Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00021
Irune Agirre
The next generation of dependable embedded systems feature autonomy and higher levels of interconnection. Autonomy is commonly achieved with the support of artificial intelligence algorithms that pose high computing demands on the hardware platform, reaching a high performance scale. This involves a dramatic increase in software and hardware complexity, fact that together with the novelty of the technology, raises serious concerns regarding system dependability. Traditional approaches for certification require to demonstrate that the system will be acceptably safe to operate before it is deployed into service. The nature of autonomous systems, with potentially infinite scenarios, configurations and unanticipated interactions, makes it increasingly difficult to support such claim at design time. In this context, the extended networking technologies can be exploited to collect post-deployment evidence that serve to oversee whether safety assumptions are preserved during operation and to continuously improve the system through regular software updates. These software updates are not only convenient for critical bug fixing but also necessary for keeping the interconnected system resilient against security threats. However, such approach requires a recondition of the traditional certification practices.
{"title":"Safe and secure software updates on high-performance embedded systems","authors":"Irune Agirre","doi":"10.1109/DSN-W50199.2020.00021","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00021","url":null,"abstract":"The next generation of dependable embedded systems feature autonomy and higher levels of interconnection. Autonomy is commonly achieved with the support of artificial intelligence algorithms that pose high computing demands on the hardware platform, reaching a high performance scale. This involves a dramatic increase in software and hardware complexity, fact that together with the novelty of the technology, raises serious concerns regarding system dependability. Traditional approaches for certification require to demonstrate that the system will be acceptably safe to operate before it is deployed into service. The nature of autonomous systems, with potentially infinite scenarios, configurations and unanticipated interactions, makes it increasingly difficult to support such claim at design time. In this context, the extended networking technologies can be exploited to collect post-deployment evidence that serve to oversee whether safety assumptions are preserved during operation and to continuously improve the system through regular software updates. These software updates are not only convenient for critical bug fixing but also necessary for keeping the interconnected system resilient against security threats. However, such approach requires a recondition of the traditional certification practices.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116878890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00011
T. D. Noia, Daniele Malitesta, Felice Antonio Merra
Deep learning classifiers are hugely vulnerable to adversarial examples, and their existence raised cybersecurity concerns in many tasks with an emphasis on malware detection, computer vision, and speech recognition. While there is a considerable effort to investigate attacks and defense strategies in these tasks, only limited work explores the influence of targeted attacks on input data (e.g., images, textual descriptions, audio) used in multimedia recommender systems (MR). In this work, we examine the consequences of applying targeted adversarial attacks against the product images of a visual-based MR. We propose a novel adversarial attack approach, called Target Adversarial Attack against Multimedia Recommender Systems (TAaMR), to investigate the modification of MR behavior when the images of a category of low recommended products (e.g., socks) are perturbed to misclassify the deep neural classifier towards the class of more recommended products (e.g., running shoes) with human-level slight images alterations. We explore the TAaMR approach studying the effect of two targeted adversarial attacks (i.e., FGSM and PGD) against input pictures of two state-of-the-art MR (i.e., VBPR and AMR). Extensive experiments on two real-world recommender fashion datasets confirmed the effectiveness of TAaMR in terms of recommendation lists changing while keeping the original human judgment on the perturbed images.
{"title":"TAaMR: Targeted Adversarial Attack against Multimedia Recommender Systems","authors":"T. D. Noia, Daniele Malitesta, Felice Antonio Merra","doi":"10.1109/DSN-W50199.2020.00011","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00011","url":null,"abstract":"Deep learning classifiers are hugely vulnerable to adversarial examples, and their existence raised cybersecurity concerns in many tasks with an emphasis on malware detection, computer vision, and speech recognition. While there is a considerable effort to investigate attacks and defense strategies in these tasks, only limited work explores the influence of targeted attacks on input data (e.g., images, textual descriptions, audio) used in multimedia recommender systems (MR). In this work, we examine the consequences of applying targeted adversarial attacks against the product images of a visual-based MR. We propose a novel adversarial attack approach, called Target Adversarial Attack against Multimedia Recommender Systems (TAaMR), to investigate the modification of MR behavior when the images of a category of low recommended products (e.g., socks) are perturbed to misclassify the deep neural classifier towards the class of more recommended products (e.g., running shoes) with human-level slight images alterations. We explore the TAaMR approach studying the effect of two targeted adversarial attacks (i.e., FGSM and PGD) against input pictures of two state-of-the-art MR (i.e., VBPR and AMR). Extensive experiments on two real-world recommender fashion datasets confirmed the effectiveness of TAaMR in terms of recommendation lists changing while keeping the original human judgment on the perturbed images.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121140090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-w50199.2020.00010
{"title":"DSN-W 2020 Commentary","authors":"","doi":"10.1109/dsn-w50199.2020.00010","DOIUrl":"https://doi.org/10.1109/dsn-w50199.2020.00010","url":null,"abstract":"","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132941930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00031
Anthony Favier, Antonin Messioux, Jérémie Guiochet, J. Fabre, C. Lesire
This paper presents a generic approach to specify a fault tolerant robot controller, and its implementation and validation with ROS and Gazebo. The main idea is to implement a fault tolerance strategy using a fault tree and an ordered set of recovery modules. A fault injection campaign has been carried out with a mobile autonomous robot for airport inspection using simulation with Gazebo and ROS. This successful experiment implements a safety-first strategy.
{"title":"A hierarchical fault tolerant architecture for an autonomous robot","authors":"Anthony Favier, Antonin Messioux, Jérémie Guiochet, J. Fabre, C. Lesire","doi":"10.1109/DSN-W50199.2020.00031","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00031","url":null,"abstract":"This paper presents a generic approach to specify a fault tolerant robot controller, and its implementation and validation with ROS and Gazebo. The main idea is to implement a fault tolerance strategy using a fault tree and an ordered set of recovery modules. A fault injection campaign has been carried out with a mobile autonomous robot for airport inspection using simulation with Gazebo and ROS. This successful experiment implements a safety-first strategy.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124328664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-w50199.2020.00003
{"title":"[Copyright notice]","authors":"","doi":"10.1109/dsn-w50199.2020.00003","DOIUrl":"https://doi.org/10.1109/dsn-w50199.2020.00003","url":null,"abstract":"","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129714632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00017
Hao Zhen, Bo-Cheng Chiou, Yao-Tung Tsou, S. Kuo, Pang-Chieh Wang
Association analysis is an important task in data analysis to find all co-occurrence relationships (i.e., frequent itemsets or confident association rules) from the transactional dataset. An association rule can help people better discover patterns and develop corresponding strategies. The process of data analysis can be highly summarized as a set of queries, where each query is a real-valued function of the dataset. However, without any restriction and protection, accessing the dataset to answer the queries may lead to the disclosure of individual privacy. In this paper, we propose and implement the association rule mining with differential privacy algorithm, which uses multiple support thresholds to reduce the number of candidate itemsets while reflecting the real nature of the items, and uses random truncation and uniform partition to lower the dimensionality of the dataset. We also stabilize the noise scale by adaptively allocating the privacy budgets, and bound the overall privacy loss. In addition, we prove that the association rule mining with differential privacy algorithm satisfies ex post differential privacy, and verify the utility of our association rule mining with differential privacy algorithm through a series of experiments. To the best of our knowledge, our work is the first differentially private association rule mining algorithm under multiple support thresholds.
{"title":"Association Rule Mining with Differential Privacy","authors":"Hao Zhen, Bo-Cheng Chiou, Yao-Tung Tsou, S. Kuo, Pang-Chieh Wang","doi":"10.1109/DSN-W50199.2020.00017","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00017","url":null,"abstract":"Association analysis is an important task in data analysis to find all co-occurrence relationships (i.e., frequent itemsets or confident association rules) from the transactional dataset. An association rule can help people better discover patterns and develop corresponding strategies. The process of data analysis can be highly summarized as a set of queries, where each query is a real-valued function of the dataset. However, without any restriction and protection, accessing the dataset to answer the queries may lead to the disclosure of individual privacy. In this paper, we propose and implement the association rule mining with differential privacy algorithm, which uses multiple support thresholds to reduce the number of candidate itemsets while reflecting the real nature of the items, and uses random truncation and uniform partition to lower the dimensionality of the dataset. We also stabilize the noise scale by adaptively allocating the privacy budgets, and bound the overall privacy loss. In addition, we prove that the association rule mining with differential privacy algorithm satisfies ex post differential privacy, and verify the utility of our association rule mining with differential privacy algorithm through a series of experiments. To the best of our knowledge, our work is the first differentially private association rule mining algorithm under multiple support thresholds.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131329062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00027
B. Sangchoolie, P. Folkesson, Pierre Kleberger, J. Vinter
Embedded electronic systems need to be equipped with different types of security mechanisms to protect themselves and to mitigate the effects of cybersecurity attacks. These mechanisms should be evaluated with respect to their impacts on dependability and security attributes such as availability, reliability, safety, etc. The evaluation is of great importance as, e.g., a security mechanism should never violate the system safety. Therefore, in this paper, we evaluate a comprehensive set of security mechanisms consisting of 17 different types of mechanisms with respect to their impact on dependability and security attributes. The results show that, in general, the use of these mechanisms have positive effect on system dependability and security. However, there are at least three mechanisms that could have negative impacts on system dependability by violating safety and availability requirements. The results support our claim that the analyses such as the ones conducted in this paper are necessary when selecting and implementing an optimal set of safety and security mechanisms.
{"title":"Analysis of Cybersecurity Mechanisms with respect to Dependability and Security Attributes","authors":"B. Sangchoolie, P. Folkesson, Pierre Kleberger, J. Vinter","doi":"10.1109/DSN-W50199.2020.00027","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00027","url":null,"abstract":"Embedded electronic systems need to be equipped with different types of security mechanisms to protect themselves and to mitigate the effects of cybersecurity attacks. These mechanisms should be evaluated with respect to their impacts on dependability and security attributes such as availability, reliability, safety, etc. The evaluation is of great importance as, e.g., a security mechanism should never violate the system safety. Therefore, in this paper, we evaluate a comprehensive set of security mechanisms consisting of 17 different types of mechanisms with respect to their impact on dependability and security attributes. The results show that, in general, the use of these mechanisms have positive effect on system dependability and security. However, there are at least three mechanisms that could have negative impacts on system dependability by violating safety and availability requirements. The results support our claim that the analyses such as the ones conducted in this paper are necessary when selecting and implementing an optimal set of safety and security mechanisms.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124767536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00019
Jimmy Le Rhun, Sylvain Girbal, D. G. Pérez
Dependable systems currently undergo a series of transformations, notably the shift to multi-core processors and the rise of concerns previously limited to the IT domain such as cybersecurity or cloud-like versatility. In this position paper we summarize the key challenges, and some promising solutions.In addition to software-based techniques devised for COTS multi-processors, emerging Open-Source processing platforms provide the ability to experiment domain-specific mitigation techniques that were previously deemed not economically feasible.
{"title":"Open Source Hardware: An Opportunity For Critical Systems","authors":"Jimmy Le Rhun, Sylvain Girbal, D. G. Pérez","doi":"10.1109/DSN-W50199.2020.00019","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00019","url":null,"abstract":"Dependable systems currently undergo a series of transformations, notably the shift to multi-core processors and the rise of concerns previously limited to the IT domain such as cybersecurity or cloud-like versatility. In this position paper we summarize the key challenges, and some promising solutions.In addition to software-based techniques devised for COTS multi-processors, emerging Open-Source processing platforms provide the ability to experiment domain-specific mitigation techniques that were previously deemed not economically feasible.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122091629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-W50199.2020.00028
M. Moradi, Bentley James Oakes, M. Saraoglu, A. Morozov, K. Janschek, J. Denil
Assessing the safety of complex Cyber-Physical Systems (CPS) is a challenge in any industry. Fault Injection (FI) is a proven technique for safety analysis and is recommended by the automotive safety standard ISO 26262. Traditional FI methods require a considerable amount of effort and cost as FI is applied late in the development cycle and is driven by manual effort or random algorithms. In this paper, we propose a Reinforcement Learning (RL) approach to explore the fault space and find critical faults. During the learning process, the RL agent injects and parameterizes faults in the system to cause catastrophic behavior. The fault space is explored based on a reward function that evaluates previous simulation results such that the RL technique tries to predict improved fault timing and values. In this paper, we apply our technique on an Adaptive Cruise Controller with sensor fusion and compare the proposed method with Monte Carlo-based fault injection. The proposed technique is more efficient in terms of fault coverage and time to find the first critical fault.
{"title":"Exploring Fault Parameter Space Using Reinforcement Learning-based Fault Injection","authors":"M. Moradi, Bentley James Oakes, M. Saraoglu, A. Morozov, K. Janschek, J. Denil","doi":"10.1109/DSN-W50199.2020.00028","DOIUrl":"https://doi.org/10.1109/DSN-W50199.2020.00028","url":null,"abstract":"Assessing the safety of complex Cyber-Physical Systems (CPS) is a challenge in any industry. Fault Injection (FI) is a proven technique for safety analysis and is recommended by the automotive safety standard ISO 26262. Traditional FI methods require a considerable amount of effort and cost as FI is applied late in the development cycle and is driven by manual effort or random algorithms. In this paper, we propose a Reinforcement Learning (RL) approach to explore the fault space and find critical faults. During the learning process, the RL agent injects and parameterizes faults in the system to cause catastrophic behavior. The fault space is explored based on a reward function that evaluates previous simulation results such that the RL technique tries to predict improved fault timing and values. In this paper, we apply our technique on an Adaptive Cruise Controller with sensor fusion and compare the proposed method with Monte Carlo-based fault injection. The proposed technique is more efficient in terms of fault coverage and time to find the first critical fault.","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115591679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-w50199.2020.00004
Abdulrahman Mahmoud
{"title":"DSN-W 2020 TOC","authors":"Abdulrahman Mahmoud","doi":"10.1109/dsn-w50199.2020.00004","DOIUrl":"https://doi.org/10.1109/dsn-w50199.2020.00004","url":null,"abstract":"","PeriodicalId":427687,"journal":{"name":"2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114808763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：