Pub Date : 2015-08-20DOI: 10.1109/Trustcom.2015.389
Rui Wang, Zhiping Jia, Lei Ju
Software-Defined Networking (SDN) and OpenFlow (OF) protocol have brought a promising architecture for the future networks. However, the centralized control and programmable characteristics also bring a lot of security challenges. Distributed denial-of-service (DDoS) attack is still a security threat to SDN. To detect the DDoS attack in SDN, many researches collect the flow tables from the switch and do the anomaly detection in the controller. But in the large scale network, the collecting process burdens the communication overload between the switches and the controller. Sampling technology may relieve this overload, but it brings a new tradeoff between sampling rate and detection accuracy. In this paper, we first extend a copy of the packet number counter of the flow entry in the OpenFlow table. Based on the flow-based nature of SDN, we design a flow statistics process in the switch. Then, we propose an entropy-based lightweight DDoS flooding attack detection model running in the OF edge switch. This achieves a distributed anomaly detection in SDN and reduces the flow collection overload to the controller. We also give the detailed algorithm which has a small calculation overload and can be easily implemented in SDN software or programmable switch, such as Open vSwitch and NetFPGA. The experimental results show that our detection mechanism can detect the attack quickly and achieve a high detection accuracy with a low false positive rate.
{"title":"An Entropy-Based Distributed DDoS Detection Mechanism in Software-Defined Networking","authors":"Rui Wang, Zhiping Jia, Lei Ju","doi":"10.1109/Trustcom.2015.389","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.389","url":null,"abstract":"Software-Defined Networking (SDN) and OpenFlow (OF) protocol have brought a promising architecture for the future networks. However, the centralized control and programmable characteristics also bring a lot of security challenges. Distributed denial-of-service (DDoS) attack is still a security threat to SDN. To detect the DDoS attack in SDN, many researches collect the flow tables from the switch and do the anomaly detection in the controller. But in the large scale network, the collecting process burdens the communication overload between the switches and the controller. Sampling technology may relieve this overload, but it brings a new tradeoff between sampling rate and detection accuracy. In this paper, we first extend a copy of the packet number counter of the flow entry in the OpenFlow table. Based on the flow-based nature of SDN, we design a flow statistics process in the switch. Then, we propose an entropy-based lightweight DDoS flooding attack detection model running in the OF edge switch. This achieves a distributed anomaly detection in SDN and reduces the flow collection overload to the controller. We also give the detailed algorithm which has a small calculation overload and can be easily implemented in SDN software or programmable switch, such as Open vSwitch and NetFPGA. The experimental results show that our detection mechanism can detect the attack quickly and achieve a high detection accuracy with a low false positive rate.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126003141","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.423
Milica Milutinovic, Italo Dacosta, A. Put, B. Decker
Incentives systems are widely adopted to encourage user actions or contributions that benefit a service provider or a community. In exchange for their actions, users receive incentive points that can be used to obtain benefits or reputation. However, these systems require users to have a central account associated with all their activities. This approach allows providers to collect vast amounts of users' private information, even if pseudonyms are used. In this paper, we present uCentive, a flexible and efficient incentives scheme that allows users to earn and redeem incentives (uCents) that cannot be linked to their identities or actions. In addition, users can prove, if requested, ownership of their incentives without breaking unlinkability guarantees. uCentive also offers perfect forward unlinkability -- even if the user's secrets are compromised, redeemed uCents cannot be linked together or to the user's identity. Even though our scheme relies on heavy cryptography, experimental evaluation shows that it is adequate for mobile devices such as smartphones. We have also made our uCentive library and prototype apps publicly available for further assessment. In short, we provide a practical privacy-preserving incentives scheme that can eliminate users' growing privacy concerns when using such systems.
{"title":"uCentive: An Efficient, Anonymous and Unlinkable Incentives Scheme","authors":"Milica Milutinovic, Italo Dacosta, A. Put, B. Decker","doi":"10.1109/Trustcom.2015.423","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.423","url":null,"abstract":"Incentives systems are widely adopted to encourage user actions or contributions that benefit a service provider or a community. In exchange for their actions, users receive incentive points that can be used to obtain benefits or reputation. However, these systems require users to have a central account associated with all their activities. This approach allows providers to collect vast amounts of users' private information, even if pseudonyms are used. In this paper, we present uCentive, a flexible and efficient incentives scheme that allows users to earn and redeem incentives (uCents) that cannot be linked to their identities or actions. In addition, users can prove, if requested, ownership of their incentives without breaking unlinkability guarantees. uCentive also offers perfect forward unlinkability -- even if the user's secrets are compromised, redeemed uCents cannot be linked together or to the user's identity. Even though our scheme relies on heavy cryptography, experimental evaluation shows that it is adequate for mobile devices such as smartphones. We have also made our uCentive library and prototype apps publicly available for further assessment. In short, we provide a practical privacy-preserving incentives scheme that can eliminate users' growing privacy concerns when using such systems.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126120005","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.611
Ayman Tarakji, Alexander Gladis, Tarek Anwar, R. Leupers
Underutilization as well as oversubscription of processing resources are common problems in current accelerator-based computing systems. Facing these challenges will require intelligent algorithms for scheduling parallel workloads on accelerators. The general aim of this paper is to achieve fair distribution of the tremendous computation power of modern devices among running applications towards enhancing resource utilization. Given a set of real applications, we evaluate our model and explore the advantages of multi-tasking and concurrency on current GPUs.
{"title":"Enhanced GPU Resource Utilization through Fairness-aware Task Scheduling","authors":"Ayman Tarakji, Alexander Gladis, Tarek Anwar, R. Leupers","doi":"10.1109/Trustcom.2015.611","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.611","url":null,"abstract":"Underutilization as well as oversubscription of processing resources are common problems in current accelerator-based computing systems. Facing these challenges will require intelligent algorithms for scheduling parallel workloads on accelerators. The general aim of this paper is to achieve fair distribution of the tremendous computation power of modern devices among running applications towards enhancing resource utilization. Given a set of real applications, we evaluate our model and explore the advantages of multi-tasking and concurrency on current GPUs.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126898033","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.554
Ezgi C. Ozan, S. Kiranyaz, M. Gabbouj
Principal Component Analysis (PCA) is widely used within binary embedding methods for approximate nearest neighbor search and has proven to have a significant effect on the performance. Current methods aim to represent the whole data using a single PCA however, considering the Gaussian distribution requirements of PCA, this representation is not appropriate. In this study we propose using Multiple PCA (M-PCA) transformations to represent the whole data and show that it increases the performance significantly compared to methods using a single PCA.
{"title":"M-PCA Binary Embedding for Approximate Nearest Neighbor Search","authors":"Ezgi C. Ozan, S. Kiranyaz, M. Gabbouj","doi":"10.1109/Trustcom.2015.554","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.554","url":null,"abstract":"Principal Component Analysis (PCA) is widely used within binary embedding methods for approximate nearest neighbor search and has proven to have a significant effect on the performance. Current methods aim to represent the whole data using a single PCA however, considering the Gaussian distribution requirements of PCA, this representation is not appropriate. In this study we propose using Multiple PCA (M-PCA) transformations to represent the whole data and show that it increases the performance significantly compared to methods using a single PCA.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114459588","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.359
Nishant Patanaik, A. Goulart
The concept of cooperative firewalls or customer edge switching (CES) has been proposed to establish secure communication sessions between public and private domains in the global Internet. It allows public (or private) domains to initiate a trusted communication session with a private domain, by using the private host's fully qualified domain name (FQDN) instead of its IP addresses. However, this concept requires further evaluation in real-world scenario deployments that could benefit from having cooperative firewalls. The scenario addressed in this paper is Internet of Things (IoT). An analytical model was developed to estimate the performance in terms of session setup delays and number of servers required for the Customer Edge Traversal Protocol (CETP) to support a large number of IP-based devices.
{"title":"Performance of Cooperative Firewalls in Real-World Deployments","authors":"Nishant Patanaik, A. Goulart","doi":"10.1109/Trustcom.2015.359","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.359","url":null,"abstract":"The concept of cooperative firewalls or customer edge switching (CES) has been proposed to establish secure communication sessions between public and private domains in the global Internet. It allows public (or private) domains to initiate a trusted communication session with a private domain, by using the private host's fully qualified domain name (FQDN) instead of its IP addresses. However, this concept requires further evaluation in real-world scenario deployments that could benefit from having cooperative firewalls. The scenario addressed in this paper is Internet of Things (IoT). An analytical model was developed to estimate the performance in terms of session setup delays and number of servers required for the Customer Edge Traversal Protocol (CETP) to support a large number of IP-based devices.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"172 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121420178","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.479
Anna Rohunen, Jouni Markkula
The collection of information on individual persons for personal data intensive systems and services poses the risk of privacy violations and raises privacy concerns. Individuals' privacy concerns and risk perceptions affect their decision-making on personal data disclosure for services. In the research presented in this paper, data subjects' privacy concerns and risk perceptions were studied by surveying drivers on the possibility of collecting driving data on their vehicles. The research sought to explore the following questions: (1) How are data subjects' risk perceptions related to their privacy concerns, (2) how do risk perceptions and privacy concerns jointly affect willingness to disclose data, (3) how should risk perceptions be incorporated into evaluation of data subjects' privacy behavior? The study's primary findings were as follows: (1) surprisingly, clear dependencies between risk perceptions and privacy concerns were not found, (2) data subjects risk perceptions and two privacy concerns-related factors independently affected their willingness to disclose data -- the two privacy concerns-related factors were the data subjects' perceptions of other drivers' privacy concerns and their discussing information privacy with other drivers, (3) risk perceptions, in combination with privacy concerns, should be incorporated into the data subjects' privacy behavior evaluations. The results of the study contribute to improving the validity of privacy behavior measurements and models.
{"title":"The Role of Risk Perceptions in Privacy Concerns Evaluation","authors":"Anna Rohunen, Jouni Markkula","doi":"10.1109/Trustcom.2015.479","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.479","url":null,"abstract":"The collection of information on individual persons for personal data intensive systems and services poses the risk of privacy violations and raises privacy concerns. Individuals' privacy concerns and risk perceptions affect their decision-making on personal data disclosure for services. In the research presented in this paper, data subjects' privacy concerns and risk perceptions were studied by surveying drivers on the possibility of collecting driving data on their vehicles. The research sought to explore the following questions: (1) How are data subjects' risk perceptions related to their privacy concerns, (2) how do risk perceptions and privacy concerns jointly affect willingness to disclose data, (3) how should risk perceptions be incorporated into evaluation of data subjects' privacy behavior? The study's primary findings were as follows: (1) surprisingly, clear dependencies between risk perceptions and privacy concerns were not found, (2) data subjects risk perceptions and two privacy concerns-related factors independently affected their willingness to disclose data -- the two privacy concerns-related factors were the data subjects' perceptions of other drivers' privacy concerns and their discussing information privacy with other drivers, (3) risk perceptions, in combination with privacy concerns, should be incorporated into the data subjects' privacy behavior evaluations. The results of the study contribute to improving the validity of privacy behavior measurements and models.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121949033","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.353
Richard S. Weiss, L. Reznik, Yanyan Zhuang, Andrew Hoffman, Darrel Pollard, Albert Rafetseder, Tao Li, Justin Cappos
Mobile devices today, such as smartphones and tablets, have become both more complex and diverse. This paper presents a framework to evaluate the trustworthiness of the individual components in a mobile system, as well as the entire system. The major components are applications, devices and networks of devices. Given this diversity and multiple levels of a mobile system, we develop a hierarchical trust evaluation methodology, which enables the combination of trust metrics and allows us to verify the trust metric for each component based on the trust metrics for others. The paper first demonstrates this idea for individual applications and Android-based smartphones. The methodology involves two stages: initial trust evaluation and trust verification. In the first stage, an expert rule system is used to produce trust metrics at the lowest level of the hierarchy. In the second stage, the trust metrics are verified by comparing data from components and a trust evaluation is produced for the combined system. This paper presents the results of two empirical studies, in which this methodology is applied and tested. The first study involves monitoring resource utilization and evaluating trust based on resource consumption patterns. We measured battery voltage, CPU utilization and network communication for individual apps and detected anomalous behavior that could be indicative of malicious code. The second study involves verification of the trust evaluation by comparing the data from two different devices: the GPS location from an Android smartphone in an automobile and the data from an on-board diagnostics (OBD) sensor of the same vehicle.
{"title":"Trust Evaluation in Mobile Devices: An Empirical Study","authors":"Richard S. Weiss, L. Reznik, Yanyan Zhuang, Andrew Hoffman, Darrel Pollard, Albert Rafetseder, Tao Li, Justin Cappos","doi":"10.1109/Trustcom.2015.353","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.353","url":null,"abstract":"Mobile devices today, such as smartphones and tablets, have become both more complex and diverse. This paper presents a framework to evaluate the trustworthiness of the individual components in a mobile system, as well as the entire system. The major components are applications, devices and networks of devices. Given this diversity and multiple levels of a mobile system, we develop a hierarchical trust evaluation methodology, which enables the combination of trust metrics and allows us to verify the trust metric for each component based on the trust metrics for others. The paper first demonstrates this idea for individual applications and Android-based smartphones. The methodology involves two stages: initial trust evaluation and trust verification. In the first stage, an expert rule system is used to produce trust metrics at the lowest level of the hierarchy. In the second stage, the trust metrics are verified by comparing data from components and a trust evaluation is produced for the combined system. This paper presents the results of two empirical studies, in which this methodology is applied and tested. The first study involves monitoring resource utilization and evaluating trust based on resource consumption patterns. We measured battery voltage, CPU utilization and network communication for individual apps and detected anomalous behavior that could be indicative of malicious code. The second study involves verification of the trust evaluation by comparing the data from two different devices: the GPS location from an Android smartphone in an automobile and the data from an on-board diagnostics (OBD) sensor of the same vehicle.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126118634","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.538
F. Jacob, Jens Mittag, H. Hartenstein
The emergence of decentralized crypto currencies such as Bitcoin and the success of the anonymizing network TOR lead to an increased interest in peer-to-peer based technologies lately - not only due to the prevalent deployment of mass network surveillance technologies by authorities around the globe. While today's application services typically employ centralized client/server architectures that require the user to trust the service provider, new decentralized platforms that eliminate this need of trust are on their rise. In this paper we critically analyze a fully decentralized alternative to today's digital ecosystem - MaidSafe - that drops most of the commonly applied principles. The MaidSafe network implements a fully decentralized personal data storage platform on which user applications can be built. The network is made up by individual users who contribute storage, computing power and bandwidth. All communication between network nodes is encrypted, yet users only have to remember a username and password. To guarantee these objectives, MaidSafe combines mechanisms such as Self-Authentication, Self-Encryption, and a P2P-based public key infrastructure. This paper provides a condensed description of MaidSafe's key protocol mechanisms, derives the underlying identity and access management architecture, and evaluates it with respect to security and privacy aspects.
{"title":"A Security Analysis of the Emerging P2P-Based Personal Cloud Platform MaidSafe","authors":"F. Jacob, Jens Mittag, H. Hartenstein","doi":"10.1109/Trustcom.2015.538","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.538","url":null,"abstract":"The emergence of decentralized crypto currencies such as Bitcoin and the success of the anonymizing network TOR lead to an increased interest in peer-to-peer based technologies lately - not only due to the prevalent deployment of mass network surveillance technologies by authorities around the globe. While today's application services typically employ centralized client/server architectures that require the user to trust the service provider, new decentralized platforms that eliminate this need of trust are on their rise. In this paper we critically analyze a fully decentralized alternative to today's digital ecosystem - MaidSafe - that drops most of the commonly applied principles. The MaidSafe network implements a fully decentralized personal data storage platform on which user applications can be built. The network is made up by individual users who contribute storage, computing power and bandwidth. All communication between network nodes is encrypted, yet users only have to remember a username and password. To guarantee these objectives, MaidSafe combines mechanisms such as Self-Authentication, Self-Encryption, and a P2P-based public key infrastructure. This paper provides a condensed description of MaidSafe's key protocol mechanisms, derives the underlying identity and access management architecture, and evaluates it with respect to security and privacy aspects.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132868190","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.577
Jesús Maillo, I. Triguero, F. Herrera
The k-Nearest Neighbor classifier is one of the most well known methods in data mining because of its effectiveness and simplicity. Due to its way of working, the application of this classifier may be restricted to problems with a certain number of examples, especially, when the runtime matters. However, the classification of large amounts of data is becoming a necessary task in a great number of real-world applications. This topic is known as big data classification, in which standard data mining techniques normally fail to tackle such volume of data. In this contribution we propose a MapReduce-based approach for k-Nearest neighbor classification. This model allows us to simultaneously classify large amounts of unseen cases (test examples) against a big (training) dataset. To do so, the map phase will determine the k-nearest neighbors in different splits of the data. Afterwards, the reduce stage will compute the definitive neighbors from the list obtained in the map phase. The designed model allows the k-Nearest neighbor classifier to scale to datasets of arbitrary size, just by simply adding more computing nodes if necessary. Moreover, this parallel implementation provides the exact classification rate as the original k-NN model. The conducted experiments, using a dataset with up to 1 million instances, show the promising scalability capabilities of the proposed approach.
{"title":"A MapReduce-Based k-Nearest Neighbor Approach for Big Data Classification","authors":"Jesús Maillo, I. Triguero, F. Herrera","doi":"10.1109/Trustcom.2015.577","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.577","url":null,"abstract":"The k-Nearest Neighbor classifier is one of the most well known methods in data mining because of its effectiveness and simplicity. Due to its way of working, the application of this classifier may be restricted to problems with a certain number of examples, especially, when the runtime matters. However, the classification of large amounts of data is becoming a necessary task in a great number of real-world applications. This topic is known as big data classification, in which standard data mining techniques normally fail to tackle such volume of data. In this contribution we propose a MapReduce-based approach for k-Nearest neighbor classification. This model allows us to simultaneously classify large amounts of unseen cases (test examples) against a big (training) dataset. To do so, the map phase will determine the k-nearest neighbors in different splits of the data. Afterwards, the reduce stage will compute the definitive neighbors from the list obtained in the map phase. The designed model allows the k-Nearest neighbor classifier to scale to datasets of arbitrary size, just by simply adding more computing nodes if necessary. Moreover, this parallel implementation provides the exact classification rate as the original k-NN model. The conducted experiments, using a dataset with up to 1 million instances, show the promising scalability capabilities of the proposed approach.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133112800","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 : 2015-08-20DOI: 10.1109/Trustcom.2015.453
A. Ometov, K. Zhidanov, S. Bezzateev, R. Florea, S. Andreev, Y. Koucheryavy
Network-assisted device-to-device (D2D) communication is a next-generation wireless technology enabling direct connectivity between proximate user devices under the control of cellular infrastructure. It couples together the centralized and the distributed network architectures, and as such requires respective enablers for secure, private, and trusted data exchange especially when cellular control link is not available at all times. In this work, we conduct the state-of-the-art overview and propose a novel algorithm to maintain security functions of proximate devices in case of unreliable cellular connectivity, whether a new device joins the secure group of users or an existing device leaves it. Our proposed solution and its rigorous mathematical implementation detailed in this work open door to a novel generation of secure proximity-based services and applications in future wireless communication systems.
{"title":"Securing Network-Assisted Direct Communication: The Case of Unreliable Cellular Connectivity","authors":"A. Ometov, K. Zhidanov, S. Bezzateev, R. Florea, S. Andreev, Y. Koucheryavy","doi":"10.1109/Trustcom.2015.453","DOIUrl":"https://doi.org/10.1109/Trustcom.2015.453","url":null,"abstract":"Network-assisted device-to-device (D2D) communication is a next-generation wireless technology enabling direct connectivity between proximate user devices under the control of cellular infrastructure. It couples together the centralized and the distributed network architectures, and as such requires respective enablers for secure, private, and trusted data exchange especially when cellular control link is not available at all times. In this work, we conduct the state-of-the-art overview and propose a novel algorithm to maintain security functions of proximate devices in case of unreliable cellular connectivity, whether a new device joins the secure group of users or an existing device leaves it. Our proposed solution and its rigorous mathematical implementation detailed in this work open door to a novel generation of secure proximity-based services and applications in future wireless communication systems.","PeriodicalId":277092,"journal":{"name":"2015 IEEE Trustcom/BigDataSE/ISPA","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132511363","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：