This paper investigates the internal secrecy and external covertness of a mixed-trust autonomous aerial vehicle (AAV) communication system assisted by rate-splitting multiple access (RSMA). In this setting, a semi-trusted user with partial decoding capability poses an internal eavesdropping threat, while multiple distributed wardens attempt to detect the transmission from the AAV to the semi-trusted user, creating an external covertness challenge. To characterize these security aspects, a unified analytical framework is developed. First, the internal eavesdropping capability of the semi-trusted user is quantified by deriving a closed-form expression for its eavesdropping success probability. Based on the outcome of the eavesdropping attempt, tractable expressions for the secrecy outage probability of the legitimate user are obtained. Furthermore, the external covertness performance is analyzed by deriving closed-form false alarm probability, missed detection probability, and detection error probability (DEP) for an individual warden, together with the optimal detection threshold and the corresponding minimum DEP. The cooperative global detection performance with multiple wardens is further characterized under conservative fusion rules. Extensive Monte Carlo simulations validate the analytical results and, through a joint evaluation of secrecy, reliability, and covertness metrics, illustrate the feasible operating regions enabled by RSMA power allocation in comparison with a NOMA baseline. The results provide a comprehensive theoretical basis for the design of secure and covert AAV communication strategies in mixed-trust environments.

Node classification is a fundamental task in hypergraph learning. Existing methods generally assume that there are a few labeled nodes given in advance. However, in a newly formed hypergraph, collecting label information is challenging and costly in practice. Besides, current approaches mainly exploit the local consistency relationship, i.e., direct neighborhood information, while ignoring the high-order consistency relationship, i.e., high-order proximity information, limiting the discrimination of the latent representations. To address these issues, we propose leveraging knowledge from an auxiliary well-labeled hypergraph (source hypergraph) to assist the learning tasks in the target hypergraph, thus studying the cross-hypergraph node classification problem. Specifically, we propose a model, namely Local and High-order Consistency Coding and Adaptation (LHCCA), which learns both discriminative and transferable node representations. On the one hand, for each hypergraph, by exploiting the local and high-order consistency relationships, LHCCA obtains two kinds of representations, which are then coded by an attention mechanism to achieve a unified representation. On the other hand, the coded source and target node representations are enforced adversarial domain adaptation and contrastive learning to discover transferable features for adaptation. Furthermore, we derive theoretical analyses to establish desirable properties of the proposed model. Extensive experiments on several real-world datasets are conducted, and the promising results demonstrate the effectiveness of the proposed model.

In recommendation systems, cold-start user recommendation is a challenging problem, where precise recommendations are required for users who have not appeared before. Several existing cold-start user recommendation models adopt domain adaptation to extract information from auxiliary source domains to assist the recommendations on the target domain. In this article, we propose that the cold-start user recommendation problem can be formulated by the heterogeneous domain adaption approach. We determine a transformation of user features, e.g., user social relations and historical interactions between warm users and their interested items, into a latent space so that the loss function is set by user feature reconstruction and by feature and distribution matching in the heterogeneous domains. The resulting optimization problem can be solved by matrix eigendecomposition, and the cold-start users’ preferences can thus be obtained. We also extend the proposed model using neural networks. We perform extensive experiments on several real-world datasets, and the results in terms of Precision, Recall, NDCG, and Hit Rate verify the effectiveness of the proposed model.

The reduced cost of DNA sequencing allows metagenomics to be applied on a larger scale. With metagenomic analysis, we have better insight into supplement usage, methane production, and feed conversion efficiency in livestock systems. Nevertheless, sequencing machines generate an enormous amount of complex data. Conventional methods used in the analysis of genomic data involve pre-processing and synchronous reconstruction by multiple systems, which is time consuming and prone to failure. Furthermore, the sequencing datasets and analysis results need to be organized and stored properly in order for scientists to search and access them. To tackle these challenges, a new workflow for metagenomic analysis with improved infrastructure is needed. The MetaPlat project supports experts in both academic and non-academic sectors dealing with challenges in the field of metagenomics by focusing on improved hardware and software platforms. High-performance, fault-tolerant, flexible, and scalable processors and analysis systems will help to increase the effectiveness and efficiency of current metagenomics studies. In this paper, we propose such as an infrastructure applying emerging technologies, such as Kafka, Docker, and Hadoop. Details of the infrastructure solution and some preliminary results are also discussed.