With the booming of cyber attacks and cyber criminals against cyber-physical systems (CPSs), detecting these attacks remains challenging. It might be the worst of times, but it might be the best of times because of opportunities brought by machine learning (ML), in particular deep learning (DL). In general, DL delivers superior performance to ML because of its layered setting and its effective algorithm for extract useful information from training data. DL models are adopted quickly to cyber attacks against CPS systems. In this survey, a holistic view of recently proposed DL solutions is provided to cyber attack detection in the CPS context. A six-step DL driven methodology is provided to summarize and analyze the surveyed literature for applying DL methods to detect cyber attacks against CPS systems. The methodology includes CPS scenario analysis, cyber attack identification, ML problem formulation, DL model customization, data acquisition for training, and performance evaluation. The reviewed works indicate great potential to detect cyber attacks against CPS through DL modules. Moreover, excellent performance is achieved partly because of several high-quality datasets that are readily available for public use. Furthermore, challenges, opportunities, and research trends are pointed out for future research.

Jun Zhang;Lei Pan;Qing-Long Han;Chao Chen;Sheng Wen;Yang Xiang

School of Science,Computing and Engineering Technologies,Swinburne University of Technology,Melbourne,VIC 3122,Australia;School of Information Technology,Deakin University,Geelong,VIC 3216,Australia;College of Science and Engineering,James Cook University,Townsville,QLD 4811,Australia

自动化学报（英文版）

[1]Jun Zhang;Lei Pan;Qing-Long Han;Chao Chen;Sheng Wen;Yang Xiang-.Deep Learning Based Attack Detection for Cyber-Physical System Cybersecurity: A Survey)[J].自动化学报（英文版）,2022(03):377-391

criminals

Deep,Learning,Based,Attack,Detection,Physical,System,Cybersecurity,Survey,With,booming,cyber,attacks,against,physical,systems,CPSs,detecting,these,remains,challenging,It,might,worst,times,but,best,because,opportunities,brought,by,machine,learning,ML,particular,deep,DL,In,general,delivers,superior,performance,its,layered,setting,effective,algorithm,extract,useful,information,from,training,models,are,adopted,quickly,this,holistic,recently,proposed,solutions,provided,detection,context,six,step,driven,methodology,summarize,analyze,surveyed,literature,applying,methods,includes,scenario,analysis,identification,problem,formulation,customization,acquisition,evaluation,reviewed,works,indicate,great,potential,through,modules,Moreover,excellent,achieved,partly,several,high,quality,datasets,that,readily,available,public,Furthermore,challenges,research,trends,pointed,out,future

0.577458