A revolution in network technology has been ushered in by software defined networking (SDN), which makes it possible to control the network from a central location and provides an overview of the network’s security. Despite this, SDN has a single point of failure that increases the risk of potential threats. Network intrusion detection systems (NIDS) prevent intrusions into a network and preserve the network’s integrity, availability, and confidentiality. Much work has been done on NIDS but there are still improvements needed in reducing false alarms and increasing threat detection accuracy. Recently advanced approaches such as deep learning (DL) and machine learning (ML) have been implemented in SDN-based NIDS to overcome the security issues within a network. In the first part of this survey paper, we offer an introduction to the NIDS theory, as well as recent research that has been conducted on the topic. After that, we conduct a thorough analysis of the most recent ML- and DL-based NIDS approaches to ensure reliable identification of potential security risks. Finally, we focus on the opportunities and difficulties that lie ahead for future research on SDN-based ML and DL for NIDS.

Numerical process modeling of continuous casting is increasingly applied to optimize production practices for higher product quality guarantees, in which real-time temperature prediction is the key factor that makes it possible to observe and optimize the solidification behavior of the continuous casting process. For offline simulations, the completeness of boundary conditions and accuracy of calculation are core concerns; as for online control, the stability of the model calculation and the correctness are equally important. This study demonstrates a process tracking and online control model of continuous casting based on IoT technology and proposes a simplified calculation method for secondary cooling boundary conditions. The proposed method calculates a comprehensive heat transfer coefficient based on offline simulated parameters and computes the total heat flux of a secondary cooling zone. The experimental results show that the temperature field calculated by this method is stable, and the difference between the actual measured temperature is within an acceptable range of 15 °C. The proposed method no longer requires an online model to consider four types of heat transfer boundary conditions, improving the versatility of the online model. The smoother surface temperature curve is beneficial for dynamic control of secondary cooling water. It can be an alternative method for implementing a more conducive online continuous casting model.

Social media has greatly accelerated the speed of information dissemination, but it has also inevitably led to a proliferation of rumors. Due to the deceptive nature of rumors, people often find it difficult to distinguish truth from falsehood, resulting in economic losses and social panic. Existing pre-trained models often focus on keywords while neglecting logic, making them prone to semantic traps. To address this, we propose the Logic-Reasoning Augmented DeBERTa (LoRAD) model. LoRAD utilizes an LLM to generate logical evidence and leverages DeBERTa’s disentangled attention mechanism to effectively integrate this evidence with the source text. We evaluate our method on three public datasets and a newly constructed dataset, Twitter26-Mini. Results show that LoRAD achieves state-of-the-art performance on all datasets. Furthermore, experiments demonstrate that LoRAD offers better performance and robustness than large language models (LLMs) while maintaining high inference speed, making it suitable for real-time rumor detection.

The widespread of Internet of Things (IoT) increases the reliance on Multi-Radio Wireless Mesh Networks (MR-WMN). This kind of networks provides a support to many IoT applications like smart grids, intelligent transportation systems and health monitoring systems. Therefore, the Quality of Service (QoS) becomes of great importance. This can be achieved by assigning the channels among links within the WMNs in a way that improvs the network throughput and capacity while maintaining a fair resource allocation. Although several studies have tried to make such trade-off, they overlook the dynamic nature of WMNs. Such dynamicity is resulted from user mobility which causes rapid changes in network topology and invalids the channel assignment that is currently in place. The static calculation of different network parameters involved in the channel assignment becomes invalid which leads to unfair channel allocation. As such, this study addresses the compatibility issue by bringing about a dynamic approach for weighing the network parameters calculation relying on the current state of the topology. The study proposes a Genetic Algorithm-Based Node Weighting scheme for channel assignment in the WMN (GA-WNR-CA). If a change is happened, the proposed algorithm can adapt and recalculate the weights accordingly. Weights calculation is carried out intelligently using the Genetic Algorithm that search for the optimal value. Once calculated, the network parameters are updated, and based on which the need for new assignment is determined. The experimental results show that the proposed algorithm outperformed those that have been proposed by related works, which conforms the efficacy of the dynamic approach adopted by the algorithm.

Due to the increasing number of students studying in universities globally, the need for effective and timely safety measures has become more critical. This study aims to provide a high tech monitoring system that can help universities realize the security they need. The main functions are mask detection. Among them, mask detection is mainly used to determine if students are wearing the right masks. This paper also carried out algorithm provinciation for two kinds of detection.In the mask detection function, YOLOV4-Tiny model is used, and SPP is added and improved on this basis. And replace the feature enhancement network with the path aggregation network (PAN). After the experiment, the accuracy was improved, Precision (P) and Recall (R) increase by 1.61% and 4.14%.and the response speed of mask detection was improved(The FPS reached 98.67) too. It greatly improves the efficiency of the system and provides security for students.

To improve people’s health safety in public places and strengthen the government’s epidemic prevention and control measures, the accuracy and response speed of mask detection in public areas need to be improved. The current real-time mask detection algorithm is developed based on a one-stage object detection algorithm in deep learning. The key to the algorithm research is how to improve the accuracy and response speed of the algorithm at the same time to meet the efficiency of real-time monitoring. Few of the current algorithms can achieve an accuracy of more than 90% under the same data set, and the FPS can also achieve the standard of real-time monitoring.