Baowen Zou

With a growing security threat in wireless communication networks, a promising method for secure next-generation networks is a zero-trust framework focusing on authentication schemes. How to analyze the risks involved in authentication is a challenge. This study quantifies authentication risks within the zero-trust framework and introduces a privacy domain prevention-control theory. The theory encompasses dynamic privacy risk assessment, intelligent risk classification, and automated selection of privacy protection schemes. First, a dynamic privacy risk assessment method, based on physical entity relationships, is proposed to evaluate all privacy risks. Second, a five-category risk classification method is designed to categorize privacy risks, facilitating the selection of prevention-control schemes, with its rationality mathematically validated. Additionally, an Analytical Hierarchy Process (AHP)-based method is introduced to guide the optimal selection of prevention-control schemes for various scenarios. Finally, the practical application of the theory in medicine multi-modal computing scene of wireless body area networks demonstrates its effectiveness. The experimental results also show the superiority and feasibility of the proposed methods.

Augmented Intelligence of Things (AIoT) combines augmented intelligence algorithms with the massive data collected by IoT devices, enabling more advanced decision-making. The typical application of AIoT is Edge Computing (EC), which provides computational and storage resources at the edge to support vehicle decision-making for computation tasks. With the development of EC, task offloading has become a hopeful paradigm for assisting the time-sensitive tasks of resource-limited vehicles, such as emergency rescue vehicles, by deploying at Roadside Units (RSUs). However, the effectiveness of task offloading for emergency vehicles is hindered by the timeliness of trajectory data and the concern regarding vehicle location. Therefore, this study introduces a Secure Task Offloading scheme relying on real-time trajectory prediction, named STODRL. Firstly, this study proposes a temporal differential privacy method to disturb vehicular location information to avoid suffering from malicious stealing. Secondly, a vehicular trajectory prediction method using the Temporal Convolutional Network (TCN) is designed to improve task offloading precision by offering supplemental trajectory information. Moreover, the scheme employs a reinforcement learning method to offload computational requests effectively and avoid dimensional disasters. Simulated results validate that STODRL outperforms existing methods, significantly reducing task completion delay and ensuring the security of location information.

Presents corrections to the article “A Tutorial-Generating Method for Autonomous Online Learning”.

Generative artificial intelligence has become the focus of the intelligent education field, especially in the generation of personalized learning resources. Current learning resource generation methods recommend customized courses based on learning styles and interests, improving learning efficiency. However, these methods cannot generate personalized tutorials according to learners’ preferences, nor can they adjust tutorial content as moods or levels of knowledge change. Therefore, this study develops an intelligent tutorial-generating system (Self-GT) for self-aid learning, integrating cognitive computing and generative learning to capture learners’ dynamic preferences. The critical components of Self-GT are the tutorial-generating model based on cyclic deep reinforcement learning (RL) and the multimodal knowledge graph containing complex relationships. Specifically, the proposed RL model dynamically explores learners’ preferences from the temporal dimension, enabling RL agents to express learning behavior characteristics accurately and generate personalized tutorials. Then, relying on the internal self-developed education base and external Internet sources, a multimodal knowledge graph with multiple self-defined relationships is designed to enhance the precision of tutorial generation. Finally, the experimental results indicate that the Self-GT performs well in generating tutorials and has been successfully applied in the generating tutorial for “Hospital Network Architecture Planning and Design.”