Search

Abstract The application of artificial neural networks (ANNs) can be found in numerous fields, including image and speech recognition, natural language processing, and autonomous vehicles. As well, intrusion detection, the subject of this paper, relies heavily on it. Different intrusion detection models have been constructed using ANNs. While ANNs are relatively mature to construct intrusion detection models, some challenges remain. Among the most notorious of these are the bloated models caused by the large number of parameters, and the non-interpretability of the models. Our paper presents Convolutional Kolmogorov-Arnold Networks (CKANs), which are designed to overcome these difficulties and provide an interpretable and accurate intrusion detection model. Kolmogorov-Arnold Networks (KANs) are developed from the Kolmogorov-Arnold representation theorem. Meanwhile, CKAN incorporates a convolutional computational mechanism based on KAN. The model proposed in this paper is constructed by incorporating attention mechanisms into CKAN’s computational logic. The datasets CICIoT2023 and CICIoMT2024 were used for model training and validation. From the results of evaluating the performance indicators of the experiments, the intrusion detection model constructed based on CKANs has an attractive application prospect. As compared with other methods, the model can predict a much higher level of accuracy with significantly fewer parameters. However, it is not superior in terms of memory usage, execution speed and energy consumption.

Abstract Greenhouse gas (GHG) emissions from rivers and lakes have been shown to significantly contribute to global carbon and nitrogen cycling. In spatiotemporal-variable and human-impacted rivers in the grassland region, simultaneous carbon dioxide, methane and nitrous oxide emissions and their relationships under the different land use types are poorly documented. This research estimated greenhouse gas (CO2, CH4, N2O) emissions in the Xilin River of Inner Mongolia of China using direct measurements from 18 field campaigns under seven land use type (such as swamp, sand land, grassland, pond, reservoir, lake, waste water) conducted in 2018. The results showed that CO2 emissions were higher in June and August, mainly affected by pH and DO. Emissions of CH4 and N2O were higher in October, which were influenced by TN and TP. According to global warming potential, CO2 emissions accounted for 63.35% of the three GHG emissions, and CH4 and N2O emissions accounted for 35.98% and 0.66% in the Xilin river, respectively. Under the influence of different degrees of human-impact, the amount of CO2 emissions in the sand land type was very high, however, CH4 emissions and N2O emissions were very high in the artificial pond and the wastewater, respectively. For natural river, the greenhouse gas emissions from the reservoir and sand land were both low. The Xilin river was observed to be a source of carbon dioxide and methane, and the lake was a sink for nitrous oxide.