Detecting sophisticated cyberattacks,mainly Distributed Denial of Service(DDoS)attacks,with unexpected patterns remains challenging in modern networks.Traditional detection systems often struggle to mitigate such atta...Detecting sophisticated cyberattacks,mainly Distributed Denial of Service(DDoS)attacks,with unexpected patterns remains challenging in modern networks.Traditional detection systems often struggle to mitigate such attacks in conventional and software-defined networking(SDN)environments.While Machine Learning(ML)models can distinguish between benign and malicious traffic,their limited feature scope hinders the detection of new zero-day or low-rate DDoS attacks requiring frequent retraining.In this paper,we propose a novel DDoS detection framework that combines Machine Learning(ML)and Ensemble Learning(EL)techniques to improve DDoS attack detection and mitigation in SDN environments.Our model leverages the“DDoS SDN”dataset for training and evaluation and employs a dynamic feature selection mechanism that enhances detection accuracy by focusing on the most relevant features.This adaptive approach addresses the limitations of conventional ML models and provides more accurate detection of various DDoS attack scenarios.Our proposed ensemble model introduces an additional layer of detection,increasing reliability through the innovative application of ensemble techniques.The proposed solution significantly enhances the model’s ability to identify and respond to dynamic threats in SDNs.It provides a strong foundation for proactive DDoS detection and mitigation,enhancing network defenses against evolving threats.Our comprehensive runtime analysis of Simultaneous Multi-Threading(SMT)on identical configurations shows superior accuracy and efficiency,with significantly reduced computational time,making it ideal for real-time DDoS detection in dynamic,rapidly changing SDNs.Experimental results demonstrate that our model achieves outstanding performance,outperforming traditional algorithms with 99%accuracy using Random Forest(RF)and K-Nearest Neighbors(KNN)and 98%accuracy using XGBoost.展开更多
Applications of Wireless Sensor devices are widely used byvarious monitoring sections such as environmental monitoring, industrialsensing, habitat modeling, healthcare and enemy movement detection systems.Researchers ...Applications of Wireless Sensor devices are widely used byvarious monitoring sections such as environmental monitoring, industrialsensing, habitat modeling, healthcare and enemy movement detection systems.Researchers were found that 16 bytes packet size (payload) requires MediaAccess Control (MAC) and globally unique network addresses overheads asmore as the payload itself which is not reasonable in most situations. Theapproach of using a unique address isn’t preferable for most Wireless SensorNetworks (WSNs) applications as well. Based on the mentioned drawbacks,the current work aims to fill the existing gap in the field area by providingtwo strategies. First, name/address solutions that assign unique addresseslocally to clustered topology-based sensor devices, reutilized in a spatialmanner, and reduce name/address size by a noticeable amount of 2.9 basedon conducted simulation test. Second, name/address solutions that assignreutilizing of names/addresses to location-unaware spanning-tree topologyin an event-driven WSNs case (that is providing minimal low latenciesand delivering addressing packet in an efficient manner). Also, to declinethe approach of needing both addresses (MAC and network) separately, itdiscloses how in a spatial manner to reutilize locally unique sensor devicename approach and could be utilized in both contexts and providing anenergy-efficient protocol for location unawareness clustered based WSNs.In comparison, an experimental simulation test performed and given theaddresses solution with less overhead in the header and 62 percent fairpayload efficiency that outperforms 34 percent less effective globally uniqueaddresses. Furthermore, the proposed work provides addresses uniquenessfor network-level without using network-wide Duplicate Address Detection(DAD) algorithm. Consequently, the current study provides a roadmap foraddressing/naming scheme to help researchers in this field of study. In general,some assumptions were taken during the work phases of this study such asnumber of Cluster Head (CH) nodes is 6% of entire sensor nodes, locationunawareness for entire sensor network and 4 bits per node address space whichconsidered as the limitation of the study.展开更多
文摘Detecting sophisticated cyberattacks,mainly Distributed Denial of Service(DDoS)attacks,with unexpected patterns remains challenging in modern networks.Traditional detection systems often struggle to mitigate such attacks in conventional and software-defined networking(SDN)environments.While Machine Learning(ML)models can distinguish between benign and malicious traffic,their limited feature scope hinders the detection of new zero-day or low-rate DDoS attacks requiring frequent retraining.In this paper,we propose a novel DDoS detection framework that combines Machine Learning(ML)and Ensemble Learning(EL)techniques to improve DDoS attack detection and mitigation in SDN environments.Our model leverages the“DDoS SDN”dataset for training and evaluation and employs a dynamic feature selection mechanism that enhances detection accuracy by focusing on the most relevant features.This adaptive approach addresses the limitations of conventional ML models and provides more accurate detection of various DDoS attack scenarios.Our proposed ensemble model introduces an additional layer of detection,increasing reliability through the innovative application of ensemble techniques.The proposed solution significantly enhances the model’s ability to identify and respond to dynamic threats in SDNs.It provides a strong foundation for proactive DDoS detection and mitigation,enhancing network defenses against evolving threats.Our comprehensive runtime analysis of Simultaneous Multi-Threading(SMT)on identical configurations shows superior accuracy and efficiency,with significantly reduced computational time,making it ideal for real-time DDoS detection in dynamic,rapidly changing SDNs.Experimental results demonstrate that our model achieves outstanding performance,outperforming traditional algorithms with 99%accuracy using Random Forest(RF)and K-Nearest Neighbors(KNN)and 98%accuracy using XGBoost.
文摘Applications of Wireless Sensor devices are widely used byvarious monitoring sections such as environmental monitoring, industrialsensing, habitat modeling, healthcare and enemy movement detection systems.Researchers were found that 16 bytes packet size (payload) requires MediaAccess Control (MAC) and globally unique network addresses overheads asmore as the payload itself which is not reasonable in most situations. Theapproach of using a unique address isn’t preferable for most Wireless SensorNetworks (WSNs) applications as well. Based on the mentioned drawbacks,the current work aims to fill the existing gap in the field area by providingtwo strategies. First, name/address solutions that assign unique addresseslocally to clustered topology-based sensor devices, reutilized in a spatialmanner, and reduce name/address size by a noticeable amount of 2.9 basedon conducted simulation test. Second, name/address solutions that assignreutilizing of names/addresses to location-unaware spanning-tree topologyin an event-driven WSNs case (that is providing minimal low latenciesand delivering addressing packet in an efficient manner). Also, to declinethe approach of needing both addresses (MAC and network) separately, itdiscloses how in a spatial manner to reutilize locally unique sensor devicename approach and could be utilized in both contexts and providing anenergy-efficient protocol for location unawareness clustered based WSNs.In comparison, an experimental simulation test performed and given theaddresses solution with less overhead in the header and 62 percent fairpayload efficiency that outperforms 34 percent less effective globally uniqueaddresses. Furthermore, the proposed work provides addresses uniquenessfor network-level without using network-wide Duplicate Address Detection(DAD) algorithm. Consequently, the current study provides a roadmap foraddressing/naming scheme to help researchers in this field of study. In general,some assumptions were taken during the work phases of this study such asnumber of Cluster Head (CH) nodes is 6% of entire sensor nodes, locationunawareness for entire sensor network and 4 bits per node address space whichconsidered as the limitation of the study.