Pub Date : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333989
Adityan Gurunarayanan, Ankit Agrawal, Ashutosh Bhatia, D. Vishwakarma
The Onion Router (TOR) networks provide anonymity, in terms of identity and location, to the Internet users by encrypting traffic multiple times along the path and routing it via an overlay network of servers. Although TOR was initially developed as a medium to maintain users’ privacy, cyber criminals and hackers take advantage of this anonymity, and as a result, many illegal activities are carried out using TOR networks. With the ever-changing landscape of Internet services, traditional traffic analysis methods are not efficient for analyzing encrypted traffic and there is a need for alternative methods for analyzing TOR traffic. In this paper, we develop a machine learning model to identify whether a given network traffic is TOR or nonTOR. We use the ISCX2016 TOR-nonTOR dataset to train our model and perform random oversampling and random undersampling to remove data imbalance. Furthermore, to improve the efficiency of our classifiers, we use k-fold cross-validation and Grid Search algorithms for hyperparameter tuning. Results show that we achieve more than 90% accuracy with random sampling and hyperparameter tuning methods.
{"title":"Improving the performance of Machine Learning Algorithms for TOR detection","authors":"Adityan Gurunarayanan, Ankit Agrawal, Ashutosh Bhatia, D. Vishwakarma","doi":"10.1109/ICOIN50884.2021.9333989","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333989","url":null,"abstract":"The Onion Router (TOR) networks provide anonymity, in terms of identity and location, to the Internet users by encrypting traffic multiple times along the path and routing it via an overlay network of servers. Although TOR was initially developed as a medium to maintain users’ privacy, cyber criminals and hackers take advantage of this anonymity, and as a result, many illegal activities are carried out using TOR networks. With the ever-changing landscape of Internet services, traditional traffic analysis methods are not efficient for analyzing encrypted traffic and there is a need for alternative methods for analyzing TOR traffic. In this paper, we develop a machine learning model to identify whether a given network traffic is TOR or nonTOR. We use the ISCX2016 TOR-nonTOR dataset to train our model and perform random oversampling and random undersampling to remove data imbalance. Furthermore, to improve the efficiency of our classifiers, we use k-fold cross-validation and Grid Search algorithms for hyperparameter tuning. Results show that we achieve more than 90% accuracy with random sampling and hyperparameter tuning methods.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"15 1","pages":"439-444"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87771704","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9334021
Rahul Kumar, Ankur Gupta, Harkirat Singh Arora, B. Raman
Gliomas originates in glial cells and recognized as one of the most malignant and dangerous brain tumors and categories into two major classes i.e., High Grade Glioma (HGG) and Low Grade Glioma (LGG). Out of both, HGG tumors are more aggressive. Classification of grade of glioma is a crucial task for deciding the treatment therapy and estimating survival period of patient. In this work, a computational approach based on Radiomics and machine learning algorithms, namely GRGE, is proposed to discriminate between HGG and LGG. The approach, GRGE, has performed better than several state-of-art methods proposed in the literature for glioma classification.
{"title":"GRGE: Detection of Gliomas Using Radiomics, GA Features and Extremely Randomized Trees","authors":"Rahul Kumar, Ankur Gupta, Harkirat Singh Arora, B. Raman","doi":"10.1109/ICOIN50884.2021.9334021","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9334021","url":null,"abstract":"Gliomas originates in glial cells and recognized as one of the most malignant and dangerous brain tumors and categories into two major classes i.e., High Grade Glioma (HGG) and Low Grade Glioma (LGG). Out of both, HGG tumors are more aggressive. Classification of grade of glioma is a crucial task for deciding the treatment therapy and estimating survival period of patient. In this work, a computational approach based on Radiomics and machine learning algorithms, namely GRGE, is proposed to discriminate between HGG and LGG. The approach, GRGE, has performed better than several state-of-art methods proposed in the literature for glioma classification.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"20 1","pages":"379-384"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84445942","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333877
Tobias Mueller
On the Internet, trust is difficult to obtain. With the rise of the possibility of obtaining gratis x509 certificates in an automated fashion, the use of TLS for establishing secure connections has significantly increased. However, other use cases, such as end-to-end encrypted messaging, do not yet have an easy method of managing trust in the public keys. This is particularly true for personal communication where two people want to securely exchange messages. While centralised solutions, such as Signal, exist, decentralised and federated protocols lack a way of conveniently and securely exchanging personal certificates.This paper presents a protocol and an implementation for certifying OpenPGP certificates. By offering multiple means of data transport protocols, it achieves robust and resilient certificate exchange between an attestee, the party whose key certificate is to be certified, and an attestor, the party who will express trust in the certificate once seen. The data can be transferred either via the Internet or via proximity-based technologies, i.e. Bluetooth or link-local networking. The former presents a challenge when the parties interested in exchanging certificates are not physically close, because an attacker may tamper with the connection. Our evaluation shows that a passive attacker learns nothing except the publicly visible metadata, e.g. the timings of the transfer while an active attacker can either have success with a very low probability or be detected by the user.
{"title":"Let’s Attest! Multi-modal Certificate Exchange for the Web of Trust","authors":"Tobias Mueller","doi":"10.1109/ICOIN50884.2021.9333877","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333877","url":null,"abstract":"On the Internet, trust is difficult to obtain. With the rise of the possibility of obtaining gratis x509 certificates in an automated fashion, the use of TLS for establishing secure connections has significantly increased. However, other use cases, such as end-to-end encrypted messaging, do not yet have an easy method of managing trust in the public keys. This is particularly true for personal communication where two people want to securely exchange messages. While centralised solutions, such as Signal, exist, decentralised and federated protocols lack a way of conveniently and securely exchanging personal certificates.This paper presents a protocol and an implementation for certifying OpenPGP certificates. By offering multiple means of data transport protocols, it achieves robust and resilient certificate exchange between an attestee, the party whose key certificate is to be certified, and an attestor, the party who will express trust in the certificate once seen. The data can be transferred either via the Internet or via proximity-based technologies, i.e. Bluetooth or link-local networking. The former presents a challenge when the parties interested in exchanging certificates are not physically close, because an attacker may tamper with the connection. Our evaluation shows that a passive attacker learns nothing except the publicly visible metadata, e.g. the timings of the transfer while an active attacker can either have success with a very low probability or be detected by the user.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"20 1","pages":"758-763"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87669339","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9334008
Youbin Jeon, Hosung Baek, Sangheon Pack
To cope with limited capabilities of mobile devices, task offloading in distributed edge computing (DEC) environments is perceived as a promising solution. However, the mobility of devices makes the task offloading a more challenging issue. In this paper, we investigate mobility-awareness for optimal task offloading in DEC environments. To this end, we formulate an optimization problem to minimize the response time of offloaded tasks. Simulation results demonstrate that the mobility-aware task offloading scheme can reduce the response time by 14% $sim 21$% compared with the conventional task offloading schemes without any mobility-awareness.
{"title":"Mobility-Aware Optimal Task Offloading in Distributed Edge Computing","authors":"Youbin Jeon, Hosung Baek, Sangheon Pack","doi":"10.1109/ICOIN50884.2021.9334008","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9334008","url":null,"abstract":"To cope with limited capabilities of mobile devices, task offloading in distributed edge computing (DEC) environments is perceived as a promising solution. However, the mobility of devices makes the task offloading a more challenging issue. In this paper, we investigate mobility-awareness for optimal task offloading in DEC environments. To this end, we formulate an optimization problem to minimize the response time of offloaded tasks. Simulation results demonstrate that the mobility-aware task offloading scheme can reduce the response time by 14% $sim 21$% compared with the conventional task offloading schemes without any mobility-awareness.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"60 1","pages":"65-68"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83592095","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333864
Kurdman Abdulrahman Rasol Rasol, J. Domingo-Pascual
In network architectures based on Software Defined Networking (SDN) the control plane (control logic) is separated from the network data plane (forwarding plane) while traditional network routers combine both. Software Defined networks facilitates a centralized networking system where a logical controller manages the global view of the network. In this paper, we first propose a new metric on the controller placement problem (CPP) that simultaneously considers the communication latency and communication reliability both between switches and controllers and between controllers. Reliability is considered for single-link failure. We model the problem of determining the optimal controller placement to provide low latencies in the control plane traffic. The objective of this study is to minimize the average accumulated latency by jointly taking into account the latency between controller to switches and inter-controller while optimizing their placement for achieving an optimal balance simultaneously. The optimization problem is formulated as a mixed-integer linear programming (MILP) model under the constraints of latency and reliability. We evaluated the performance of our proposed metric by using the Internet2 OS3E network topology. Different from previous work, we focus on the control traffic exchanged among controllers to synchronize their shared data structure. Results demonstrate that the proposed method is promising.
{"title":"Joint Latency and Reliability-Aware Controller Placement","authors":"Kurdman Abdulrahman Rasol Rasol, J. Domingo-Pascual","doi":"10.1109/ICOIN50884.2021.9333864","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333864","url":null,"abstract":"In network architectures based on Software Defined Networking (SDN) the control plane (control logic) is separated from the network data plane (forwarding plane) while traditional network routers combine both. Software Defined networks facilitates a centralized networking system where a logical controller manages the global view of the network. In this paper, we first propose a new metric on the controller placement problem (CPP) that simultaneously considers the communication latency and communication reliability both between switches and controllers and between controllers. Reliability is considered for single-link failure. We model the problem of determining the optimal controller placement to provide low latencies in the control plane traffic. The objective of this study is to minimize the average accumulated latency by jointly taking into account the latency between controller to switches and inter-controller while optimizing their placement for achieving an optimal balance simultaneously. The optimization problem is formulated as a mixed-integer linear programming (MILP) model under the constraints of latency and reliability. We evaluated the performance of our proposed metric by using the Internet2 OS3E network topology. Different from previous work, we focus on the control traffic exchanged among controllers to synchronize their shared data structure. Results demonstrate that the proposed method is promising.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"73 2 1","pages":"197-202"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83622105","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333906
Seunghyeok Oh, Jaeho Choi, Jong-Kook Kim, Joongheon Kim
Convolutional Neural Network (CNN) is a breakthrough learning model that shows outstanding performance in computer vision and deep learning applications. However, it is a relatively burdened model in terms of learning speed and resource usage compared to other learning models when the learning scale becomes large. Quantum Convolutional Neural Network (QCNN) is a novel model as a potential solution using quantum computers to handle this problem. Quantum computers with a limited number of usable qubits needs a resource-efficient method to process large-scale data at once. In addition, Quantum Random Access Memory (QRAM) can store the large data to qubits logarithmically using superposition and entanglement. The QRAM algorithm can design a new QCNN model that can efficiently process in massive data. This paper proposes a more resource and depth efficient model for larger-sized input data and the number of output channels using the QRAM algorithm and efficiently extracting features.
{"title":"Quantum Convolutional Neural Network for Resource-Efficient Image Classification: A Quantum Random Access Memory (QRAM) Approach","authors":"Seunghyeok Oh, Jaeho Choi, Jong-Kook Kim, Joongheon Kim","doi":"10.1109/ICOIN50884.2021.9333906","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333906","url":null,"abstract":"Convolutional Neural Network (CNN) is a breakthrough learning model that shows outstanding performance in computer vision and deep learning applications. However, it is a relatively burdened model in terms of learning speed and resource usage compared to other learning models when the learning scale becomes large. Quantum Convolutional Neural Network (QCNN) is a novel model as a potential solution using quantum computers to handle this problem. Quantum computers with a limited number of usable qubits needs a resource-efficient method to process large-scale data at once. In addition, Quantum Random Access Memory (QRAM) can store the large data to qubits logarithmically using superposition and entanglement. The QRAM algorithm can design a new QCNN model that can efficiently process in massive data. This paper proposes a more resource and depth efficient model for larger-sized input data and the number of output channels using the QRAM algorithm and efficiently extracting features.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"147 1","pages":"50-52"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91448407","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333895
Soyi Jung, Won Joon Yun, Joongheon Kim, Jae-Hyun Kim
This paper proposes a cooperative multi-agent deep reinforcement learning (MADRL) algorithm for energy trading among multiple unmanned aerial vehicles (UAVs) in order to perform big-data processing in a distributed manner. In order to realize UAV-based aerial surveillance or mobile cellular services, seamless and robust wireless charging mechanisms are required for delivering energy sources from charging infrastructure (i.e., charging towers) to UAVs for the consistent operations of the UAVs in the sky. For actively and intelligently managing the charging towers, MADRL-based energy management system (EMS) is proposed and designed for energy trading among the energy storage systems those are equipped with charging towers. If the required energy for charging UAVs is not enough, the purchasing energy from utility company is desired which takes high consts. The main purpose of MADRL-based EMS learning is for minimizing purchasing energy from outside utility company for minimizing operational costs. Our data-intensive performance evaluation verifies that our proposed framework achieves desired performance.
{"title":"Infrastructure-Assisted Cooperative Multi-UAV Deep Reinforcement Energy Trading Learning for Big-Data Processing","authors":"Soyi Jung, Won Joon Yun, Joongheon Kim, Jae-Hyun Kim","doi":"10.1109/ICOIN50884.2021.9333895","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333895","url":null,"abstract":"This paper proposes a cooperative multi-agent deep reinforcement learning (MADRL) algorithm for energy trading among multiple unmanned aerial vehicles (UAVs) in order to perform big-data processing in a distributed manner. In order to realize UAV-based aerial surveillance or mobile cellular services, seamless and robust wireless charging mechanisms are required for delivering energy sources from charging infrastructure (i.e., charging towers) to UAVs for the consistent operations of the UAVs in the sky. For actively and intelligently managing the charging towers, MADRL-based energy management system (EMS) is proposed and designed for energy trading among the energy storage systems those are equipped with charging towers. If the required energy for charging UAVs is not enough, the purchasing energy from utility company is desired which takes high consts. The main purpose of MADRL-based EMS learning is for minimizing purchasing energy from outside utility company for minimizing operational costs. Our data-intensive performance evaluation verifies that our proposed framework achieves desired performance.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"99 1","pages":"159-162"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79486441","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333879
K. Hwang, Sang-Chul Kim
In this paper, we studied a model that can distinguish several different human behaviors. We trained data [1] using the Convolutional Neural Network algorithm. The suggested model showed 94.597% accuracy in distinguishing seven different human activities.
{"title":"A Study of CNN-Based Human Behavior Recognition with Channel State Information","authors":"K. Hwang, Sang-Chul Kim","doi":"10.1109/ICOIN50884.2021.9333879","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333879","url":null,"abstract":"In this paper, we studied a model that can distinguish several different human behaviors. We trained data [1] using the Convolutional Neural Network algorithm. The suggested model showed 94.597% accuracy in distinguishing seven different human activities.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"300 1","pages":"749-751"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79694952","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 : 2021-01-13DOI: 10.1109/ICOIN50884.2021.9333994
A. Stefanov
The paper considers the route distortion for underwater acoustic mobile networks consisting of autonomous underwater vehicles (AUV’s). The AUV’s transmit the information along a multihop route through the network. The simple stop and wait automatic repeat request (ARQ) protocol is implemented on a hop-by-hop basis. The mobility model is direction persistent. Each AUV-to-AUV channel experiences frequency dependent path loss, Ricean fading and interference. Numerical examples are presented to demonstrate the impact of ARQ on the average route distortion.
{"title":"Impact of ARQ on the Distortion Performance of Underwater Acoustic Mobile Networks","authors":"A. Stefanov","doi":"10.1109/ICOIN50884.2021.9333994","DOIUrl":"https://doi.org/10.1109/ICOIN50884.2021.9333994","url":null,"abstract":"The paper considers the route distortion for underwater acoustic mobile networks consisting of autonomous underwater vehicles (AUV’s). The AUV’s transmit the information along a multihop route through the network. The simple stop and wait automatic repeat request (ARQ) protocol is implemented on a hop-by-hop basis. The mobility model is direction persistent. Each AUV-to-AUV channel experiences frequency dependent path loss, Ricean fading and interference. Numerical examples are presented to demonstrate the impact of ARQ on the average route distortion.","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"102 1","pages":"429-431"},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76688670","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 : 2021-01-13DOI: 10.1109/icoin50884.2021.9333887
{"title":"ICOIN 2021 Front Matter","authors":"","doi":"10.1109/icoin50884.2021.9333887","DOIUrl":"https://doi.org/10.1109/icoin50884.2021.9333887","url":null,"abstract":"","PeriodicalId":6741,"journal":{"name":"2021 International Conference on Information Networking (ICOIN)","volume":"108 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79399566","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：