Although AI has been extensively adopted and has profoundly transformed our lives, it is not feasible to directly deploy large AI models on edge devices with limited resources. To enhance the performance of Edge Intelligence (EI), model split inference has been proposed. In this approach, an AI model is segmented into sub-models, with the most resource-intensive parts offloaded wirelessly to the edge server. This reduces the resource demands and inference latency on the device. However, previous studies have primarily focused on enhancing and optimizing system Quality of Service (QoS), often overlooking Quality of Experience (QoE), which is another crucial aspect for users. Even though QoE has been extensively studied in Edge Computing (EC), the distinct differences between task offloading in EC and split inference in EI, along with specific QoE issues that remain unaddressed in both fields, render these algorithms ineffective for edge split inference scenarios. Therefore, this paper introduces an effective resource allocation algorithm, dubbed ERA, which aims to: 1) expedite split inference in EI, and 2) balance inference delay, QoE, and resource consumption. ERA incorporates resource consumption, QoE, and inference latency to determine the most optimal model split and resource allocation strategies. Given that it is impossible to simultaneously minimize inference delay and resource consumption while maximizing QoE, we employ a gradient descent-based algorithm to find the best possible compromise. Furthermore, to address the complexity arising from parameter discretization in the gradient descent algorithm, we have developed a pipeline gradient descent approach, known as PipGD. We have also examined the properties of the proposed algorithms, including their convergence, complexity, and approximation error. The experimental results clearly show that ERA outperforms previous studies significantly in terms of performance.

Abstract In the pursuit of advancing computer vision, this manuscript addresses the complex challenge of underwater object detection through a novel YOLOv9-Side scan Network (YOLOv9-SN) model. The research encapsulates the integration of Negative Sample Refinement, Attention and Convolution mix strategy, and Spatial and Channel reconstruction Convolution convolutional layers, thus enhances the model’s discriminative learning and efficiency. The incorporation of Bidirectional Feature Pyramid Network and Multi-Path Distance Intersectionover Union metrics significantly improves the performance of feature integration and object localization. Comparative analysis with established models such as Faster region-based convolutional neural network, DEtection TRansformer, and YOLOv5, along with rigorous ablation studies, demonstrates the superiority of the proposed YOLOv9-SN model. Evaluating on the zero-shot learning-sonar submarine simulation dataset, this model achieves a [email protected]:0.95 of 72.1%, surpassing the baseline of YOLOv9 by 3.5%. This research contributes to the enhancement of detection metrics and the advancement of side scan sonar imaging for underwater targets, emphasizing the model’s high precision and accuracy in underwater target detection.

The edge intelligence (EI) has been widely applied recently. Splitting the model between device, edge server, and cloud can significantly improve the performance of EI. The model segmentation without user mobility has been investigated in detail in previous studies. However, in most EI use cases, the end devices are mobile. Few studies have been conducted on this topic. These works still have many issues, such as ignoring the energy consumption of mobile device, inappropriate network assumption, and low effectiveness on adapting user mobility, etc. Therefore, to address the disadvantages of model segmentation and resource allocation in previous studies, we propose mobility and cost aware model segmentation and resource allocation algorithm for accelerating the inference at edge (MCSA). Specifically, in the scenario without user mobility, the loop iteration gradient descent (Li-GD) algorithm is provided. When the mobile user has a large model inference task that needs to be calculated, it will take the energy consumption of mobile user, the communication and computing resource renting cost, and the inference delay into account to find the optimal model segmentation and resource allocation strategy. In the scenario with user mobility, the mobility aware Li-GD (MLi-GD) algorithm is proposed to calculate the optimal strategy. Then, the properties of the proposed algorithms are investigated, including convergence, complexity, and approximation ratio. The experimental results demonstrate the effectiveness of the proposed algorithms.

DNS resolvers are an important measurement targets in the IPv4/IPv6 Internet for cybersecurity and network management. However, due to the vast address space of IPv6, it is infeasible to discover IPv6 DNS resolvers using brute-force Internet-wide scanning as in IPv4. To address this issue, researchers have developed target generation algorithms (TGAs) to discover active targets in the IPv6 address space. However, most TGAs utilize ICMP as the probing protocol, and depend on large, high-quality ICMP seed address datasets. When the same TGA methods are applied to the UDP/53 protocol, which has a limited number of seed addresses, the efficiency of discovering DNS resolvers is low. To solve this issue, we developed 6Hound to efficiently discover DNS resolvers in the IPv6 Internet. To mitigate the scarcity of UDP/53 seed addresses, we proposed the Pattern-merged Tree, which strategically expands the scanning space by utilizing ICMP seed addresses. To efficiently discover de-aliased active addresses within these merged patterns, we proposed a hierarchical multi-armed bandit to control the distribution of probe packets. We introduced the Sliced Address Generation algorithm and a dynamic alias detection mechanism to enhance the hit rate of each detection round and avoid the misleading effects of aliased addresses. In the experiments conducted in the native IPv6 Internet, we discovered about a million de-aliased active DNS resolver addresses under a budget scale of 50M, which is 110% to 465% higher than the state-of-the-art baseline methods.

Wireless Power Transfer technology, which can charge IoT devices over the air, has become a promising technology for IoT networks. In wireless-powered IoT networks, broadcasting is a fundamental networking service for disseminating messages to the whole network. To seek a fast and collision-free broadcast schedule, the problem of Minimum Latency Broadcast Scheduling (MLBS) has been well studied when nodes are energy-abundant. However, in wireless-powered networks, a node can only receive or transmit packets after it has harvested enough energy. In such networks, it is of great importance to exploit the divergent harvested energy to reduce the broadcast latency. Unfortunately, existing works always assume a predetermined tree and a fixed transmission power for broadcast scheduling, which greatly limits their performance. Thus, in this article, we investigate the first work for the MLBS problem in wireless-powered networks without relying on predetermined trees. First, the problem is formulated and proved to be NP-hard. Then, two structure-adaptive scheduling algorithms are proposed with a theoretical bound, which can intertwine the construction of broadcast tree with the computation of an energy-aware schedule simultaneously. Furthermore, a power-aware scheduling method is also proposed to take the structure of the broadcast tree, the adjustment of nodes’ transmission powers, and the interference during transmissions into account simultaneously. Additionally, the algorithm for the MLBS problem under the physical interference model is also studied. Finally, the theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of latency.

Abstract Robust and fast image recognition and matching is an important task in the underwater domain. The primary focus of this work is on extracting subsea features with sonar sensor for further Autonomous Underwater Vehicle navigation, such as the robotic localization and landmark mapping applications. With the assistance of high-resolution underwater features in the Side Scan Sonar (SSS) images, an efficient feature detector and descriptor, Speeded Up Robust Feature, is employed to seabed sonar image fusion task. In order to solve the nonlinear intensity difference problem in SSS images, the main novelty of this work is the proposed Underwater Wireless Sensor Network-based Delaunay Triangulation (UWSN-DT) algorithm for improving the performances of sonar map fusion accuracy with low computational complexity, in which the wireless nodes are considered as underwater feature points, since nodes could provide sufficiently useful information for the underwater map fusion, such as the location. In the simulated experiments, it shows that the presented UWSN-DT approach works efficiently and robustly, especially for the subsea environments where there are few distinguishable feature points.

As many countries have promulgated laws to protect users’ data privacy, how to legally use users’ data has become a hot topic. With the emergence of federated learning (FL) (also known as collaborative learning), multiple participants can create a common, robust, and secure machine learning model while addressing key issues in data sharing, such as privacy, security, accessibility, etc. Unfortunately, existing research shows that FL is not as secure as it claims, gradient leakage and the correctness of aggregation results are still key problems. Recently, some scholars try to address these security problems in FL by cryptography and verification techniques. However, there are some issues in this scheme that remain unsolved. First, some solutions cannot guarantee the correctness of the aggregation results. Second, existing state-of-the-art FL schemes have a costly computational and communication overhead. In this article, we propose SVFLC, a secure and verifiable FL scheme with chain aggregation to solve these problems. We first design a privacy-preserving method that can solve the problem of gradient leakage and defend against collusion attacks by semi-honest users. Then, we create a verifiable method based on a homomorphic hash function, which can ensure the correctness of the weighted aggregation results. Besides, the SVFLC can also track users who encounter calculation errors during the aggregation process. Additionally, the extensive experiment results on real-world data sets demonstrate that the SVFLC is efficient, compared with other solutions.

Even the artificial intelligence (AI) has been widely used and significantly changed our life, deploying the large AI models on resource limited edge devices directly is not appropriate. Thus, the model split inference is proposed to improve the performance of edge intelligence (EI), in which the AI model is divided into different sub-models and the resource-intensive sub-model is offloaded to edge server wirelessly for reducing resource requirements and inference latency. Unfortunately, with the sharp increasing of edge devices, the shortage of spectrum resource in edge network becomes seriously in recent years, which limits the performance improvement of EI. Refer to the NOMA-based edge computing (EC), integrating non-orthogonal multiple access (NOMA) technology with split inference in EI is attractive. However, the NOMA-based communication aspect and the influence of intermediate data transmission fail to be considered properly in model split inference of EI in previous works, and the sophistication in resource allocation caused by NOMA scheme makes it further complicated. Thus, the Effective Communication and Computing resource allocation algorithm is proposed in this paper for accelerating the split inference in NOMA-based EI, shorted as ECC. Specifically, the ECC takes the energy consumption and the inference latency into account to find the optimal model split strategy and resource allocation strategy (subchannel, transmission power, computing resource). Since the minimum inference delay and energy consumption cannot be satisfied simultaneously, the gradient descent (GD) based algorithm is adopted to find the optimal tradeoff between them. Moreover, the loop iteration GD approach (Li-GD) is developed to reduce the complexity of the GD algorithm caused by parameter discretization. The key idea of Li-GD is that: the initial value of the$i\mathrm {th}$layer’s GD procedure is selected from the optimal results of the former$(i-1)$layers’ GD procedure whose intermediate data size is the closest to$i\mathrm {th}$layer. Additionally, the properties of the proposed algorithms are investigated, including convergence, complexity, and approximation error. The experimental results demonstrate that the performance of ECC is much better than that of the previous studies.

Abstract The sketch is one of the typical and widely used data structures for estimating the frequencies of items in data streams. However, since the counter sizes in traditional rectangular sketch (r-sketch) are the same, it is hard to achieve small space usage, high capacity (i.e. the maximum frequency can be recorded) and high estimated accuracy simultaneously. Moreover, when considering the high skewness of data streams, this problem will become even worse. Consequently, we propose the trapezoidal sketch (t-sketch) in this paper. In the t-sketch, different from the r-sketch, the counter sizes in different layers are different. Therefore, the low space usage and high capacity can be achieved simultaneously in the t-sketch. Moreover, based on the basic t-sketch, we propose the space-saving t-sketch and the capacity-improvement t-sketch and analyze the properties of these two t-sketches. Finally, for improving the estimation accuracy of the t-sketch further, we propose the probabilistic-based estimation error-reducing algorithm. Compared with the CM sketch, CU sketch, C sketch and A sketch, the simulation results show that the performances on space usage, capacity and estimation accuracy are improved successfully by the space-saving t-sketch and the capacity-improvement t-sketch.

Abstract Due to the limitations of physical resources, if a large-scale emulation network environment composed of millions of vertices and edges is constructed by virtualization technology, the whole network topology should be partitioned into a set of subnets. The topology partition is a work of graph partition. The existing topology partition methods have shortcomings, such as low efficiency and poor practicability, especially for large-scale network topology. The emulation network is a kind of complex network and has the characteristics of community structure. Therefore, we proposed LENTP (large-scale emulation network topology partition) based on the community detection with the weight of the vertex similarity for large-scale topology partition. In the first stage, the tree-structured area compression reduces the topology scales significantly to improve partition efficiency. And then, the improved Louvain algorithm is used to topology partitioning and obtain an initial set of subnets with the minimum number of subnets and remote links. Finally, after repartitioning and merging for the initial subnets, the result of subnets is the final topology partition that reaches the optimization objectives with the conditions of the virtual resources. In the experiment, the method is tested in five groups of network topology with different scales. The results demonstrate that LENTP can partition the network topology over 1 000 000 nodes and significantly improve the running-time efficiency of the network topology partition.

The sketch is one of the typical and widely-used data structures for estimating the frequencies of items in data streams. However, since the counter sizes in traditional rectangular sketch (r-sketch) are the same, it is hard to achieve small space usage, high capacity (i.e., the maximum frequency can be recorded), and high estimated accuracy simultaneously. Moreover, when considering the high skewness of data streams, this problem will become even worse. Consequently, we propose the trapezoidal sketch (t-sketch) in this paper. In the t-sketch, different from the r-sketch, the counter sizes in different layers are different. Therefore, the low space usage and high capacity can be achieved simultaneously in the t-sketch. Moreover, based on the basic t-sketch, we propose the space-saving t-sketch and the capacity-improvement t-sketch, and analyze the properties of these two t-sketches. Compared with the CM sketch, CU sketch, C sketch, and A sketch, the simulation results show that the performances on space usage, capacity, and estimation accuracy are improved successfully by the space-saving t-sketch and the capacity-improvement t-sketch.

In wireless network applications, such as routing decision, network selection, etc., the Multi-Attribute Decision Making (MADM) is widely used. The MADM approach can address the multi-objective decision making issues effectively. However, when the parameters vary greatly, the traditional MADM algorithm is not effective anymore. To solve this problem, in this paper, we propose the pairwise-based MADM algorithm. In the PMADM, only two nodes' utilities are calculated and compared at each time. The PMADM algorithm is much more accurate than the traditional MADM algorithm. Moreover, we also prove that the PMADM algorithm is sensitive to the parameters which vary seriously and insensitive to the parameters which change slightly. This property is better than that of the traditional MADM algorithm. Additionally, the PMADM algorithm is more stable than traditional MADM algorithm. For reducing the computational complexity of the PMADM algorithm, we propose the low-complexity PMADM algorithm. For analyzing the computational complexity of the l PMADM algorithm, we propose the tree-based decomposing algorithm in this paper. The l PMADM algorithm has the same properties and performances as that of the PMADM algorithm; however, it is simpler than the PMADM algorithm. The simulation results show that the PMADM and l PMADM algorithms are much more effective than the traditional MADM algorithm.

The link failure recovery is important to the Internet. For improving the performance of link failure recovery in the IP network, the software defined networking (SDN) is applied to achieve this target. The SDN is effective on solving this kind of issue. However, considering the deployment cost, only a few IP routers can be replaced by the SDN switches. Thus, to minimize the number of SDN switches, the greedy-based approach is proposed to select the most appropriate deployment locations. But the greedy-based approach has disadvantages. For addressing these disadvantages, in this paper, we proposed the search-tree based SDN candidate selection (SCS) algorithm. In this algorithm, for achieving better performance than the greedy-based approach, three algorithms are proposed, which are the search-tree based feasible solutions calculation algorithm, the most appropriate feasible solution selection algorithm, and the most appropriate designated SDN switch selection algorithm. Based on these algorithms, the performance of the search-tree based SCS algorithm is improved greatly compared with the greedy-based algorithms.

Mobile edge computing (MEC) has emerged for reducing energy consumption and latency by allowing mobile users to offload computationally intensive tasks to the MEC server. Due to the spectrum reuse in the network of MEC, the inner-cell interference has a great effect on MEC's performance. In this paper, for reducing the energy consumption and latency of MEC, we propose a game theory based approach to join task offloading decision and resource allocation together in the MEC system. In this algorithm, the offloading decision, the CPU capacity adjustment, the transmission power control, and the network interference management of mobile users are regarded as a game. In this game, based on the best response strategy, each mobile user makes their own utility maximum rather than the utility of the whole system. We prove that this game is an exact potential game and the Nash equilibrium (NE) of this game exists. We also investigate the properties of this algorithm, including the convergence, the computational complexity, and the Price of anarchy (PoA). We evaluate the performance of this algorithm by simulation. The simulation results illustrate that this algorithm is effective in improving the performance of the multi-user MEC system.

Due to the different transmission media, the underwater environment poses a more severe situation for routing algorithm design than that of the terrestrial environment. In this paper, we propose a weight based fuzzy logic (WBFL) relay node selection algorithm for the underwater Ad-hoc network, which can get the most balanced result than the previous algorithms without increasing the computation complexity. In this algorithm, the parameter scatters instead of the parameter values are inputted into the fuzzy logic inference system. By this innovation, the algorithm can take as much cross-layer parameters into account as possible during the relay node selection. Moreover, taking the node mobility into account, we propose a geographic based link lifetime prediction algorithm for underwater Ad-hoc network. The simulation results show that the WBFL algorithm can improve the network throughput at most 50% compare with the ExOR algorithm; moreover, the WBFL is effective and accurate on selecting relay nodes and the computation complexity is less than the previous algorithms.