Search Results (396)

Adversarial evasion against learning-based malware detectors has shifted from feature-space perturbations to semantic-preserving, problem-space manipulations. In this paradigm, attackers inject semantic NOPs—functionally NOP instructions that shift the static feature distribution—into assembly code to suppress detection confidence. Existing defenses primarily recalibrate classifier decision boundaries, leaving the adversarially modified malware intact and thereby hindering downstream tasks including but not limited to malicious API localization and capability attribution. We introduce MalRefiner, a reinforcement-learning agent that automatically identifies and removes adversarially inserted semantic NOPs to restore the original malicious representation. The recovery process is formulated as a Markov Decision Process, where a policy network sequentially decides whether to retain or remove each opcode. The agent is trained with a composite reward function that balances detection confidence recovery with semantic preservation, guided by a lightweight 1D causal convolutional environment providing compact state representations and delayed rewards. Extensive evaluation on the PEMML and RawMal-TF datasets against four state-of-the-art detectors (1D CNN, MalConv, TCN, and MALIGN) demonstrates that MalRefiner restores F1 to within 3.18 ± 0.94% of the clean baseline and achieves a recovery rate exceeding 90% across all models and datasets, without requiring retraining or architectural modification of the target classifier. Full article

This work introduces Uni-Removal, an innovative two-stage framework that effectively addresses the critical challenge of domain adaptation in unified image restoration. Contemporary approaches often face significant performance degradation when transitioning from synthetic training environments to complex real-world scenarios due to the substantial domain discrepancy. Our proposed solution establishes a comprehensive pipeline that systematically bridges this gap through dual-phase representation learning. In the first stage, we implement a structured multi-teacher knowledge distillation mechanism that enables a unified student architecture to assimilate and integrate specialized expertise from multiple pre-trained degradation-specific networks. This knowledge transfer is rigorously regularized by our novel Instance-Grained Contrastive Learning (IGCL) objective, which explicitly enforces representation consistency across both feature hierarchies and image spaces. The second stage introduces a groundbreaking output distribution calibration methodology that employs Cluster-Grained Contrastive Learning (CGCL) to adversarially align the restored outputs with authentic real-world image characteristics, effectively embedding the student model within the natural image manifold without requiring paired supervision. Comprehensive experimental validation demonstrates Uni-Removal’s superior performance across multiple real-world degradation tasks including dehazing, deraining, and deblurring, where it consistently surpasses existing state-of-the-art methods. The framework’s exceptional generalization capability is further evidenced by its competitive denoising performance on the SIDD benchmark and, more significantly, by delivering a substantial 4.36 mAP improvement in downstream object detection tasks, unequivocally establishing its practical utility as a robust pre-processing component for advanced computer vision systems. Full article

This paper proposes a novel multi-dimensional reliability evaluation method for distribution network under renewable energy installation and load installation, considering fault reconfiguration and multiple constraints. Firstly, the related output models of distribution network are presented. Secondly, a post-fault load restoration model of the distribution network by considering fault reconfiguration and multiple constraints under distributed generation integration is established in our study. Subsequently, the reliability evaluation indices for the distribution network, including system average interruption duration indicator, customer average interruption duration indicator, system average interruption frequency indicator, and average service availability indicator, are introduced. And the process of reliability evaluation founded on the sequential Monte Carlo simulation is also explained. Finally, simulations were performed on an improved IEEE 69-node distribution system to verify the proposed method, and the reliability of the distribution network from the distributed generation installation, types of distributed generation installation, distributed generation installation capacity, and new load installation is discussed in-depth. Full article

The red-crowned crane (Grus japonensis) is a flagship species for wetland biodiversity in East Asia. The Songnen Plain is a vital wetland and habitat for rare and endangered birds in Northeast China. However, rapid land use changes have raised urgent concerns about habitat loss and the survival of these populations. We combined 30 years (1990–2020) of field surveys with ensemble species distribution models (SDMs) to analyze the spatio-temporal changes in suitable habitats for all three key life stages—spring migration, breeding, and autumn migration—across the Songnen Plain. We also assessed how well the current protected-area (PA) network covers suitable habitats and identified conservation gaps. Land use type was the most significant predictor of habitat suitability. Over this period, suitable habitats decreased sharply by 60% (spring migration), 72% (breeding), and 76% (autumn migration), with severe fragmentation and a clear northward shift. Core suitable areas are now mainly found within a few nature reserves, including Zhalong, Wuyu’er River, and Xianghai. We identified three significant conservation gaps: Lindian–Anda, Tailai–Dumeng, and Meilisi Daur–Fuyu. Our results show widespread habitat reduction and demonstrate the inadequacy of the current PA network in supporting the long-term survival of red-crowned crane populations. We recommend expanding protections and restoring wetland connectivity within these gaps to maintain critical habitats and improve landscape resilience for this endangered species. Full article

Electric vehicle charging stations (EVCSs) are critical interfaces between urban mobility and distribution grids and are increasingly exposed to false data that can mislead operations and degrade voltage quality. This study proposes a defense-planning framework that models how cyber manipulation propagates to physical impacts in a coupled transport–power system. The interaction is modeled as a tri-level defender–attacker–operator problem in which a defender hardens a subset of charging stations, an attacker forges measurements and demand, and an operator redispatches resources to keep the system secure. We solve this problem with a method that embeds corrective operation into the evaluation and uses improved implicit enumeration (IIE) with pruning to identify a small set of high-value stations to protect with far fewer trials than an exhaustive search. On a benchmark feeder coupled to a road network, protecting a few traffic-critical stations restores compliance with voltage limits under tested attack levels while requiring roughly an order of magnitude fewer evaluations than complete enumeration. Sensitivity analysis shows that the loss of reactive power from PV inverters (PV VARs) harms voltage profiles more than an equivalent reduction in distributed storage, indicating that maintaining local reactive capability reduces the number of stations that must be hardened to meet a given voltage target. These results guide utilities and city planners to prioritize protection at traffic-critical EVCSs and co-plan local Volt/VAR capability, achieving code-compliant voltage quality under adversarial conditions with markedly lower planning effort. Full article

Under the dual pressures of rapid urbanization and intense human socioeconomic activities, habitat fragmentation and poor landscape connectivity have become critical issues in cities. Constructing ecological networks is essential for maintaining urban ecosystem health and promoting sustainable environmental development. It represents an effective approach to balancing regional economic growth with ecological conservation. This study focused on the Shenzhen Special Economic Zone. Ecological sources were identified using Morphological Spatial Pattern Analysis (MSPA) and landscape connectivity assessment. Circuit theory was applied to extract ecological corridors, ecological pinch points, and ecological barriers. The importance levels of ecological corridors were classified to form an ecological network. The network was optimized by adding ecological sources, stepping stones, and restoring breakpoints. Its structure and functionality were evaluated before and after optimization. The results indicate the following: (1) The core area in Shenzhen City Area covers 426.67 km², the largest proportion among landscape types. It exhibits high fragmentation, low connectivity, and a spatial pattern characterized as “dense in the east and west, sparse in the center.” (2) Seventeen ecological sources were identified, consisting of 8 key sources, 5 important sources, and 4 general sources, accounting for 17.62% of the total area. Key sources are mainly distributed in forested regions such as Wutong Mountain, Maluan Mountain, Paiya Mountain, and Qiniang Mountain in the southeast. (3) Twenty-six ecological corridors form a woven network, with a total length of 127.44 km. Among these, 13 key corridors are concentrated in the eastern region, while 7 important corridors and 6 general corridors are distributed in the western and central parts. Few corridors exist in the southwest and southeast, leading to ecological flow interruption. (4) The optimized ecological network includes 12 newly added ecological source areas, 20 optimized ecological corridors, 120 ecological pinch points, and 26 ecological barriers. The maximum current value increased from 10.60 to 20.51, indicating significantly enhanced connectivity. The results provide important guidance for green space planning, biodiversity conservation, and ecosystem functionality enhancement in Shenzhen City Area. Full article

Urban agglomerations in plateau regions often face severe landscape fragmentation and cross-boundary ecological pressures, highlighting the need for coordinated eco-logical planning for sustainable urban development. We coupled species–landscape interactions and multi-ecological services to construct sustainable ecological security patterns (ESPs) and establish a collaborative optimization framework. Specifically, we integrated MaxEnt-derived habit suitability with InVEST-based ecosystem services to identify ecological sources (ESs) and analysis the environmental impacts on species distribution. Based on this, we built a multi-factor resistance surface and employed circuit theory to extract ecological corridors (ECs) and critical nodes (pinch points and barrier points). Then, we quantitatively compared two simulated scenarios (barrier points restoration and stepping stone augmentation) to assess the spatial priority of ecological nodes. We identified 48 ESs (26,410.48 km², mainly distributed in Chuxiong, Yuxi, Honghe, and Kunming), 115 ECs (2670.02 km, with a west-dense and east-sparse spatial pattern), 43 pinch points, and 39 barrier points. Scenario simulation shows that repairing 39 barrier nodes increases network connectivity by an average of 33.52% and global network efficiency by 19.44%, whereas adding steeping stones yields improvements of 20.09% and 5.56%, respectively, indicating that barrier-node restoration produces larger contribution in both connectivity and efficiency at the global scale. Leveraging EN construction and scenario simulation, we developed an ESP-based sustainable framework for collaborative optimization in plateau urban agglomerations. The framework specifies agglomeration-specific coordination pathways, which are expected to provide a transferable blueprint for biodiversity conservation, ecosystem optimization, and sustainable development. Full article

Misiones province covers < 1% of Argentina’s land area yet harbors > 50% of the country’s biodiversity, with a significant remnant of Atlantic Forest, a global biodiversity hotspot. Approximately 540,000 ha of this native forest is protected, with the remaining areas facing threats from ongoing land conversion, an expanding road network, and a growing rural population. A prior study incorporated noninvasive data on five carnivores into a multifaceted cost analysis to define the optimal location for a multispecies biological corridor, with the goal of enhancing landscape connectivity among protected areas. Subsequent analyses, with an updated framework, emphasized management strategies that balanced human–wildlife coexistence and habitat needs. Building on these efforts, our study applied ecological niche modeling to data located by conservation detection dogs, with genetics used to confirm species identity, and two land-use scenarios, to predict potential distributions of three game species—lowland tapir (Tapirus terrestris), white-lipped peccary (Tayassu pecari), and collared peccary (Pecari tajacu)—that are not only threatened by poaching, road mortality, and habitat loss but also serve as essential prey for carnivores. We assessed the suitability of unique and overlapping vegetation types, within and outside of protected areas, as well as within this multispecies corridor, identifying zones of high conservation concern that underscore the need for integrated planning of predators and prey. These results highlight that ensuring the long-term viability of wildlife across the heterogeneous land-use matrices of Misiones requires going beyond protected areas to promote functional connectivity, restore degraded habitats, and balance human–wildlife needs. Full article

Urban agglomerations are key to balancing carbon emissions (CEs) and ecosystem services (ESs), yet structural imbalances exist between LUCE and ESs due to the lack of standardized frameworks and clear governance strategies. This study investigates the relationship between LUCE and ESs in the Chang-Zhu-Tan urban agglomeration using multi-source data from 2010 to 2023. The study aims to address three main research questions: (1) How do LUCE and ES networks evolve over time? (2) What factors drive their heterogeneity? (3) How do urbanization and ecological restoration impact LUCE and ES network dynamics? To answer these, we apply centrality metrics and develop heterogeneity indices to evaluate connectivity, accessibility, and driving factors. The findings show that both LUCE and ES networks exhibit corridor-like structures, with asymmetric node distributions. The LUCE-Network’s degree centrality increased from 0.16 to 0.29, while the ES-Network’s rose from 0.16 to 0.23. Heterogeneity was initially positive but turned negative by 2023, indicating a shift from LUCE dominance to an increased emphasis on ES. This transition was influenced by urbanization, land use changes, and ecological restoration efforts. Notably, the proportion of built-up land (X11) grew from 0.0187 in 2010 to 0.1500 in 2023, intensifying the disparity between LUCE and ESs. Similarly, urbanization (X7) surged to 0.1558 in 2023, increasing CEs and the demand for ESs. A collaborative pathway is proposed to address these challenges, involving controlled urban development, restoration of green spaces, and prioritizing multimodal transport and energy efficiency. This framework offers actionable diagnostics for improving low-carbon and ecological governance in urban agglomerations. Full article

In post-disaster scenarios, Unmanned Aerial Vehicles (UAVs) acting as Mobile Aerial Base Stations (MABSs) offer a flexible means of restoring communication for isolated user equipment (UE) when conventional infrastructure is unavailable. More broadly, clustering is a fundamental tool for organizing spatially distributed entities in wireless, IoT, and sensor networks. However, static algorithms such as Affinity Propagation Clustering (APC) often fail to generalize across diverse environments and user densities. This study introduces a hybrid clustering framework that dynamically selects between APC and density-based clustering (DBSCAN), guided by a neural classifier trained on spatial distribution features. The chosen centroids then seed a Genetic Algorithm (GA) that evolves UAV trajectories under multiple performance indicators, including coverage, capacity, and path efficiency. Simulation results demonstrate that the hybrid clustering approach improves the adaptability and effectiveness of UAV deployments by learning context-aware clustering strategies. Beyond UAV-assisted disaster recovery, the proposed framework illustrates how intelligent clustering selection can enhance performance in heterogeneous, real-time applications such as IoT connectivity, smart city monitoring, and large-scale sensor coordination. Full article

Escalating human–land conflicts have exacerbated ecosystem degradation, threatening regional sustainable development. As the largest sandy land in China, the Horqin Sandy Land (HSL) in eastern Inner Mongolia exhibits high ecological fragility. Delineating ecological restoration zones (ERZ) is critical to transition from localized restoration to system-wide stability, thereby enhancing regional ecological security, which reflects ecosystem health and integrity. Ecological security patterns (ESP), as spatial configurations that support and maintain ecological security, serve as the foundational framework for ERZ planning. Unlike conventional applications of InVEST and MSPA, this study integrates an ecosystem service assessment with morphological spatial pattern analysis (MSPA) under a “Source–Resistance–Corridor–Note” paradigm to develop a novel “ecological network–zoning” approach. This framework transforms ecological connectivity analysis into actionable restoration zoning, bridging theoretical ESP construction with practical management needs. Key findings include the following: (1) In total, 76 vital ecological source regions were mapped, representing about 10,204.38 km² of ecologically significant land, with primary distribution in the northwestern mountainous regions; (2) A total of 169 ecological corridors were extracted, spanning 4071.94 km in length. Ecological pinch points with 239.91 km² and barrier points with 568.85 km² were systematically identified; (3) A “Five Zones, Three Belts, One Core” spatial strategy was proposed, aligning with regional ecological conditions and development goals. This study provides a transferable methodology for ecosystem restoration in global arid and semi-arid regions, bridging theoretical frameworks with actionable zoning practices. Full article

This study aims to evaluate the restoration effect of artificially mixed-sown grasslands by investigating the characteristics of plant communities and soil fungal communities in long-term (22-year-established) artificial grasslands under six Poaceae mixture combinations. The experiment took mixed-sown grasslands of grass species established in 2002 on the Qinghai–Tibet Plateau as the research object. It employed ITS gene high-throughput sequencing technology to construct a fungal community distribution map and combined it with FUNGuild (Functional Guilds of Fungi) functional predictions to analyze fungal species abundance, structural diversity, molecular co-occurrence networks, and functional characteristics. By integrating Mantel test and RDA (redundancy analysis), we identified key environmental factors driving soil microbial community structure in mixed-sown grasslands and revealed the plant–soil–microbe interaction mechanisms in a Poaceae mixture grassland. The results showed that the HC treatment (a mixture of three grass species) significantly enhanced plant biomass and soil nutrient accumulation. In 2023 and 2024, its aboveground biomass increased by 66.14% and 60.91%, respectively, compared to the HA treatment (monoculture). Soil organic matter increased by 52.32% and 48.35%, while electrical conductivity decreased by 48.99% and 51.72%, respectively. The fungal community structure improved under the HD treatment (a mixture of four grass species), with an increased abundance of the dominant phylum Ascomycota and a 14.44% rise in the Shannon index compared to the HA treatment. The network complexity under the HF treatment (a mixture of six grass species) increased (with edge numbers reaching 494), while the functional abundance of plant pathogen was significantly lower than that under the HA treatment. Mantel test and RDA revealed that SEC (soil electrical conductivity) was significantly positively correlated with pH, while both exhibited negative correlations with other plant and soil physicochemical indicators. Moreover, SEC emerged as the core factor driving fungal community assembly. Mixed sowing of three to four grass species effectively regulated soil electrical conductivity, simultaneously enhancing plant biomass, soil nutrients, and fungal community diversity, representing an optimal strategy for artificial restoration of degraded grasslands. Full article

To address the limitations of conventional reliability assessment methods in handling mesh distribution networks with flexible operation characteristics and complex topologies, namely their poor adaptability and low computational efficiency, this paper proposes a matrix-based analytical approach for reliability assessment of mesh distribution networks. First, a network configuration centered on the soft open points (SOP) is established. Through multi-feeder interconnection and flexible power flow control, a topology capable of fast fault transfer and service restoration is formed. Second, based on the restoration modes of load nodes under fault scenarios, three types of fault incidence matrices (FIM) are proposed. By means of matrix algebra, explicit analytical expressions are derived for the relationships among equipment failure probability, duration, impact range, and reliability indices. This overcomes the drawbacks of iterative search in conventional reliability assessments, significantly improving efficiency while ensuring accuracy. Finally, a modified 44 bus Taiwan test system is used for reliability assessment to verify the effectiveness of the proposed method. The results demonstrate that the proposed matrix-based analytical reliability assessment method enables explicit analytical calculation of both system-level and load-level reliability indices in mesh distribution networks, providing effective support for planning and operational optimization to enhance reliability. Full article

Soil microorganisms are important decomposers in soil, and they play important roles in litter degradation, nutrient cycle and balance, soil physicochemical property improvement, and soil fertility maintenance. To understand the influence of Eucalyptus plantations on the growth, reproduction, and activity of soil microorganisms in severely degraded land, the Leizhou Peninsula in tropical China was selected as the research area. The vegetation restoration types of Eucalyptus urophylla × grandis planted in its severely degraded red soil areas (ES: Eucalyptus–shrub, EG: Eucalyptus–grass, and ED: Eucalyptus–Dicranopteris pedata (Houtt.) Nakaike) were studied, and the nearby natural vegetation types (S: shrub, G: grass, and D: Dicranopteris pedata) served as control groups. The microbial characteristics of different vegetation restoration types were compared, and the influence of Eucalyptus plantations on the growth, reproduction, and activity of soil microorganisms in severely degraded red soil areas was discussed by setting up sample plots for investigation, sample determination, and statistical analysis. The structure of soil microorganisms differed significantly between Eucalyptus vegetation restoration (ER) and natural vegetation restoration without Eucalyptus (NER). Key organic decomposers, including bacterial genera such as Candidatus Solibacter (ER: 1.2 ± 0.4% vs. NER: 0.9 ± 0.1%), Candidatus Koribacter (ER: 1.0 ± 0.4% vs. NER: 0.7 ± 0.1%), and Edaphobacter (ER: 0.9 ± 0.1% vs. NER: 0.4 ± 0.1%), as well as fungal genera such as Rhizophagus (ER: 0.1 ± 0.0% vs. NER: 0.0 ± 0.0%), Paxillus (ER: 0.1 ± 0.0% vs. NER: 0.0 ± 0.0%), and Pisolithus (ER: 0.1 ± 0.0% vs. NER: 0.0 ± 0.0%), exhibited a significantly higher relative richness and a broader distribution in ER compared to NER (p < 0.05). Soil microbial biomass carbon, nitrogen and phosphorus (MBC, MBN, MBP), community structure (keystone taxa and symbiosis network complexity), and functional genes (for growth, reproduction, and decomposition) in ER, especially in ES, were significantly higher than in NER. This study illustrated that Eucalyptus plantations, especially ES types, can promote the growth and reproduction of soil organic decomposers, improve microbial metabolic and biological activities, and increase functional diversity and interactions among microorganisms, thus accelerating the cycle of soil carbon, nitrogen, and phosphorus nutrients, improving soil quality and fertility, and accelerating the recovery of degraded soil fertility. In areas with serious soil degradation and where natural vegetation restoration is difficult, planting Eucalyptus, especially while guiding the understory vegetation to develop into the shrub vegetation type, is an effective vegetation restoration model. Full article

With the rapid advancement of mobile edge computing and Internet of Things (IoT) technologies, device-to-device (D2D) cooperative computing has garnered significant attention due to its low latency and high resource utilization efficiency. However, workflow task scheduling in D2D networks poses considerable challenges, such as severe heterogeneity in device resources and complex inter-task dependencies, which may result in low resource utilization and inefficient scheduling, ultimately breaking the computational symmetry—a balanced state of computational resource allocation among terminal devices and load balance across the network. To address these challenges and restore system-level symmetry, a novel workflow task scheduling method tailored for D2D cooperative environments is proposed. First, a Non-dominated Sorting Genetic Algorithm (NSGA) is employed to optimize the allocation of computational resources across terminal devices, maximizing the overall computing capacity while achieving a symmetrical and balanced resource distribution. A scoring mechanism and a normalization strategy are introduced to accurately assess the compatibility between tasks and processors, thereby enhancing resource utilization during scheduling. Subsequently, task priorities are determined based on the calculation of each task’s Shapley value, ensuring that critical tasks are scheduled preferentially. Finally, a hybrid algorithm integrating Q-learning with Asynchronous Advantage Actor–Critic (A3C) is developed to perform precise and adaptive task scheduling, improving system load balancing and execution efficiency. Extensive simulation results demonstrate that the proposed method outperforms state-of-art methods in both energy consumption and response time, with improvements of 26.34% and 29.98%, respectively, underscoring the robustness and superiority of the proposed method. Full article