks remains a complex challenge. To address these issues, we propose a user-scheduling-based partial task offloading scheme supported by a heterogeneous task prioritization mechanism, where delay-sensitive tasks are offloaded to autonomous aerial vehicles (AAVs) and computation-intensive tasks are directed to low earth orbit (LEO) satellites. Our goal is to minimize the average weighted sum of task execution time and energy consumption by optimizing the use of computational and communication resources from both AAVs and LEO satellites. To this end, we introduce a twin delayed deep deterministic policy gradient-based partial task Offloading (TD3-PTO) algorithm that enables real-time high-efficiency offloading decisions while managing trajectory planning. Simulation results validate the effectiveness of the proposed TD3-PTO scheme in jointly optimizing task offloading strategies, handling dynamic satellite transitions, and improving user association policies, thereby significantly reducing both latency and energy consumption.

In the platform of Low Earth Orbit (LEO) satellite constellations combined with terrestrial edge servers, both satellite and edge nodes have limited storage and computing resources, and each content item may be available in multiple bitrate–quality representations, making cache allocation a challenging task. In this paper, we study QoE-aware multi-representation content caching in a heterogeneous LEO satellite– edge architecture, where a serving LEO satellite cooperates with multiple edge servers to deliver encoded video representations to users. We model the expected user Quality of Experience (QoE) by combining a content popularity distribution with a representation-dependent quality metric and a hierarchical delivery path that prioritizes nearby edge caches, followed by serving satellites and inter-satellite links or gateways. We formulate cache placement decisions at satellite and edge nodes as a joint optimization problem under storage and computing constraints and show that the resulting utility function is monotone and submodular. This allows us to cast the problem as a submodular knapsack maximization and to design a greedy caching strategy that iteratively selects cache actions with the largest marginal QoE gain per unit resource cost. Simulation results show that the proposed approach yields higher QoE than conventional ones.

Visible-infrared person re-identification (VI-ReID) aims to match pedestrian images captured under visible and infrared modalities, which suffer from significant domain discrepancies. Existing approaches either synthesize cross-modal images or learn modality-invariant representations, yet often encounter semantic degradation or limited alignment capacity. Recent vision-language models leverage textual semantics for modality bridging; however, CLIP-based frameworks typically rely on learnable token proxies with limited expressiveness. In this article, we propose a novel semantic-driven framework that explicitly generates rich, modality-agnostic textual descriptions from images as alignment cues. Specifically, we design a dual-branch Text Semantic Generation Module that includes: (1) an Attribute-Aware text description Generation module using prompt-based templates to capture modality-invariant identity cues, and (2) a Contour-Aware text prompt Module that provides complementary structural information often missing in textual descriptions. To reconcile semantic heterogeneity, a Text Re-definition Module (TRM) fuses instance-level and class-level semantics into unified representations, enabling fine-grained alignment with image features. Furthermore, we construct category-level textual prototypes as global semantic anchors to enhance cross-modal consistency. Extensive experiments on two standard VI-ReID benchmarks demonstrate that our method achieves superior performance, validating its effectiveness in semantic-guided modality alignment.

Silent packet drops and packet corruptions, which are caused by faulty network elements and hurt performances of cloud applications, are common in data centers but hard to detect and localize. Existing solutions based on active probes introduce additional probe traffic and are constrained by probe rate, while solutions based on passive traffic monitoring measure the entire network traffic, and are generally unable to pinpoint the locations where packet losses happen. In this paper, we presentCanary, a system for detecting and localizing network faults with partial traffic monitoring. Canary employs a lightweight and adaptive mechanism to detect packet losses by monitoring a small set of large-sized network flows, and it ensures that on each network path, a sufficient number of packets are monitored by upstream and downstream switches. In addition, Canary encodes information of the path that a packet travels along within its header, and by leveraging path information of the lost packets, Canary is capable to localize network faults with high accuracy. We theoretically prove the effectiveness of our proposed method, and prototype Canary with P4 on commodity hardware programmable switches. Results from extensive experiments driven by real-world traffic show that Canary is lightweight regarding measurement overhead, robust under traffic dynamics, and is accurate in detecting and localizing faulty network links. In particular, comparing with the state-of-the-art solutions, Canary reduces the memory overhead by over 97% under$10^{-2}$link loss rate, and increases the F1-score in localizing the faulty links by over 20% on a$k=8$fat-tree data center network.

Chain-of-thought (CoT) prompting has achieved remarkable success in natural language processing (NLP). However, its vast potential remains largely unexplored for graphs. This raises an interesting question: How can we design CoT prompting for graphs to guide graph models to learn step by step? On one hand, unlike natural languages, graphs are non-linear and characterized by complex topological structures. On the other hand, many graphs lack textual data, making it difficult to formulate language-based CoT prompting. %Therefore we cannot directly adopt the CoT prompting methods used in the language domain. In this work, we propose the first CoT prompt learning framework for text-free graphs, GCoT. Specifically, we decompose the adaptation process for each downstream task into a series of inference steps, with each step consisting of prompt-based inference, ''thought'' generation, and thought-conditioned prompt learning. While the steps mimic CoT prompting in NLP, the exact mechanism differs significantly. Specifically, at each step, an input graph, along with a prompt, is first fed into a pre-trained graph encoder for prompt-based inference. We then aggregate the hidden layers of the encoder to construct a ''thought'', which captures the working state of each node in the current step. Conditioned on this thought, we learn a prompt specific to each node based on the current state. These prompts are fed into the next inference step, repeating the cycle. To evaluate and analyze the effectiveness of GCoT, we conduct comprehensive experiments on eight public datasets, which demonstrate the advantage of our approach.

Abstract A sketch-based method is promising for traffic monitoring in data center networks. Existing data plane programming model (e.g. P4) assumes target switch as one single pipeline, while state-of-the-art programmable switches actually contain multiple independent pipelines. The status quo approach for deploying a sketch-based measurement application on a multi-pipeline switch is to deploy a sketch instance in each pipeline individually. However, under multi-path routing, such a naive approach leads to poor accuracy. To overcome this problem, in this paper, we present Confluence, a sketch-based network measurement system for multi-pipeline switches. For monitoring network flows that have packets arrived in bursts and spread over multiple pipelines, Confluence introduces novel data structures to collect short-term traffic statistics in ingress pipelines, and converge the measurement data to egress pipelines. Confluence is carefully designed under the switch hardware constraints, and in particular, to resolve the circular dependency in querying and updating a flow’s measurement data from sketch buckets, we propose a novel algorithm and theoretically prove its effectiveness. Both theoretical analysis and experiments driven by real-world traffic traces show that Confluence delivers higher measurement accuracies than existing solutions, especially in the critical task of detecting heavy hitters. Assessment on hardware switch suggests that Confluence is practical for real-world deployment.

In conventional vehicular edge computing (VEC), vehicles at edge nodes often face issues such as congestion and overhead, particularly when numerous vehicles offload their tasks to a single edge node. This scenario results in heightened processing delays and increased energy consumption. Additionally, the unpredictability of the task offloading process at the edge node presents a major challenge for vehicles in determining their offloading strategies within a dynamic environment. In this paper, we propose a jointed approach for task offloading and resource allocation aimed at minimizing the overall latency and energy consumption of all vehicles. This is accomplished through task profiling, channel allocation, and resource distribution for both vehicles and roadside units (RSUs). We introduce a framework for partial task offloading and resource allocation based on a decentralized learning algorithm known as the Distributed Partial Task Offloading and Resource Allocation (DPTORA) scheme. This approach provides flexibility in task processing, allowing each vehicle to choose whether to execute its task locally, partially offload it, or distribute it among multiple RSUs using vehicle-to-infrastructure (V2I) connections in a dense network. We develop algorithms based on the Generalist Pursuit Learning Algorithm (GPLA) and the Distributed Partial Task Offloading (DPTO) scheme to effectively address the optimization problem. Additionally, we provide a sub-optimal solution with low computational complexity. Extensive simulations validate the effectiveness of our proposed scheme in decreasing time latency and energy consumption while facilitating partial task offloading and resource allocation through a decentralized learning Approach in dynamic VEC networks.

In VANETs, millimeter wave (mmWave) is capable of providing ultra-high bandwidth, large data throughput, and very low transmission latency. However, the difficulty of mmWave transmission is further exacerbated by its limited transmission range and its physical properties of low diffraction and low penetration, coupled with real-time dynamic network topology. In future 6G networks, digital twins (DTs) are considered a promising enabling technology as it can provide seamless interaction between the virtual network world and the real world. In this paper, we propose a DT-empowered multi-hop V2X routing scheme using mmWave as the transmission link in urban scenarios. First, we propose a novel DT-assisted mmWave multi-hop relay network architecture, which uses the global traffic flow information possessed by the DTs to assist vehicles in selecting the optimal relay nodes. Second, the uncrewed aerial vehicles (UAVs) are deployed at intersections to improve the packet forwarding efficiency of the intersections. Finally, a reinforcement learning algorithm is used to make forwarding decisions between streets. In addition, we also consider vehicle density and network load as key indicators for optimal street selection. The experimental results indicate that our solution has significant advantages in terms of key indicators such as packet delivery rate and latency.

To overcome the inherent limitation of the optical circuit switch (OCS) while utilizing its high bandwidth, circuit/packet hybrid networks are widely proposed for modern data centers. However, as today’s OCS has reduced the reconfiguration delay to microseconds, the circuit from a source rack to a destination rack typically lasts fewer than 10 RTTs. Such a short circuit time brings a critical challenge to TCP, as it is difficult for a TCP sender to sufficiently grow its congestion window (CWND) and utilize the optical bandwidth. To address this problem, in this work, we presentAccelToR, a top-of-rack switch for improving TCP performance in circuit/packet hybrid data center networks. AccelToR leverages end-host congestion control to “accelerate” a blocked TCP flow by temporarily scheduling its packets to be transferred through the packet network a few RTTs before its circuit is established, and after enlarging the flow’s CWND with the acceleration, the switch buffers the last window of packets. During the circuit time, the switch sends out the buffered packets, and the accelerated flows, which have their CWNDs already grown large, continue to send packets at high rates to achieve a high optical bandwidth utilization. Experiment results show that AccelToR achieves high throughputs for elephant flows and utilizes over$\mathbf {90\%}$of the optical bandwidth, and it preserves short flow completion times for mice flows at the same time. In addition, AccelToR is robust under unexpected packet losses, and can benefit a wide range of TCP congestion control algorithms.

Large spread and high persistence are widely observed in malicious activities such as botnets and DDoS attacks in high-speed networks, while how to identify persistent-spreaders is still a challenging issue. In this work, we presentPS-Sketch, a system for estimating persistent-spreads of network flows and detecting persistent-spreaders in network data streams. PS-Sketch is based on our definitions of persistence and persistent-spread, which overcome the limitations of the conventional definitions by better capturing network flows’ behaviors, and being difficult for attackers to bypass. Within a switch’s pipeline, PS-Sketch processes packets with two adjacent sketch data structures, namely the P-sketch and the S-sketch. In the P-sketch, we employ low-pass filter (LPF) to trace an element’s persistence that incorporates occurrences in its entire history, and in the S-sketch, we extend the HyperLogLog (HLL) algorithm, and integrate an element’s persistence to the spread estimation of the flow that the element belongs to, for estimating the flow’s persistent-spread. We present theoretical analysis on the error bound of PS-Sketch. Trace-driven evaluation shows that PS-Sketch achieves high accuracy in estimating network flows’ persistent-spreads, and outperforms the existing solutions in detecting persistent-spreaders. We further prototype PS-Sketch in P4 and show that the system is deployable on commodity hardware switches.

Graph neural networks (GNNs) and heterogeneous graph neural networks (HGNNs) are prominent techniques for homogeneous and heterogeneous graph representation learning, yet their performance in an end-to-end supervised framework greatly depends on the availability of task-specific supervision. To reduce the labeling cost, pre-training on self-supervised pretext tasks has become a popular paradigm, but there is often a gap between the pre-trained model and downstream tasks, stemming from the divergence in their objectives. To bridge the gap, prompt learning has risen as a promising direction especially in few-shot settings, without the need to fully fine-tune the pre-trained model. While there has been some early exploration of prompt-based learning on graphs, they primarily deal with homogeneous graphs, ignoring the heterogeneous graphs that are prevalent in downstream applications. In this paper, we propose HGPROMPT, a novel pre-training and prompting framework to unify not only pre-training and downstream tasks but also homogeneous and heterogeneous graphs via a dual-template design. Moreover, we propose dual-prompt in HGPROMPT to assist a downstream task in locating the most relevant prior to bridge the gaps caused by not only feature variations but also heterogeneity differences across tasks. Finally, we thoroughly evaluate and analyze HGPROMPT through extensive experiments on three public datasets.

It is important to cooperatively share perception data among vehicles for the safety of autonomous driving systems. However, since vehicles may experience long data transmission delays, especially in case that the number of vehicles participating in cooperative perception is large, the dissemination of large-sized perception data remains a challenging problem. Considering the high transmission cost of perception data in the overlapping perception area of nearby vehicles, we propose a novel fusion-tree-based perception data dissemination scheme in millimeter-wave (mmWave) vehicular networks. Vehicles first aggregate their perception data at a preselected leader along a dynamically constructed fusion tree, and a merge-and- forward policy is adopted to reduce the amount of redundant data. After the aggregation stage ends, the leader broadcasts the merged data back to all vehicles through multi-hop broadcast links. The performance evaluation results show that the proposed scheme yields the best transmission performance among the three comparison schemes in terms of dissemination delay.

Graph Neural Networks (GNNs) have emerged as a mainstream technique for graph representation learning. However, their efficacy within an end-to-end supervised framework is significantly tied to the availability of task-specific labels. To mitigate labeling costs and enhance robustness in few-shot settings, pre-training on self-supervised tasks has emerged as a promising method, while prompting has been proposed to further narrow the objective gap between pretext and downstream tasks. Although there has been some initial exploration of prompt-based learning on graphs, they primarily leverage a single pretext task, resulting in a limited subset of general knowledge that could be learned from the pre-training data. Hence, in this paper, we propose MultiGPrompt, a novel multi-task pre-training and prompting framework to exploit multiple pretext tasks for more comprehensive pre-trained knowledge. First, in pre-training, we design a set of pretext tokens to synergize multiple pretext tasks. Second, we propose a dual-prompt mechanism consisting of composed and open prompts to leverage task-specific and global pre-training knowledge, to guide downstream tasks in few-shot settings. Finally, we conduct extensive experiments on six public datasets to evaluate and analyze MultiGPrompt.

While the state-of-the-art network embedding approaches often learn high-quality embeddings for high-degree nodes with abundant structural connectivity, the quality of the embeddings for low-degree ortailnodes is often suboptimal due to their limited structural connectivity. While many real-world networks are long-tailed, to date little effort has been devoted to tail node embeddings. In this article, we formulate the goal of learning tail node embeddings as afew-shot regressionproblem, given the few links on each tail node. In particular, since each node resides in its own local context, we personalize the regression model for each tail node. To reduce overfitting in the personalization, we propose alocality-awaremeta-learning framework, calledmeta-tail2vec, which learns to learn the regression model for the tail nodes at different localities. Moreover, to address the heterogeneity in nodes and edges on heterogeneous information networks (HINs), we further extend the proposed model and formulatemeta-tail2vec+, which is based on a dual-adaptation mechanism to facilitate the locality-aware tail node embeddings on HINs. Finally, we conduct extensive experiments and demonstrate the promising results of both meta-tail2vec and its extension meta-tail2vec+.

Graphs can model complex relationships between objects, enabling a myriad of Web applications such as online page/article classification and social recommendation. While graph neural networks (GNNs) have emerged as a powerful tool for graph representation learning, in an end-to-end supervised setting, their performance heavily relies on a large amount of task-specific supervision. To reduce labeling requirement, the “pre-train, fine-tune” and “pre-train, prompt” paradigms have become increasingly common. In particular, prompting is a popular alternative to fine-tuning in natural language processing, which is designed to narrow the gap between pre-training and downstream objectives in a task-specific manner. However, existing study of prompting on graphs is still limited, lacking a universal treatment to appeal to different downstream tasks. In this paper, we proposeGraphPrompt, a novel pre-training and prompting framework on graphs.GraphPromptnot only unifies pre-training and downstream tasks into a common task template, but also employs a learnable prompt to assist a downstream task in locating the most relevant knowledge from the pre-trained model in a task-specific manner. In particular,GraphPromptadopts simple yet effective designs in both pre-training and prompt tuning: During pre-training, a link prediction-based task is used to materialize the task template; during prompt tuning, a learnable prompt vector is applied to theReadOutlayer of the graph encoder. To further enhanceGraphPromptin these two stages, we extend it intoGraphPrompt+with two major enhancements. First, we generalize a few popular graph pre-training tasks beyond simple link prediction to broaden the compatibility with our task template. Second, we propose a more generalized prompt design that incorporates a series of prompt vectors within every layer of the pre-trained graph encoder, in order to capitalize on the hierarchical information across different layers beyond just the readout layer. Finally, we conduct extensive experiments on five public datasets to evaluate and analyzeGraphPromptandGraphPrompt+.

Millimeter-wave (mmWave) directional communication has been deemed as a promising means to achieve the goal of multi-gigabit data rate with low latency in vehicular ad-hoc networks (VANETs). However, due to the frequent transmission errors caused by the lossy nature of mmWave wireless links and high dynamic of VANETs, the channel resources are not fully exploited. In order to enhance the resilience to transmission errors, we introduce the symbol level network coding (SLNC) to mmWave wireless communications. We propose a greedy network coding strategy based on graph-theoretic approach, named GTNC, so as to take full advantage of SLNC. Moreover, considering the directionality of mmWave and the effect of GTNC on degrading transmission redundancy, it has great potential to realize concurrent transmission in mmWave vehicular networks. Therefore, we propose a cooperative concurrent distribution scheme with GTNC, named CCDS-GTNC, to coordinate transmissions between vehicles, which carries out collaborative vehicle-to-vehicle (V2V) and vehicle-to-Infrastructure (V2I) communications and achieves concurrent transmissions of multiple transmitters simultaneously. The simulation results show that our proposed scheme can effectively improve the efficiency of content dissemination in view of reducing the transmission delay and transmission redundancy.

Weighted fair queueing (WFQ) is an essential strategy for enforcing bandwidth guarantee and isolation in high-speed networks. Unfortunately, implementing the original WFQ packet scheduling algorithm on today’s commodity switch hardware is challenging due to the prohibitive complexity. Approximate WFQ packet schedulers, which work with the cheap and widely available First-In First-Out (FIFO) queues, have been proposed as an alternative in recent years. In this paper, we show that both the ideal and the approximate WFQ packet schedulers are unable to allocate bandwidths to TCP flows fairly, because of the bursty nature of the TCP traffic. Furthermore, we find that the representative approximate WFQ schedulers further degrade the scheduling fairness, due to their excessive packet drops. To address these issues, we present novel approximate WFQ packet scheduling algorithms in this paper. Our initial design, namely SQ-WFQ, imposes the minimum hardware requirement by using one single FIFO queue, and effectively reduces the excessive packet drops. Extended from SQ-WFQ, we propose the SQ-EWFQ packet scheduling algorithm. SQ-EWFQ inherits all the merits of SQ-WFQ, and is adaptive to the bursty TCP traffic by tolerating short-term packet bursts, while enforcing a long-term fairness among the TCP flows. We have implemented our proposed schedulers on commodity hardware programmable switches, and achieve line rate packet scheduling with them. Experiment results from a real-world testbed and large-scale simulations show that SQ-WFQ and SQ-EWFQ outperform the state-of-the-art approximate schedulers regarding the scheduling fairness, and SQ-EWFQ allocates bandwidths to TCP flows more fairly than SQ-WFQ and other existing solutions.

Sketch-based method has emerged as a promising direction for per-flow measurement in data center networks. Usually in such a measurement system, a sketch data structure is placed as a whole at one switch for counting all passing packets, but when summarizing measurement results from multiple switches, the overall accuracy is generally constrained by a few individual switches with small-sized sketches due to their limited memory resources. To address this problem, in this paper, we present Distributed Sketch, a new method for per-flow network measurement in data center networks. In Distributed Sketch, each network path is associated with a logical sketch, whose data structure is collectively maintained by all the switches along the path; meanwhile, each switch multiplexes its physical sketch to the constructions of the logical sketches of all the paths it belongs to. With Distributed Sketch, switches collaborate to measure network flows, and the network-wide measurement workload is fairly distributed among all the switches in the network. We implement Distributed Sketch with P4 on commodity hardware programmable switch, and in particular, to overcome the limitation that hardware switches do not support float-point computation, we present an optimal approximation method that involves only integer operations. We also propose an In-band Network Telemetry (INT) based method for addressing the challenges in deploying Distributed Sketch in large-scale data centers. Experiment results and theoretical analysis show that our proposed method is lightweight regarding measurement overhead, and by aggregating and making fair uses of resources from all the switches in the network, Distributed Sketch achieves a higher measurement accuracy compared with the state-of-the-art solutions.

Subgraph isomorphism counting is an important problem on graphs, as many graph-based tasks exploit recurring subgraph patterns. Classical methods usually boil down to a backtracking framework that needs to navigate a huge search space with prohibitive computational cost. Some recent studies resort to graph neural networks (GNNs) to learn a low-dimensional representation for both the query and input graphs, in order to predict the number of subgraph isomorphisms on the input graph. However, typical GNNs employ a node-centric message passing scheme that receives and aggregates messages on nodes, which is inadequate in complex structure matching for isomorphism counting. Moreover, on an input graph, the space of possible query graphs is enormous, and different parts of the input graph will be triggered to match different queries. Thus, expecting a fixed representation of the input graph to match diversely structured query graphs is unrealistic. In this paper, we propose a novel GNN called Count-GNN for subgraph isomorphism counting, to deal with the above challenges. At the edge level, given that an edge is an atomic unit of encoding graph structures, we propose an edge-centric message passing scheme, where messages on edges are propagated and aggregated based on the edge adjacency to preserve fine-grained structural information. At the graph level, we modulate the input graph representation conditioned on the query, so that the input graph can be adapted to each query individually to improve their matching. Finally, we conduct extensive experiments on a number of benchmark datasets to demonstrate the superior performance of Count-GNN.

In point cloud analysis, point-based methods have rapidly developed in recent years. These methods have recently focused on concise MLP structures, such as Point-NeXt, which have demonstrated competitiveness with Convolutional and Transformer structures. However, standard MLPs are limited in their ability to extract local features effectively. To address this limitation, we propose a Vector-oriented Point Set Abstraction that can aggregate neighboring features through higher-dimensional vectors. To facilitate network optimization, we construct a transformation from scalar to vector using independent angles based on 3D vector rotations. Finally, we develop a PointVector model that follows the structure of PointNeXt. Our experimental results demonstrate that PointVector achieves state-of-the-art performance 72.3% mIOU on the S3DIS Area 5 and 78.4% mIOU on the S3DIS (6-fold cross-validation) with only 58% model parameters of PointNeXt. We hope our work will help the exploration of concise and effective feature representations. The code will be released soon.

Current Scene text image super-resolution approaches primarily focus on extracting robust features, acquiring text information, and complex training strategies to generate super-resolution images. However, the upsampling module, which is crucial in the process of converting low-resolution images to high-resolution ones, has received little attention in existing works. To address this issue, we propose the Pixel Adapter Module (PAM) based on graph attention to address pixel distortion caused by upsampling. The PAM effectively captures local structural information by allowing each pixel to interact with its neighbors and update features. Unlike previous graph attention mechanisms, our approach achieves 2-3 orders of magnitude improvement in efficiency and memory utilization by eliminating the dependency on sparse adjacency matrices and introducing a sliding window approach for efficient parallel computation. Additionally, we introduce the MLP-based Sequential Residual Block (MSRB) for robust feature extraction from text images, and a Local Contour Awareness loss (ℒlca) to enhance the model's perception of details. Comprehensive experiments on TextZoom demonstrate that our proposed method generates high-quality super-resolution images, surpassing existing methods in recognition accuracy. For single-stage and multi-stage strategies, we achieved improvements of 0.7% and 2.6%, respectively, increasing the performance from 52.6% and 53.7% to 53.3% and 56.3%. The code is available at https://github.com/wenyu1009/RTSRN.

Graphs can model complex relationships between objects, enabling a myriad of Web applications such as online page/article classification and social recommendation. While graph neural networks (GNNs) have emerged as a powerful tool for graph representation learning, in an end-to-end supervised setting, their performance heavily relies on a large amount of task-specific supervision. To reduce labeling requirement, the “pre-train, fine-tune” and “pre-train, prompt” paradigms have become increasingly common. In particular, prompting is a popular alternative to fine-tuning in natural language processing, which is designed to narrow the gap between pre-training and downstream objectives in a task-specific manner. However, existing study of prompting on graphs is still limited, lacking a universal treatment to appeal to different downstream tasks. In this paper, we propose GraphPrompt, a novel pre-training and prompting framework on graphs. GraphPrompt not only unifies pre-training and downstream tasks into a common task template, but also employs a learnable prompt to assist a downstream task in locating the most relevant knowledge from the pre-trained model in a task-specific manner. Finally, we conduct extensive experiments on five public datasets to evaluate and analyze GraphPrompt.

Model extraction replicates the functionality of machine learning models deployed as a service. Recently, generative adversarial networks (GANs)-based methods have achieved remarkable performance in data-free model extraction. However, previous methods generate random data in every training batch, resulting in slow convergence and redundant queries. We propose to tackle the task with a much simpler paradigm. Specifically, we steal a data set with GAN before training the clone model rather than during every training batch. Benefiting from full use of the generated data, the proposed paradigm needs less training time and query cost. To improve the class distribution of data, a balancing strategy is applied. Furthermore, the balanced data set is filtered based on adversarial robustness for better quality. Combining the above strategies, we propose an efficient model extraction by data set stealing, balancing, and filtering (DSBF). Experiments on three widely used data sets show that DSBF outperforms previous methods while converging faster and costing fewer queries.

It is very important to detect and prevent bogus messages (here called rumors) in vehicular ad hoc networks (VANETs) because such misbehaviors could cause road safety problems and even casualties. In this paper, we first propose an evolutionary public goods game (EPGG) model to detect bogus messages and stimulate nodes to implement data verification. When a node detects a rumor, it immediately broadcasts an anti-rumor. We here present a susceptible, immune, and neutral (SIN) rumor diffusion model to characterize the spread of rumors and anti-rumors. Moreover, we propose a dynamic, hierarchical public goods game (DHPGG) model to analyze the symbiosis and confrontation of rumors and anti-rumors. Simulation results show these models could detect and suppress rumors effectively in VANETs.

In vehicular ad hoc networks (VANETs), quick and reliable multi-hop broadcasting is important for the dissemination of emergency warning messages. By scheduling multiple nodes to transmit messages concurrently and cooperatively, cooperative transmission based broadcast schemes may yield much better broadcast performance than conventional broadcast schemes. However, a cooperative transmission requires multiple relays to achieve strict synchronization on both time and frequency, which may induce high cost for a cooperative transmission process. In this paper, we analyze the cost and benefit of a cooperative transmission for data broadcasting in vehicular networks, and introduce a new metric called the single hop broadcast efficiency (SBE) to evaluate the overall broadcast performance. We propose an efficient, non-deterministic cooperation mechanism to reduce the cooperation cost. The mechanism maximizes the expected broadcast performance by selecting cooperators with the largest expected SBE value for a lead relay, and initiates cooperative broadcasting process when the expected SBE value is larger than that of a single-relay based broadcasting. Based on the non-deterministic mechanism, we propose an efficient, cooperative transmission based opportunistic broadcast (ECTOB) scheme which further utilizes rebroadcast to improve the reliability of the broadcast scheme. Simulation results show that the proposed scheme outperforms the conventional ones.

In VANETs, it is important to support fast and reliable multi-hop broadcast for safety-related applications. The performance of multi-hop broadcast schemes is greatly affected by relay selection strategies. However, the relationship between the relay selection strategies and the expected broadcast performance has not been fully characterized yet. Furthermore, conventional broadcast schemes usually attempt to minimize the waiting time difference between adjacent relay candidates to reduce the waiting time overhead, which makes the relay selection process vulnerable to internal interference, occurring due to retransmissions from previous forwarders and transmissions from redundant relays. In this paper, we jointly take both of the relay selection and the internal interference mitigation into account and propose a fast, reliable, opportunistic multi-hop broadcast scheme, in which we utilize a novel metric called the expected broadcast speed in relay selection and propose a delayed retransmission mechanism to mitigate the adverse effect of retransmissions from previous forwarders and an expected redundancy probability based mechanism to mitigate the adverse effect of redundant relays. The performance evaluation results show that the proposed scheme yields the best broadcast performance among the four schemes in terms of the broadcast coverage ratio and the end-to-end delivery latency.

Cloud computing is critical for today's information society. In this paper, we shed light on two radically different cloud design philosophies: theDC-cloudbuilt around massive data centers, and theISP-cloudbuilt upon a large ISP. With extensive measurements on Alibaba, Tencent, and CTYun, we find that both designs have strengths and weaknesses: the ISP-cloud of CTYun has less inflated paths to users within the same ISP, but its paths to external users are more inflated comparing with the DC-clouds of Alibaba and Tencent. By analyzing the clouds’ routing policies, we reveal the reasons behind the path inflations: Alibaba and Tencent adopt anearly-exitpolicy to use more inflated public Internet paths as early as possible; while CTYun follows aglobal and location-agnosticpolicy to detour traffic to remote PoPs, leading to highly inflated paths. Based on the insights, we suggest alternative policies and averagely reduce 11.0% latency to 30.5% destinations for Alibaba, 9.8% latency to 18.6% destinations for Tencent, and 54.1% latency to external destinations for CTYun. The results suggest that both cloud designs have rooms for improvement, and an ISP-cloud has the potential to achieve a superior performance, thanks to its inherited advantages from the ISP infrastructure.

In millimeter wave (mmWave) vehicular communications, beam scanning is usually used for beam alignment. However, beam scanning takes a long time, which makes it unsuitable in the mmWave vehicular communications with high-speed nodes. Moreover, the existing research on mmWave vehicular communications is usually based on one-hop transmission. However, due to a limited transmission range of mmWave, in many cases, multi-hop transmissions are required to transmit information from a source node to a destination node. To solve these problems, we propose an opportunistic routing protocol based on position information for beam alignment in mmWave vehicular communications. Beam alignment is performed based on the location information of neighbor nodes instead of beam scanning, which avoids the increase of network load and delay caused by beam scanning. In addition, the interference in concurrent transmissions is considered in order to optimize beamwidth and forwarding priorities. When the opportunistic routing is used, even if one candidate forwarder (CF) fails to forward a packet, the other CFs receiving the packet can still forward it, avoiding an increase in the network load and unicast delay caused by a large number of retransmissions. In an experiment, we extend the existing one-hop communication protocols into multi-hop communication protocols. Compared with these protocols, the proposed protocol yields lower end-to-end delay, higher packet delivery ratio and network throughput.

In this paper, we investigate how to efficiently schedule vehicle-to-vehicle (V2V) requests in millimeter wave (mmWave) vehicular networks. Considering the vulnerability of mmWave signals to blockages and the disadvantages of traditional relay mode, we firstly propose the cooperation of intelligent reflecting surfaces (IRSs) and relays to enable the scheduling of non-line-of-sight (NLOS) requests. To maximize the resource utilization and consequently improve the success ratio of link scheduling, we design a centralized scheduling algorithm based on spatial-time division multiple access (STDMA) and the cooperation scheme. The proposed algorithm can provide more opportunities for scheduling NLOS requests by the cooperation of IRSs and relays, and achieve the goal of collision-free concurrent transmissions by fully leveraging spatial reuse, so that as many V2V requests as possible can be fulfilled. Simulation results demonstrate that the proposed scheme can effectively enhance the success ratio of scheduling V2V requests especially NLOS requests.

With the prevalence of accessible multi-modal sensors and the maturity of pose estimation algorithms, skeleton-based action recognition has gradually become the mainstream of human action recognition (HAR). The key issue is to mine the correlations and dependencies between different joints and bones. In this paper, we propose a cross view learning approach. First, the static and dynamic representations of skeletons, from two different views (joints and bones), are calculated and aggregated respectively. Then, the integrated representations of these two views are used as parallel inputs to the cross view learning model, which mainly includes two blocks, namely a multi-scale learning block and a multi-view fusion block. The former is used to excavate the intra-view’s discriminative and comprehensive features, and the latter is utilized to capture the complementary representations of the inter-view. Finally, the fused representations are input to the classifier for action recognition. It has been experimentally proven that our proposed approach outperforms several state-of-the-art baseline methods and achieves a very competitive performance.

Millimeter-wave (mmWave) directional communication has been viewed as a means to achieve the goal of multiple gigabit data rate with low latency in vehicular networks, so as to fulfill the requirement of dissemination of a huge amount of sensory data. However, most of the existing researches focus on implementing the vehicular mmWave communications by point-to-point (P2P) transmissions between vehicles or base stations, which is not conductive to efficient dissemination of messages in a whole road segment. In this paper, we propose a novel gigabit message dissemination scheme called P2MP in mmWave/Dedicated Short Range Communications (DSRC) multi-hop vehicle-to-vehicle (V2V) network for highway scenario. P2MP includes: 1) a clustering-based transmitter beamforming method which helps a transmitter to search a relatively optimal transmission beam according to the collected environmental information. 2) a positioning-errors-based receiver beamforming which is used to determine the beam direction and beamwidth at every receiver. 3) a candidate forwarder selection mechanism which is devised to select reliable relays for next hop broadcast. Based on these three components, the proposed P2MP transmission scheme is able to broadcast massive messages to vehicles in the road. The simulation results show that our proposed scheme can effectively improve the efficiency of broadcasting large messages through mmWave.

Given that heterogeneous information networks (HIN) encompass nodes and edges belonging to different semantic types, they can model complex data in real-world scenarios. Thus, HIN embedding has received increasing attention, which aims to learn node representations in a low-dimensional space, in order to preserve the structural and semantic information on the HIN. In this regard, metagraphs, which model common and recurring patterns on HINs, emerge as a powerful tool to capture semantic-rich and often latent relationships on HINs. Although metagraphs have been employed to address several specific data mining tasks, they have not been thoroughly explored for the more general HIN embedding. In this paper, we leverage metagraphs to learn relationship-preserving HIN embedding in a self-supervised setting, to support various relationship mining tasks. In particular, we observe that most of the current approaches often under-utilize metagraphs, which are only applied in a pre-processing step and do not actively guide representation learning afterwards. Thus, we propose the novel framework of mg2vec, which learns the embeddings for metagraphs and nodes jointly. That is, metagraphs actively participates in the learning process by mapping themselves to the same embedding space as the nodes do. Moreover, metagraphs guide the learning through both first- and second-order constraints on node embeddings, to model not only latent relationships between a pair of nodes, but also individual preferences of each node. Finally, we conduct extensive experiments on three public datasets. Results show that mg2vec significantly outperforms a suite of state-of-the-art baselines in relationship mining tasks including relationship prediction, search and visualization.

In-band Network Telemetry (INT) is a novel framework for monitoring network health in real-time, and its recent variant, Probabilistic INT (PINT), reduces its bandwidth consumption with a probabilistic approach. However, as we show in this paper, a PINT task can be successfully accomplished only when it is allocated a sufficient number of packets, and if there are many tasks executed in parallel, packets become a scarce resource. Meanwhile, today’s production network generally executes multiple measurement tasks for tracing different network states simultaneously. Therefore, in such a context, scheduling parallel PINT tasks on one single INT flow that has a limited number of packets becomes a critical problem. In this paper, we address this problem for the first time. We propose an algorithm that efficiently schedules multiple parallel PINT tasks on a flow by allocating the flow’s packets to the tasks and showing that the allocation is optimal. We realize the algorithm with a packet processing pipeline and implement it on software and hardware-programmable switches. Comprehensive evaluation on a FatTree testbed shows that at a low scheduling overhead, our algorithm can conduct parallel PINT tasks to detect various network faults in a timely and accurate manner. Additionally, the algorithm accomplishes more PINT tasks with higher quality than the alternative solutions.

Current self-supervised depth estimation algorithms mainly focus on either stereo or monocular only, neglecting the reciprocal relations between them. In this paper, we propose a simple yet effective framework to improve both stereo and monocular depth estimation by leveraging the underlying complementary knowledge of the two tasks. Our approach consists of three stages. In the first stage, the proposed stereo matching network termed StereoNet is trained on image pairs in a self-supervised manner. Second, we introduce an occlusion-aware distillation (OA Distillation) module, which leverages the predicted depths from StereoNet in non-occluded regions to train our monocular depth estimation network named SingleNet. At last, we design an occlusion-aware fusion module (OA Fusion), which generates more reliable depths by fusing estimated depths from StereoNet and SingleNet given the occlusion map. Furthermore, we also take the fused depths as pseudo labels to supervise StereoNet in turn, which brings StereoNet’s performance to a new height. Extensive experiments on KITTI dataset demonstrate the effectiveness of our proposed framework. We achieve new SOTA performance on both stereo and monocular depth estimation tasks.

Wireless energy charging has emerged as a very promising technology for prolonging sensor lifetime in wireless rechargeable sensor networks (WRSNs). Existing studies focused mainly on the one-to-one charging scheme that a single sensor can be charged by a mobile charger at each time, this charging scheme however suffers from poor charging scalability and inefficiency. Recently, another charging scheme, the multi-node charging scheme that allows multiple sensors to be charged simultaneously by a mobile charger, becomes dominant, which can mitigate charging scalability and improve charging efficiency. However, most previous studies on this multi-node energy charging scheme focused on the use of a single mobile charger to charge multiple sensors simultaneously. For large scale WRSNs, it is insufficient to deploy only a single mobile charger to charge many lifetime-critical sensors, and consequently sensor expiration durations will increase dramatically. To charge many lifetime-critical sensors in large scale WRSNs as early as possible, it is inevitable to adopt multiple mobile chargers for sensor charging that can not only speed up sensor charging but also reduce expiration times of sensors. This however poses great challenges to fairly schedule the multiple mobile chargers such that the longest charging delay among sensors is minimized. One important constraint is that no sensor can be charged by more than one mobile charger at any time due to the fact that the sensor cannot receive any energy from either of the chargers or the overcharging will damage the recharging battery of the sensor. Thus, finding a closed charge tour for each of the multiple chargers such that the longest charging delay is minimized is crucial. In this paper we address the challenge by formulating a novel longest charging delay minimization problem. We first show that the problem is NP-hard. We then devise the very first approximation algorithm with a provable approximation ratio for the problem. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results demonstrate that the proposed algorithm is promising, and outperforms existing algorithms in various settings.

Network embedding is an active research area due to the prevalence of network-structured data. While the state of the art often learns high-quality embedding vectors for high-degree nodes with abundant structural connectivity, the quality of the embedding vectors for low-degree or tail nodes is often suboptimal due to their limited structural connectivity. While many real-world networks are long-tailed, to date little effort has been devoted to tail node embedding. In this paper, we formulate the goal of learning tail node embeddings as a few-shot regression problem, given the few links on each tail node. In particular, since each node resides in its own local context, we personalize the regression model for each tail node. To reduce overfitting in the personalization, we propose a locality-aware meta-learning framework, called meta-tail2vec, which learns to learn the regression model for the tail nodes at different localities. Finally, we conduct extensive experiments and demonstrate the promising results of meta-tail2vec. (Supplemental materials including code and data are available at https://github.com/smufang/meta-tail2vec.)

The performance of multi-hop broadcast in vehicular ad hoc networks can be greatly affected by the highly varying link dynamics caused by mobility of vehicles and complex wireless channel environment. The conventional receiver-based and sender-based broadcast protocols usually decide the relay sequences only based on distance, neglecting the influence of links qualities. In this paper, we consider the impact of link qualities and vehicular mobility and investigate how to select broadcast relays to minimize the broadcast delay and maximize the broadcast efficiency. We first propose a novel broadcast scheme based on the prediction of dynamics (BPD), which utilizes the dynamic information to achieve the model-based prediction and combines the sender-based and receiver-based relay selection schemes for multi-hop broadcast. Then, we propose a novel metric called the expected remaining delay (D) and implement it in BDP (BPD-D) for minimizing the broadcast delay. We also propose a novel metric called the expected rebroadcast efficiency (E) and implement it in BDP (BPD-E) for maximizing the broadcast efficiency. The simulation results show that our proposed BPD-D and BPD-E broadcast protocols outperform the conventional protocols, while BPD-D can achieve the least delay and BPD-E has the highest dissemination efficiency.

For opportunistic routing in independent duty-cycled wireless sensor networks (WSNs), a sender dynamically determines a relay candidate set depending on the real-time network conditions. Due to independent and varying duty cycle length, the sender may handle different waking-up orders of the potential forwarders when it tries to forward data packets at different time instants. Conventional opportunistic routing protocols overlook the time-varying property of the waking-up order of the candidate nodes. In this paper, we theoretically analyze how to obtain an optimal candidate set for each node in order to minimize the end-to-end latency. Then, considering the realistic scenarios, we propose an opportunistic routing which jointly considers global and localized optimizations. Based on the relatively stable topology and duty-cycle length information, an original candidate set is constructed. Then, by considering the real-time link and duty cycle information in the local context, a further optimization for the original candidate set can be achieved. Simulation results show that our proposed schemes can significantly improve the end-to-end latency compared with the benchmarks.

Leading e-commerce providers have built large and complicated systems to provide countrywide or even worldwide services. However, there have been few substantive studies on e-commerce systems in real world. In this paper, we investigate the systems of Tmall and JD, the top-two most popular e-commerce websites in China, with a measurement approach. By analyzing traffics from campus network, we present a characterization study that covers several features, including usage patterns and shopping behaviors, of the e-commerce workload; in particular, we characterize the massive flash crowd in the Double-11 Day, which is the biggest online shopping festival in the world. We also reveal Tmall and JD's e-commerce infrastructures, including content delivery networks (CDNs) and clouds, and evaluate their performances under the flash crowd. We find that Tmall's CDN proactively throttles bandwidths for ensuring low but guaranteed throughputs, while JD still follows the best-effort way, leading to poor and unstable performances; both providers do not have sufficient capacities in their private clouds, resulting in extraordinarily long transaction latencies. Based on the insights obtained from measurement, we discuss the design choices of e-commerce CDNs, and investigate the potential benefits brought by incorporating client-side assistances in offloading massive flash crowd of e-commerce workloads.

Wireless energy charging has emerged as a very promising technology for prolonging sensor lifetime in Wireless Rechargeable Sensor Networks (WRSNs). Existing studies focused mainly on the 'one-to-one' charging scheme that a sensor can be charged by a single mobile charger at each time, this charging scheme however suffers from poor charging scalability and inefficiency. Recently, another charging scheme - the 'multiple-to-one' charging scheme that allows multiple sensors to be charged simultaneously by a single charger, becomes dominant and can mitigate charging scalability and improve the charging efficiency. Most research studies on this latter scheme focused on the use of a mobile charger to charge multiple sensors simultaneously. However, for large scale WRSNs, it is insufficient to deploy just a single mobile charger to charge many lifetime-critical sensors, and consequently sensor expiration durations will increase dramatically. Instead, in order to charge as many as lifetime-critical sensors, the use of multiple mobile chargers for charging sensors can speed up sensor charging significantly, thereby reducing their expiration durations and improving the monitoring quality of WRSNs. However, this poses great challenges to schedule multiple mobile chargers for sensor charging at the same time such that the longest delay among the chargers is minimized due to multiple critical constraints. One such an important constraint in multiple mobile chargers is that each sensor cannot be charged by more than one mobile charger at each time; otherwise, the sensor cannot receive any energy from either of the chargers. In this paper we address this challenge by first formulating a novel longest delay minimization problem that is NP-hard. We then devise the very first approximation algorithm with a provable approximation ratio for the problem. We finally evaluate the performance of the proposed algorithm through experimental simulations. Simulation results demonstrate that the proposed algorithm is very promising, which outperforms the other heuristics in various settings.

Crowdsourced live video broadcasting services, such as Twitch and YouTube Live, are becoming increasingly popular. In such a service, viewers are allowed to perform rich interactions, such as posting comments and donating monetary virtual gifts, while watching videos. Understanding viewer interactions is essential for people to comprehend the production and consumption of the crowdsourced live video content and improve the service. However, the basic characteristics of the viewer interactions are still unknown. In this paper, we present a comprehensive measurement study of the viewer interactions on Douyu, a popular crowdsourced live video broadcasting website in China. Our measurement spans four months and contains comment posting and virtual gift donating interactions from tens of millions of viewers in hundreds of thousands of channels. Based on the measurement data, we carry out a content analysis on danmu comments and characterize the patterns of the viewer interactions. We build a suite of models for capturing the gift donating process, viewer activity, and channel popularity. We further analyze the influences of the broadcaster's behavioral factors on a channel's popularity and present methodologies for popularity predicting. Our measurement and analysis have important implications on the design and business policy of the crowdsourced live video broadcasting services.

There are high demands for cooperation among vehicle nodes to disseminate data in vehicular ad hoc networks (VANETs). Game theory approaches have been applied to analyzing and regulating the behavior of nodes in complex networks. However, game theory-based cooperation in VANETs still poses a research challenge due to continuous changes in the network topology. In this paper, we focus on the issue of the common behavior of cooperation in group vehicular interactions. We first present a dynamic member public goods game (PGG) model for real VANETs in which games are carried out within each group. Then, a dynamic grouping PGG (DGPGG) model is proposed to model VANETs with high-node density. In addition, a greedy neighbor selection strategy is proposed to replace the typical random strategy. Finally, the individual reputation of a vehicle is integrated into the DGPGG model. The simulation results show that the DGPGG model can greatly increase the proportion of cooperative nodes in high-density VANETs no matter whether the network is static or dynamic. When the network topology changes quickly, the greedy neighbor selection strategy can promote cooperation between nodes more effectively than the random strategy.

Network coding and schedule-based flooding trees have been widely deployed to improve the performance of a flooding process in low-duty-cycle wireless sensor networks (WSNs). However, the conventional schemes induce too long flooding delay because there is only one static sink node as a single source. In this paper, we use a mobile sink to travel around the monitoring area. When the mobile sink travels to a position, it serves as a new source. Since multiple sources can simultaneously forward data packets, flooding speed can be accelerated. We investigate the new problems when both a mobile sink and network coding are integrated with a schedule-tree-based flooding scheme. We model the process to deliver multiple encoded packets from the source to a node as a Markov process, and we theoretically investigate how to minimize the total expected flooding delay with a mobile sink by considering all the possible traveling orders of the branches. We propose a heuristic algorithm to obtain an appropriate trajectory by accounting for the effect of traveling delay of the mobile sink. Simulation results show that our proposed flooding scheme with a mobile sink can significantly reduce the flooding delay.

Many applications in vehicular ad hoc networks (VANETs) are based on broadcast to disseminate information among vehicles. In conventional broadcast protocols, only one vehicle is scheduled to rebroadcast a message at a certain time to avoid collisions among vehicles. In this paper, we derive the maximum temporal displacement required by the concurrent transmissions in VANETs and propose a Concurrent Transmission based Broadcast (CTB) protocol. The CTB includes two parts, broadcast in a street and broadcast at intersections. We divide the transmission range in the broadcast direction into segments and schedule the concurrent transmissions of the forwarders in the same segment. In our protocol, even if some selected forwarders fail to receive the message, other forwarders having received the message can still rebroadcast it, which reduces the broadcast delay and increases the broadcast reliability. Simulation results show that our protocol is faster and more reliable, compared to conventional broadcast protocols.

Although the travel time is the most important information in road networks, many spatial queries, e.g.,$k$-nearest-neighbor ($k$-NN) and range queries, for location-based services (LBS) are only based on the network distance. This is because it is costly for an LBS provider to collect real-time traffic data from vehicles or roadside sensors to compute the travel time between two locations. With the advance of web mapping services, e.g., Google Maps, Microsoft Bing Maps, and MapQuest Maps, there is an invaluable opportunity for using such services for processing spatial queries based on the travel time. In this paper, we propose a server-sideSpatialMashupService (SMS) that enables the LBS provider to efficiently evaluate$k$-NN queries in road networks using the route information and travel time retrieved from an external web mapping service. Due to the high cost of retrieving such external information, the usage limits of web mapping services, and the large number of spatial queries, we optimize the SMS for a large number of$k$-NN queries. We first discuss how the SMS processes a single$k$-NN query using two optimizations, namely,direction sharingandparallel requesting. Then, we extend them to process multiple concurrent$k$-NN queries and design a performance tuning tool to provide a trade-off between the query response time and the number of external requests and more importantly, to prevent a starvation problem in the parallel requesting optimization for concurrent queries. We evaluate the performance of the proposed SMS using MapQuest Maps, a real road network, real and synthetic data sets. Experimental results show the efficiency and scalability of our optimizations designed for the SMS.

In urban vehicular ad hoc networks (VANETs), due to the high mobility and uneven distribution of vehicles, how to select an optimal relaying node in an intra-street and how to determine a street selection at the intersection are two challenging issues in designing an efficient routing protocol in complex urban environments. In this paper, we build a link model with a Wiener process to predict the probability of link availability, which considers the stable and unstable vehicle states according to the behavior of vehicles. We introduce a novel concept called the link correlation which represents the influence of different link combinations in network topology to transmit a packet with less network resource consumption and higher goodput. Based on this concept, we design an opportunistic routing metric called the expected transmission cost over a multi-hop path (ETCoP) implemented with our link model as the selection guidance of a relaying node in intra-streets. This metric can also provide assistance for the next street selection at an intersection. Finally, we propose a street-centric opportunistic routing protocol based on ETCoP for VANETs (SRPE). Simulation results show that our proposed SRPE outperforms the conventional protocols in terms of packet delivery ratio, average end-to-end delay, and network yield.

In wireless sensor networks (WSNs), the broadcasting is used for remote network configuration, information dissemination, route discovery and so on by propagating messages across the whole network. To reduce the energy consumption of the asynchronous WSNs, sensor nodes usually adopt independent wake/sleep scheduling. However, this may induce unpredictable waiting delay. In some real-time applications, the broadcasting should be accomplished within a limited time. In this paper, we propose a broadcasting protocol to save energy under a constrained end-to-end (E2E) delay. We estimate the average one-hop delay and use an adjusting factor to balance the remaining delay of continuous hops. Then, we set up a sender candidate node set for each node according to its waiting delay constraint, and choose the least-cost energy node to forward the packet under the constrained waiting delay to save energy. The simulation results show that our broadcasting protocol is more energy efficient than conventional protocols under E2E delay constraint.

As the foundation of routing, topology control should minimize the interference among nodes, and increase the network capacity. With the development of mobile ad hoc networks (MANETs), there is a growing requirement of quality of service (QoS) in terms of delay. In order to meet the delay requirement, it is important to consider topology control in delay constrained environment, which is contradictory to the objective of minimizing interference. In this paper, we focus on the delay-constrained topology control problem, and take into account delay and interference jointly. We propose a cross-layer distributed algorithm called interference-based topology control algorithm for delay-constrained (ITCD) MANETs with considering both the interference constraint and the delay constraint, which is different from the previous work. The transmission delay, contention delay and the queuing delay are taken into account in the proposed algorithm. Moreover, the impact of node mobility on the interference-based topology control algorithm is investigated and the unstable links are removed from the topology. The simulation results show that ITCD can reduce the delay and improve the performance effectively in delay-constrained mobile ad hoc networks.

Bandwidth-delay product (BDP) is an important indicator of the network capacity and usually used to measure the quality of a connection. In wireless ad hoc networks over IEEE 802.11, packets may be blocked due to link layer contention, and delay triggered by the contention called contention delay has been revealed has nothing to do with BDP. BDP in wireless ad hoc networks has been studied a lot to improve the transmission control protocol (TCP) performance, however, in the calculation of BDP, most of the previous studies neglected to remove the contention delay and thus obtained an overestimated value which may cause congestion window (CWND) overshooting problem. In this paper, we propose a new method to accurately measure the congestion delay which is the effective “delay” of BDP and calculate more accurate BDP. Then, based on the BDP, we set suitable congestion window limit (CWL) to alleviate the CWND overshooting problem and enhance TCP performance. Simulation results show that TCP with CWL based on our accurate BDP works more efficiently.

In mobile ad hoc networks, node mobility may cause frequent route failures and route rediscoveries, which induce large overhead. Flooding used in a route discovery may cause a large number of unnecessarily redundant forwardings of route request (RREQ) packets, resulting in significant routing overhead. Conventional broadcast schemes, which aim to reduce the overhead in the route discovery by limiting the forwardings of RREQ packets, will also reduce the number of the underlying paths that route discovery can find, thus may miss the stablest path for data packets transmission. In this paper, we propose a novel routing discovery scheme called probabilistic broadcasting algorithm (PBA) based on the connectivity information of predictable rendezvous nodes (PRN) by considering node mobility and link stability in order to predict the possible connection of moving nodes. The PBA can find a stable path while limiting the number of RREQ packets forwardings in the route discovery.

Localization of nodes is a very crucial issue in many wireless sensor network (WSN) applications. In this paper, we consider a WSN initially consisting of a single location-aware mobile anchor nodes and a number of location-unaware, stationary nodes, and propose two algorithms to localize the unknown nodes using the mobile anchor nodes. We can obtain the location information of the unknown nodes through the transmitted location information of the mobile anchor node, even in the presence of obstacles. Received Signal Strength (RSS) is elaborately used in both the two algorithms. The first algorithm controls the trajectory of the mobile anchor node and uses a geometric property to simplify the estimation of the location of the unknown nodes; the second algorithm exploits to use directional information. The proposed algorithms outperform the conventional algorithm based on RSS in terms of localization accuracy. Simulation results show the effectiveness of the proposed algorithms.

Packet loss in multi-hop wireless ad hoc networks is an important factor in network performance optimization. Typically, MAC frame collisions and queue overflow caused by MAC contentions and interference are the main reasons of packet loss. Recently, MAC frame collisions had been paid much attentions to in mitigating packet loss. Whereas, queue overflow among inter-flow has rarely attracted attentions. In this paper, we first analyze how contentions among inter-flows will cause queue overflow and further packet loss. We show network local spatial reuse is also an important factor which will cause queue overflow at a forwarding station. With the information of queue status, each forwarding station determines whether to reserve the transmission of subsequent frames in the waiting queue. The reserved data frames will not suffer from contentions and interference and network local channel utilization is also improved. By theory analysis and simulations, we show our proposed channel reservation scheme can effectively reduce packet loss and improve network local channel efficiency in multi-hop wireless networks.

In asynchronous wireless sensor networks, finding an optimal candidate set can significantly improve the efficiency of an opportunistic routing protocol by balancing the one-hop cost from a sender to a relay and the remaining cost from the relay to the sink. A larger candidate set indicates smaller one-hop waiting cost to wait for a candidate to wake up, however, a packet is more likely to deviate from the shortest path, which will induce larger remaining cost. Remaining cost of each candidate can be used to construct a coordinate system. Neighbors of a node with high coordinate value is excluded from the candidate set, thus, we can select the next-hop relay by simply taking the first wake-up candidate. In this paper, we propose a distributed algorithm which is similar to the Bellman-Ford algorithm to construct a coordinate system and find an optimal candidate set for each node. Moreover, we propose an algorithm which induces acceptable overhead to maintain the coordinate system and each node's optimal candidate set. We also use simulation results to show that our proposed opportunistic routing protocol is more efficient than conventional strategies in terms of delay, energy efficiency, and miss ratio.

Physical carrier sensing (PCS) is an effective mechanism to reduce the collision and interference caused by hidden terminals in wireless ad hoc networks, which allows nodes to be separated by a silent margin. However, potential simultaneous transmissions may be suppressed in this silent margin due to exposed terminal problem. As a consequence, PCS mechanism affects the network performance in two opposite ways. In this paper, we present an analytical model for deriving the optimal carrier sensing threshold to make a tradeoff between hidden terminal and exposed terminal problems for given network topology. Our model takes into account the potential throughput loss of exposed terminals, which previous studies have ignored. Based on the model, an adaptive and iterative algorithm is proposed to obtain the optimal sensing threshold and dynamically adjust the physical carrier sensing threshold through periodic estimation of channel condition. Simulation results demonstrate that the proposed scheme outperforms IEEE 802.11 protocol in terms of average network throughput.

Broadcast is one of the most fundamental services in WSNs. In asynchronous WSNs which every node has different wake/sleep schedule, the broadcast protocol may be achieved by multiple unicasts. Moreover, in the duty-unaware asynchronous networks which the nodes do not know when their neighbors wake up, it is difficult to realize the broadcast protocol. In our paper, we propose a new broadcast protocol to reduce the data redundancy and total transmission energy under the condition of covering the whole networks. We first set up a global optimal transmission cost tree, then based on this tree we set up a local delay coordinate for each node to decrease the total waiting delay. The proposed protocol combines the global optimization and local optimization to balance the transmission cost and waiting cost in broadcasting. The experiment evaluation shows that our proposed protocol can be more efficient in duty-unaware WSNs than the ADB protocol.

Carrier sensing schemes have been recognized as a key knob for improving network performance. The distributed coordination function (DCF) and its modifications focus on the feature of the ongoing transmission links to enhance link throughput, however, a lack of coordinated adjustment of the surrounding nodes may reduce the aggregate throughput of entire network. In this paper, we are concerned with the delayed transmissions from exposed terminals, and interference and collisions from hidden terminals. We classify the neighboring areas for a given transmission link into three areas: a hidden area, an exposed area and an overlapped area, and then propose a coordinated dynamic physical carrier sense (CDPCS) scheme in which the carrier sensing threshold (CST) values of the neighbors are adjusted at the same time to achieve local optimization by utilizing the exchanged information and states of neighboring nodes. Simulation results show that the proposed schemes can work effectively.

Connectivity is one of the most fundamental properties in wireless ad hoc networks. Due to fading effects, the received power at certain distance becomes a random variable and a mass of unidirectional links emerged which is reflected by the shadowing model. Thus, research for network connectivity based on geometric model is inaccurate. Additionally, interference determines whether a link can be regarded as existence when the signal to interference and noise ratio (SINR) is larger than some threshold. In this paper, we propose a mathematical model to depict the impact of interference on the connectivity in a lognormal shadow fading environment. Based on the physical model of interference and mechanism of carrier sense, we first formulate the link connection probability, and obtain the probability of network connection with the random graph theory. Simulation shows effective carrier sense threshold has more important influence on the probability of network connection than other factors.

Interference is one of key factors of performance degradation in wireless communications. Various interference models have been proposed in the literature. However, most of them did not take node mobility into account. In fact, as an interferer approaches a carrier sensing range due to node mobility, the transmitted frame may also encounter a collision. Due to the interfering source, it is necessary to make an effective interference model to characterize the interference and signal-to-interference ratio(SIR). In this paper, we propose a novel interference model through interference and SIR distribution functions obtained from mobile nodes. In particular, we focus on a random direction (RD) model. Based on the probability distribution function of distance between any node pairs, we theoretically estimate the distribution of the accumulated interference contributed by concurrent transmissions and the corresponding SIR values. Through this result, we investigate the probability of successful transmissions and the probability of carrier sensing failures.

Due to high mobility of nodes in mobile ad hoc networks (MANETs), there exist frequent link breakages which lead to frequent path failures and route discoveries. The overhead of a route discovery cannot be neglected. In a route discovery, broadcasting is a fundamental and effective data dissemination mechanism, where a mobile node blindly rebroadcasts the first received route request packets unless it has a route to the destination, and thus it causes the broadcast storm problem. In this paper, we propose a neighbor coverage-based probabilistic rebroadcast protocol for reducing routing overhead in MANETs. In order to effectively exploit the neighbor coverage knowledge, we propose a novel rebroadcast delay to determine the rebroadcast order, and then we can obtain the more accurate additional coverage ratio by sensing neighbor coverage knowledge. We also define a connectivity factor to provide the node density adaptation. By combining the additional coverage ratio and connectivity factor, we set a reasonable rebroadcast probability. Our approach combines the advantages of the neighbor coverage knowledge and the probabilistic mechanism, which can significantly decrease the number of retransmissions so as to reduce the routing overhead, and can also improve the routing performance.

Recently, the hybrid Fiber-Wireless (FiWi) access network integrating passive optical networks (PONs) and wireless mesh networks (WMNs) has been proposed to provide the high bandwidth, low cost and ubiquitous Internet access. Typically, for the PON subnetwork of FiWi networks, a hybrid TDM/WDM architecture has been applied due to its cost efficiency and the high bandwidth provided by multiple wavelengths carried in the optical trunk. In such hybrid TDM/WDM FiWi networks, we aim to maximize the overall network throughput and meanwhile to find an appropriate wavelength assignment scheme of ONUs to achieve the approximate maximal throughput with the minimum number of wavelengths to be used. In this paper, we first obtain the maximal achievable throughput at ONUs and then get the minimal number of wavelength to be used. Given the number of wavelengths, we first use LP-based formulations to estimate the expected traffic load range at each ONU and then proposed the Approximate Wavelength Assignment(AWA) algorithm to calculate the appropriate traffic load at each ONU and obtain the proper wavelength assignment scheme of ONUs. We also verify the efficiency of the wavelength assignment scheme obtained through AWA algorithm. Simulation results strongly validate our algorithm.

Energy efficient asynchronous duty cycle MAC protocols are crucial to the success of wireless sensor networks (WSNs). In asynchronous protocols, each node operates its active/sleep schedule independently and enjoys a very low duty cycle when there is no traffic. However, when a node has data to send, it has to keep active and wait until the receiver wakes up due to the lack of schedule knowledge of the receiver. In low duty cycle networks, where nodes wake up infrequently, a sender usually suffers a long period of waiting, which consumes more energy than transmitting data packet itself. In this paper, we propose a new asynchronous MAC protocol, called OCMAC, which decreases the waiting time of sender by exploring opportunistic cooperation among senders. In OC-MAC, neighboring active senders are permitted to exchange data with each other aggressively when waiting for receivers to wake up. After delegating data to another sender, a sender can go to sleep before its receiver wakes up. Though this cooperation works only when there are multiple neighboring active senders, itself almost incurs no additional overhead. Simulation results and measurements on a testbed have shown that OC-MAC helps decrease idle listening, collision and end-to-end delay further.

In recent integrated MAC/routing solutions for wireless sensor networks (WSNs), hop-count is exploited to build a coarse-grained logical coordinates to help forward packets towards the direction of sink. This method can retain the merits of geographic routing at the absence of exact location knowledge. However, these solutions may present low energy-efficiency and unacceptable delays in real networks since they seldom consider the impacts of low duty-cycling and link unreliability on routing. Furthermore, geographic advancement of forwarding in hop- count based coordinates is very unreliable and even towards the reverse direction. In this work, average power cost to the sink of each node is considered together with hop-count to build a fine grained logical coordinates, which can help forward packets towards the direction of sink more accurately. Then we propose an energy-efficient integrated MAC/routing (EEMR) protocol for event-driven and time-critical applications based on new logical coordinates. The optimal relay is elected in each hop dynamically, where the objective is to optimize forwarding energy-efficiency on the premise that end-to-end delay is restricted under the predefined upper bound. Analysis and extensive simulations are given to demonstrate the superiority of EEMR by comparing its performance against existing solutions.

In this paper, we present a joint power control, link scheduling and rate control (PSR) algorithm for wireless ad hoc networks by using the convex optimization theory. This algorithm practically considers the power control problem in the interference-based link scheduling process, and provides a congestion control on the transport layer. Both our theoretical analysis and simulation results prove that the PSR algorithm can converge quickly and is possible for distributed implementation in the wireless ad hoc networks.