With the emergence of the Internet of Things (IoT) technology, the number of related devices has been increasing at a very rapid speed. The security of IoT systems has become a crucial issue. Due to the complex IoT environment and users' unawareness, such issues are usually hard to resolve. Many IoT systems lack proper access control mechanisms and suffer from various large scale attacks. We need a robust and effective secure access control to build IoT systems that retain user privacy and data integrity with high availability. In this paper, we propose an access control framework based on OAuth 2.0, with which we constructed a remote control system for various devices. The secured authentication schemes prevent possible private data leaks. The proposed framework provides flexibility for further functional extensions with new IoT devices.
Authors: Min-Zheng Shieh (Information Technology Service Center, National Chiao Tung University), Jui-Chun Liu (Information Technology Service Center, National Chiao Tung University), Yi-Chih Kao (Information Technology Service Center, National Chiao Tung University), Shi-Chun Tsai (Department of Computer Science, National Chiao Tung University), Yi-Bing Lin (Department of Computer Science, National Chiao Tung University),
Hide Authors & Abstract

Device-to-device (D2D) communications can improve the overall network performance, including low latency, high data rates, and system capability for the fifth generation (5G) wireless networks. The system capability can even be improved by reusing resource between D2D user equipment (DUE) and cellular user equipment (CUE) without bring harmful interference to the CUEs. A D2D re-source allocation method is expected to have the characteristic that one CUE can be allocated with variable number of resource blocks (RBs), and the RBs can be reused by more than one CUE. In this study, Multi-Player Multi-Armed Bandit (MPMAB) rein-forcement learning method is employed to model such problem by establishing preference matrix. A fair resource allocation method is then proposed to achieve fairness, prevent wasting resource, and alleviate starvation. This method even has better throughput if there are not too many D2D pairs.
Authors: Fang-Chang Kuo (National Ilan University), Hwang-Cheng Wang (National Ilan University), Chih-Cheng Tseng (National Ilan University), Jia-Hao Xu (Dept. of Electronic Engineering, National Ilan University), Jung-Shyr Wu (National Central University),
Hide Authors & Abstract

Despite the importance of customers’ perception and behavior intention in online travel agency, there has been little theoretically research on perceived severity of negative publicity for tourism purposes. This study is based on the online com-parison price website: Trivago, and explores the impact of consumer satisfaction, E-service quality, perceived severity of negative publicity and perceived risk on consumers’ advocacy intention to embrace after the impact of Trivago’s negative publicity. This research collected a total of 300 valid questionnaires, and verified the research results: E-service quality was positively related to consumer satisfac-tion and negatively related to perceived risk; perceived severity of negative pub-licity was positively related to perceived risk; consumer satisfaction mitigates the positively influence of perceived severity of negative publicity on perceived risk; perceived severity of negative publicity was negatively related to advocacy inten-tion. When detailing the effect of perceived risk on advocacy intention, psycho-logical risk, performance risk and financial risk were negatively related to advo-cacy intention; social risk and time risk were not negatively related to advocacy intention.
Authors: Kuan-Yang Chen (National Taipei University of Nursing and Health Sciences), Chih-Hu Hsiao (National Chiayi University), Yi-Tsen Hsien (National Chiayi University),
Hide Authors & Abstract

Smart Grid is one of the critical technologies that provide essential services to sustain social and economic developments. There are various cyber attacks on the Smart Grid system in recent years. Therefore, understanding the characteristics and evaluating the consequences of an attack on the Smart Grid system is essential. The combination of Graphical Security Model, including Attack Tree and Attack Graph, and the Common Vulnerability Score System is a potential technology to analyze attack on Smart Grid system. However, there are a few research works about Smart Grid attack analysis using GrSM and CVSS. In this research, we first conduct a comprehensive study of the existing research on attack analysis using GrSM and CVSS, ranging from Traditional Networks, Emerging Technologies, to Smart Grid. We indicate that the framework for automating security analysis of the Internet of Things is a promising direction for Smart Grid attack analysis using GrSM and CVSS. The framework has been applied to assess security of the Smart Grid system. A case study using the PNNL Taxonomy Feeders R4-12.47-2 and Smart Grid network model with gateways was conducted to validate the utilized framework. Our research is enriched by capturing all potential attack paths and calculating values of selected security metrics during the vulnerability analysis process. Furthermore, AG can be generated automatically. The research can potentially be utilized in Smart Grid cybersecurity training.
Authors: Tan Le Duy (Japan Advanced Institute of Science and Technology), Mengmeng Ge (Deakin University), Duy Phan The (University of Information Technology - VNU-HCM), Hien Do Hoang (University of Information Technology - VNU-HCM), Adnan Anwar (Deakin University), Seng Loke (Deakin University), Razvan Beuran (Japan Advanced Institute of Science and Technology), Yasuo Tan (Japan Advanced Institute of Science and Technology),
Hide Authors & Abstract

Real-time monitoring of large-scale solar farms is one important aspect of re-liable and secure deployment of 100% renewable energy-based grids. The real-time ability to observe sensors on solar panels using Internet of Things (IoT) technologies make it possible to study behavior of solar panels under various conditions and to detect anomalous behaviors in real-time. This technology will allow grid administrators make to informed decisions in re-acting to anomalies such as panel damage, electrical errors, monitoring hard-ware decay, or malicious data injection attacks. Smart edge devices offer an opportunity to reduce the cost of continuously sending data for anomaly de-tection by performing analytics on local edge device within a given farm and sending only the result of the analysis back to datacenters. This paper pre-sents the design and evaluation of a low-cost edge anomaly detection system for remote solar power using Raspberry Pi and deep learning. The design was implemented and tested using real-life observations from a solar monitoring system under soiling conditions. The experiments showed that it is possible to run real-time anomaly detection algorithms on edge devices with little overhead in terms of power consumption and utilization of computational resources.
Authors: Salsabeel Shapsough (American University of Sharjah), Imran Zualkernan (American University of Sharjah), Rached Dhaouadi (American University of Sharjah),
Hide Authors & Abstract

The online store can keep the track of customers’ purchasing records and personal information. By analyzing these customers’ records, online store can have a better understanding of their customers’ profile and purchasing behavior. In this paper, we define a standard product loyalty status, or SPLS, using customers’ purchasing records to evaluate each customer’s loyalty to a certain product. SPLS is incorporated with loyal customers’ personal backgrounds as the input of cluster analysis that divides loyal customers into different groups. Loyal customers in the same groups have similar purchasing behavior and personal backgrounds. Similarity analysis measures the similarity of backgrounds between a non-loyal customer and groups of loyal customers in order to find this customer’s belonged group. Then, an expected SPLS value is assigned to this non-loyal customer to estimate his/her probability of purchasing a certain product. Customers who have expected SPLS value larger than a threshold are regarded as potential customers. Marketing specialists should recommend the product to potential customers. Loyal customers, on the other hand, are analyzed with the reference map to identify their preference for current product line. The reference map is used to provide suggestions under the condition of new-product-launch. Overall, this paper proposes a systematic model to construct online customer profile and a recommendation strategy for loyal customers and potential customers.
Authors: Chia-Ling Ho (National Taipei University of Nursing and Health Sciences), Peng-Yu Chou (Taipei City University of Science and Technology), Pei-Syuan Li (Taipei City University of Science and Technology), Ying-Ching Wang (Taipei City University of Science and Technology), Yan-Ling Pan (Taipei City University of Science and Technology), Shi-Ting Chen (Taipei City University of Science and Technology),
Hide Authors & Abstract

High-efficiency artificial incubation technology is the basis of the development of the poultry industry, and good chicks can be obtained through excellent egg breeding and good incubation technology. In Taiwan, the control of important parameters for waterfowl hatchery is still based on the inheritance of experience. The manual intervention of the cold egg operation during the hatching process will also affect the stability of the hatching environment and the risk of poultry biological safety infection. Therefore, in addition to discussing the current factors affecting waterfowl hatching, this study will also establish a set of IoT sensing systems suitable for waterfowl hatching. We use thermal imaging cameras and air quality sensors to collect the key factors that affect the hatching of waterfowl during the hatching process, and use the machine learning analysis framework to analyze the collected big data of waterfowl hatching. Although the application of thermal imaging technology has limitations, due to the non-invasive characteristics and the cost of technology, the application of poultry science has gradually received attention. Combining poultry science and information science, we have reintegrated a complete set of intelligent detection and application improvement solutions, which can enhance the digitalization and intelligence of the hardware and software of the waterfowl industry chain.
Authors: Shun-Chieh Chang (Department of Business Administration, Shih Hsin University), Chih-Hsiang Cheng (Ilan Branch, Livestock Research Institute, Council of Agriculture, Executive Yuan), Tse-Yung Huang (Cybersecurity Technology Institute, Institute for Information Industry), Liou-Yuan Li (Department of international Business Administration Rajamangala University of Technology Thanyaburi), Yu-Liang Liu (Department of Multimedia and Game Design, Overseas Chinese University),
Hide Authors & Abstract

Every aspect of tea planting requires a lot of manpower. Applying intelligent agricultural planting technology will effectively reduce the burden of farm operations, reduce labor demand, and enable tea farmers to operate more ef-ficiently. The research introduces the most popular artificial intelligence (AI) technology, Internet of Things (IoT) and the AgriTalk technology to a tea farm that is located in Nantou, Taiwan. Through remotely monitoring the growth of tea and various environmental data, intelligent tea farm manage-ment platform will ensure the healthy growth of tea and increase tea produc-tion. In addition, the key patterns of tea cultivation will be established through long-term data collection and analysis. The key patterns for tea cul-tivation include suitable temperature, humidity, sunshine, fertilization time, etc.
Authors: Mei-Yu Wu (National Taichung University of Science and Technology), Chih-Kun Ke (National Taichung University of Science and Technology),
Hide Authors & Abstract

We proposed a multi-criteria artificial bee colony (MABC) algorithm to optimize the problem of energy consumption in wireless sensor networks. The method used the artificial bee colony algorithm to discover sensor nodes in the network as a cluster head combination. Different nodes were dynamically selected according to their current status in the network. The purpose is to group the sensor nodes in the network first, so other nodes can transmit the packet to the belong group head, and then find the best energy-saving packet routing path for the group heads to the Internet of Things (IoT) base station, where multiple attributes in the path including residual energy, energy consumption, path distance, number of hops, and frequency make decision scores to help the algorithm discover a better solution. The experimental results showed that the MABC algorithm providing a more energy-efficient packet routing path than the basic ABC and made the wireless sensor network survive longer, which confirmed the evaluation of the candidate path through multi-criteria analysis. The contribution of this work is to the use of group intelligence algorithms in wireless sensor network routing evaluation. The artificial bee colony algorithm based on multi-criteria is used in a wireless sensor network to improve the problem of fast convergence of the algorithm.
Authors: Chih-Kun Ke (National Taichung University of Science and Technology), Mei-Yu Wu (National Taichung University of Science and Technology), Chia-Yu Chen (National Taichung University of Science and Technology),
Hide Authors & Abstract

Dynamic random access memory (DRAM) products are the key parts in consumer products. To fulfill the current market’s strict specifications, vari-ous customers have asked DRAM manufacturers to continue improving the quality of DRAM products. The resistance of the Ti film directly affects the electrical quality of DRAM products. At present, the DRAM products devel-oped by the case company have caused customer returns due to abnormal re-sistance value of Ti film. Process engineers always adjust the engineering pa-rameters based on experience, which resulted in slow improvement and inability to determine the setting of engineering parameters. Consequently, shipments of DRAM products are delayed. This study adopts the Ti film resistance of DRAM products as the main research object for improvement and applies the response surface method, neural networks, and genetic algorithms to help process engineers analyze and improve DRAM products. This work assists the case company in achieving a significant improvement in Ti film resistance from 210.33 Ω (the origin made by the case company) to 185.28 Ω (the improvement made by this work) where the specified target value is 185 Ω. The results are effective in shortening the improvement time and reducing customer returns.
Authors: Yun-Wei Lin (College of Artificial Intelligence, National Chiao Tung University, Tainan, Taiwan), Chia-Ming Lin (College of Artificial Intelligence, National Chiao Tung University, Tainan, Taiwan),
Hide Authors & Abstract

The tsunami of aging is coming making the population structure in Taiwan to change drastically. The elderly population with disability quickly rising which is accompanied by the doubling of the problem care. Family members who cannot be there with their elders due to economic burden send them to boarding typing institutions. Traditional medical care model cannot effectively manage due to limitations in deficient professional medical labor and medical resources resulting in the enigma of increasing social cost. In light of the fast development of information technology, that many innovative and cross-field applications can now be effectively placed into clinical institutions to enhance medical treatment efficiency and expand the scope of policy have become an important key.The research will bring in the basic concept of IoT and deploy medical grade IoT modules combined with communication transmission technology through edge computing connecting phasal warning mechanism to attempt to deploy deeply into clinical situation simulation to help the elderly self-examine their own health status on a regular basis in order to construct an appropriate model for health care and to reduce the waste of overcentralized social resources in large medical institutions for creating a life-protecting, protection, health care elderly care life environment so the institutional caring quality can be increased and that the combination of medical care for establishing a healthy elderly society.
Authors: Lun-Ping Hung (National Taipei University of Nursing and Health Sciences), Zong-JIe Wu (National Taipei University of Nursing and Health Sciences), Shih-Chieh Li (National Taipei University of Nursing and Health Sciences), Chien-Liang Chen (Overseas Chinese University), Chiang-Shih Chun (National Taipei University of Nursing and Health Sciences),
Hide Authors & Abstract

This work proposed the Claim-based Channel Access (CCA) method for IEEE 802.11ah, which is designed for Internet of Things (IOT). The proposed CCA method uses the newly devised Claiming RAW to let those stations having up-linked data frames to claim their intentions for uplinking data frames to reduce collisions. In addition, this work adopted the registered backoff time mechanism, for which a station registers its next backoff time to AP when its current channel access is finished. In this way, AP is able to schedule stations according to their registered backoff time in advance to avoid collisions more effectively. Compar-ing with the traditional IEEE 802.11ah, the proposed CCA method has the lower collision rate and higher throughput in the network environment having the more number of stations in each time slot for accessing the channel.
Authors: Rung-Shiang Cheng (Department of Information Technology, Overseas Chinese University), Chung-Ming Huang (Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan), Yan-Jia Pan (Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan),
Hide Authors & Abstract

With the development of the e-commerce and the logistics industry, more and more personal information has been collected by the third-party logistics. The personalized privacy protection problem with multiple sensitive attributes is seldom considered in data publishing. To solve this problem, a method of Multi-sensitive attributes Weights Clustering and Dividing (MWCD) is pro-posed. Firstly, set the corresponding weight for each sensitive attribute value considering the different requirements of users and then cluster the data based on the weights. Secondly, divide the records by level rule to select record for l-diversity. Finally, publish data based on the idea of Multi-Sensitive Bucketiza-tion. The experimental results indicate that the release ratio of the important da-ta though the proposed algorithm is above 95%, and the execution time is shorter.
Authors: haiyan Kang (Beijing Information Science and Technology University),
Hide Authors & Abstract

The wide application of smart grid improves the energy utilization rate and improves the power market, but at the same time, it also introduces many security problems, such as data storage, transmission, theft and other security problems in the process of smart grid data communication. Due to the special position of power system, how to ensure the security of data storage in smart grid is of great significance for the safe and stable operation of power grid system. This paper first analyzes the smart grid and its data characteristics, combined with the relevant technologies of cloud computing, gives a data security storage model of smart grid to strengthen the reliability and storage capacity of smart grid. Then, in order to ensure the security of user data storage in the cloud in smart grid, this paper studies the data encryption algorithm, and proposes a hybrid encryption scheme for smart grid data security storage. Finally, the scheme is compared with the traditional method. The experimental results show that the scheme has the advantages of good encryption and decryption effect, fast execution speed and high security. It is an ideal scheme for smart grid data security storage.
Authors: haiyan Kang (Beijing Information Science and Technology University), Jie Deng (Beijing Information Science and Technology University),
Hide Authors & Abstract

In wireless sensor networks, the classification of uncertain data reported by sensor nodes is an open issue because the given attribute information can be insufficient for making a correct specific classification of the objects. Although the traditional Evidential k-Nearest Neighbor (EkNN) algorithm can effectively model the uncertainty, it is easy to misjudge the target data to the incorrect class when the observed sample data is located in the feature overlapping region of training samples of different classes. In this paper, a novel Evidential k-Nearest Neighbor (NEkNN) algorithm is proposed based on the evidential editing method. The main idea of NEkNN algorithm is to consider the expected value and standard deviation of various training sample data sets, and use normalized Euclidean distance to assign class labels with basic belief assignment (BBA) structure to each training sample, so that training samples in overlapping region can offer more abundant and diverse class information. Further, EkNN classification of the observation sample data is carried out in the training sample sets of various classes, and mass functions of the target to be tested under this class are obtained, and Redistribute Conflicting Mass Proportionally Rule 5 (PCR5) combination rule is used to conduct global fusion, thus obtaining the global fusion results of the targets. The experimental results show that this algorithm has better performance than other classification methods。
Authors: Yang Zhang (Beijing Jiaotong University), Yun Liu (Beijing Jiaotong University), Zhenjiang Zhang (Beijing Jiaotong University),
Hide Authors & Abstract

In this paper, we propose a security authentication scheme based on machine learning algorithms to detect spoofing attacks in the Internet of Things (IoT) network. This authentication method exploits the physical layer properties of the wireless channel to identify sensors and applies neural networks to learn channel fingerprints without being aware of the communication network model. We propose a channel differences-based security framework to provide lightweight authentication and a Long Short-Term Memory (LSTM) network-based detection approach to further enhance the authentication performance for sinks that support intelligent algorithms. Experiments and simulations were carried out in an indoor conference room. The results show that our strategy improves the authentication accuracy rate compared with the existing non-learning security authentication methods.
Authors: xiaoying qiu (Beijing Information Science and Technology University), Xuan Sun (Beijing Information Science and technology University), Xiameng Si (Beijing information Science and Technology University),
Hide Authors & Abstract

With the fast development of Internet of Things, more and more applications are deployed in this “connecting everything” network. Edge computing and cloud computing are two paradigms to implement the Internet of Things. To utilize the advantages of both these two computing forms, edge-cloud computing was proposed. In this paper, we construct a secure edge-cloud computing (SECC) framework. Sensor nodes and applications can interact with the framework through unified interfaces. We implement the edge server as a collection of services, including edge device orchestration, data processing and storage, communication management, authentication and authorization, environment sensing and situation analysis. Through a daisy-chain approach, our framework can be secured for heterogeneous security needs of different parts of the system. We also build a prototype of the proposed framework and demonstrate the efficacy of the SECC framework through experiments and comprehensive analysis.
Authors: Yao Zhao (Beijing Jiaotong University), Zhenjiang Zhang (School of Software Engineering, Beijing Jiaotong University, Beijing, China), Jian Li (Beijing Jiaotong University),
Hide Authors & Abstract

Applying automatic text summarization technology to Internet of Things can save network cost and improve computing speed. Current abstractive text summarization models are always based on sequence-to-sequence model with attention mechanism. Unfortunately, this approach to abstractive sum-marization suffers from two problems: they can’t solve the problem of un-known words, and their generated summaries are not very readable because of repetition. In this work, we aim to improve the semantic coherence of generated summaries for Chinese social media texts. To achieve this goal, we propose a novel model that augments the traditional sequence-to-sequence model in two ways. First, we apply semantic relevance to pointer-generator network to encourage high similarity between source texts and summaries. Second, we change the mechanism of coverage and use it to pointer-generator network to discourage repetition. Following previous work, we ap-ply our model to a popular Chinese social media dataset LCSTS. Our exper-iments show that our model outperforms the current abstractive baseline sys-tems on the dataset.
Authors: Lijun Wei (Beijing Jiao Tong University), Yun Liu (Key Laboratory of Communication and Information Systems; Beijing Municipal Commission of Education, Department of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China), jian li (Beijing Jiaotong University),
Hide Authors & Abstract

Edge computing could play an important role in Internet of Things(IoT). Computing capacity allocation has been researched a lot in mobile edge computing, which is task oriented. However, hierarchical edge computing also needs computing capacity allocation which is node oriented. This paper focuses on capacity allocation of nodes in hierarchical edge computing. We take energy efficiency and loss in high concurrency scenarios into consideration, and work out a method to do allocation by weighing loss and energy efficiency. Simulation is under circumstances that nodes overload, which means that loss is inevitable. A new inspiration of deployment is also given after simulation.
Authors: Ziheng Zhou (Beijing Jiaotong University), Zhenjiang Zhang (Beijing Jiaotong University), Jianjun Zeng (Beijing Li’antong Information Technology Company), Jian Li (Beijing Jiaotong University),
Hide Authors & Abstract

The increasing scale of the IOT poses challenges to the energy consumption, transmission bandwidth and processing delay of centralized cloud computing data centers. The cloud computing data centers is moving from the center of the network to edge nodes with lower latency, namely, edge computing. Meanwhile, it can meet the needs of users for real-time services. In the field of edge computing, offloading decision and resource scheduling are the hot-spot issues. As for offloading decision and resource scheduling problems of single-cell multi-user partial offloading, the system model is also firstly es-tablished from four aspects: network architecture, application type, local computing and offloading computing. Based on the system model, the optimization problem of resource scheduling is modeled, where the solution is hard to be found. Thus, the deep reinforcement learning method based on policy gradient is selected to establish the SPBDDPG algorithm that can solve the problem. Then, in order to solve the practical problems, the SPBDDPG algorithm is set up with the state and action for iteration, as well as the environment for generating new state and feedback reward value. Finally, an appropriate iteration step is written for the edge computing resource scheduling problem by combining with the original deep reinforcement learning algorithm. We also evaluate the proposed approaches by relevant experiments. The complexity and effectiveness of the results are validated.
Authors: Zhenjiang Zhang (School of Software Engineering, Beijing Jiaotong University, Beijing, China), Tong Wu (School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China), Zhiyuan Li (School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China), Bo Shen (Key Laboratory of Communication and Information Systems; Beijing Municipal Commission of Education, Department of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China;), Naiyue Chen (School of Computer and Information Technology, Beijing Jiaotong University, Beijing, China), Jian Li (School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China),
Hide Authors & Abstract

In recent years, the Internet of Things technology has developed rapidly. Due to the large number of devices at the edge, the wide distribution range, and the complex environment, cloud computing and edge computing failed to fully consider security risks at the beginning of the combination, and traditional protection methods can no longer fully meet their security requirements. The establishment of a new cloud-edge integrated security system is of great significance for ensuring the data and privacy of Internet users. This article first investigates the current status of traditional network security and analyzes its inherent shortcomings, and analyzes the organizational structure and main advantages of the zero-trust network. Designed a security certification system that meets the needs of cloud-edge integrated applications. The API security access gateway part of the system is designed and implemented. According to the type of client access request, it is equipped with multiple authentication methods. It also realized the functions of reverse proxy, load balancing, flow control, log audit, analysis and monitoring of microservices, and finally developed a supporting UI management tool based on Vue. This design provides a new set of secure access solutions for clients and microservices, which has produced good industrial benefits. It is of great significance to promote the development and popularization of cloud-edge integration.
Authors: Liu Sai (Beijing Jiaotong University), Zhang jiang (Beijing Jiaotong University), Cui Yong (Beijing Thunisoft information technology co. LTD), Zhang Yang (Beijing Jiaotong University),
Hide Authors & Abstract

This paper proposes an Open Shortest Path First (OSPF) based Backhaul Protocol for 5th Generation (5G) mobile communication millimeter wave network (OBPG). This protocol includes the establishment stage and the transmission stage of the millimeter wave backhaul network. During the establishment stage, all millimeter wave micro base stations use fixed time division. The method is based on the OSPF protocol to establish the backhaul path between millimeter wave micro base stations. In the transmission stage, based on the network topology formed by millimeter wave micro base stations, the coloring method is used to allocate data transmission time slots. After that, the micro base station polls and schedules users for data transmission in the allocated data transmission time slot. The simulation results show that when the blocking probability of the millimeter-wave directional link is 30%, and along with the number of multiple associations between users and millimeter-wave micro base stations increases, the network throughput performance is about 20% higher than the performance of the number which is one less.
Authors: Zhanyu Zhang (Northwestern Polytechnical University), Xindai An (Northwestern Polytechnical University), Zhongjiang Yan (Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University),
Hide Authors & Abstract

Wireless Mesh Networks (WMNs) has the advantages of high bandwidth, flexible networking, wide coverage and low investment risk. Routing metrics have a great impact on network. Appropriate routing metrics can reduce intra-stream and inter-stream interference, improve throughput and reliability, achieve load balancing and eliminate network hot spots. At present, research on routing metrics for WMNs has made some progress. Relevant scholars have proposed various routing metrics, but no scholars have compared and classified these routing metrics. In this paper, the classical routing metrics in WMNs and the routing metrics proposed in the last ten years are studied. The following conclusions are drawn from these investigations. Firstly, delay, packet loss rate and bandwidth are the most commonly considered factors in routing metrics. Secondly, routing metrics separately describe the types of disturbances that lead to the introduction of variable constants. Thirdly, routing metrics often ignore the choice of gateway nodes. Finally, delay is the most important parameter of routing metrics. For example, the introduction of bandwidth and bottleneck channels is for more accurate calculation of delay. NS3 is used to simulate Hop Routing Metric (HOP) and Distance Routing Metric. The simulation results show that in a small network, Distance Routing Metric can effectively reduce the delay and increase the network throughput.
Authors: Yunlong Wang (Northwestern Polytechnical University), Zhongjiang Yan (Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University),
Hide Authors & Abstract

To guarantee low energy consumption and high efficiency for the nodes of the Internet of Things (IoT), designing and implementing the network protocols, e.g., routing and multiple access control (MAC) protocols, on FPGA become significant issues. Aiming at designing and implementing FPRP protocol on FPGA, this paper proposes a design scheme of FPGA implementation of FPRP protocol. The overall architecture includes clock counting module, reservation module, data transmission module, algorithm module and reservation result generation module, which are connected and restricted with each other. In order to verify, we divide it into two parts, i.e., single node verification and double node verification. Finally, the basic implementation of the five-step reservation is initially completed which show that the design scheme is correct and feasible.
Authors: Shuning Lei (Northwestern Polytechnical University), Zhongjiang Yan (Northwestern Polytechnical University), Xiaojiao Hu (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University),
Hide Authors & Abstract

To solve the problem that multi-round and multi-slot Hello-Reply scheme takes a long time to discover all the neighbors due to its multiple parameters and difficulty in optimization, this paper proposes an optimal multi-round and multi-slot (o-MRMS) Hello-Reply algorithm, which theoretically proves and reduces the total time and number of rounds of the algorithm for neighbor discovery, effectively reducing the neighbor discovery time, and combined with the existing node discovery protocol. Simulation results show that compared with a fixed slots number Hello-Reply algorithm, the total neighbor discovery time in the proposed optimal multi-round and multi-slot Hello-Reply algorithm is reduced by about 50%. Compared with the DANDi protocol [3], discovery efficiency increased with the nodes number, when the number of nodes is 256, the discovery time reduced by approximately 25%. It can be combined with the existing neighbor discovery protocol, which verifies its feasibility and efficiency.
Authors: Xinru Li (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University), Hang Zhang (Science and Technology on Communication Networks Laboratory),
Hide Authors & Abstract

In recent years, with the development and application of technologies such as the Internet of Things, cloud computing, and big data, file management, the originally complicated task, has become more intelligent and flexible. However, when recording files, the existing technology still uses the method of scanning one file at a time. Therefore, when massive file information needs to be recorded at the same time, additional time cost or equipment cost will be incurred. In order to shorten the time of file entry as much as possible and reduce unnecessary equipment costs, we propose to mark the key information of files with the QR code technology in the Internet of Things, and then design a special image segmentation method according to the characteristics of the file bag or file box. Thus, the segmentation and extraction of different archive information within the visual range of the input device can be realized, and the effect of simultaneously inputting multiple archive information can be achieved. The experimental results show that in an ideal situation, our method can input the information of 20 files at one time, which greatly improves the efficiency of file entry. It is expected that if this method is applied to the current file management system, it can solve the problem of time consuming and energy consuming to some extent, and realize a more intelligent file management system at the same time.
Authors: Genwang Wang (Beijing Jiaotong University),
Hide Authors & Abstract

Although there are a large number of available channels that can be bonded together for data transmission in the next generation WLAN, i.e., IEEE 802.11be protocol, it may cause long data transmission time to transmit large files due to the inefficient channel bonding strategies. This paper proposes an optimal channel bonding strategy based on the optimal stopping theory. Firstly, under the constraint of the number of available channels, the problem of minimizing the transmission time of large files is formulated as an optimal stopping problem, where the time duration of large file transmission is defined as the sum of channel accessing time and data transmission time after successful access into the channel, Secondly, the threshold of successful bonded channel number is derived based on the optimal stopping theory. When the channel access is successful, data transmission is performed if the number of bondable channels is larger than the threshold. Otherwise this data transmission opportunity is dropped and channel competition is resumed. The simulation results show that, compared with the traditional EDCA if access-success then-transmit strategy and the fixed bonding channel number threshold strategy, the data transmission completion time of large file is shortened by more than 40%.
Authors: Ke Sun (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University),
Hide Authors & Abstract

Neighbor node discovery is one of the important steps in a wireless directed ad hoc network. Improving the efficiency of neighbor node discovery can not only reduce the collision during node communication, but also improve the performance of the wireless ad hoc network as a whole. In the Ad-hoc network of directional antennas, by analyzing and summarizing the deficiencies of the neighbor discovery algorithm, this paper proposes a probing and $p$-probability based two round directional neighbor discovery algorithm (PPTR). The second round adjusts the probability of neighboring neighbor nodes competing for slots based on the number of free slots, successful slots, and collision slots in the first round, thereby reducing the collision of neighboring nodes to reach the maximum number of neighbors discovered within a fixed time. We verified the protocol through network simulation. The simulation results show that the PPTR algorithm and the traditional neighbor discovery algorithm have the same neighbor discovery efficiency when the number of network nodes and the number of time slots are consistent. But The neighbor discovery efficiency increases on average 81.3$\%$ when the number of nodes increases to five times the number of timeslots.
Authors: Xiaojiao Hu (School of Electronics and Information, Northwestern Polytechnical University), Qi Yang (Northwestern Polytechnical University), Zhongjiang Yan (Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University),
Hide Authors & Abstract

A data scheduling algorithm based on link distance is proposed in this paper, aiming at the problem of low data transmission throughput caused by the unequal link distance between aircraft relay nodes and ground nodes in a directional aviation relay network. Firstly, the aircraft node acquires the data transmission request of the ground node during the data transmission request collection stage, and measures the transmission distance and the data transmission delay with the ground node. Secondly, a data scheduling algorithm based on downlink first uplink, long distance first and short distance is designed, which fully utilizes the communication delay expansion gain brought by the unequal link distance. Finally, the simulation results show that compared with the distributed scheduling (DS) algorithm, the algorithm named link distance data scheduling (LDDS) proposed in this paper improves the network throughput by 7.4$\%$.
Authors: WeiLing Zhou (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (Northwestern Polytechincal University), Zhongjiang Yan (Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

Real-time applications (RTA) develop rapidly these days. Latency sensitive traffic guarantee becomes increasingly important and challenging in wireless local area network (WLAN). In order to solve this problem, this paper proposes a latency oriented random access scheme, which is compatible with IEEE 802.11 standards, based on orthogonal frequency division multiple access (OFDMA) in the next generation WLAN: IEEE 802.11be. AP utilizes trigger frame (TF) based OFDMA random access and reserves several resource units (RUs) for latency sensitive traffic only. We theoretically analyze the estimated number of STAs who have latency sensitive data to send in the next TF according to the collision status in past TF as well as the traffic arrival features. Thus, AP will allocate appropriate RU number for latency sensitive STAs dynamically. The simulation results show that the proposed dynamic RU adjusting algorithm outperforms the other schemes in both average throughput and delay. The throughput utility of proposed algorithm is 28.65% higher than that of IEEE 802.11ax. And the delay utility is 37.80% lower than that of IEEE 802.11ax. Besides, it is interesting and valuable that the performance of latency insensitive STA can also be improved.
Authors: Zhaozhe Jiang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

In order to solve the problem of identification collision in multi-reader Radio Frequency Identification (RFID) systems, this paper proposes a multi-channel anti-collision algorithm based on grouping strategy named McAnCo. In this algorithm, the interference types among readers are classified and modeled, and then the vertex coloring algorithm in graph theory is used to group the readers with mutual interference. The readers in the same group can work at the same frequency and time slot, while the readers in different groups in the same set can work at the same time slot in different frequency, so that the maximum number of readers can work simultaneously without interference. Simulation results show that, compared with distributed color selection (DCS) algorithm, hierarchical Q-learning (HiQ) algorithm and neighborhood friendly anti-collision scheme (NFRA) algorithm, the proposed algorithm effectively prevent reader collision and raise the identification efficiency of the system.
Authors: Zhiyong DING (School of Electronics and Information, Northwestern Polytechnical University), Jianying LI (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University), Wenhui CHEN (Northwestern Polytechnical University),
Hide Authors & Abstract

In 2014, the 3rd Generation Partnership Project (3GPP) proposed the "Licensed-Assisted Access using LTE" (LAA) research project, with the intention to further improve LTE network capacity by deploying LTE system in 5GHz unlicensed frequency band. But deploying LAA in unlicensed band will inevitably affect other existing wireless systems such as WiFi and increase the risk of data collision. Therefore, the research of coexistence between LAA and WiFi is of great significance to facilitate the fair sharing and efficient utilization of unlicensed spectrum sources. However, the existing studies consider the rela-tionship between LAA and WiFi in the coexistence scenario from the perspective of contention. In this paper, we proposed a cell cooperation based channel access mechanism. The key idea of the proposed scheme is let an LAA cell and a WiFi cell establish a pair of cooperative relationship. Once LAA node accesses the channel, it will notify its cooperated pair WiFi node to share the transmission opportunity, and vice versa. After the transmission of the LAA node, the WiFi node can access the channel without performing backoff again. Simulation results show that the proposed cell cooperation scheme improves the communication performances of WiFi system as well as the LAA system and the spectrum efficiency in the coexistence scenarios with high channel resources demand.
Authors: Peilin Liu (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

In recent years, people have put forward higher and higher requirementsforhigh-speedcommunicationswithinalocalarea.Millimeterwave WLAN has attracted much attention from academia and industry by virtue of its ultra-large bandwidth and short-range coverage. Beam training is a key technology of millimeter wave WLAN. The quality of beam training is related to communication performance and even communication. However, as the number of nodes continues to increase, the beam training efficiency of the traditional millimeter wave WLAN is very low, which affects system performance.This paper proposes a beamforming training method based on dynamic time slot adjustment for the next generation millimeter wave WLAN. The beam training slot in the subsequent beacon interval (BI) can be adjusted according to the completion of the beam training in the previous BI, thereby improving the efficiency of beamtraining. The simulation results prove that the method proposed in this paper caneffectively improve the beam training efficiency and has a small impact on the performance of the system.
Authors: Zhaotun Feng (School of information engineering, Chang'an University), Yong Fang (Chang'an University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

A rising demand for larger network capacity is leading to the rapid development of the 5th Generation Mobile Communications System (5G). Due to the scarcity of spectrum resources of the conventional licensed band, in September 2014, 3GPP launched the research project of Licensed-Assisted Access using LTE (LAA), aiming to design a single global solution framework and protocol to ensure the efficient operation of LAA in unlicensed band (e.g. 5GHz spectrum band). However, existing studies indicate the difficulty to improve the performance of LAA and WiFi at the same time. In this paper, we propose a Power Control based Spacial Reuse scheme (PC-based SR) aiming to increase the probability of concurrent transmissions in coexisting scenarios of LAA and WiFi. For LAA, it should raise its energy detection (ED) threshold after recognizing existing WiFi signal and transmit with adjusted power, and vice versa. Our simulation results show that the performance of LAA and WiFi are improved at the same time and the fairness to WiFi is enhanced using the proposed method, with respect to throughput and latency.
Authors: Duoduo Hang (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

With the advent of the era of artificial intelligence and big data, people have higher and higher requirements for quality of service(QOS). In the next-generation wireless local area network (WLAN), the Institute of Electrical and Electronics Engineers (IEEE) 802.11 regards low-latency guarantee as one of the main technical goals. Multi-link operation(MLO) technology can effectively guarantee network delay. Based on this, this paper proposes a low-latency guarantee protocol based on multi-link scheduling random access. The specific process of the low-latency guarantee protocol is proposed and designed, and the frame format of the protocol is designed to make it compatible with the 802.11 frame format. The simulation results show that the random access protocol based on multi-link scheduling can effectively improve the quality of service of the network compared with the traditional single link. It can not only increase the throughput of the network, but also effectively reduce the data transmission delay. The addition of scheduling algorithms can effectively reduce the delay of high-priority networks and reduce the packet loss rate of the entire network.
Authors: LuoTing Gan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

With the rapid development of wireless communications, Wireless Local Area Network (WLAN) has entered thousands of households and has become one of the most important ways of carrying data business. In recent years, delay sensitive applications and services such as real-time game and wireless meeting have been increasing sharply. These applications and services require low latency. However, the traditional WLAN can hardly satisfy the needs of this requirement. The next generation WLAN standard: IEEE 802.11be which was established in 2019 regards low latency as one important technical objective. In this paper, we propose a delay sensitive priority adjustment scheme for the next generation WLAN to improve the latency performance of delay sensitive services. In the Media Access Control (MAC) layer, if there is still some time left before deadline, the priority of delay sensitive packets will be normal. Conversely, as the deadline approaches, the priority of delay sensitive packets will be upgraded. The priority related mechanisms in MAC layer include queuing policy, channel access parameters, and Modulation and Coding Scheme (MCS) selection.
Authors: Ning Wang (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

The application of Wireless Local Area Networks (WLAN) technology makes the devices connected to the network mobile, and can quickly and conveniently solve the communication problems that are not easy to implement in the wired network. The proposal of millimeter wave WLAN technology enables wireless communication to support the transmission of services such as video information with super-high data rates. Among them, beamforming (BF) is a key issue of millimeter wave WLAN technology, but there is a problem of conflicts when nodes perform BF training during the association beamforming training (A-BFT) period. To solve the above problems, this paper proposes a grouping based beamform training scheme for the next generation millimeter wave WLAN. The scheme groups different station (STA) and stipulates that different groups of STA enter a designated A-BFT slot to perform BF training. The simulation result shows that grouping based BF training scheme can effectively reduce the collision probability of STA during BF training in A-BFT slot.
Authors: Guo Linlong (School of Information Engineering, Chang'an University, Xi'an, China), Yong Fang (Chang'an University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

RFID is the critical technology of the Internet of Things, During the Tag identification process, an effective anti-collision method will make a significant contribution to the RFID system in accelerating the identification speed. This paper proposes a method to improved RFID anti-collision protocol that incorporates a NOMA technique, which is based on ISO 18000-6C standard. This paper simulates the method and compares it with traditional scheme of 180006C18000-6C. Through our simulations, NOMA-based RFID Tag identification method outperforms Traditional schemes in both the average access slot efficiency and time efficiency. It can solve serious collision under massive Tags numbers and improve system efficiency. It also can be conveniently applied to engineering implementations.
Authors: Qingyuan Miao (School of Information Engineering, Chang’ an University), Yong Fang (Chang'an University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

The next generation Wireless Local Area Network (WLAN) standard: IEEE 802.11be focuses on achieving extremely high throughput (EHT). High-dense deployment network is still one challenging and important scenario for IEEE 802.11be. In order to improve quality of experience (QoS) of users in high-dense scenario, access point (AP) collaboration is considered as a promising technology for IEEE 802.11be. Many researchers have proposed different specific AP cooperative solutions, but the work on the performance verification of the proposed protocol is not yet fully validated. This article focuses on the performance verification of the Coordinated AP Time Sharing in a Transmit Opportunity (co-TDMA) protocol proposed in the standard proposal. Aiming at the co-TDMA protocol flow, we design a specific simulation protocol implementation scheme, and perform simulations based on the system-level and link-level integrated simulation platform, and evaluate the performance of co-TDMA. Simulation results show that the co-TDMA protocol can improve the throughput and packet loss rate performance under certain network scenarios.
Authors: Huanhuan Cai (School of Electronics and Information, Northwestern Polytechnical University), Bo LI (School of Electronics and Information, Northwestern Polytechnical University), Mao Yang (School of Electronics and Information, Northwestern Polytechnical University), Zhongjiang Yan (School of Electronics and Information, Northwestern Polytechnical University),
Hide Authors & Abstract

The security of sharing sensitive information through a distributed in- formation storage and sharing platform is required strictly in many situations. In this paper, we present a novel approach with the aim of increasing security of sharing sensitive information. The model improves security by exploiting Attrib- ute-Based Encryption. In addition, we incorporate network coding to our model to improve the efficiency of transmitting information which can be applied in IoT.
Authors: 徐 志强 (Beijing Jiaotong University), 沈 波 (Key Laboratory of Communication and Information Systems, Beijing Municipal Commission of Education, Beijing, China), 张 致远 (Key Laboratory of Communication and Information Systems, Beijing Municipal Commission of Education, Beijing, China),
Hide Authors & Abstract