With the development of Flying Ad-Hoc Networks (FANETs) technology, FANET routing protocols have attracted more researchers’ attention. Moreover, FANET traffic is usually task-related, so it is crucial to design a task-oriented FANET routing protocol. Named Data Networking (NDN) is a content-centric communication architecture. Compared with the traditional end-to-end communication model, NDN is more concerned about the content and has more potential in task-oriented FANET. However, existing NDN schemes lack awareness of node capabilities, consequently leading to blindly forwarding requests. In addition, traditional broadcast-based forwarding strategies in NDN face fierce competition. To address these issues, we propose a Service-Aware Flying Ad-Hoc Network Routing Protocol with NDN, abbreviated as SARP, which expands the concept of traditional NDN content into services. SARP comprises two components: i) A local-scale service availability information announcement mechanism to perceive Unmanned Aerial Vehicles (UAVs) service capabilities. ii) A task-oriented two-phase forwarding strategy to reduce UAVs’ load and transmission overhead. We compared SARP with existing schemes in ndnSIM and deployed it in the real world. Experimental results show that SARP reduces service response time by 60% and traffic consumption per response by 50% and increases the service response ratio by 20%.

Deep learning (DL) has been applied in billions of mobile devices due to its astonishing performance in image, text, and audio processing. However, limited by the computing capability of mobile devices, a large amount of DL inference tasks need to be offloaded to edge or cloud servers, which makes powerful GPU servers are struggling to ensure the quality of service(QoS). To better utilize the highly parallel computing architecture of GPU to improve the QoS, we propose BatOpt, a framework that uses dynamic batch processing to strike a good balance between service latency and GPU memory usage in DL inference services. Specifically, BatOpt innovatively models the DL inference service as a$M/G(a,b)/1/N$queue, with the consideration of stochastic task arrivals, which enables it to predict the service latency accurately in different system states. Furthermore, we propose an optimization algorithm to trade off the service latency and GPU memory usage in different system states by analyzing the queueing model. We have implemented BatOpt on Pytorch and evaluated it on an RTX 2080 GPU using real DL models. BatOpt brings up to 31x and 4.3x times performance boost in terms of service latency, compared to single-input and fixed-batch-size strategies, respectively. And BatOpt's maximum GPU memory usage is only 0.3x that of greedy-dynamic-batch-size strategy on the premise of the same service latency.

The last decades are marked by massive and diverse image data, which shows increasingly high resolution and quality. However, some images we obtained may be corrupted, affecting the perception and the application of downstream tasks. A generic method for generating a high-quality image from the degraded one is in demand. In this paper, we present a novel GAN inversion framework that utilizes the powerful generative ability of StyleGAN-XL for this problem. To ease the inversion challenge with StyleGAN-XL, Clustering \& Regularize Inversion (CRI) is proposed. Specifically, the latent space is firstly divided into finer-grained sub-spaces by clustering. Instead of initializing the inversion with the average latent vector, we approximate a centroid latent vector from the clusters, which generates an image close to the input image. Then, an offset with a regularization term is introduced to keep the inverted latent vector within a certain range. We validate our CRI scheme on multiple restoration tasks (i.e., inpainting, colorization, and super-resolution) of complex natural images, and show preferable quantitative and qualitative results. We further demonstrate our technique is robust in terms of data and different GAN models. To our best knowledge, we are the first to adopt StyleGAN-XL for generating high-quality natural images from diverse degraded inputs. Code is available at https://github.com/Booooooooooo/CRI.

Background: Uterine Leiomyoma (UL) is one of the most common benign tumors of the female reproductive system. A Traditional Chinese Medicine (TCM) treatment known as Lichong Decoction (LD) can effectively alleviate the clinical symptoms of UL and reduce adverse reactions and complications. Methods: The active ingredients and targets of LD were retrieved in the TCMSP database and related literature, and the action targets of UL were obtained through the DisGeNET, Genecards, and OMIM databases. The intersection of LD and UL targets was extracted by R software, and the “ingredient-target” network was constructed by Cytoscape software. At the same time, a potential target protein interaction network was constructed by means of a protein interaction database (STRING), and information visualization and topology analysis were completed by Cytoscape software. R software was used for GO enrichment analysis and KEGG pathway enrichment analysis of intersection targets. Molecular docking technology was used for docking verification of key ingredients and core targets. Results: A total of 93 LD active ingredients and 143 LD-UL intersecting targets involving 2,660 GO analysis items and 178 KEGG signaling pathways were obtained. It is predicted that the key active ingredients of LD in the treatment of UL may be quercetin, luteolin, hederagenin, bis-demethoxycurcumin, and kaempferol, which mainly involve SRC, TP53, HSP90AA1, AKT1, MAPK1, and other targets, and may play a role in the treatment of UL through multiple KEGG pathways such as PI3K−Akt signaling pathway, AGE-RAGE pathway in diabetic complications, and cancer-related pathway. Molecular docking results indicated that quercetin, luteolin, hederagenin, bis-demethoxycurcumin, and kaempferol had good docking effects with SRC, TP53, HSP90AA1, AKT1, and MAPK1, respectively. Conclusions: This study analyzed the key ingredients, core targets and related pathways of LD in the treatment of UL, confirmed the therapeutic effect of LD in the treatment of UL, and provided a reference for subsequent clinical practice and basic research in traditional Chinese medicine.

Abstract Recently, visual object tracking has become a hot topic in computer vision community owing to its extensive applications and fundamental research significance. Many researchers have paid attention to the discriminative correlation filter (DCF) due to its excellent tracking performance. The background-aware CF (BACF) is developed to handle the inevitable boundary effects of DCFs that shows superior performance compared with other CF-based trackers. However, BACF has a poor occlusion handling ability. To overcome this defect, we propose a saliency-aware CF tracking (SACF) model. In SACF, a saliency map of the target is introduced on CFs to strengthen the ability to extract the target from a complex background. Meanwhile, to adjust to the rapid variation of the target appearance, we adaptively modify the learning rate of the filter template using adaptive learning rate. By incorporating the saliency map into the BACF objective function and adaptive learning rate adjustment, the robustness and accuracy of the tracker are boosted significantly. Extensive experiments validate that our method can efficaciously solve the occlusion problem and achieves state-of-the-art performance in terms of accuracy and speed on standard benchmarks, i.e. OTB-2013, OTB-2015 and LaSOT. Compared with other state-of-the-art CF tracking methods, the proposed method is competitive considering accuracy and success rate.

Convolutional neural networks (CNNs) are highly successful for super-resolution (SR) but often require sophisticated architectures with heavy memory cost and computational overhead significantly restricts their practical deployments on resource-limited devices. In this paper, we proposed a novel contrastive self-distillation (CSD) framework to simultaneously compress and accelerate various off-the-shelf SR models. In particular, a channel-splitting super-resolution network can first be constructed from a target teacher network as a compact student network. Then, we propose a novel contrastive loss to improve the quality of SR images and PSNR/SSIM via explicit knowledge transfer. Extensive experiments demonstrate that the proposed CSD scheme effectively compresses and accelerates several standard SR models such as EDSR, RCAN and CARN. Code is available at https://github.com/Booooooooooo/CSD.

Higher-order statistics (HOS) estimation hinges on the availability of a huge amount of data records, which causes exceedingly high sampling rates and overwhelming energy consumption for the sampling devices, especially when dealing with wideband signals. To overcome these challenges, this paper develops a novel compressive cumulant slice sensing (CCSS) method that aims to efficiently reconstructing the 1D diagonal slice of higher-order cumulants under the compressive sensing framework. It first parsimoniously collects data with a properly designed sampler based on the minimal sparse ruler principle, and then accurately reconstructs the 1D diagonal cumulant slice from those compressive measurements. The reconstructed HOS can be useful in many signal processing tasks, which is illustrated through the task of line spectrum estimation with resilience to colored Gaussian noise at reduced sampling rates.

The evaluation and optimization methods are presented for energy architecture of distributed electrical propulsion Fuel-cell UAV in this paper. DC power flow model of hybrid power source based on fuel-cell and battery are derived, which is considered with the flighting air dynamics coupling with the multi-propeller powertrain. Based on hybrid power architecture, power flow calculation analysis of the fuelcell DC power system are given, evaluation software for architecture of fuel-cell UAV powertrain is designed, parameterized models of typical powertrain components are build, and multi-objective convex optimization model with some constraints of hybrid power system in the some optimization environment are derived and the optimal solver is performed under different flight stage of UAV in this paper. It is expected to provide optimization and evaluation methods tools for the design and evaluation of distributed electric aircraft. Some preliminary research results are presented in this paper.

Visual object tracking remains a challenging task in computer vision although important progress has been made in the past decades. Particle filter (PF) is now a standard framework for solving non‐linear/non‐Gaussian problems, especially in visual object tracking. This study proposes an ant colony optimisation (ACO)‐based iterative PF for object tracking. In the proposed method, the basic idea of ACO is used to simulate the behaviour of a particle moving toward the posterior distribution. Such idea is incorporated into the particle filtering framework in order to overcome the well‐known particle impoverishment problem. An iterative unscented Kalman filter is used to design a proposal distribution for particle generation in order to generate better predicted sample states. For the likelihood model, the authors adopt the locality sensitive histogram to model the appearance of the target object, which can better handle the illumination variation during tracking. The experimental results demonstrate that the proposed tracker shows better performance than the other tracking methods.

Software testing aims to assure the quality of software under test. To improve the efficiency of software testing, especially regression testing, test-case prioritization is proposed to schedule the execution order of test cases in software testing. Among various test-case prioritization techniques, the simple additional coverage-based technique, which is a greedy strategy, achieves surprisingly competitive empirical results. To investigate how much difference there is between the order produced by the additional technique and the optimal order in terms of coverage, we conduct a study on various empirical properties of optimal coverage-based test-case prioritization. To enable us to achieve the optimal order in acceptable time for our object programs, we formulate optimal coverage-based test-case prioritization as an integer linear programming (ILP) problem. Then we conduct an empirical study for comparing the optimal technique with the simple additional coverage-based technique. From this empirical study, the optimal technique can only slightly outperform the additional coverage-based technique with no statistically significant difference in terms of coverage, and the latter significantly outperforms the former in terms of either fault detection or execution time. As the optimal technique schedules the execution order of test cases based on their structural coverage rather than detected faults, we further implement the ideal optimal test-case prioritization technique, which schedules the execution order of test cases based on their detected faults. Taking this ideal technique as the upper bound of test-case prioritization, we conduct another empirical study for comparing the optimal technique and the simple additional technique with this ideal technique. From this empirical study, both the optimal technique and the additional technique significantly outperform the ideal technique in terms of coverage, but the latter significantly outperforms the former two techniques in terms of fault detection. Our findings indicate that researchers may need take cautions in pursuing the optimal techniques in test-case prioritization with intermediate goals.

ABSTRACT A challenge faced by many system administrators in utilizing the intrusion detection system (IDS) is to sift out genuine alerts buried with overwhelming alerts of benign activities generated by the IDS, especially for IDS deployed in large networks. Existing methods propose to identify the real alerts to aid the administrators. In our paper, we extend the idea of filtering irrelevant alerts based on alert volumes. And we formulate the flow estimation of abrupt changes in feature distribution caused by anomalies, by computing Kullback–Leibler distance of alert feature values under observation in comparison with a reference distribution, which is the mixture of a distribution drawing a tread from historical alerts, and a distribution derived from expertise provided by administrators. Experimental studies on the Defense Advanced Research Projects Agency dataset as well as real‐life data gathered from the IDS of a large network show that our method is able to distinguish and highlight genuine anomalies arising from the tremendous number of intrusion alerts, including different kinds of attacks and network failures. Application of this technique to alerts greatly helps the administrators in identifying real alerts and then reduces the alert load in the future. Copyright © 2013 John Wiley & Sons, Ltd.