Efficient adversarial attack methods are critical for assessing the robustness of computer vision models.In this paper, we reconstruct the optimization objective for generating adversarial examples as "maximizing the difference between the non-true labels' probability upper bound and the true label's probability," and propose a gradient-based attack method termed Sequential Difference Maximization (SDM).SDM establishes a three-layer optimization framework of "cycle-stage-step." The processes between cycles and between iterative steps are respectively identical, while optimization stages differ in terms of loss functions: in the initial stage, the negative probability of the true label is used as the loss function to compress the solution space; in subsequent stages, we introduce the Directional Probability Difference Ratio (DPDR) loss function to gradually increase the non-true labels' probability upper bound by compressing the irrelevant labels' probabilities.Experiments demonstrate that compared with previous SOTA methods, SDM not only exhibits stronger attack performance but also achieves higher attack cost-effectiveness.Additionally, SDM can be combined with adversarial training methods to enhance their defensive effects.The code is available at https://github.com/X-L-Liu/SDM.

Due to the highly dynamic nature of UAV swarms, the ad-hoc network of UAV swarms usually experiences frequent changes, which poses significant challenges to maintaining persistent connectivity. Especially, the vulnerable nodes with a topological degree of one in the network topology of a UAV swarm, have the weakest network connectivity and are most susceptible to disconnection. To eliminate the vulnerable nodes, this paper proposes a midpoint guided topology control optimization method. This approach maintains the swarm’s steady-state operation while dynamically adjusting the positions of vulnerable nodes to increase their network topology degrees, thereby eliminating vulnerable nodes and enhancing the robustness and fault tolerance of the ad-hoc network. Furthermore, the validity of the proposed method is proved to be efficient through mathematical analysis, and the experimental results in three-dimensional space consistently demonstrate the superiority of the proposed approach.

Modern decentralized deep learning methods for network intrusion detection in Software-Defined Networking (SDN)-enabled Industrial Internet of Things (IIoT) environments encounter significant challenges, particularly for IIoT data heterogeneity and privacy leakage. To this end, we propose a novel framework for network intrusion detection, dubbed SFLNID, that improves Federated Learning (FL) to ensure both efficient training and privacy preservation in SDN-enabled IIoT. Specifically, we firstly design joint optimization mechanism for unbalanced and non-IID data, which introduces a Focal loss as the loss function, and leverages the Wasserstein distance between global and local models as the regularization term. In addition, we improve adaptive differential privacy with dynamic gradient clipping techniques, adjusting the clip-threshold based on Holt exponential smoothing to achieve privacy protection during the local model training. Moreover, we develop a customized CNN-GRU model tailored for FL-based network intrusion detection to make a tradeoff between model accuracy and overheads. Theoretical analysis confirms the convergence and privacy guarantees of SFLNID. Extensive experiments, conducted on well-known IIoT datasets including ToN-IoT, RT-IoT and Edge-IIoT, demonstrate that SFLNID outperforms the state-of-the-art methods in terms of detection accuracy, communication overhead, and cooperative privacy preservation.

Neural networks are vulnerable to meticulously crafted adversarial examples, resulting in high-confidence misclassifications in image classification tasks. Due to their stealthiness and difficulty in detection, black-box transfer attacks have become a significant focus of defense. In this article, we propose a purification defense based on probabilistic scheduling algorithm of pre-trained models (ProbSched-PTM) to counter diverse transfer attacks. We first quantify the differences among various models based on their output scores and verify the linear negative correlation between adversarial transferability and model difference. Subsequently, guided by the model difference probability, we integrate the negative momentum probability as a regularization factor to construct ProbSched-PTM. It selects the most appropriate substitute model from multiple pre-trained models to generate strong-transferability adversarial examples for training the purification model, which enables the purification model to effectively eliminate diverse adversarial perturbations. The ProbSched-PTM-based purification defense provides robust defense against unseen adversarial attacks from different substitute models. In a black-box attack scenario, utilizing ResNet-34 as the target model, our approach achieves average defense rates of over 94.8% on CIFAR-10 and over 71.2% on Mini-ImageNet, demonstrating state-of-the-art performance.

Due to the high mobility of unmanned aerial vehicles (UAVs), the network topology may change frequently, making persistent connectivity and fault tolerance difficult. Deep reinforcement learning (DRL) offers the opportunity to make proper actions in a large decision space, which could be utilized for the complicated topology control of flying ad-hoc networks. However, how to train and deploy the DRL algorithms on resource-limited and hardware-constrained UAVs to ensure network connectivity and fault tolerance still faces huge challenges. In this work, a topology control method based on positional movement and DRL is proposed, which is suitable for topology construction and topology adjustment. First, a centroid-guided target-driven method is designed to transform arbitrary graphs into 2-connected graphs by connecting each node with its two designated target nodes in a specific order. Then, a topology control method based on the centroid-guided target-driven method and soft actor-critic (CGTD-SAC) is proposed. CGTD-SAC trains agents to keep connectivity with two target agents and keep a safe distance from surrounding agents. CGTD-SAC generates 2-connected topologies in a distributed manner. CGTD-SAC is a tiny algorithm with low computational complexity and less communication overhead. Finally, experiments demonstrate that CGTD-SAC has an excellent ability to obtain network topologies with 2-connectivity, suitable link length, and appropriate number of links.

As an important part of intrusion detection, feature selection plays a significant role in improving the performance of intrusion detection. Krill herd (KH) algorithm is an efficient swarm intelligence algorithm with excellent performance in data mining. To solve the problem of low efficiency and high false positive rate in intrusion detection caused by increasing high-dimensional data, an improved krill swarm algorithm based on linear nearest neighbor lasso step (LNNLS-KH) is proposed for feature selection of network intrusion detection. The number of selected features and classification accuracy are introduced into fitness evaluation function of LNNLS-KH algorithm, and the physical diffusion motion of the krill individuals is transformed by a nonlinear method. Meanwhile, the linear nearest neighbor lasso step optimization is performed on the updated krill herd position in order to derive the global optimal solution. Experiments show that the LNNLS-KH algorithm retains 7 features in NSL-KDD dataset and 10.2 features in CICIDS2017 dataset on average, which effectively eliminates redundant features while ensuring high detection accuracy. Compared with the CMPSO, ACO, KH, and IKH algorithms, it reduces features by 44%, 42.86%, 34.88%, and 24.32% in NSL-KDD dataset, and 57.85%, 52.34%, 27.14%, and 25% in CICIDS2017 dataset, respectively. The classification accuracy increased by 10.03% and 5.39%, and the detection rate increased by 8.63% and 5.45%. Time of intrusion detection decreased by 12.41% and 4.03% on average. Furthermore, LNNLS-KH algorithm quickly jumps out of the local optimal solution and shows good performance in the optimal fitness iteration curve, convergence speed, and false positive rate of detection.

Network functional combination is a promising direction in enhancing Internet adaptability. It decomposes the current layered network into fine-grained building blocks and combines them on demand. However, what legacy functions should be decomposed and how to combine them in an optimal way are unclear. We propose a novel adaptive architecture called reconstructive network architecture (RECON) based on the principles of the Complex Adaptive System. This study has three main contributions. First, RECON decomposes functions of the protocol stack at layers 3 and 4 into fine-grained building blocks, called atomic capabilities to open the network core functions unlike existing solutions. Second, RECON can customize different service chains on demand by combining atomic capabilities in an optimal way. We formulate the atomic capability combination into a nonlinear integer optimization problem with the proposed algorithm to reach an appropriate tradeoff between the optimal solution and computation cost. Finally, we implement a proof-of-concept for RECON in the network node. Results are corroborated by several numerical simulations.

How to provide quality of service (QoS) guarantees in the routing and switching devices has become one of the key research topics in the areas of routing and switching technologies. Differentiated services architecture (DiffServ), which forwards packets based on their per-hop behaviours, is known as a promising way for supporting QoS in a high-speed backbone network scenario. Motivated by the load-balanced Birkhoff-von Neumann switch, which has received much attention recently because of its outstanding extensibility performance, in this study, the authors propose a new switch architecture called load-balanced DiffServ support switch (LBDS). LBDS consists of pipeline traffic bandwidth control, priority-based load balancing, and bandwidth-based priority scheduling to implement fast forwarding for expedited forwarding (EF) traffic and assured forwarding for AF classes of traffic. The authors evaluate LBDS by extensive theoretical analysis and comprehensive simulations. As expected, LBDS can provide bandwidth guarantees for both EF and AF traffic. We also show that under burst traffic arrivals, especially when the traffic arrival pattern is non-uniform and the load is above 0.6, LBDS performs much better delay performance than existing DiffServ support schemes.

A novel two stage load-balanced multipath self-routing switch structure is introduced in this paper. Both stages use a multipath self-routing fabric. With simple algorithms and small buffers, the first stage fabric transforms the incoming traffic into uniform and the second stage fabric forwards the data in a self-routing manner to their final destinations. Compared with other similar structures, this structure outstands with no queuing delay and zero jitter, its component complexity and propagation delay are significantly reduced. Mathematical analysis and simulations show this structure can achieve 100% throughput under admissible traffic pattern, which is a common presumption for incoming traffic. For statistically admissible traffic, by stacking up a few copies of this structure, it is suitable to support QoS application for building super large scale switching fabric in next generation network(NGN).

The combined input-crosspoint-queued (CICQ) crossbar switch is very appealing because it can obtain high throughput with simple scheduling mechanisms. However, in order to support multiple priority levels, separate queues per priority are required at each crosspoint, hence there needs much more memories and many priority schedulers to be implemented in a buffered crossbar, which is of great complexity. In this paper we propose a scheme that uses a hierarchical priority queuing mechanism in the input queues and a simple queue per crosspoint to effectively support multiple priorities. We present a priority weighted double round robin (PWDRR) scheduling algorithm in input scheduler to implement bandwidth allocation among multiple priorities and a simple compensation priority round robin (CPRR) scheduling policy in crosspoint scheduler to transfer cells to the output. The simulation results verify a preferable performance of our scheme.

Though output-queued (OQ) switch can provide quality-of-service (QoS) guarantees, the high-speed memory requirements limit its use for large capacity switching architecture. A recent result indicates that a buffered crossbar switch with three times speedup can perfectly emulate an OQ switch. However, to achieve a speedup factor of three, a conventional buffered crossbar switch requires the switch fabric and the memory to operate three times faster than the line rate. In this paper, we propose a buffered crossbar switch with space- division multiplexing expansion (denoted as SDM CICOQ switch) which only requires the switch fabric and the memory to operate at the line rate. Using fluid model, we prove that an SDM CICOQ switch with an expansion factor 2 can achieve 100% throughput for any admissible traffic. It is also proved that an SDM CICOQ switch with an expansion factor 2 can exactly emulate an OQ switch. All the scheduling algorithms used in emulating OQ switch are distributed, thus are practical. Finally, for the simplicity of implementation, we present a hierarchical priority scheduling (HPS) scheme for the SDM CICOQ switch. Simulation results show that the SDM CICOQ switch operating under the HPS scheduling scheme can obtain good performance.

The parallel packet switch (PPS) attracts a lot of attention from the communications equipment vendors. However, a lack of the detailed analysis and understanding of the issues involved in congestion management has slowed down its recognition and deployment. We analyze the conditions of stable working for a PPS with buffered input demultiplexors. By comparing a PPS with a reference switch, we present a definition of stable working for the PPS architecture. The necessary and sufficient condition of stable working for a PPS is proposed and proved. Then we describe a family of dispatch algorithms for the PPS, and give the restriction of algorithms for guaranteeing stable working. Finally the minimum bound of PPS layers and core speedup are analyzed.

While several analytical studies of parallel packet switch (PPS) are presented, the findings publicized can not be considered final. In this paper, the necessary and sufficient conditions of the stability for parallel packet switch with bufferless input de-multiplexers are analyzed. We firstly introduce a reference switch and inventory theory, and based upon aggravating no congestion of the switch structure, a definition of PPS stability is proposed. Then the necessary and sufficient conditions of the stability for the PPS and algorithms for traffic dispatch are analyzed. The minimum of the PPS layers and speedup are also given.