An Adaptive DDoS Detection and Classification Method in Blockchain Using an Integrated Multi-Models

With the rising adoption of blockchain technology due to its decentralized,secure,and transparent features,ensuring its resilience against network threats,especially Distributed Denial of Service(DDoS)attacks,is crucial.This research addresses the vulnerability of blockchain systems to DDoS assaults,which undermine their core decentralized characteristics,posing threats to their security and reliability.We have devised a novel adaptive integration technique for the detection and identification of varied DDoS attacks.To ensure the robustness and validity of our approach,a dataset amalgamating multiple DDoS attacks was derived from the CIC-DDoS2019 dataset.Using this,our methodology was applied to detect DDoS threats and further classify them into seven unique attack subcategories.To cope with the broad spectrum of DDoS attack variations,a holistic framework has been pro-posed that seamlessly integrates five machine learning models:Gate Recurrent Unit(GRU),Convolutional Neural Networks(CNN),Long-Short Term Memory(LSTM),Deep Neural Networks(DNN),and Support Vector Machine(SVM).The innovative aspect of our framework is the introduction of a dynamic weight adjustment mechanism,enhancing the system’s adaptability.Experimental results substantiate the superiority of our ensemble method in comparison to singular models across various evaluation metrics.The framework displayed remarkable accuracy,with rates reaching 99.71%for detection and 87.62%for classification tasks.By developing a comprehensive and adaptive methodology,this study paves the way for strengthening the defense mechanisms of blockchain systems against DDoS attacks.The ensemble approach,combined with the dynamic weight adjustment,offers promise in ensuring blockchain’s enduring security and trustworthiness.

A DFIM Sensor Faults Multi-Model Diagnosis Approach Based on an Adaptive PI Multiobserver—Experimental Validation

This paper studies the problem of diagnosis strategy for a doubly fed induction motor (DFIM) sensor faults. This strategy is based on unknown input proportional integral (PI) multiobserver. Thecontribution of this paper is on one hand the creation of a new DFIM model based on multi-model approach and, on the other hand, the synthesis of an adaptive PI multi-observer. The DFIM Volt per Hertz drive system behaves as a nonlinear complex system. It consists of a DFIM powered through a controlled PWM Voltage Source Inverter (VSI). The need of a sensorless drive requires soft sensors such as estimators or observers. In particular, an adaptive Proportional-Integral multi-observer is synthesized in order to estimate the DFIM’s outputs which are affected by different faults and to generate the different residual signals symptoms of sensor fault occurrence. The convergence of the estimation error is guaranteed by using the Lyapunov’s based theory. The proposed diagnosis approach is experimentally validated on a 1 kW Induction motor. Obtained simulation results confirm that the adaptive PI multiobserver consent to accomplish the detection, isolation and fault identification tasks with high dynamic performances.

基于自适应多模型扩展卡尔曼滤波的感应电机转速估计方法

本文提出了一种自适应多模型扩展卡尔曼滤波(AME)的转速估计方法。该算法建立了基于速度和磁链估计的多模型EKF,模型的转换遵循马尔可夫链,通过融合不同模型在不同权重下的输出,得到了估计值,并对加权计算进行了研究;同时,利用残差序列可以连续地自适应调整转移概率和系统噪声矩阵,利用后验信息对先验信息进行修正,得到模型间更精确的匹配和转换情况。本文提出的方法提高了模型对实际系统和环境变化的适应性,有效降低了速度估计误差。对基于AMM-EKF的感应电机无速度传感器矢量控制系统进行了实验验证,实验结果验证了算法的正确性和有效性。