{"title":"Knowledge-defined networking using in-band network telemetry","authors":"J. Hyun, N. V. Tu, J. W. Hong","doi":"10.1109/NOMS.2018.8406169","DOIUrl":"https://doi.org/10.1109/NOMS.2018.8406169","url":null,"abstract":"Today's network is rapidly growing in terms of the number of connected devices and becomes more complex, making the complexity of network management increasing. Self-driving network concept is arising to solve those problems and a good candidate for the closed-loop network management solution. Knowledge-defined Networking (KDN), network telemetry and Software-defined Networking (SDN) are fundamental building blocks to realize self-driving network. Using the self-driving network, OPEX can be reduced and the performance of the network can be grown. In this paper, we propose an architecture for collecting network telemetry and combining KDN to work with SDN and P4 INT. We also suggest several use-cases for the self-driving network architecture.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129558681","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}
{"title":"Distributed analytics in fog computing platforms using tensorflow and kubernetes","authors":"P. Tsai, Hua-Jun Hong, A. Cheng, Cheng-Hsin Hsu","doi":"10.1109/APNOMS.2017.8094194","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094194","url":null,"abstract":"Modern Internet-of-Things (IoT) applications produce large amount of data and require powerful analytics approaches, such as Deep Learning to extract useful information. Existing IoT applications transmit the data to resource-rich data centers for analytics. However, it may congest networks, overload data centers, and increase security vulnerability. In this paper, we implement a platform, which integrates resources from data centers (servers) to end devices (IoT devices). We launch distributed analytics applications among the devices without sending everything to the data centers. We analyze challenges to implement such a platform and carefully adopt popular open-source projects to overcome the challenges. We then conduct comprehensive experiments on the implemented platform. The results show: (i) the benefits/limitations of distributed analytics, (ii) the importance of decisions on distributing an application across multiple devices, and (iii) the overhead caused by different components in our platform.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117335769","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}
{"title":"Threshold estimation in self-destructing scheme using regression analysis","authors":"Young Ki Kim, C. Hong","doi":"10.1109/APNOMS.2017.8094192","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094192","url":null,"abstract":"As technologies and services that leverage cloud computing evolve, more and more businesses or individuals are using them. However, as users increasingly use and store personal information in cloud storage, research on privacy protection models in the cloud environment is becoming more important. A self-destructing scheme has been proposed to prevent the decryption of encrypted user data after a certain period of time using a DHT network. However, the existing privacy protection model does not mention the method of setting the threshold value considering the availability and security of the data. Therefore, in this paper, we propose an optimal threshold finding method considering both data availability and security of privacy protection model by applying regression analysis.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115444219","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}
Jee-Tae Park, Kyu-Seok Shim, Sung-Ho Lee, Myung-Sup Kim
{"title":"Classification of application traffic using tensorflow machine learning","authors":"Jee-Tae Park, Kyu-Seok Shim, Sung-Ho Lee, Myung-Sup Kim","doi":"10.1109/APNOMS.2017.8094156","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094156","url":null,"abstract":"Applications are becoming more complicated and diverse as the network environment grows day by day. So, it is important to classify application traffic accurately. Although there are many ways to classify applications traffic, machine learning based approaches are becoming more efficient in nowadays. This is because machine learning methods are more appropriate than existing methods for accurate and efficient applications traffic classification. Payload signature methods have limitations to deal with various patterns and increasing application traffic complexity. In this paper, we propose a method for extracting flow features and a system for classifying applications traffic based on Machine Learning.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115452338","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}
W. Jung, Jeong-Han Yun, Sin-Kyu Kim, Kyu-Seok Shim, Myung-Sup Kim
{"title":"Structured whitelist generation in SCADA network using PrefixSpan algorithm","authors":"W. Jung, Jeong-Han Yun, Sin-Kyu Kim, Kyu-Seok Shim, Myung-Sup Kim","doi":"10.1109/APNOMS.2017.8094163","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094163","url":null,"abstract":"SCADA system works in repeated or periodic used of only limited communication devices. Because of this feature, whitelist based security techniques are widely used and access restriction method using whitelist based static ACL is most commonly applied in security field. Static ACL have advantages in security, but their expressiveness is too simple to express communication using dynamic allocated port. In addition, it does not reflect all the communication characteristics of the control device, and the generated static ACL should always be open regardless of the frequent use. We propose a structured ACL that extends the fixed generation sequence information between the communication and communication-specific periodicity to reflect the mechanical and repetitive communication characteristics of the SCADA system in the static ACL. We demonstrate the feasibility of the proposed Structured ACL model in this paper by applying the real SCADA network traffic.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117063421","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}
{"title":"Scalable TDMA cluster-based MAC (STCM) for multichannel vehicular networks","authors":"N. Shahin, Young-Tak Kim","doi":"10.1109/APNOMS.2017.8094171","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094171","url":null,"abstract":"Recently, several cluster-based time-division multiple access (TDMA)-based MAC protocols have been proposed for VANETs to ensure that all vehicles can deliver safety messages with high reliability and within a bounded delay. The existing TDMA-based MAC protocols for VANETs, however, are analyzed to be poor in scalability in terms of the number of vehicles. In this paper, we propose a TDMA-based MAC protocol that provides reliable delivery of safety messages, high throughput of non-safety messages, and high scalability where the traffic and communication densities are high in a single-hop cluster. The newly proposed scalable TDMA Cluster-based MAC (STCM) is an enhanced version of our previous Enhanced TDMA Cluster-based MAC (ETCM) that supports only up to 156 vehicles in one cluster. With simulations, we confirmed that the scalability of STCM is enhanced in various different density up to 300 vehicles in one cluster. The simulation results show that the STCM also outperforms existing MAC-layer protocols with respect to packet delivery ratio and average end-to-end delay of safety messages.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127455941","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}
{"title":"Effect of the penalty on the QoS received by multihomed clients","authors":"Rohit Tripathi","doi":"10.1109/APNOMS.2017.8094138","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094138","url":null,"abstract":"The number of people with the internet access is increasing. Many people now want a reliable internet service. An internet service provider (ISP) wants to maximize his profit. Whenever an ISP has to make a choice between profit and quality of service (QoS) provided to clients, an ISP is likely to choose profit. It is possible to put pressure on an ISP to provide a good internet service by imposing penalties when bad services are provided. So an ISP will try to maximize his profit and also minimize the penalties paid to clients. The question here is in choosing the appropriate penalty value. In this paper, a system of ISPs and clients is considered. Clients have access to multiple ISPs using multiple SIM cards. An ISP provides internet service guarantee and whenever clients do not get the assured service, a predefined penalty is paid to the clients. The ISP uses best available solution to maximize profit. The system is simulated to find the effect of the penalty value on the average QoS that clients receive. Simulations show that during congestion, the average minimum QoS depends on only the penalty value. This shows that irrespective of the other factors such as ISP size, number of customers of each ISP; the penalty value is the only factor that decides the average minimum QoS that clients get. Based on simulations, a mathematical equation is obtained that shows the relation between the average minimum QoS and the penalty value. This can be used to find the appropriate penalty value that can provide the desired average minimum QoS to clients.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126161354","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}
{"title":"Firmware over the air for home cybersecurity in the Internet of Things","authors":"Che-Chun Teng, Jia-Wei Gong, Ya-Shian Wang, Chin-Ping Chuang, Mei-Chun Chen","doi":"10.1109/APNOMS.2017.8094190","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094190","url":null,"abstract":"With rising of Internet of Things (IoT) technology, billions of IoT devices and sensors interconnect and exchange large amount of data. Along with new IoT applications, there are many emerging important issues, such as device security and data protection. Recently, there are several global distributed denial-of-service (DDoS) attacks against home routers and network cameras. These IoT embedded devices are vulnerable to cyber-attacks because of existing vulnerabilities, such as network access, constrained resources, lack of update, and system heterogeneity. In this paper, we propose an over-the-air firmware update system for home routers cooperating with ISPs' network management system (NMS) and operations support system (OSS). This system has been applied to at least one million routers in Chunghwa telecom and has reached completion rate of firmware update around 96.32%. It keeps firmware of routers up-to-date, assures home cybersecurity of customers and reduces the inconvenience caused to customers while updating.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122196403","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}
{"title":"Ruin theory based modeling of fair spectrum management in LTE-U","authors":"Aunas Manzoor, Nguyen H. Tran, C. Hong","doi":"10.1109/APNOMS.2017.8094199","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094199","url":null,"abstract":"Long Term Evolution-UnUcensed (LTE-U) is a 5G enabling technology in which both LTE and Wi-Fi systems operate together using the same frequency spectrum. LTE can co-exist with Wi-Fi in both 2.5GHz and 5GHz bands for fully utilizing the spectrum. Aggregating these two technologies while implementing a fair allocation of resources among them is a challenge. Moreover, the LTE technology provides better spectral efficiency and is more bandwidth hungry, thus, it can consume more spectrum which is unfair for Wi-Fi systems. Therefore, appropriate spectrum management scheme is needed to be developed that can maintain the fairness among the two coexisting technologies. This paper proposes a ruin theory based redundant spectrum allocation to LTE-U users while providing sufficient fairness to Wi-Fi systems.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128751372","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}
{"title":"Identification of user behavior from flow statistics","authors":"S. Ata, Yusuke Iemura, N. Nakamura, I. Oka","doi":"10.1109/APNOMS.2017.8094176","DOIUrl":"https://doi.org/10.1109/APNOMS.2017.8094176","url":null,"abstract":"Recently, network traffic is becoming strongly biased by user's action taken in an application. In this paper, we propose a method to infer such action (we define as “user behavior”) from the monitored traffic. The proposed method firstly composes a set of traffic features (statistical features of measured traffic flows) and then applies a Supervised Machine Learning (ML) algorithm to identify the user behavior from the statistical features. Through experimental results by using actual traffic, we show that the proposed method achieves around 91% accuracy of identification for 9 major applications, and around 81% accuracy of identification for 43 user behaviors.","PeriodicalId":188636,"journal":{"name":"2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121222412","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}