The development of intrusion detection systems (IDS) that are adapted to allow routers and network defence systems to detect malicious network traffic disguised as network protocols or normal access is a critical challenge. This paper proposes a novel approach called SCDNN, which combines spectral clustering (SC) and deep neural network (DNN) algorithms. First, the dataset is divided into k subsets based on sample similarity using cluster centres, as in SC. Next, the distance between data points in a testing set and the training set is measured based on similarity features and is fed into the deep neural network algorithm for intrusion detection. Six KDD-Cup99 and NSL-KDD datasets and a sensor network dataset were employed to test the performance of the model. These experimental results indicate that the SCDNN classifier not only performs better than backpropagation neural network (BPNN), support vector machine (SVM), random forest (RF) and Bayes tree models in detection accuracy and the types of abnormal attacks found. It also provides an effective tool of study and analysis of intrusion detection in large networks.
Meiyu is a unique rainy season over the Yangtze and Huaihe River valleys (YHRV), which persists from mid-June to mid-July and accounts for about 50% of the accumulated June-July-August (summer) rainfall in the region (Ding & Chan, 2005;Tao & Chen, 1987). The 2020 summer was marked by the longest Meiyu season over the past 60 years, which started on June 1 and ended on August 2 with a duration of 62 days, about twice of the climatology (Ding et al., 2021). The record-long Meiyu season caused accumulated rainfall averaged over the YHRV (28°-33°N, 110°-122°E) exceeding +720 mm (Figure S1a), setting the highest record since 1961 (Figure S1b). The associated severe floods affected about 45.5 million people and caused a direct economic loss of more than 100 billion Chinese Yuan (Wei et al., 2020). An accurate prediction of this kind of extreme event is imperative for agriculture, the economy, and human health.
Integrated circuits have to be robust to manufacturing variations. This paper presents a new statistical methodology to determine the worst-case corners for a set of circuit performances. Our methodology first estimates response surfaces for circuit performances as quadratic functions of the process parameters with known statistical distributions. These response surface models are then used to extract the worst-case corners in the process parameter space as the points where the circuit performances are at their min/max values corresponding to a specified statistical level. Corners in the process parameter space close to each other are clustered to reduce their number, which reduces the number of simulations required for design verification. We introduce the novel concept of relaxation coefficient to ensure that the corners capture the min/max values of all the circuit performances at the desired statistical level. The corners are realistic since they track the multivariate statistical distribution of the process parameters. Expected worstcase circuit performances can thus be extracted with a small number of simulations suitable for subsequent design verifications. The methodology is demonstrated with examples showing extraction of corners from digital standard cells and also the corners for analog/RF blocks found in typical communication ICs.
In order to protect computing systems from malicious attacks, network intrusion detection systems have become an important part in the security infrastructure. Recently, hybrid models that integrating several machine learning techniques have captured more attention of researchers. In this paper, a novel hybrid model was proposed with the purpose of detecting network intrusion effectively. In the proposed model, Gini index is used to select the optimal subset of features, the gradient boosted decision tree (GBDT) algorithm is adopted to detect network attacks, and the particle swarm optimization (PSO) algorithm is utilized to optimize the parameters of GBDT. The performance of the proposed model is experimentally evaluated in terms of accuracy, detection rate, precision, F1-score, and false alarm rate using the NSL-KDD dataset. Experimental results show that the proposed model is superior to the compared methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.