Pub Date : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1139
Hassan Ismail Fawaz, D. Zeghlache, Tran Anh Quang Pham, Jérémie Leguay, P. Medagliani
With the goal of meeting the stringent throughput and delay requirements of classified network flows, we propose a Deep Q-learning Network (DQN) for optimal weight selection in an active queue management system based on Weighted Fair Queuing (WFQ). Our system schedules flows belonging to different priority classes (Gold, Silver, and Bronze) into separate queues, and learns how and when to dequeue from each queue. The neural network implements deep reinforcement learning tools such as target networks and replay buffers to help learn the best weights depending on the network state. We show, via simulations, that our algorithm converges to an efficient model capable of adapting to the flow demands, producing thus lower delays with respect to traditional WFQ.
{"title":"Deep Reinforcement Learning for Smart Queue Management","authors":"Hassan Ismail Fawaz, D. Zeghlache, Tran Anh Quang Pham, Jérémie Leguay, P. Medagliani","doi":"10.14279/TUJ.ECEASST.80.1139","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1139","url":null,"abstract":"With the goal of meeting the stringent throughput and delay requirements of classified network flows, we propose a Deep Q-learning Network (DQN) for optimal weight selection in an active queue management system based on Weighted Fair Queuing (WFQ). Our system schedules flows belonging to different priority classes (Gold, Silver, and Bronze) into separate queues, and learns how and when to dequeue from each queue. The neural network implements deep reinforcement learning tools such as target networks and replay buffers to help learn the best weights depending on the network state. We show, via simulations, that our algorithm converges to an efficient model capable of adapting to the flow demands, producing thus lower delays with respect to traditional WFQ.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125226766","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1149
Paul Walther, Markus Richter, T. Strufe
In wireless network security, physical layer security provides a viable alternative to classical cryptography, which deliver high security guarantees with minimal energy expenditure. Nevertheless, these cryptograhpic primitives are based on assumptions about physical conditions which in practice may not be fulfilled. In this work we present a ray-tracing based attack, which challenges the basic assumption of uncorrelated channel properties for eavesdroppers. We realize this attack and evaluate it with real world measurement, and thereby show that such attacks can predict channel properties better than previous attacks and are also more generally applicable.
{"title":"Ray-tracing based Inference Attacks on Physical Layer Security","authors":"Paul Walther, Markus Richter, T. Strufe","doi":"10.14279/TUJ.ECEASST.80.1149","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1149","url":null,"abstract":"In wireless network security, physical layer security provides a viable alternative to classical cryptography, which deliver high security guarantees with minimal energy expenditure. Nevertheless, these cryptograhpic primitives are based on assumptions about physical conditions which in practice may not be fulfilled. In this work we present a ray-tracing based attack, which challenges the basic assumption of uncorrelated channel properties for eavesdroppers. We realize this attack and evaluate it with real world measurement, and thereby show that such attacks can predict channel properties better than previous attacks and are also more generally applicable.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"243 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127695547","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1160
Quentin Delooz, Andreas Festag, A. Vinel
This paper addresses collective perception for connected and automated driving. It proposes the adaptation of filtering rules based on the currently available channel resources, referred to as Enhanced DCC-Aware Filtering (EDAF).
{"title":"Congestion Aware Objects Filtering for Collective Perception","authors":"Quentin Delooz, Andreas Festag, A. Vinel","doi":"10.14279/TUJ.ECEASST.80.1160","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1160","url":null,"abstract":"This paper addresses collective perception for connected and automated driving. It proposes the adaptation of filtering rules based on the currently available channel resources, referred to as Enhanced DCC-Aware Filtering (EDAF).","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"193 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116463889","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1150
Christoph Funda, K. Hielscher, R. German
Hardware-in-the-loop test benches are distributed computer systems including software, hardware and networking devices, which require strict real-time guarantees. To guarantee strict real-time of the simulator the performance needs to be evaluated. To evaluate the timing performance a discrete event simulation model is built up. The input modeling is based on measurements from the real system in a prototype phase. The results of the simulation model are validated with measurements from a prototype of the real system. The workload is increased until the streaming source becomes unstable, by either exceeding a certain limit of bytes or exceeding the number of parallel software processes running on the cores of the central processing unit. To evaluate the performance beyond these limits, the discrete event simulation model needs to be enriched by a scheduler and a hardware model. To provide real-time guarantees an analytical model needs to be built up.
{"title":"Discrete event simulation for the purpose of real-time performance evaluation of distributed hardware-in-the-loop simulators for autonomous driving vehicle validation","authors":"Christoph Funda, K. Hielscher, R. German","doi":"10.14279/TUJ.ECEASST.80.1150","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1150","url":null,"abstract":"Hardware-in-the-loop test benches are distributed computer systems including software, hardware and networking devices, which require strict real-time guarantees. To guarantee strict real-time of the simulator the performance needs to be evaluated. To evaluate the timing performance a discrete event simulation model is built up. The input modeling is based on measurements from the real system in a prototype phase. The results of the simulation model are validated with measurements from a prototype of the real system. The workload is increased until the streaming source becomes unstable, by either exceeding a certain limit of bytes or exceeding the number of parallel software processes running on the cores of the central processing unit. To evaluate the performance beyond these limits, the discrete event simulation model needs to be enriched by a scheduler and a hardware model. To provide real-time guarantees an analytical model needs to be built up.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125472246","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1131
Jens Dede, A. Udugama, Anna Förster
Repeatability, reproducibility, and replicability are essential aspects of experimental and simulation-driven research. Use of benchmarks in such evaluations further as sists corroborative performance evaluations. In this work, we present a demonstrator of a simulation service, called ”OPS on the bench” which tackles these challenges in performance evaluations of opportunistic networks.
{"title":"Demo: Simulation-as-a-Service to Benchmark Opportunistic Networks","authors":"Jens Dede, A. Udugama, Anna Förster","doi":"10.14279/TUJ.ECEASST.80.1131","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1131","url":null,"abstract":"Repeatability, reproducibility, and replicability are essential aspects of experimental and simulation-driven research. Use of benchmarks in such evaluations further as sists corroborative performance evaluations. In this work, we present a demonstrator of a simulation service, called ”OPS on the bench” which tackles these challenges in performance evaluations of opportunistic networks.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127802788","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1178.1114
Ahmed M. Elmisery, Mirela Sertovic
Phishing attacks are deceptive types of social engineering techniques that attackers use to imitate genuine websites in order to steal the login credentials and private data of the end-users. The continued success of these attacks is heavily attributed to the prolific adoption of online services and the lack of proper training to foster a security awareness mindset of online users. In addition to the financial and reputational damages caused by data breaches of individual users and businesses, cyber adversaries can further use the leaked data for various malicious purposes. In this work, a modular platform was introduced that facilitates accurate detection and automatic evaluation of websites visited by employees of a company or organization. The basis for this approach is a preceding website analysis, which is essential when hunting for potential threats from proxy logs. The platform contains three modules. Characterization of suspicious websites relies on a set of pre-defined features and a multi-stage threat intelligence technique, the functionality of which has been ascertained in initial tests on real data sets
{"title":"Modular Platform for Detecting and Classifying Phishing Websites Using Cyber Threat Intelligence","authors":"Ahmed M. Elmisery, Mirela Sertovic","doi":"10.14279/TUJ.ECEASST.80.1178.1114","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1178.1114","url":null,"abstract":"Phishing attacks are deceptive types of social engineering techniques that attackers use to imitate genuine websites in order to steal the login credentials and private data of the end-users. The continued success of these attacks is heavily attributed to the prolific adoption of online services and the lack of proper training to foster a security awareness mindset of online users. In addition to the financial and reputational damages caused by data breaches of individual users and businesses, cyber adversaries can further use the leaked data for various malicious purposes. In this work, a modular platform was introduced that facilitates accurate detection and automatic evaluation of websites visited by employees of a company or organization. The basis for this approach is a preceding website analysis, which is essential when hunting for potential threats from proxy logs. The platform contains three modules. Characterization of suspicious websites relies on a set of pre-defined features and a multi-stage threat intelligence technique, the functionality of which has been ascertained in initial tests on real data sets","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127656589","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1151
Sebastian Reuter, Jens Hiller, J. Pennekamp, A. Panchenko, Klaus Wehrle
Website fingerprinting (WFP) attacks on the anonymity network Tor have become ever more effective. Furthermore, research discovered that proposed defenses are insufficient or cause high overhead. In previous work, we presented a new WFP defense for Tor that incorporates multipath transmissions to repel malicious Tor nodes from conducting WFP attacks. In this demo, we showcase the operation of our traffic splitting defense by visually illustrating the underlying Tor multipath transmission using LED-equipped Raspberry Pis.
{"title":"Demo: Traffic Splitting for Tor - A Defense against Fingerprinting Attacks","authors":"Sebastian Reuter, Jens Hiller, J. Pennekamp, A. Panchenko, Klaus Wehrle","doi":"10.14279/TUJ.ECEASST.80.1151","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1151","url":null,"abstract":"Website fingerprinting (WFP) attacks on the anonymity network Tor have become ever more effective. Furthermore, research discovered that proposed defenses are insufficient or cause high overhead. In previous work, we presented a new WFP defense for Tor that incorporates multipath transmissions to repel malicious Tor nodes from conducting WFP attacks. In this demo, we showcase the operation of our traffic splitting defense by visually illustrating the underlying Tor multipath transmission using LED-equipped Raspberry Pis.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"2019 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114740469","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1177.1119
Patrick Krämer, Andreas Blenk
Traditional routing protocols such as Open Shortest Path First cannot incorporate fast-changing network states due to their inherent slowness and limited expressiveness. To overcome these limitations, we propose COMNAV, a system that uses Reinforcement Learning (RL) to learn a distributed routing protocol tailored to a specific network. COMNAV interprets routing as a navigational problem, in which flows have to find a way from source to destination. Thus, COMNAV has a close connection to congestion games. The key concept and main contribution is the design of the learning process as a congestion game that allows RL to effectively learn a distributed protocol. Game Theory thereby provides a solid foundation against which the policies RL learns can be evaluated, interpreted, and questioned. We evaluate the capabilities of the learning system in two scenarios in which the routing protocol must react to changes in the network state, and make decisions based on the properties of the flow. Our results show that RL can learn the desired behavior and requires the exchange of only 16 bits of information.
{"title":"Navigating Communication Networks with Deep Reinforcement Learning","authors":"Patrick Krämer, Andreas Blenk","doi":"10.14279/TUJ.ECEASST.80.1177.1119","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1177.1119","url":null,"abstract":"Traditional routing protocols such as Open Shortest Path First cannot incorporate fast-changing network states due to their inherent slowness and limited expressiveness. To overcome these limitations, we propose COMNAV, a system that uses Reinforcement Learning (RL) to learn a distributed routing protocol tailored to a specific network. COMNAV interprets routing as a navigational problem, in which flows have to find a way from source to destination. Thus, COMNAV has a close connection to congestion games. The key concept and main contribution is the design of the learning process as a congestion game that allows RL to effectively learn a distributed protocol. Game Theory thereby provides a solid foundation against which the policies RL learns can be evaluated, interpreted, and questioned. We evaluate the capabilities of the learning system in two scenarios in which the routing protocol must react to changes in the network state, and make decisions based on the properties of the flow. Our results show that RL can learn the desired behavior and requires the exchange of only 16 bits of information.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131112816","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1166.1118
M. Cooney, E. Järpe, A. Vinel
What if an autonomous vehicle (AV) could secretly warn of potential threats? “Steganography”, the hiding of messages, is a vital way for vulnerable populations to communicate securely and get help. Here, we shine light on the concept of vehicular steganography (VS) using a speculative approach: We identify some key scenarios, highlighting unique challenges that arise from indirect perception, message generation, and effects of perspective–as well as potential carrier signals and message generation considerations. One observation is that, despite challenges to transmission rates and robustness, physical signals such as locomotion or sound could offer a complementary, currently-unused alternative to traditional methods. The immediate implication is that VS could help to mitigate some costly safety problems–suggesting the benefit of further discussion and ideation.
{"title":"\"Vehicular Steganography\"?: Opportunities and Challenges","authors":"M. Cooney, E. Järpe, A. Vinel","doi":"10.14279/TUJ.ECEASST.80.1166.1118","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1166.1118","url":null,"abstract":"What if an autonomous vehicle (AV) could secretly warn of potential threats? “Steganography”, the hiding of messages, is a vital way for vulnerable populations to communicate securely and get help. Here, we shine light on the concept of vehicular steganography (VS) using a speculative approach: We identify some key scenarios, highlighting unique challenges that arise from indirect perception, message generation, and effects of perspective–as well as potential carrier signals and message generation considerations. One observation is that, despite challenges to transmission rates and robustness, physical signals such as locomotion or sound could offer a complementary, currently-unused alternative to traditional methods. The immediate implication is that VS could help to mitigate some costly safety problems–suggesting the benefit of further discussion and ideation.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124627064","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 : 2021-09-08DOI: 10.14279/TUJ.ECEASST.80.1158.1102
Heiko Bornholdt, David Jost, Philipp Kisters, Michel Rottleuthner, Sehrish Shafeeq, W. Lamersdorf, T. Schmidt, Mathias Fischer
In smart cities, the number of citizen-operated sensor devices continues to increase. This development multiplies the amount of collectible data about our urban environment. Currently, many of those sensors are connected to proprietary cloud services with questionable effects on user privacy. Simply avoiding interconnecting those devices misses out on opportunities to collectively use data as a shared source of information. The city of Hamburg sponsored project Smart Networks for Urban Citizen Participation (SANE) aims at enabling citizens to connect existing sensors to create a city-wide data space while keeping control of devices and gathered data. Interconnecting these sensors enables all citizens to generate higher value information and thus improve their urban environmental awareness. In this paper, we describe the demo of our SANE architecture. The demo presents how citizens connect to the city-wide data space, how data can be collected, processed, and shared in a privately manner.
{"title":"Smart Urban Data Space for Citizen Science","authors":"Heiko Bornholdt, David Jost, Philipp Kisters, Michel Rottleuthner, Sehrish Shafeeq, W. Lamersdorf, T. Schmidt, Mathias Fischer","doi":"10.14279/TUJ.ECEASST.80.1158.1102","DOIUrl":"https://doi.org/10.14279/TUJ.ECEASST.80.1158.1102","url":null,"abstract":"In smart cities, the number of citizen-operated sensor devices continues to increase. This development multiplies the amount of collectible data about our urban environment. Currently, many of those sensors are connected to proprietary cloud services with questionable effects on user privacy. Simply avoiding interconnecting those devices misses out on opportunities to collectively use data as a shared source of information. The city of Hamburg sponsored project Smart Networks for Urban Citizen Participation (SANE) aims at enabling citizens to connect existing sensors to create a city-wide data space while keeping control of devices and gathered data. Interconnecting these sensors enables all citizens to generate higher value information and thus improve their urban environmental awareness. In this paper, we describe the demo of our SANE architecture. The demo presents how citizens connect to the city-wide data space, how data can be collected, processed, and shared in a privately manner.","PeriodicalId":115235,"journal":{"name":"Electron. Commun. Eur. Assoc. Softw. Sci. Technol.","volume":"137 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122826046","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：