Transformer-based large language models (LLMs) are increasingly being adopted in networking research to address domain-specific challenges. However, their quadratic time complexity and substantial model sizes often result in significant computational overhead and memory constraints, particularly in resource-constrained environments. Drawing inspiration from the efficiency and performance of the Deepseek-R1 model within the knowledge distillation paradigm, this paper introduces Mamba4Net, a novel cross-architecture distillation framework. Mamba4Net transfers networking-specific knowledge from transformer-based LLMs to student models built on the Mamba architecture, which features linear time complexity. This design substantially enhances computational efficiency compared to the quadratic complexity of transformer-based models, while the reduced model size further minimizes computational demands, improving overall performance and resource utilization. To evaluate its effectiveness, Mamba4Net was tested across three diverse networking tasks: viewport prediction, adaptive bitrate streaming, and cluster job scheduling. Compared to existing methods that do not leverage LLMs, Mamba4Net demonstrates superior task performance. Furthermore, relative to direct applications of transformer-based LLMs, it achieves significant efficiency gains, including a throughput 3.96 times higher and a storage footprint of only 5.48% of that required by previous LLM-based approaches. These results highlight Mamba4Net’s potential to enable the cost-effective application of LLM-derived knowledge in networking contexts. The source code is openly available to support further research and development.

Time series analysis plays a fundamental role in revealing data evolution, trends, and cyclical patterns. However, existing studies often fail to effectively address the dynamic dependencies between variables in multidimensional time series and the temporal evolution patterns within variables, thereby limiting the effectiveness of complex time series feature analysis. In this paper, we propose a dual-branch frequency-time interactive representation learning model (FreqTime) that captures the correlations between variables and the temporal dependencies within variables through a collaborative architecture in the time domain and frequency domain. The time domain branch uses an inverse Transformer architecture to model cross-variable interactions, while the frequency domain branch utilizes multi-scale gated convolutions to capture features and map them back to the time domain. Finally, global representations are obtained by interactively fusing the representations learned from the two branches in the time domain. Experiments demonstrate that FreqTime achieves state-of-the-art performance on long sequence prediction, classification, and anomaly detection tasks, and exhibits strong robustness in noisy environments.

Autoencoders, as a generative self-supervised learning, have received more and more attention in recent years in image, video, and other media-related information processing. However, Graph AutoEncoder (GAE) has yet to achieve the capability demonstrated by contrastive learning in the task-centered on attribute networks. The main limitation lies in the fact that traditional autoencoder architectures require pretext tasks that align with downstream tasks, resulting in limited expressive power of the encoder. In this paper, we propose a novel separable-task generative self-supervised learning framework capable of providing high-quality representations, Split Masked AutoEncoder (SMAE), which unleashes the encoder’s ability to extract representations through an intelligent design. Our approach focuses on unlocking the potential of the encoder by introducing encoding transfer and feature replacement strategies, thereby enabling self-supervised pretext tasks to achieve atomic separation and fully unleash the encoder’s feature representation potential. We conducted extensive experiments on widely-used graph classification datasets, and the results demonstrate that SMAE outperforms state-of-the-art baselines in terms of graph classification accuracy and generation quality. Furthermore, our experimental findings show that prediction at the representation layer is more effective than original graph layer reconstruction in the field of masked graph autoencoders.

Biometric and password-based two-factor authentication has received attention from the community over the past decades because of its simplicity, portability, and robustness. In wireless body area networks (WBANs), dozens of authentication and key agreement schemes have been proposed. Despite well-studied security issues, preserving user privacy in these schemes is still challenging. In this work, we propose biometric-based authentication and key agreement (BAKA), a scheme based on a fuzzy extractor for WBAN, where a novel biometric and password-based authentication algorithm is proposed by using a fuzzy extractor to achieve anonymous identity authentication, a privacy-preserving key agreement algorithm for session key security, and finally, we deploy blockchain to record biometric information using its noncomparability and distributed storage to protect users’ privacy to a large extent. BAKA is secure as per formal security proof and informal security analysis, symmetric encryption is utilized to reduce computation overhead to improve the efficiency of BAKA, where security is not compromised. Extensive experiments to validate the performance of BAKA are performed, and the results demonstrate the security efficacy proposed.

With the vigorous development of the Internet of Things (IoT), mobile users need to access data from other domains in edge computing. To achieve secure data sharing, mobile users first need to be authenticated by servers from different domains and then negotiate session keys among them. However, traditional schemes cannot solve cross-domain identity authentication and key agreement problems well due to the limited computational resources of IoT devices. In this work, we propose a Blockchain-based Cross-domain Authentication scheme for Edge computing, namely BCAE. First, to achieve secure identity verification, we design a novel cross-domain mutual identity authentication algorithm based on digital certificates and digital signatures. Next, to improve efficiency, we utilize the blockchain to share information among different domains to reduce the computation overhead. To realize quick key agreement, we apply the elliptic curve cryptography technique to design a lightweight key agreement algorithm and obtain secure session keys. Extensive experiments conducted on an actual smart healthcare issue to validate the performance of BCAE and formal security analysis confirmed the potential of the proposed work.

Large-scale UAV switching formation tracking control has been widely applied in many fields such as search and rescue, cooperative transportation, and UAV light shows. In order to optimize the control performance and reduce the computational burden of the system, this study proposes an event-triggered optimal formation tracking controller for discrete-time large-scale UAV systems (UASs). And an optimal decision - optimal control framework is completed by introducing the Hungarian algorithm and actor-critic neural networks (NNs) implementation. Finally, a large-scale mixed reality experimental platform is built to verify the effectiveness of the proposed algorithm, which includes large-scale virtual UAV nodes and limited physical UAV nodes. This compensates for the limitations of the experimental field and equipment in real-world scenario, ensures the experimental safety, significantly reduces the experimental cost, and is suitable for realizing large-scale UAV formation light shows.

With the advent of convolutional neural networks, stereo matching algorithms have recently gained tremendous progress. However, it remains a great challenge to accurately extract disparities from real-world image pairs taken by consumer-level devices like smartphones, due to practical complicating factors such as thin structures, non-ideal rectification, camera module inconsistencies and various hard-case scenes. In this paper, we propose a set of innovative designs to tackle the problem of practical stereo matching: 1) to better recover fine depth details, we design a hierarchical network with recurrent refinement to update disparities in a coarse-to-fine manner, as well as a stacked cascaded architecture for inference; 2) we propose an adaptive group correlation layer to mitigate the impact of erroneous rectification; 3) we introduce a new synthetic dataset with special attention to difficult cases for better generalizing to real-world scenes. Our results not only rank 1ston both Middlebury and ETH3D benchmarks, outperforming existing state-of-the-art methods by a notable margin, but also exhibit high-quality details for real-life photos, which clearly demonstrates the efficacy of our contributions.

Wireless sensor network enhances the classic features of wireless communication with cognitive capabilities for efficient spectrum usage. This work focuses on the dynamic correlation between the secondary users (SUs) based on their statistical behaviour while performing the cooperative communication in cognitive radio sensor network. The proposed approach addresses the problem of uneven and repetitive communication between the SUs in a cooperative communication scenario. The authors’ objective is to use a novel approach based on the Gaussian copula theory and advanced particle swarm optimisation algorithm to analyse the dependencies of time‐varying spectrum sensing behaviour of multiple SUs. Here, time delay in prediction reduces due to the analysis of the dynamic correlation between the time delay in spectrum sensing results for the same set of channels. The simulation results show the performance of the proposed approach outperforming the other well‐known techniques.