IEEE HotICN 2018 Beta

Accepted Papers of IEEE HotICN 2018

Coupling DAS, SVC and NDN: an SVC-aware cache and forwarding policy for NDN routers
Authors: Wei Liu, Guoqiang Zhang, and Qian Gao
Abstract: Video streaming traffic has grown sharply in recent years. Dynamic Adaptive Streaming (DAS) is widely used in video streaming because of its ability to cope with the network dynamics and terminal heterogeneity. Compared with AVC(advanced video coding), SVC(scalable video coding) can better adapt to network conditions, reduce server storage and improve users’ playback QoE. However, the present TCP/IP network is communication-oriented, focusing on addressing rather than content, which cannot fully realize the potential of DAS and SVC. Named Data Networking (NDN) is a new network architecture that centers around content. Its name based routing, in-network caching, hop-by-hop forwarding and built-in multicast support naturally fits the pattern of DAS and SVC. As the last piece of jigsaw puzzle, coupled with DAS and SVC, NDN is believed to be able to provide better video streaming performance. In this paper, we proposed a customized SVC-aware cache and forwarding policy for ordinary NDN routers to support video streaming applications. Simulation results based on ndnSIM.3 showed that the proposed policy can improve the average playback bitrate and reduce the video stall frequency as well as duration.
Cluster-based Selective Cooperative Caching Strategy in Vehicular Named Data Networking
Authors: Wanying Huang, Tian Song, Yating Yang and Yu Zhang
Abstract: Vehicular named data networking (VNDN) is a potential candidate to be deployed in information rich applications of vehicular communication. Due to vehicle mobility, it is difficult to construct stable and reliable connections between nodes. Consequently, VNDN is suffering lowq uality of experience and heavy network load in data transmission. In order to solve these problems and improve network performance, we propose a Cluster-based Selective Cooperative Caching (CSCoC) strategy in VNDN. The main idea of CSCoC is to cache popular data at a set of vehicles that may establish stable communication with others. In our strategy, cluster dividing method is used to divide vehicles and select cluster heads as caching vehicles based on their mobility patterns. To mitigate the impact from fast vehicle mobility, we design a cooperative caching strategy which allows cluster members to request and fetch data from their cluster heads. Additionally, we put forward a method of computing the local data popularity, and a cache replacement strategy based on the data popularity. Compared with existing strategies, simulation results demonstrate how CSCoC increases VNDN performance by around 70% in terms of average access delay, average hop count and cache hit ratio.
Performance Comparison and Evaluation of WebSocket Frameworks: Netty, Undertow, Vert.x, Grizzly and Jetty
Authors: Yukun Wang, Lei Huang, Xiaoyou Liu, Tao Sun, Kai Lei
Abstract: The WebSocket protocol emerges to supersede existing bidirectional communication technologies that use HTTP as a transport layer. Currently, there are many network application frameworks that support the WebSocket protocol, but have different behaviors in performance of various aspects. To study and compare the performance of common WebSocket frameworks, say Netty, Undertow, Vert.x, Grizzly and Jetty, in this paper, we use concurrency test, flow test, connection test and resource occupancy test. The experiment results show that Netty and Undertow perform better in highly concurrent environments, while Grizzly is suitable for large flow conditions. The results also show that with persistent connection, Netty far outperforms other frameworks, and that Vert.x and Undertow can handle most requests within relatively shorter time. Besides, Netty and Vert.x occupy less CPU and memory resources in comparison with other frameworks.
A QoS-supported Multi-constrained Routing Strategy Based on Ant-Colony Optimization for Named Data Networking
Authors: Yong Zheng, Min Han, Liuting He, Ya Li, Guanglin Xing, Rui Hou
Abstract: Named data networking (NDN) is a typical representation and implementation of content-centric networking (CCN) and serves as a basis for the next-generation Internet. To the best of our knowledge, very few studies have considered multi-metric constraints on routing strategies. Therefore, This paper proposes a multi-constrained routing strategy based on ant-colony optimization (ACO) to satisfy the multi-metric requirements for supporting the QoS of NDN. In the proposed routing strategy, namely ACO-McQoS, bandwidth, delay, and cost are defined as metrics. An optimal combination of these metrics is chosen to accommodate different QoS requests. An extensive simulation is conducted among the shortest path routing, maximum bandwidth path routing, random path routing, and ACO-McQoS. The simulation results show that the proposed routing strategy outperforms other.
DATE: A Decentralized, Anonymous, and Transparent E-voting System
Authors: Wei-Jr Lai, Yung-chen Hsieh, Chih-Wen Hsueh, Ja-Ling Wu
Abstract: A trusted electronic election system requires that all the involved information must go public. However, it focuses not only on transparency but also on privacy issues. In other words, each ballot should be counted anonymously, correctly, and efficiently. In this work, an effective e-voting system is proposed for voters to minimize their trust in the authority or government. We ensure the transparency of election by putting all messages on the Ethereum blockchain; in the meantime, the privacy of individual voter is protected via an effective ring signature mechanism. Besides, the attractive self-tallying feature is also built in our system, which guarantees that everyone who can access the blockchain network is able to tally the result on his own, i.e., no third party is required after the voting phase. More importantly, we ensure the correctness of voting results and keep the Ethereum gas cost of individual participant as low as possible, at the same time. Moreover, the pre-described characteristics of stealth address in our system makes it more suitable for large-scale election on line.
A Competitiveness-driven and Secure Incentive Mechanism for Competitive Organizations Data Sharing: A Contract Theoretic Approach
Authors: Bingyi Guo, Xiaofang Deng, Quansheng Guan and Jie Tian
Abstract: In the era of big data and artificial intelligence, data sharing is desirable for vigorous development of data driven intelligent services. Although data sharing is supported to a certain extent by current mechanisms and technologies, organizations especially with potential competitive relationships might refuse to share their data. One reason is that data holders worry that data sharing improves competitors’ competitiveness. The other reason is that data sharing suffers huge privacy security risk. To address these problems, in this paper, the concept of competitiveness is introduced as a data sharing transaction driving force to eliminate the competitiveness worry of data holders while differential privacy is adopt to protect their privacy. As there is an information asymmetry between data sharers and data demanders, a contract theoretic approach is proposed to motivate data holders to share data with privacy protection, which is expected to achieve a target of win-win and data sharing security. By designing optimal contracts, the data demander can decide rationally how to pay the data holders given the privacy parameter. Moreover, data holders can choose the contract that maximize their utilities. Numerical results substantiate the effectiveness of the the proposed scheme.
A Private Data Protection Scheme Based on Blockchain under Pipeline Model
Authors: Qianyi Dai∗, Kaiyong Xv, Song Guo, Leyu Dai, Zhicheng Zhou
Abstract: In order to solve the security problem that private data can be easily disclosed in the centralized storage mode of the Trusted Third Party (TTP), and to realize the dynamic trusted security storage of private data in distributed environment, we designed a distributed user private data protection scheme based on blockchain technology. In the scheme, secondary sharding algorithm is used to encrypt individual private data and the blockchain storage model under the designed channel system improves the sharding data storage scheme and solves the problems caused by the private data centralization storage. Through the theoretical analysis and experiment of the scheme, the designed scheme of this paper has high security, data synthesis efficiency and accuracy.
An in-network collaborative verification mechanism for defending content poisoning in Named Data Networking
Authors: Haohao Kang, Yi Zhu, Yu Tao, Jianlong Yang
Abstract: The verification mechanism is the key to ensuring the content security in Name Data Networking (NDN). However, due to the limited computational capacity of NDN router, it is difficult to complete the verification task of all received data packets under heavy traffic. As a consequence, content poisoning has become one of the important security risks of current NDN. To solve this problem, we introduce the concept of data packet credibility and propose an in-network cooperative verification mechanism. In our design, the router calculates the credibility of received data packet from two aspects, one is the internal-evaluation estimated by itself, another is the external-evaluation from its upstream routers. After completing the combined evaluation, router further performs a probabilistic verification according to the credibility. For the data packet with high credibility, it will be verified with low probability. Then router forwards the combined evaluation result to its downstream routers on the reverse path by modifying the structure of data packet. Through building a collaborative verification relationship, this mechanism tries to avoid repeatedly verifying the data packets verified by upstream routers. Simulation results show that it can effectively defend content poisoning while significantly reducing content verification overhead.
Towards Application Portability on Blockchains
Authors: Kazuyuki Shudo, Reiki Kanda, Kenji Saito
Abstract: We discuss the issue of what we call incentive mismatch, a fundamental problem with public blockchains supported by economic incentives. This is an open problem, but one potential solution is to make application portable. Portability is desirable for applications on private blockchains. Then, we present examples of middleware designs that enable application portability and, in particular, support migration between blockchains.
Identity Based Approach Under a Unified Service Model for Secure Content Distribution in ICN
Authors: Jiangtao Luo, Guoliang Xu, Chen He, Edmond Jonckheere
Abstract: Various schemes have been proposed for secure content delivery and access control in Information-Centric Networks (ICN). However, it is not trivial to compare their performances due to the lack of unified service model and consistent implementation methods. In this paper, a general service model merging publish-subscribe pattern with ICN framework is proposed to enable performance evaluation of different access control schemes or various implementations. In addition, an identity based hybrid approach under this model is designed and analyzed, in which the content to be delivered is encrypted using a symmetric secret key, which is then protected by an identity-based encryption scheme together with the license to play the content, and later distributed to consumers as requested. Finally, this approach and two existing schemes are implemented on a common cryptography library, and evaluated. Test results show that the proposed approach exhibits better performance and higher energy-efficiency in mobile terminals than those existing ones.
Real-Time Data Retrieval in Named Data Networking
Authors: Spyridon Mastorakis, Peter Gusev, Alexander Afanasyev, Lixia Zhang
Abstract: The Named Data Networking (NDN) architecture names and secures data directly at the network layer, thus enabling in-network data caching, which in turn facilitates large scale data dissemination. Applications fetch the desired data by names, and the data can come from either the original data producers or router caches. This data retrieval design works seamlessly when the applications know the exact names of data, but poses challenges for realtime applications, such as video and audio conferencing, where new participants may not know the exact names of the latest data production when they join. In this paper we present Realtime Data Retrieval (RDR), a simple protocol that enables applications to discover the latest data. Through prototyping and simulation-based study, we show that RDR can effectively retrieve realtime data with minimal additional delays.
A Probability-based Caching Strategy with Consistent Hash in Named Data Networking
Authors: Yang Qin, Weihong Yang, Wu Liu
Abstract: In-network caching is one of prominent features of Named Data Networking (NDN), which greatly improves the performance of data transmission. In this paper, we propose a probability-based caching strategy with consistent hash (Prob-CH). Prob-CH makes caching decision based on the probability that calculated by jointly considering content’s popularity, node’s betweenness, and distance to consumers. The consistent hashing algorithm is used to guarantee that there is at most one copy cached in the network, which can reduce the redundancy of cache. Moreover, Prob-CH forwards a newly arrived Interest according to a dual forwarding strategy, in which the Interest packet will be guided to the Data’s producer and its cached node calculated by consistent hashing, respectively. The simulation results show that the proposed Prob-CH caching strategy can achieve better performance in terms of cache hit ratio, hop counts and server load.
A Top-k Concast Service for Multiple Tiny Data Retrieval in NDN
Authors: Zhuhua Liao,Jian Zhang, Zengde Teng, Yizhi Liu, Aiping Yi
Abstract: Offering a paradigm for retrieving and aggregating multiple data from multiple sources is a crucial requirement in a large content-centric network. However, the major hindrances to this paradigm are network’s dynamic feature, traffic balance, wired forwarding and the absence of cooperation between communications and computations. In this paper, we present a scalable and top-k Concast service on Named Data Networking (NDN). The service enables cooperation between top-k tiny data discovering and aggregating among multiple routers and paths for a user’s Interest that contained a hierarchical name and other constraints. Specifically, multiple types and strategies of tiny data aggregation for merging and processing the positive data and suppressing the negative, futile data, as well as a determination of response completeness are introduced for enhancing relevant results recall and sharing. The experimentation demonstrated the top-k Concast service can effectively improve the service quality, reduce network traffic and shorten response time.
WinCM: A Window based Congestion Control Mechanism for NDN
Authors: Minxiao Wang, Meng Yue, Zhijun Wu
Abstract: Named Data Networking (NDN) architecture inherits the hourglass shape of IP, whereas the narrow waist part is changed from IP addresses to content names. In NDN architecture, consumers can retrieve data from multiple content chunks and through multiple paths by sending Interest packets carrying a given name rather than data objects’ location. These new features lead to the invalidation of end-to-end congestion control mechanisms. In this paper, we propose a novel window based congestion control mechanism (WinCM) to support high-throughput applications in NDN. WinCM contains three modules: Active Queue Management (AQM) module, Consumer Window Adjustment module and Forwarding Strategy module. AQM module detects congestion by monitoring packet-sojourn time and notices consumers and downstream nodes along delivery path. AQM keeps marking each Data packets that are dequeued continuously for the duration of a queue delay, so that consumers can decrease their windows every time they receive a congestion tag. Simultaneously, each downstream node’s Forwarding Strategy module can accurately and instantly adjust the per-interface Unsatisfied Interest Window, which is one of parameters recorded in Measurements Table to decide how to forward Interest packets and avoid congestion. Simulations based on ndnSIM show that WinCM can exploit available bandwidth faster and maximize bandwidth utilization while maintaining lower queue delay.
Chameleon: A Scalable and Adaptive Permissioned Blockchain Architecture
Authors: Guobiao He, Wei Su, Shuai Gao
Abstract: Abstract—Designing blockchain architecture is still an open question and encounters a lot of challenges such as scalability, security, high utilization and so on. In this paper, we propose Chameleon, a scalable and adaptive permissioned blockchain architecture. We adopt the principles of non-forking, high security, scalable and high utilization to design the Chameleon to be suitable for next generation blockchain architecture. In Chameleon, we introduce credit value which can only be acquired through honest behavior to enhance the security of the consensus algorithm. We also introduce the QoS of transactions to meet the various needs of different users. While previous work using sharding to improve the scalability, they either store all transactions in every area or just work independently, it either causes too much redundant data or low resource utilization. Combined with the cloud storage, Chameleon partition the nodes into different area according to different scenario, every area process their own transactions and can also cooperate with each other dynamically to improve transaction throughput and resource utilization.
Name Weighted Round Robin (NWRR) Algorithm for Named Data Networking
Authors: Yiming Jiang, Jiangtao Luo
Abstract: Named Data Networking(NDN)is an emerging communication paradigm to resolve a traffic explosion problem due to repeated and duplicated delivery of large multimedia content. To make NDN being useful more widely, however, it should support various types of traffic and their Quality of Service (QoS)requirements. In this paper, we propose a queue scheduling algorithm named Name Weighted Round Robin (NWRR) that can work on top of the diffserv model in NDN. The proposed algorithm performs combined scheduling of Interest packets and Data packets and modifies the weight of each queue according to the average packet length of different queues. Results in ndnSIM simulator demonstrate that NWRR can provide different levels of service for different priority applications as well as can cope with the issue of lack of fairness caused by different average packet length in each queue.
Improving Traffic Information Retrieval in VANET with NDN
Authors: Zhongda Xia, Peng Yu, Yu Zhang
Abstract: Traffic Information System(TIS)is a major application over vehicular networks. Decentralized TIS relies on vehicle2-vehicle communications, and efficient traffic information dissemination is a major challenge. In this paper, we propose a preliminary design to support the traffic information retrieval in vehicular ad hoc networks (VANET). Our design adopts an information-centric communication model, and a probabilistic forwarding strategy based on names and geolocations. We implemented a prototype of our design and evaluated its performance in ndnSIM.
Bandwidth-based QoS-aware Multisource Architecture for Information-Centric Wireless Multihop Networks
Authors: Jian Kuang, Shun-Zheng Yu
Abstract: Information-Centric Networking (ICN) is one of the representative architectures for Future Internet that focuses on content rather than IP address. Different from traditional TCP/IP networking, any node that caching content can respond to a request for the content. This property makes it convenient that multiple providers deliver different chunks of a content to a requester concurrently, which can offer more reliable network services and better load balancing. However, it may impact quality of service (QoS) because traditional end-to-end QoS control cannot be performed in this scenario. In this paper, a Bandwidth-based QoS-aware Multisource Architecture (BQMA) is proposed for Information-Centric wireless multihop networks, which performs QoS control for multisource routing, forwarding, reservation procedure, and provider selection functions. Simulation results demonstrate the effectiveness of the proposed architecture.
Research and Application of BFT Algorithms Based on the Hybrid Fault Model
Authors: Qichao Zhang, Zhuyun Qi, Xiaoyou Liu, Tao Sun, Kai Lei
Abstract: The recent explosion of interest in blockchain led to a plethora of researches on consensus algorithms. Compared with bitcoin-derived consensus mechanisms, Byzantine FaultTolerance (BFT) algorithms are more applicable for consortium blockchain. However, these algorithms work lies in the foundation that 3f +1nodes are required to tolerate only f faults, which results in high network traffic and cost. The hybrid fault model with Trusted Components (TC) assistance is proposed to reduce the minimum to 2f +1and improve overall performance. In this paper, we firstly introduce the basic principles of BFT algorithms based on the hybrid fault model. Then we analyze different TC abstractions and implementation scenarios, and express our own choices. The general framework for BFT algorithms design choices and the applications of this model in blockchain are also discussed. Therefore, this paper aims to provide guidance for reasonable design of efficient BFT algorithms and application of TC assistance mechanism.
An Insightful Experimental Study of a Sophisticated Interest Flooding Attack in NDN
Authors: Lixia Zhao, Guang Cheng, Xiaoyan Hu, Hua Wu, Jian Gong, Wang Yang, Chengyu Fan
Abstract: NDN (Named Data Networking), a promising next-generation architecture, puts named content in the first place of the network and is resilient to many existing DDoS attacks. However, Interest Flooding Attack (IFA), a typical NDN-specific DDoS attack, has been widely recognized as a serious threat to the development of NDN. The existing countermeasures against IFA mainly aim at the scenario that attackers send spoofed Interests at a fairly high rate and intermediate routers near the attackers can timely detect the attack by themselves. Instead, this work focuses on a more sophisticated scenario that carefully-crafted attackers send Interests at a respectively lower rate at the beginning but gradually speed up to keep the victims’ PIT sizes increasing to eventually deplete the PIT resource for legitimate users. We conduct an insightful experimental study of such sophisticated IFAs on a real-world network topology and our experimental results demonstrate that the statistics of intermediate routers near the attackers change more gradually and slightly in such an attack, which makes it more difficult for an intermediate router near the attackers to detect by itself. Based on the analytical results of this study, we discuss a potential detection and countermeasure mechanism against such a sophisticated IFA in which a central controller monitors the network from a global view.
Per-Packet Protection (PPP) Scheme for Named Data Networking
Authors: Chen He, Jiangtao Luo, Fei Zhang, Zuoqi Jiang, Mengnan Wang
Abstract: Named Data Networking (NDN) is regarded as a promising architecture for the future Internet. Due to the characteristics of in-network caching and name-based routing in NDN, access control cannot be tied to a particular location, and traditional channel-based access control mechanisms are no longer viable, which brings a major challenge to the access control enforcement. To enhance content-based access control in NDN, this paper presents a per-packet protection (PPP) scheme based on a combination of public key encryption and symmetric key cryptography, which adopts one-way hash functions to generate random cipher keys for different data packets. Furthermore, PPP using secret sharing method provides efficient and flexible access control, which supports scalability and collusion resistance. The experimental results prove that our solution introduces acceptable overheads and reduces the computation time at the users.
Joint Faces Scheduling and Bitrate Switching for Dynamic Adaptive Streaming over NDN Based on Stochastic Optimization
Authors: Xi Wei, Xiaobin Tan, Xiangyang Wu, Lei Xu
Abstract: With the fast increase of video traffic transmitting over the Internet, dynamic adaptive streaming (DAS) has emerged as a conventional way for video streaming distribution. At the same time, Named Data Network (NDN) has been proposed to deal with various problems existed in the Internet. However, existing researches in DAS over NDN did not pay much attention to multihoming capability of NDN. Consumers who request data using multiple faces can aggregate bandwidth of different links which results in choosing higher layers of the video, increasing the quality of video requested. Multiple faces streaming can also smoothen the throughput seen by the client, leading to less fluctuation of the layer switching. Moreover, applying multiple faces can decrease the risk of port failure, increase reliability of the system. In this paper, we propose a dynamic adaptive algorithm of joint faces scheduling and bitrate switching for video streaming request over NDN aiming at optimizing long-term quality of Quality of Experience (QoE) and the cost of consumers under a constraint of playback smoothness and fluctuation. The proposed algorithm is based on Lyapunov Optimization which does not need priori knowledge of the dynamic network state information for each port. Both our theoretical analysis as well as experiments show that this algorithm is effective.
A Blockchain-based key Management Scheme for Named Data Networking
Authors: Junjun Lou, Qichao Zhang, Zhuyun Qi, Kai Lei
Abstract: Named Data Networking is built with security which requires each named Data object to be digitally signed by its producer. Thus, the NDN project has proposed a key management model on NDN testbed for verification of the Data packet to be immune to distributing poisoned content. However, in practice, this model poses two challenges for verifying fake content: (1) the centralized architecture easily leads to a single point of failure, especially when the root key fails, its difficult to verify the keys across sites due to the lack of trust between them, and (2) excessive overhead of certificate chain traversal when verifying signature. This paper first proposes a blockchain based key management scheme in NDN to address the problem of lack of mutual trust between sites without trust anchors. Specifically, all site nodes form a permissioned blockchain for storing public key hashes to ensure the authenticity, and the proxy gateway participates in verifying to reduce excessively frequent communication between the router and the blockchain. In addition, the NDN public key content object and the scheme of their storage, verification, and revocation are redesigned. The result of our analysis and evaluation shows that the proposed scheme is capable of supporting less verification numbers and higher verification efficiency.
MANET for Disaster Relief based on NDN
Authors: Yang Jin, Xiaobin Tan, Weiwei Feng, Jinyang Lv, Aerman Tuerxun, Kunpeng Wang
Abstract: During disaster relief situation, there is an acute need for reliable/effective communication and content distribution, which can be supported by Mobile Ad hoc Networks (MANET) and Delay Tolerant Networks (DTN). Traditional IP protocol performs ineffectively in MANET, while Named Data Networking (NDN) can be more beneficial because of many inbuilt features. In this paper, we propose a solution to adopt NDN for communication in the specific context of disaster relief. Firstly, we propose a proactive routing protocol which is also reactive-routing-enabled. Every node broadcasts their existence to neighbors and synchronizes the Network Information Base in order to update its own FIB. A universal entry set in FIB can be used to probe potential paths. It also remains multiple next-hops for every prefix in FIB in order to support multipath forwarding. Secondly, by duplicating or splitting the queue of Interests at a node, it’s able to utilize multipath so as to improve transmission reliability or efficiency. For the convenience of future work, our experiments are carried out in a real platform consisting of some smart cars equipped with Raspberry Pi. Evaluation results show our solution is able to build a communication network by nodes themselves, and supports multipath forwarding with high reliability/efficiency in a disaster relief scenario.
Delay-aware Power Saving Mechanism for 802.11 Wireless LANs via NDN
Authors: Xinfang Xie, Wang Yang*, Kaijin Tian
Abstract: Idle listening (IL) is the major source of power consumption in Wi-Fi interface for mobile station. The reason is that station (STA) does not know when there would be data for it and it has to stay in IL to wait for the data. With the advantages of named data networking, this problem can be solved well, However, stations may have different requirements on delay, the existing power saving mechanism lacks consideration of delay. In order to improve the quality of service (QoS), we apply an innovative network architecture of NDN and mark the real-time requirements of stations in the Interest packets using a flag bit. And we use a priority sending queue to further reduce the latency of the delay-sensitive packets. The requirements of delay and energy efficiency can be satisfied with different settings of the flag bit. The simulation results show that the proposal can reduce the average delay up to 20% compared with the power saving mode in NDN and gain the commendable energy efficiency.
CoinMingle: A Decentralized Coin Mixing Scheme with a Mutual Recognition Delegation Strategy
Authors: Mixue Xu, Chao Yuan, Xueming Si, Gang Yu, Jinhua Fu, Feng Gao
Abstract: Bitcoin is the first decentralized cryptocurrency proposed by SatoshiNakamoto. Although transactions are conducted between pseudonyms, Bitcoin cannot offer strong privacy guarantees. Mixing coins, to some extent, can address this drawback through different technologies, but also meets flexibility, storage and anonymity problems. This paper proposes a decentralized coin mixing scheme CoinMingle, based on ring signature, onetime output address, CoinJoin and our mutual recognition delegation strategy. Our mutual recognition delegation strategy is simple but effective: when a user prepares to submit messages, he will delegate it to other users for anonymity. Users in CoinMingle don’t need pre-compute. And CoinMingle can tolerate different input amounts and plug the leak of personal information in shuffling phase. In addition, CoinMingle owns parallel structure which is more efficient than other coin mixing schemes.
Fast Handover for High-Speed Railway via NDN
Authors: Runtong Chen, Wang Yang*, Fan Wu, Muhua Sun
Abstract: Handover is crucial for quality of communication service in high-speed railway. With the speed increases, the handover becomes more frequent in high-speed railway. However, the hard handover of GSM-R will break down the communication service for a short time. To reduce the handover latency, we propose the Fast Handover Mechanism (FHM) in Named Data Networking-Railway (NDN-R). Named Date Networking (NDN) is content oriented, FHM does not need to establish an end-to-end connection after the handover occurs. Therefore, FHM can fast handover to the next-hop node without interrupt the communication service. In addition, the FHM can avoid the ping-pong handover which further improves the quality of communication. Simulation results show that the FHM can reduce about 72% of the handover latency compared to the traditional handover mechanism. The packet loss rate, average delay can be reduced and the throughput can be improved.
Improve Blockchain Performance using Graph Data Structure and Parallel Mining
Authors: Jia Kan, Shangzhe Chen, Xin Huang
Abstract: Blockchain technology is ushering in another breakout year, the challenge of blockchain still remains to be solved. This paper analyzes the features of Bitcoin and Bitcoin-NG system based on blockchian, proposes an improved method of implementing blockchain systems by replacing the structure of the original chain with the graph data structure. It was named GraphChain. Each block represents a transaction and contains the balance status of the traders. Additionally, as everyone knows all the transactions in Bitcoin system will be baled by only one miner that will result in a lot of wasted effort, so another way to improve resource utilization is to change the original way to compete for miner to election and parallel mining. Researchers simulated blockchain with graph structure and parallel mining through python, and suggested the conceptual new graph model which can improve both capacity and performance.
Intelligent Eco Networking (IEN): an Advanced Future Internet of intelligence for Digital Social Economic Ecosystem
Authors: Scott Turing
Abstract: Intelligent Eco Networking (IEN) aims to gradually evolve to be an advanced infrastructure for future Internet of intelligence which treats valuable content data as the first-class entity, motivated by data exchange, manipulated by artificial intelligence, based on the approaches of software defined, virtualized and programmable devices, additionally by comprehensively measuring the cost and gain of storage, computing and network resources, it also integrates decentralized consensus trust preservation and tokenized fine-grained allocation mechanism of blockchain to establish a prosperous mutual-contributed and mutualbeneficial value-oriented networking industry ecosystem.
Design and Evaluation of a Multi-source Multi-destination Real-time Application on Content Centric Network
Authors: Asit Chakraborti, Syed Obaid Amin, Aytac Azgin and Ravishankar Ravindran
Abstract: Multi-source multi-destination class of applications can range from interactive AR/VR games and high-definition video conferencing to non real-time file sharing applications; and also includes similar applications in other environments including IoT and data center. These applications require efficient synchronization among the end points with varying level of loss and delay tolerance. Also application specific requirements like throughput, end-to-end latency and mobility of end points impose even more stringency. IP based solutions often fail to meet these requirements due to lack of efficient multicast. In this paper we study this problem in the context of multi-party video conferencing system, and propose an information-centric solution based on content centric networking (CCN). CCN, with its innate support for multicast delivery and service centricity, is used to deliver content and control state in a secure and reliable manner. We address the challenge of synchronizing participant states for a real-time experience through a media agnostic notification framework aided by service functions in the CCN network. Then, we implement a proof-of-concept testbed appropriate for this architecture and demonstrate its impact using scalability, QoE and reliability metrics.
Complex Network Based Knowledge Graph Ontology Structure Analysis
Authors: Yuehang Ding, Hongtao Yu, Ruiyang Huang, Yunjie Gu
Abstract: Ontology is the core of knowledge graph. Traditional ontology description and ontology representation rely on ontology descriptional language. This kind of representation method makes it difficult for people to quickly grasp ontology's structure and then reuse it or segment it. To solve this problem, we proposed a method to transform ontologies into complex networks. This paper analyses ontologies’ structural characteristics through ontology visualization and ontologies’ degree distribution, clustering coefficient, average path length and eigenvector centrality. We observed that many ontologies have tree-like structures. Our analyses further revealed that a concept’s importance is positively related to its degree and eigenvector centrality. Experiments in university ontology shows that our method has a good effect in intuitively understanding the ontology structure.
A Survey of Traffic Classification in Software Defined Networks
Authors: Jinghua Yan and Jing Yuan
Abstract: Traffic classification has been widely used in network management, service measurements, network design, security monitoring and advertising. Software defined networks (SDN) is an newly-developing technology, which is capable of address problems in the traditional network by simplifying network management, introducing network programmability, and providing a global view of a network. Recent years, SDN has brought new opportunity to classify traffic. Traffic classification techniques in SDN have been investigated, proposed and developed. This paper looks at emerging research into the traffic classification techniques in SDN. We first introduce SDN and related work of traffic classification, and then review several representative works of traffic classification in SDN. These works are reviewed in line with the choice of classification strategies and contribution to the literature. Research challenges and future directions for SDN traffic classification are also discussed.
Education-Industry Cooperative System Based on Blockchain
Authors: Qin Liu, Qingchen Guan, Xiaowen Yang, Hongming Zhu, Gill Green and Shaohan Yin
Abstract: Blockchain is the underlying technology of Bitcoin. It is considered a prospect by industries such as finance, education and healthcare, due to its decentralization and non-tampering features. In the current Education-Industry cooperation, substantive information asymmetry exists between universities and employing companies. Student credit systems and the historical information data chain need urgent establishment and maintenance. The paper makes use of the transparency and non-tampering features of the blockchain system to implement an Education-Industry cooperative system based on the blockchain Hyperledger framework. Using the Certificate Authority service and transactions in the Hyperledger framework, the system simulates the roles of universities and companies in the system, and enables universities and enterprises to share information transparently, which achieves information symmetry among student skill and knowledge information, demands of enterprise recruitment, and current market trends. This offers a significant attempt on the application of blockchain technology to Education-Industry cooperation as a pilot development for technology deployment.
A Longitude Analysis on Bitcoin Issue Repository
Authors: Chelsea Hinds-Charles, Jenelee Adames, Ye Yang, Yusong Shen, Yong Wang
Abstract: As one of the most successful Blockchain systems, Bitcoin evolved over the past 8 years. The collaborative contribution of its online software development community gradually shaped the functionality and performance of Bitcoin. To date, most discussions around Bitcoin are from technologies underlying the product, as well as market applications. There are very few studies on the development and evolution processes of the Bitcoin software. It is important to investigate on such developmental issues, in order to better understand the development methodologies and lessons learnt from such a spearheaded Blockchain system.This paper serves this purpose by examining the issues data extracted from the Bitcoin GitHub repository from 2011 to 2018. It reports the trends of the major development issues from a longitude perspective. The main results include: 1) the average lifespan of an issue in Bitcoin issue repository is approximately 57 days; and 2) the Top-7 issue types including refactoring, tests, doc, RPC.REST.ZMQ, GUI, bugs, and wallet, accounting for 64.3% of all issues; 3) topic modeling techniques are beneficial in mining popularity and evolution of key issue topics and most problematic architecture components. Using data analysis and visualization techniques, this paper suggests the insights for significant development decisions such as better managing issue repository and strategic allocating of bug resolution effort.
Blockchain-inspired Event Recording System for Autonomous Vehicles
Authors: Hao Guo, Ehsan Meamari, Chien-Chung Shen
Abstract: Autonomous vehicles are capable of sensing their environment and navigating without any human inputs. However, when autonomous vehicles are involved in accidents between themselves or with human subjects, liability must be indubitably decided based on accident forensics. This paper proposes a blockchain-inspired event recording system for autonomous vehicles. Due to the inefficiency and limited usage of certain blockchain features designed for the traditional cryptocurrency applications, we design a new “proof of event” mechanism to achieve indisputable accident forensics by ensuring that event information is trustable and verifiable. Specifically, we propose a dynamic federation consensus scheme to verify and confirm the new block of event data in an efficient way without any central authority. The security capability of the proposed scheme is also analyzed against different threat and attack models.
Towards Using Public Blockchain in Information-Centric Networks: Challenges Imposed by the European Union’s General Data Protection Regulation
Authors: Dominik Schmelz, Gerald Fischer, Phillip Niemeier, Lei Zhu, Thomas Grechenig
Abstract: Advances in computer science have raised concerns about the privacy of personal information. Therefore solutions to address these concerns have to be found. Blockchain enables new approaches to solve privacy issues in distributed systems, but at the same time also raises new concerns with its openness and immutability. The European Union has taken steps towards addressing information privacy concerns and define rights of data subjects and obligations of controllers and processors of personal data. We will apply and discuss these in light of current Blockchain implementations. This will result in a guideline for GDPR compliant Blockchain developments in the future.
DAS for Smart Communities: Challenges and Approaches
Authors: Yi Liu, Haixia Cui, Yide Wang
Abstract: Distributed antenna system (DAS), one of the key technologies to realize smart decisions in wireless communication systems, can settle network smart coverage problem and improve system energy/spectrum efficiency significantly. This paper summarizesandanalyzestheexistingDASresearchforsmartwireless communities. Furthermore, some future research challenges and directions are highlighted.
Evaluating the Reliability of Blockchain Based Internet of Things Applications
Authors: Ying Liu, Kai Zheng, Paul Craig, Yuexuan Li, Yangkai Luo, Xin Huang
Abstract: In this paper we describe an evaluation to test the reliability of a blockchain based Internet of Things application using a continuous-time Markov chain model. The factors affecting the reliability in our system include the number of devices, the reliability of individual devices, and the underlying consensus algorithm etc. The effect of some factors on overall system reliability is tested, and we find that the total number of devices has the most significant factor to affect overall system reliability. So that we can improve system reliability according to the affecting factors.
The Exchange Center: A Case Study of Hybrid Decentralized and Centralized Applications in Blockchai
Authors: Yung-Chen Hsieh, Ja-Ling Wu, Chih-Wen Hsueh
Abstract: Blockchain and smart contract provide a convenient vehicle to build decentralize applications. Among varies Decenrtalized applications (DAPPs), exchange is one of the most mentioned ones. To fully-decentralized an exchange looks somehow like an utopia. Most works tried to use different methods to solve different problems on this subject. According to our observations, the problems caused by blockchain include 1) Long confirmation time, 2) Vulnerable to front running attack, 3) wasting on-chain resource, all these problems make a blockchain based exchange not user friendly. To deal with the above-mentioned problems, a hybrid centralized and decentralized exchange is presented in this work. By simulating the transaction execution processes and eliminating all the unwanted uncertainties, we are confident that the transactions in the proposed exchange center (EC) will be confirmed and executed well on the blockchain, that is, our system allows users to conclude their trades without need of waiting for a long blockchain confirmation time. In conclusion, a Hybrid Centralized and Decentralized Application (HCDAPP) provides a better-quality service, just like a centralized server; at the same time, it also provides a higher-level security, like a decentralized smart contract.
Blockchain-based Crowd-sensing System
Authors: Junqin Huang, Lingkun Kong, Linghe Kong, Zhen Liu, Zhiqiang Liu, Guihai Chen
Abstract: Crowd-sensing systems have gained considerable interests and adoption in recent years. However, most existing crowd-sensing systems rely on central servers, which are subject to low reliability due to the traditional centralized architecture and high service fee. Motivated by constructing a crowdsensing system with security and low service fee, we propose a blockchain-based crowd-sensing system (BCS), which replaces traditional triangle architecture by decentralized blockchain system. Also, in order to accelerate the formation of the fabric of trust, BCS implements applications of smart contracts to reward sensing-task workers, which offers reliable anonymity in the meantime. By leveraging blockchain technology, BCS has good performance in privacy protection and system robustness.
Edge Cache-based Intelligent Content Delivery in Information-Centric Wireless Networks
Authors: Chao Fang, Haipeng Yao, Zhuwei Wang, Yanhui Tu, and Yixin Chen
Abstract: As mobile data traffic increases explosively, content delivery issue in the Internet is a growing concern. In order to fundamentally solve this problem, information-centric networking has recently been proposed and applied to wireless networks. However, the influence of edge caches and related decision schemes on the performance of content delivery is largely ignored in the existing works. In this poster, we propose an edge cache based intelligent content delivery solution to improve network performance in information-centric wireless networks, where a smart router makes caching decisions by analyzing the information of content popularity, users’ mobility and social relation, and its connected access points update cache status of network contents according to the adopted decisions.
Edge Computing and Caching based Blockchain IoT Network
Authors: Fangmin Xu, Fan Yang, Chenglin Zhao,Chao Fang
Abstract: With the development of internet of things (IoT) network in recent years, while the centralization data management brings longer delay while has more safety troubles. Moreover, once a single device is attacked, the system crashes. So, blockchain technique which could realize safety distributed data management is introduced. However, the consensus mechanism of blockchain consumes huge computation resources. And the storage of entire ledger also consumes huge storage resources of the blockchain nodes. Due to the limited capacity of IoT devices, the nodes cannot afford the computation resources to reach consensus and caching resources. This paper introduces an architecture of edge computing based blockchain network which makes use of the computation and caching capacity of edge server to help the IoT devices in reaching consensus and storing data.
Medical Images Sharing System Based on Blockchain and Smart Contract of Credit Scores
Authors: Huanrong Tang, Ning Tong and Jianquan Ouyang
Abstract: At present, medical images account for nearly 70% of medical diagnostic data, which is an important basis for disease diagnosis. However, medical data leakage incidents have occurred in more than 90% medical institutions, the protection of patients' medical data is of great urgency. At present, all types of medical institutions involved in the medical imaging business use the PACS to archive, manage, and use the collected medical images, but only sharing the managed video resources within the organization. This method applies only the traditional data protection strategy and cannot guarantee a higher level security protection for patients’ private information. And patients have no control over medical information at the time of treatment. For this reason, this paper proposes a method of secure sharing of medical images based on smart contracts of block chain and credit scores. Through a block chain based, distributed, reliable database of recording image sharing process, we realize a cross-organizational, cross-regional, trustworthy, and supervisory medical image sharing system. And the establishment of smart contracts based on credit scores of patients and medical institutions guarantee intelligent sharing by rules and conditions. Compared with traditional PACS, the method proposed in this paper extends its scope of application on the basis of PACS, increases its robustness, and provides new ideas for more extensive, multi-level, safe and reliable medical image sharing.
Multiple Attributes Based Spoofing Detection Using an Improved Clustering Algorithm in Mobile Edge Network
Authors: Shida Xia,Na Li, Xiaofeng, Tao, and Chao Fang
Abstract: Information centric network (ICN) based Mobile Edge Computing (MEC) network has drawn growing attentions in recent years. The distributed network architecture brings new security problems, especially the identity security problem. Because of the cloud platform deployed on the edge of the MEC network, multiple channel attributes can be easily obtained and processed. Thus this paper proposes a multiple channel attributes based spoofing detection mechanism. To further reduce the complexity, we also propose an improved clustering algorithm. The simulation results indicate that the proposed spoofing detection method can provide near-optimal performance with extremely low complexity.
WISChain: An Online Insurance System based on Blockchain and DengLu1 for Web Identity Security
Authors: Yurong Guo, Zongcheng Qi, Xiangbin Xian, Hongwen Wu, Zhenguo Yang, Jialong Zhang and Wenyin Liu
Abstract: An insurance system based on blockchain is proposed for web identity security, which provides two insurance service models for personal web identity security of end users and data security of commercial websites, respectively. Claim evidences are uploaded automatically to the blockchain to keep their authenticity. Smart contracts are automatically applied between insurers and policyholders to build trust between them.
CoderChain: A BlockChain Community for Coders
Authors: Yuzhi Lin, Zongcheng Qi, Hongwen Wu, Zhenguo Yang, Jialong Zhang and Wenyin Liu
Abstract: An online community based on blockchain is proposed for software developers to share, assess, and learn codes and other code or software related knowledge. It involves three modules or roles, namely: developer (or coder, or more generally, knowledge contributor), code (or knowledge contribution), and jury (or assessor, who is usually a developer with advanced skills), in addition to the blockchain based database. Each full node of the blockchain hosts a copy of all activities of developers in such community, including uploading contributions, assessing others’ contributions, and conducting transactions. Smart contracts are applied to automate transactions after code assessment or other related activities. The system aims to assess and improve the value of codes accurately, simulate the creativity of the developers, and improve software development efficiency, so as to establish a virtuous cycle of a software development community.
Research on Task Scheduling Strategy: Based on Smart Contract in Vehicular Cloud Computing Environment
Authors: Jun Fan, Ru Li and Shuo Li
Abstract: With the development of computer and network technology, on-board computers have more stronger capability of computing and network communication. Multiple vehicles can form an Autonomous Vehicular Cloud (AVC) for the certain need to provide cloud computing services for customers. However, the infrastructure of AVC is open completely, how to guarantee the non-repudiation of task execution information is a very important issue during the process of task execution. Blockchain technology is a proven technology with information security and non-repudiation in distributed environments. Smart contract of Ethereum used to design a kind of task scheduling strategy which is suitable for the AVC environment in this paper. The strategy can guarantee the non-repudiation of task execution information. This paper builds the private chain of Ethereum on simulation vehicles node and creates and deploys the smart contract. Experiments show that the task scheduling strategy is very effective.
Camera Position Estimation for UAVs Using SolvePnP with Kalman Filter
Authors: Dhong Hun Lee, Sang Su Lee, Hyun Ho Kang and Choon Ki Ahn
Abstract: In this paper, we propose an algorithm for obtaining camera coordinates for unmanned aerial vehicles (UAVs) using SolvePnP and Kalman filter (KF). The data and motion of circle can be recognized and tracked through the KF with high accuracy. Then, the recognized circle is used as an image point of the SolvePnP algorithm. The translation and rotation vectors can be extracted from reference of the world coordinate system using the image point. The position of UAVs can be obtained from the extracted information and the results demonstrate that the position can be derived from only camera. The robust performance of the proposed algorithm is demonstrated through experimental results
A Cascade Structure for Blockchain
Authors: Zhuyun Qi, Yan Zhang, Yi Wang, Jinfan Wang and Yu Wu
Abstract: Blockchain as a novel technology which consists of peer-to-peer network, distributed consensus, cryptography and other fundamental knowledge. Blockchain can help people build trust among untrusted people, which will help people establish new application models in multiple areas. But there are several legacy problems which obstruct the deployment and development of blockchain, the biggest problem is blockchain performance that blockchain cannot process the transactions as much as the centralized server per second. We propose acascade structure for blockchain, which can accelerate the block generation, enlarge the capacity of the block, reduce the risk of fork and increase the difficulty to launch the 51% attack.
An Agri-product Traceability System Based on IoT and Blockchain Technology
Authors: Weigbin Hong, Yefan Cai, Ziru Yu and Xiangyang Yu
Abstract: This paper starts from the food quality problem and proposes a traceability system for agricultural products based on Iot and blockchain technology. The architecture of the system is given and some existing problems are discussed. By using the consortium blockchain as the basic network and the IoT devices as the recorder, a more reliable, trustable and extendable traceability system can be achieved.
Design of Miniature Spectrometer Based on Blockchain Technology
Authors: Jianwei Yin, Weibin Hong, Ziru Yu and Xiangyang Yu
Abstract: The data measured by traditional miniature spectrometers can easily be tampered with, and even if it is provided to others, the authenticity of the data cannot be ensured, and the data value is low. Blockchain network gives each registered miniature spectrometer a unique identity. It guarantees the authenticity of the data, can be fully utilized, and increases the value of the data.
A security architecture of VANET based on blockchain and mobile edge computing
Authors: Xiaodong Zhang, Ru Li and Bo Cui
Abstract: The development of Vehicular Ad-hoc NETwork (VANET) has brought many conveniences to human beings. It also brings a very prominent security problem. The traditional solution to the security problem is based on centralized approach which requires a trusted central entity. However, there is a single point of failure problem. Moreover, there is no approach of technical level to ensure security of data. Therefore, this paper proposes a security architecture of VANET based on blockchain and mobile edge computing. The architecture includes three layers, namely perception layer, edge computing layer and service layer. The perception layer ensures the security of VANET data in the transmission process through the blockchain technology. The edge computing layer provides computing resources and edge cloud services to the perception layer. The service layer uses the combination of traditional cloud storage and blockchain to ensure the security of data.
A High Performance Blockchain Platform for Intelligent Device
Authors: Shitang Yu, Kun Lv, Zhou Shao, Yingchen Guo, Jun Zou and Bo Zhang
Abstract: In the area of IoT (Internet of Things), more and more intelligent devices are being connected to the Internet. These intelligent devices have been producing a huge amout of useful data over time, however there is still a lack of a platform which can efficiently transfer and utilize the value of the massive IoT data. Blockchain is able to transfer value with a relative low cost, which makes it possible for the data from smart devices to create economic value. This work of this paper is to design a high performance blockchain platform, using technologies such as distributed network architecture, intelligent devices node mapping, as well as PBFT-DPOC consensus algorithm to realize the decentralized autonomy of intelligent devices.
3D-DAG: A High Performance DAG Network with Eventual Consistency and Finality
Authors: Joe Zou, Zhongli Dong, Allen Shao, Peng Zhuang4, Wei Li and Albert Y. Zomaya
Abstract: Blockchain has been widely recognized as the trust machine underpinning the Internet of Things (IoT). However, the poor performance of the existing mainstream platforms renders this expectation unattainable. New technologies like DAG and Hash graph emerge as promising candidates to address the pressing issues, but their abilities in decentralized consensus are still in question. We propose a 3D-DAG model to address the trilemma of decentralization, scalability and security. Experiment results show that, by adopting the separation of concerns (SOC) architectural principle and a software defined chain(SDC) approach, 3D-DAG can meet challenging requirements expected by IoT applications.
End-to-end Answer Selection via Attention-Based Bi-LSTM Network
Authors: Yuqi Ren, Tongxuan Zhang, Hongfei Lin and Xikai Liu
Abstract: Many people ask medical questions online, finding the most suitable answer from candidate answers is an important research area in health care. The IEEE HotICN Knowledge Graph Academic Competition given a question and several candidate answers, then sort the candidate answers to get the best answer. We treated this subtask as a binary classification task, sorted the answers by calculating similarity between the question and each answer. In this work, we proposed a neural selection model trained on the training dataset. Our network architecture is based on the combination of Bi-LSTM and Attention mechanism, extended with biomedical word embeddings. Based on this fact, our model achieve state-of-the-art results on answer selection of medical community.
ICNRL: An Initiative Framework Towards Information Centric Network Representatio
Authors: Yuming Lu and Weichao Li, Xiaojun Wang
Abstract: The exponentially growing demand for computational resources prevents the Information Centric Networking (ICN) being deployed in practice due to the high dimensional sparse data computation. However, we argue that Network Representation Learning (NRL) can help to solve the problem by transforming the raw network information data into lowdimensional dense adjacency matrix representation. In this paper, we propose ICNRL, a novel task-based NRL scheme for ICN. Based on the adjacency matrix generated by NRL, ICNRL can calculate the index threshold value to support the networking decision making of content-store (CS), pending information table (PIT), and forwarding information base (FIB), and therefore improves the management and processing capabilities of ICN