Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864925
Fatih Acun, Sinan Kalkan, Ebru Aydin Gol
In this study, traffic speed prediction on a large-scale traffic network in Ankara City is performed using deep neural networks. For this purpose, a spatiotemporal deep learning model consisting of Graphical Convolutional Networks and Gated Recurrent Units used as the baseline, and (i) the input space is expanded by temporal embedding to better take into account temporal information, and (ii) to increase the performance for the peak hours of traffic, the loss function is extended with a novel weighting mechanism. Our comprehensive experiments have shown that the proposed method is significantly more successful in peak hours than ARIMA (Autoregressive Integrated Moving Average) and deep learning-based methods.
{"title":"Traffic Prediction with Peak-Aware Temporal Graph Convolutional Networks","authors":"Fatih Acun, Sinan Kalkan, Ebru Aydin Gol","doi":"10.1109/SIU55565.2022.9864925","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864925","url":null,"abstract":"In this study, traffic speed prediction on a large-scale traffic network in Ankara City is performed using deep neural networks. For this purpose, a spatiotemporal deep learning model consisting of Graphical Convolutional Networks and Gated Recurrent Units used as the baseline, and (i) the input space is expanded by temporal embedding to better take into account temporal information, and (ii) to increase the performance for the peak hours of traffic, the loss function is extended with a novel weighting mechanism. Our comprehensive experiments have shown that the proposed method is significantly more successful in peak hours than ARIMA (Autoregressive Integrated Moving Average) and deep learning-based methods.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"53 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113937113","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864719
Usama Derbashi, E. Aptoula
Attention networks enable neural networks to focus on the most beneficial parts of their input. In the context of remote sensing image classification, studies about spatial, spectral and spatial-spectral attention networks have already been reported. In this paper, a network integrating a scale-based attention module, in addition to spatial-spectral attention is proposed. The scale-space has been produced via alpha-trees, in order for the network to focus on the most useful scales. It is tested with two real hyperspectral datasets, where it achieves a performance improvement.
{"title":"Scale-Spectral-Spatial Attention Network for Hyperspectral Image Classification","authors":"Usama Derbashi, E. Aptoula","doi":"10.1109/SIU55565.2022.9864719","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864719","url":null,"abstract":"Attention networks enable neural networks to focus on the most beneficial parts of their input. In the context of remote sensing image classification, studies about spatial, spectral and spatial-spectral attention networks have already been reported. In this paper, a network integrating a scale-based attention module, in addition to spatial-spectral attention is proposed. The scale-space has been produced via alpha-trees, in order for the network to focus on the most useful scales. It is tested with two real hyperspectral datasets, where it achieves a performance improvement.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122463212","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864839
Desdina Kof Ündar, Zeynep Gul Pehlivanli, Ali Arsal
Software-defined networking is in the midst of information security challenges. In the case that distributed denial of service attacks occur against a software-defined network controller, network traffic data becomes vulnerable because of the overload. In this study, distributed denial of service attacks in software defined network are detected by using machine learning based models with different datasets. First, certain features are obtained on the software-defined network for the dataset under normal conditions and under attack traffic. Subsequently, a new data set was generated by using the proposed hybrid feature selection method on the existing data set. The proposed feature selection method algorithm is trained and tested with logistic regression, artificial neural network, k-nearest neighbors and Naive Bayes models. The results show that the use of the hybrid method has positive impacts on the performance metrics and provides lower processing time.
{"title":"A Feature Selection Method Based On Software Defined Networks","authors":"Desdina Kof Ündar, Zeynep Gul Pehlivanli, Ali Arsal","doi":"10.1109/SIU55565.2022.9864839","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864839","url":null,"abstract":"Software-defined networking is in the midst of information security challenges. In the case that distributed denial of service attacks occur against a software-defined network controller, network traffic data becomes vulnerable because of the overload. In this study, distributed denial of service attacks in software defined network are detected by using machine learning based models with different datasets. First, certain features are obtained on the software-defined network for the dataset under normal conditions and under attack traffic. Subsequently, a new data set was generated by using the proposed hybrid feature selection method on the existing data set. The proposed feature selection method algorithm is trained and tested with logistic regression, artificial neural network, k-nearest neighbors and Naive Bayes models. The results show that the use of the hybrid method has positive impacts on the performance metrics and provides lower processing time.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123037068","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864929
Murat Karayaka, Arda Bayer, Semih Balki, E. Anarim, M. Koca
Currently, web-based applications have become a part of every piece of our daily lives. The rapid advancements in these applications which have found its use in variety of sectors has made it necessary for the respective security systems to adapt as fast in order to identify these applications. In this work, some up-to-date and commonly used applications’ web traffic data have been collected for network traffic classification problem and they are presented for the use of researchers. In addition, an analysis of these data with respect to this problem is performed using features based on statistical protocol identification. It has been shown that for the traffic classification, training a Random Forest classifier with these features is more effective than using the mean KL divergence which was used in previous work with these features.
{"title":"Application Based Network Traffic Dataset and SPID Analysis","authors":"Murat Karayaka, Arda Bayer, Semih Balki, E. Anarim, M. Koca","doi":"10.1109/SIU55565.2022.9864929","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864929","url":null,"abstract":"Currently, web-based applications have become a part of every piece of our daily lives. The rapid advancements in these applications which have found its use in variety of sectors has made it necessary for the respective security systems to adapt as fast in order to identify these applications. In this work, some up-to-date and commonly used applications’ web traffic data have been collected for network traffic classification problem and they are presented for the use of researchers. In addition, an analysis of these data with respect to this problem is performed using features based on statistical protocol identification. It has been shown that for the traffic classification, training a Random Forest classifier with these features is more effective than using the mean KL divergence which was used in previous work with these features.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127768711","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864853
Ilhan Selcuk Mert, E. Anarim, M. Koca
While the internet is gaining more and more importance in our daily life, the number of applications used via the internet are increasing at the same speed. Today, fast and accurate classification of data packets transmitted over the network based on the applications has become an important issue in terms of security as well as network management. In this study, with the proposed classification approach, it is aimed to determine which application these network packets belong to, by inspecting their payloads. To classify packets, a multi-phase method based on majority voting is proposed. This method is based on training deep learning-based classifiers using different numbers of packets and updating the classification prediction as the number of packets in the network flow increases. This updated prediction is achieved by majority voting by using the predictions of previous classifiers trained by smaller number of packets from flows. With this approach, more accurate classifications can be made with less number of packages and this allows an early classification without waiting for more packages to arrive. This approach has been tested on real data collected for various applications.
{"title":"Multi-Phase Traffic Classification Based on Payload","authors":"Ilhan Selcuk Mert, E. Anarim, M. Koca","doi":"10.1109/SIU55565.2022.9864853","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864853","url":null,"abstract":"While the internet is gaining more and more importance in our daily life, the number of applications used via the internet are increasing at the same speed. Today, fast and accurate classification of data packets transmitted over the network based on the applications has become an important issue in terms of security as well as network management. In this study, with the proposed classification approach, it is aimed to determine which application these network packets belong to, by inspecting their payloads. To classify packets, a multi-phase method based on majority voting is proposed. This method is based on training deep learning-based classifiers using different numbers of packets and updating the classification prediction as the number of packets in the network flow increases. This updated prediction is achieved by majority voting by using the predictions of previous classifiers trained by smaller number of packets from flows. With this approach, more accurate classifications can be made with less number of packages and this allows an early classification without waiting for more packages to arrive. This approach has been tested on real data collected for various applications.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128579362","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864682
Begüm Yetskn, Sibel Arslan
The Colebrook equation, which calculates the flow friction, is used to calculate pressure loss in ventilation ducts with turbulent flow, pipes with water or oil. The computational complexity of the equation increases when the friction factor occurs on both sides of the equation. In this study, a new Colebrook approach to compute flow friction with lower cost is proposed based on the Immune Plasma Programming (IPP) automatic programming method based on the stages of immune plasma therapy. The success of IPP was compared with Artificial Bee Colony Programming (ABCP), quick ABCP, semantic ABCP, quick semantic ABCP. The simulation results show that IPP can be used to effectively solve real-world problems.
{"title":"Approximation of the Colebrook Equation for Flow Friction with Immune Plasma Programming","authors":"Begüm Yetskn, Sibel Arslan","doi":"10.1109/SIU55565.2022.9864682","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864682","url":null,"abstract":"The Colebrook equation, which calculates the flow friction, is used to calculate pressure loss in ventilation ducts with turbulent flow, pipes with water or oil. The computational complexity of the equation increases when the friction factor occurs on both sides of the equation. In this study, a new Colebrook approach to compute flow friction with lower cost is proposed based on the Immune Plasma Programming (IPP) automatic programming method based on the stages of immune plasma therapy. The success of IPP was compared with Artificial Bee Colony Programming (ABCP), quick ABCP, semantic ABCP, quick semantic ABCP. The simulation results show that IPP can be used to effectively solve real-world problems.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"2006 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128845394","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864692
Y. Can, Koray Büyükoğuz, Efe Batur Giritli, Mustafa Sisik, Fatih Alagöz
Airline ticket price is influenced by several elements, such as flight distance, purchasing time, number of transfers, etc. Furthermore, every carrier has its own proprietary rules and techniques to determine the ticket price accordingly. With recent improvements in Machine Learning (ML), these rules could be inferred and the price variation could be modeled. In this study, we first created the first dataset containing flight prices for Turkey. The flight price dataset consists of over 1000 domestic flights towards the touristic cities in Turkey. We then use machine learning algorithms to model the ticket price based on different origin and destination pairs. We achieved promising results for predicting the flight ticket price on our dataset.
{"title":"Predicting Airfare Price Using Machine Learning Techniques: A Case Study for Turkish Touristic Cities","authors":"Y. Can, Koray Büyükoğuz, Efe Batur Giritli, Mustafa Sisik, Fatih Alagöz","doi":"10.1109/SIU55565.2022.9864692","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864692","url":null,"abstract":"Airline ticket price is influenced by several elements, such as flight distance, purchasing time, number of transfers, etc. Furthermore, every carrier has its own proprietary rules and techniques to determine the ticket price accordingly. With recent improvements in Machine Learning (ML), these rules could be inferred and the price variation could be modeled. In this study, we first created the first dataset containing flight prices for Turkey. The flight price dataset consists of over 1000 domestic flights towards the touristic cities in Turkey. We then use machine learning algorithms to model the ticket price based on different origin and destination pairs. We achieved promising results for predicting the flight ticket price on our dataset.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116659176","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864804
Berat Erdemkilic, Mehmet Akif Yazici
Network units in FANET systems work at high speed with high mobility capability. This increases the frequency of disconnection. Traditional routing algorithms can not perform well for this problem. In this paper, the SDN Based Routing Protocol, which is designed to enhance the communication performances of FANET systems is proposed. Position-based protocols and topology-based reactive and proactive protocols were investigated, and the protocol designed on the basis of SDN technology is compared with them in terms of delay, throughput and control packet overhead. The proposed protocol has been shown to perform better.
{"title":"SDN Based Routing Protocol for FANETs","authors":"Berat Erdemkilic, Mehmet Akif Yazici","doi":"10.1109/SIU55565.2022.9864804","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864804","url":null,"abstract":"Network units in FANET systems work at high speed with high mobility capability. This increases the frequency of disconnection. Traditional routing algorithms can not perform well for this problem. In this paper, the SDN Based Routing Protocol, which is designed to enhance the communication performances of FANET systems is proposed. Position-based protocols and topology-based reactive and proactive protocols were investigated, and the protocol designed on the basis of SDN technology is compared with them in terms of delay, throughput and control packet overhead. The proposed protocol has been shown to perform better.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117036229","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864939
Usama Mirza, Onat Dalmaz, T. Çukur
Magnetic Resonance Imaging (MRI) is an imaging technique used to produce detailed anatomical images. Acquiring multiple contrast MRI images requires long scan times forcing the patient to remain still. Scan times can be reduced by synthesising unacquired contrasts from acquired contrasts. In recent years, deep generative adversarial networks have been used to synthesise contrasts using one-to-one mapping. Deeper networks can solve more complex functions, however, their performance can decline due to problems such as overfitting and vanishing gradients. In this study, we propose adding skip connections to generative models to overcome the decline in performance with increasing complexity. This will allow the network to bypass unnecessary parameters in the model. Our results show an increase in performance in one-to-one image synthesis by integrating skip connections.
{"title":"Skip Connections for Medical Image Synthesis with Generative Adversarial Networks","authors":"Usama Mirza, Onat Dalmaz, T. Çukur","doi":"10.1109/SIU55565.2022.9864939","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864939","url":null,"abstract":"Magnetic Resonance Imaging (MRI) is an imaging technique used to produce detailed anatomical images. Acquiring multiple contrast MRI images requires long scan times forcing the patient to remain still. Scan times can be reduced by synthesising unacquired contrasts from acquired contrasts. In recent years, deep generative adversarial networks have been used to synthesise contrasts using one-to-one mapping. Deeper networks can solve more complex functions, however, their performance can decline due to problems such as overfitting and vanishing gradients. In this study, we propose adding skip connections to generative models to overcome the decline in performance with increasing complexity. This will allow the network to bypass unnecessary parameters in the model. Our results show an increase in performance in one-to-one image synthesis by integrating skip connections.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"154 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115580173","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 : 2022-05-15DOI: 10.1109/SIU55565.2022.9864969
Kübra Aygün, A. Özen
In this study, BER performance of Flip-OFDM, an efficient unipolar version of OFDM for optical wireless communication systems, and classical VLC-OFDM waveforms are analyzed under the effect of multipath fading. Ceiling bounce and Lambertian channel models are used to generate the optical channel impulse response considering the diffused optical wireless channel configuration. Flip-OFDM and classical VLC-OFDM waveforms, in which M-CSK and M-QAM signal constellations are used, are compared on the BER-SNR performance criterion in the 5-taps diffused wireless optical channel environment. From the obtained numerical results, it is seen that approximately 1 dB SNR difference occurs between M-CSK and M-QAM signal constellations. In addition, it is understood from the results that the Flip-OFDM waveform provides approximately 12 dB more SNR gain than the classical VLC-OFDM waveform at 1E-4 BER level.
{"title":"BER Performance Analysis of M-CSK Modulated Flip-OFDM System in Multipath Optical Channel Environment","authors":"Kübra Aygün, A. Özen","doi":"10.1109/SIU55565.2022.9864969","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864969","url":null,"abstract":"In this study, BER performance of Flip-OFDM, an efficient unipolar version of OFDM for optical wireless communication systems, and classical VLC-OFDM waveforms are analyzed under the effect of multipath fading. Ceiling bounce and Lambertian channel models are used to generate the optical channel impulse response considering the diffused optical wireless channel configuration. Flip-OFDM and classical VLC-OFDM waveforms, in which M-CSK and M-QAM signal constellations are used, are compared on the BER-SNR performance criterion in the 5-taps diffused wireless optical channel environment. From the obtained numerical results, it is seen that approximately 1 dB SNR difference occurs between M-CSK and M-QAM signal constellations. In addition, it is understood from the results that the Flip-OFDM waveform provides approximately 12 dB more SNR gain than the classical VLC-OFDM waveform at 1E-4 BER level.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114272416","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：