Accurately detecting super host that establishes connections to a large number of distinct peers is significant for mitigating web attacks and ensuring high quality of web service. Existing sketch-based approaches estimate the number of distinct connections called flow cardinality according to full IP addresses, while ignoring the fact that a malicious or victim super host often communicates with hosts within the same subnet, resulting in high false positive rates and low accuracy. Though hierarchical-structure based approaches could capture flow cardinality in subnet, they inherently suffer from high memory usage. To address these limitations, we propose SegSketch, a segmented cardinality estimation approach that employs a lightweight halved-segment hashing strategy to infer common prefix lengths of IP addresses, and estimates cardinality within subnet to enhance detection accuracy under constrained memory size. Experiments driven by real-world traces demonstrate that, SegSketch improves F1-Score by up to 8.04× compared to state-of-the-art solutions, particularly under small memory budgets.

The Industrial Internet of Things (IIoT) leverages Federated Learning (FL) for distributed model training while preserving data privacy, and meta-computing enhances FL by optimizing and integrating distributed computing resources, improving efficiency and scalability. Efficient IIoT operations require a trade-off between model quality and training latency. Consequently, a primary challenge of FL in IIoT is to optimize overall system performance by balancing model quality and training latency. This paper designs a satisfaction function that accounts for data size, Age of Information (AoI), and training latency for meta-computing. Additionally, the satisfaction function is incorporated into the utility function to incentivize IIoT nodes to participate in model training. We model the utility functions of servers and nodes as a two-stage Stackelberg game and employ a deep reinforcement learning approach to learn the Stackelberg equilibrium. This approach ensures balanced rewards and enhances the applicability of the incentive scheme for IIoT. Simulation results demonstrate that, under the same budget constraints, the proposed incentive scheme improves utility by at least 23.7% compared to existing FL schemes without compromising model accuracy.

Sketch has gained wide deployment and application for approximate flow estimation, because of its ability to maintain good accuracy and high throughput with limited memory resources. However, most existing sketch approaches ignore the distinctions between flow priorities, though the high-priority flows are relatively scarce but hold significant information. Therefore, a class of priority-aware sketches has appeared recently to provide differentiated measurement accuracy for flows with different priorities. Unfortunately, it is challenging for these priority-aware sketches to strike a good balance between accuracy and throughput. To address this issue, we propose a priority-adaptive architecture PA-Sketch, which utilizes priority-aware hash to dynamically allocate appropriate numbers of hash functions for different flows according to their priorities. Moreover, to further achieve good accuracy in extremely small memory space, we introduce the priority-aware sampling into PA-Sketch. The test results show that PA-Sketch significantly improves accuracy while minimizing the hash overhead. Compared to the state-of-the-art priority-aware sketches, PA-Sketch reduces the ARE of high-priority flows by 86% and improves the F1 score by 1.83 times, meanwhile maintaining slight accuracy loss for low-priority flows.

Synchronous stochastic gradient descent (SSGD) has been widely used in distributed deep learning. However, since the local gradients need to be shared among workers at every iteration, SSGD performance is significantly influenced by network bottlenecks caused by either heterogeneous environment or bandwidth contention. To solve this problem, asynchronous parallel (ASP) strategy allows each worker to update parameters independently without synchronization, while suffering from accuracy loss and convergence inefficiency. In this paper, we propose a novel similarity-based synchronization scheme called SIM, which mitigates the impact of network bottlenecks and ensures convergence efficiency. Specifically, SIM reduces the number of aggregation workers based on the gradient similarity between global and local gradients, therefore shrinking the waiting time for the stragglers. We provide a theoretical analysis of convergence efficiency and conduct large-scale testbed experiments on CIFAR-10 and SQUAD dataset. The experimental results show that SIM reduces the convergence time of four classical deep learning models by up to 40%.

Quantiles over data stream have been recognized to be an essential feature in network traffic. To provide accurate estimation results, current quantile estimation approaches are preconfigured according to traffic distributions such as log-normal and Pareto distributions. In most practical applications, however, the traffic distributions are not known a priori or are highly dynamic, disturbing estimation results. In this paper, we propose SplitSketch, a sketch-based mechanism that aims to accurately estimate quantiles over data stream without any prior knowledge of traffic distributions. SplitSketch adjusts its estimation granularity according to the changing process of the traffic distribution. The granularities with denser distribution will be recorded with the finer granularity to provide more accurate estimation results. Experimental results demonstrate that, compared to existing approaches, SplitSketch reduces the absolute error in quantile estimation by 59.1% for heavy-tailed distributions and 79.2% for general distributions.

With the rapid growth scale of dataset and model, the training of deep neural networks (DNN) tends to be deployed in a distributed manner. In the large-scale distributed training, the bottlenecks have gradually moved from computational resources to communication process. Recent researches adopt in-network aggregation (INA) that offloads the gradient aggregation process to programmable switches, thereby reducing network traffic amount and transmission latency. Unfortunately, due to the bandwidth competition in shared training clusters, the straggler will slow down the training efficiency of INA. To address this issue, we propose an Asynchronous Control based Aggregation Transport Protocol (AC-ATP), which makes full use uncongested links to transmit gradients and the switch memory to cache gradients from the fast workers to accelerate the gradient aggregation. Meanwhile, AC-ATP performs congestion control according to the transmission progress of worker and the remaining completion time of the job. The evaluation results of real testbed and large-scale simulations show that AC-ATP reduces the aggregate time by up to 68% and speeds up training in real-world benchmark models.

Each output port in a shared memory switch can compete for shared memory pool resources. The allocation strategy of the shared buffer directly affects the ability of each output port to absorb network traffic. Due to the unpredictability of traditional network traffic, existing switch buffer management strategies take a passive approach, allocating buffers to each port only after traffic arrives. This passive response has the problem of untimely buffer allocation and cannot effectively absorb burst traffic. Distributed deep learning follows a specific training pattern, and network traffic exhibits obvious periodic characteristics during transmission. Thus, we propose a Proactive Dynamic Threshold (PDT) strategy, which realizes the pre-adjustment of switch port threshold by detecting the traffic characteristics of distributed training.

By using the intelligent edge computing technologies, a large number of computing tasks of end devices in Industrial Internet of Things (IIoT) can be offloaded to edge servers, which can effectively alleviate the burden and enhance the performance of IIoT. However, in large-scale multi-service-oriented IIoT scenarios, offloading service resources are heterogeneous and offloading requirements are mutually exclusive and time-varying, which reduce the offloading efficiency. In this paper, we propose a cloud-edge-end collaboration intelligent service computation offloading scheme based on Digital Twin (DT) driven Edge Coalition Formation (DECF) approach to improve the offloading efficiency and the total utility of edge servers, respectively. Firstly, we establish a DT model to obtain accurate digital representations of heterogeneous end devices and network state parameters in dynamic and complex IIoT scenarios. The DT model can capture time-varying requirements in a low latency manner. Secondly, we formulate two optimization problems to maximize the offloading throughput and total system utility. Finally, we convert the multi-objective optimization problems to a Stackelberg coalition game model and develop a distributed coalition formation approach to balance the two optimizing objectives. Simulation results indicate that, compared with the nearest coalition scheme and non-coalition scheme, the proposed approach achieves offloading throughput improvements of 11.5% and 148%, and enhances the overall utility by 12% and 170%, respectively.

In recent years, motivated by new datacenter applications and the well-known shortcomings of TCP in data center, many receiver-driven transport protocols have been proposed to provide ultra-low latency and zero packet loss by using the proactive congestion control. However, in the scenario of mixed short and long flows, the short flows with ON/OFF pattern generate micro-burst traffic, which significantly deteriorates the performance of existing receiver-driven transport protocols. Firstly, when the short flows turn into ON mode, the long flows cannot immediately concede bandwidth to the short ones, resulting in queue buildup and even packet loss. Secondly, when the short flows change from ON to OFF mode, the released bandwidth cannot be fully utilized by the long flows, leading to serious bandwidth waste. To address these issues, we propose a new receiver-driven transport protocol, called SAR, which predicts the micro burst generated by short flows and adjusts the sending rate of long flows accordingly. With the aid of micro-burst prediction mechanism, SAR mitigates the bandwidth competition due to the arrival of short flows, and alleviates the bandwidth waste when the short flows leave. The testbed and NS2 simulation experiments demonstrate that SAR reduces the average flow completion time (AFCT) by up to 66% compared to typical receiver-driven transport protocols.

In order to cope with the fluctuation of network bandwidth and provide smooth video services, adaptive video streaming technology is proposed. In particular, the adaptive bitrate (ABR) algorithm is widely used in dynamic adaptive streaming over HTTP (DASH) to improve quality of experience (QoE). However, existing ABR algorithms still ignore the inherent visual sensitivity of human visual system (HVS). As the final receiver of video, HVS has different sensitivity to the quality distortion of different video content, and video content with high visual sensitivity needs to allocate more bitrate resources. Therefore, existing ABR algorithms still have limitations in reasonably allocating bitrate and maximizing QoE. To solve this problem, this paper designs an adaptive bitrate strategy from the perspective of user vision, studies the modeling of visual sensitivity, and proposes a visual sensitivity aware ABR algorithm. We extract a set of content features and attribute features from the video, and consider the simulation of HVS to establish a total masking effect model that reflects the visual sensitivity more accurately. Further, the network status, buffer occupancy, and visual sensitivity are comprehensively considered under a deep reinforcement learning framework to select the appropriate bitrate for maximizing QoE. We implement the proposed algorithm over a realistic trace-driven evaluation and compare its performance with several latest algorithms. Experimental results show that our algorithm can align ABR strategy with visual sensitivity to achieve better QoE in high visual sensitivity content, and improves the average perceptual video quality and overall user QoE by 18.3% and 22.8%, respectively. Additionally, we prove the feasibility of our algorithm through subjective evaluation in the real environment.

To reduce the communication overhead in distributed sketch system, it is desirable to compress sketches before uploading. However, current sketch compression approaches hardly achieve high speed of compression procedure and low error of compressed sketches at the same time. In this paper, we take a clean slate approach to design a sketch compression scheme called as Fast-Mapping that achieves both fast compression speed and high accuracy. Based on the prior statistics knowledge of bucket data distribution, Fast-Mapping compresses the similar buckets to obtain high accuracy. We also theoretically derive the compression error bound of Fast-Mapping. The experimental results show that, Fast-Mapping achieves higher speed and lower error than the-state-of-art works.

Most existing multi-unmanned aerial vehicle (multi-UAV) systems focus on fly path or energy consumption for task assignment, while little attention has been paid to the dynamic feature of the task, resulting in poor task completion ratio. The machine learning (ML) paradigm provides new methodologies for task assignment. However, ML methods are usually of heavy resource-consumption that cannot be directly applied in the UAV. In this paper, a digital twin (DT) assisted task assignment approach is proposed to improve the resource-intensive utilization and the efficiency of deep reinforcement learning (DRL) in multi-UAV system. The approach has a three-layer network structure which can dynamically assign tasks based on the task time constraints. Moreover, the approach is divided into two stages of initial task-assignment and task-reassignment. In the first stage, airship divides a task into multiple subtasks according to the shortest distance based on genetic algorithm and assigns them to UAVs. In the second stage, the DT can be leveraged to enable the airships to learn from the features of tasks and to generate the Q-value of the estimated value network of DRL for UAVs via pre-train of DT. The Q-value can be directly applied for deep Q-learning network (DQN) in the UAVs to reduce the training episode. Furthermore, the DQN is adopted to train task-reassignment strategy. Simulation results indicate that the DQN with DT can significantly reduce the training episode, improving 30% of the task completion ratio and 19% of the system energy efficiency compared with that of the baseline methods.

Identifying heavy flows is essential for network management. However, it is challenging to detect heavy flow quickly and accurately under the highly dynamic traffic and rapid growth of network capacity. Existing heavy flow detection schemes can make a trade-off in efficiency, accuracy and speed. However, these schemes still require memory large enough to obtain acceptable performance. To address this issue, we propose ChainSketch, which has the advantages of good memory efficiency, high accuracy and fast detection. Specifically, ChainSketch uses the selective replacement strategy to mitigate the over-estimation issue. Meanwhile, ChainSketch utilizes the hash chain and compact structure to improve memory efficiency. We implement the ChainSketch on OVS platform, P4-based testbed and large-scale simulations to process heavy hitter and heavy changer detection. The results of trace-driven tests show that, ChainSketch greatly improves the F1-score by up to$3.43\times $compared with the state-of-the-art solutions especially for small memory.

Assay of single cell transposase-accessible chromatin with high-throughput sequencing (scATAC-seq) have enabled massively profiling of the chromatin accessibility landscape at the single-cell level. The essential step in analyzing scATAC-seq data is to cluster the cells into different clusters and utilize the clustering information in the subsequent downstream analysis. However, there are some challenges in the clustering analysis of scATAC-seq data. For example, scATAC-seq data are often high-dimensional and extremely sparse, as well as featuring high loss rate or noise. In this study, we proposed the scIAC to address these challenges of scATACseq data. In particular, scIAC combines the Student’s t-distribution similarity imputation and the denoising autoencoder based on the Zero-inflated Negative Binomial (ZINB) distribution. The Student’s t-distribution similarity imputation is used to solve the problem of high sparsity and high loss rate. The denoising autoencoder is employ to extract features which are useful for clustering and to reduce data noises. In addition, the self-training soft K-means and pairwise constraints are utilized in the clustering phase to enhance clustering performance. The experimental validation on several datasets shows that the proposed method performed better than other state-of-the-art methods. In conclusion, scIAC is an effective method to accurately cluster and identify cell types in scATAC-seq data.

Heavy flow detection in enormous network traffic is a critical task for network measurement. Due to the limited memory size and high link capacity, accurate detection of heavy flows becomes challenging in large-scale networks. Almost all existing approaches of detecting heavy flows use single-dimension statistics of flow size to make flow-replacement decisions. However, under the mass number of small flows, the heavy flows are prone to be frequently and mistakenly replaced, resulting in unsatisfactory accuracy. To solve this problem, we reveal that the number of uninterrupted arrival packets is a useful metric in identifying flow types. We further propose UA-Sketch that expels small flows and protects heavy ones according to the multiple-dimension statistics including both estimated flow size and number of uninterrupted arrival packets. The test results of trace-driven simulations and OVS experiments show that, even under small memory, UA-Sketch achieves higher accuracy than the existing works, with the F1 Score by up to 2.1 ×.

Client-side adaptive bitrate (ABR) algorithms based on deep reinforcement learning (RL) can continuously improve its adaptability to network conditions. However, most existing methods adopt fixed reward functions to train the ABR policy, which leads the results being not consistent with user-perceived quality of experience (QoE) in a long duration under various network conditions. In order to optimize the QoE, this paper proposes a novel ABR algorithm considering user preference based on short trajectory segments. The user-specific preference feedback, which is selected by the user from a pair of short track segments in advance, is collected and applied to define the training goal of RL. Specifically, we train a deep neural network to define the RL reward and integrate it with A3C-based ABR algorithm. The experiment results show that the accuracy of the proposed reward model outperforms most existing fixed reward functions by 13.6% in user preference prediction, and the optimized ABR algorithm improves QoE by 16.4% on average.

Simultaneous wireless information and power transfer (SWIPT) becomes more and more popular in cognitive radio (CR) networks, as it can increase the resource reuse rate of the system and extend the user’s lifetime. Due to the deployment of energy harvesting nodes, traditional secure beamforming designs are not suitable for SWIPT-enabled CR networks as the power control and energy allocation should be considered. To address this problem, a dedicated green edge power grid is built to realize energy sharing between the primary base stations (PBSs) and cognitive base stations (CBSs) in SWIPT-enabled mobile edge computing (MEC) systems with CR. The energy and computing resource optimal allocation problem is formulated under the constraints of security, energy harvesting, power transfer, and tolerable interference. As the problem is nonconvex with probabilistic constraints, approximations based on generalized Bernstein-type inequalities are adopted to transform the problem into solvable forms. Then, a robust and secure artificial noise- (AN-) aided beamforming algorithm is presented to minimize the total transmit power of the CBS. Simulation results demonstrate that the algorithm achieves a close-to-optimal performance. In addition, the robust and secure AN-aided CR based on SWIPT with green energy sharing is shown to require a lower transmit power compared with traditional systems.

The major challenge of Real Time Protocol is to balance efficiency and fairness over limited bandwidth. MPTCP has proved to be effective for multimedia and real time networks. Ideally, an MPTCP sender should couple the subflows sharing the bottleneck link to provide TCP friendliness. However, existing shared bottleneck detection scheme either utilize end-to-end delay without consideration of multiple bottleneck scenario, or identify subflows on switch at the expense of operation overhead. In this paper, we propose a lightweight yet accurate approach, EMPTCP, to detect shared bottleneck. EMPTCP uses the widely deployed ECN scheme to capture the real congestion state of shared bottleneck, while at the same time can be transparently utilized by various enhanced MPTCP protocols. Through theory analysis, simulation test and real network experiment, we show that EMPTCP achieves higher than 90% accuracy in shared bottleneck detection, thus improving the network efficiency and fairness.

The fair allocation of the resources among different nodes is one of the critical problems in wireless mesh networks. Existing solutions mainly focus on rate-limitation policies or distributed fair MAC schemes at the potential expense of total network utilization. This paper investigates a special starvation problem among TCP flows that are different hops away from the BS, as well as the recently proposed solution, the "Minimum Content Window" policy based on IEEE 802.11e. It is found that the aggregate throughput degrades sharply because the effect of this policy on the TCP congestion mechanism has been overlooked. This paper proposes a priority-based congestion control by using "Cross-Layer Explicit Congestion Notification". Analysis and simulation results demonstrate that our scheme can improve the fairness of TCP flows while the aggregate throughput is at least 20% higher than the "Minimum Content Window" policy.

Many of packet loss as a result of factors other than congestion impact the performance of TCP in wired/wirelss hybrid networks. Firstly, this paper proposes one concept of congestion probability (CP) and analyzes the correlation of CP and network state. Then a transport control protocol named as TCP-PCP is proposed, which is based on the prediction of congestion probability instead of single loss event. Depending on the ECN mechanism, TCP-PCP calculates the value of CP by analyzing some latest loss events. The TCP-PCP sender's behavior is controlled by the change of CP value. Simulation results show that TCP-PCP can improve TCP performance more efficiently than Westwood and Jersey.