Mobile users can open program in new tab for better viewing.

Open program in new tab

Day 1 15/08/2020
Room #1

08:30 - Organizing Committee Welcome Message 08:30 - 08:45

Conference & Community Managers' message 08:45 - 09:00

Keynote 1: Professor Dusit Niyato 09:00 - 10:00

Reliable Federated Learning for Mobile Networks

Coffee Break #1 10:00 - 10:10

Session #1 10:10 - 12:00

10:10 - 10:25
An Outage Probability Based Channel Bonding Algorithm for 6G Network

Multiple input multiple output (MIMO) can greatly improve the throughput and frequency utilization of wireless transmission systems. In the IEEE 802.11ax protocol, 4G, 5G, and even future 6G communications, MIMO has greatly increased the throughput of transmissions. Proper channel bonding strategies can improve channel utilization and transmission throughput. This paper proposes an outage probability based channel bonding algorithm (OP-CB). OP-CB calculates the outage probability according to the average signal-to-noise ratio (SNR) and the threshold signal-to-noise ratio. SNR is affected by transmission power and transmission distance. And then OP-CB performs channel bonding according to the outage of data transmission. The algorithm proposed in this paper can be used in massive MIMO (mMIMO) in 5G and future 6G communications. Combing the outage probability with the channel bonding technique can improve the success rate of data transmissions. The simulation results show that OP-CB can improve the throughput of data transmission by about 40% when the transmission power is low or the transmission distance is large. OP-BC can also reduce the bit error rate (BER).
Authors: Weifeng Sun (Dalian University of Technology), Guanghao Zhang (Dalian University of Technology), Yiming Zhou (Dalian University of Technology), Rui Guo (Dalian University of Technology),
Hide Authors & Abstract

Show Authors & Abstract
10:25 - 10:40
Two Attacking Strategies of Coordinated Cyber-Physical Attacks for Cascading Failure Analysis in Smart Grid

As a classic Cyber-Physical System (CPS), smart grids often suffer from various types of attacks, one of which the most threatening attacks is Coordinated Cyber-Physical Attack (CCPA). In order to improve the robustness of the smart grid under CCPA, two attack strategies are proposed to analyze the cascading failure of smart grids. Firstly, we define attack goals (AGs) function to identify important cyber and physical nodes as possible targets. Secondly, based on these targets, the algorithm of optimal attack and saturation attack strategy is designed and applied to CCPA for analyzing the effect of those attack strategies on smart grids. Finally, node loss is used as an evaluation index to compare the attack effect of CCPA, Cyber Attack (CA) and Physical Attack (PA). The experimental results show that when the same proportion of nodes are removed, the CCPA has more node losses than the CA and PA, regardless of based on the optimal attack strategy or the saturation attack strategy.
Authors: WenJie Kang (Changsha University),
Hide Authors & Abstract

Show Authors & Abstract
10:40 - 10:55
Deep Reinforcement Learning-based joint task offloading and radio resource allocation for platoon-assisted vehicular edge computing

Platoons, formed by smart vehicles driving in the same patterns, bring potential benefits to road traffic efficiency while providing a promising paradigm to execute computation tasks with onboard computing resources. However, constrained resources of individual vehicles(IV), limited wireless coverage of vehicular communication nodes as well as high mobility of running platoons pose critical challenges on task scheduling and resource management. To address these challenges, we propose a platoon-based vehicular edge computing mechanism, which exploits computation capabilities of both platoons and edge computing enabled Roadside Units (RSUs), and jointly optimizes task offloading target selection and resource allocation. Taking aim at minimize delay cost and energy consumption of the platoon-based task execution, we leverage deep deterministic policy gradient (DDPG) to design a learning algorithm, which efficiently determines target computation servers and obtains optimized resource scheduling strategies. Numerical results demonstrate that our algorithm significantly reduces delay and energy costs in comparing its performance to that of benchmark schemes.
Authors: Yi Chen (Research Institute of Highway ministry of Transport), Xinyu Hu (School of Information and Communication Engineering, University of Electronic Science and Technology of China), Haoye Chai (School of Information and Communication Engineering, University of Electronic Science and Technology of China), Ke Zhang (School of Information and Communication Engineering, University of Electronic Science and Technology of China), Fan Wu (School of Information and Communication Engineering, University of Electronic Science and Technology of China), Lisha Gu (Research Institute of Highway ministry of Transport),
Hide Authors & Abstract

Show Authors & Abstract
10:55 - 11:10
A Joint Scheduling Scheme for Relay-Involved D2D Communications with Imperfect CSI in Cellular Systems

The introduction of device-to-device (D2D) communication brings many benefits to cellular systems, especially when relay-assisted (RA) D2D mode is included. This paper focuses on the joint scheduling problem when some channel state information (CSI) is no longer complete, involving relay selection, probabilistic access control, power coordination, mode selection and resource allocation. Since the fading components of the channel appear in the model as random variables in imperfect channels, we make some modifications to access control and channel allocation based on the previous work. By using the existing algorithm and the corresponding mathematical optimization theory, the transformed integer programming (IP) problem is decomposed into two stages to be solved separately. At the same time, in order to improve the solution efficiency, we transform the original NP-hard problem into an easily solved linear programming (LP) problem. Simulation results verify the performance of the joint scheduling scheme and the influence of imperfect CSI from the perspective of cell capacity.
Authors: Zhixiang Hu (Harbin Institute of Technology (Weihai)), Zhiliang Qin (Beiyang Electric Group Co. Ltd.), Ruofei Ma (Harbin Institute of Technology), Gongliang Liu (Harbin Institute of Technology),
Hide Authors & Abstract

Show Authors & Abstract
11:10 - 11:25
Research on coal mine gas safety evaluation based on D-S evidence theory data fusion

In order to improve the accuracy of coal mine gas safety evaluation results, a gas safety evaluation model based on D-S evidence theory data fusion is proposed. First, the prediction result of weighted least squares support vector machine is used as the input of D-S evidence theory, and the basic probability distribution function of each sensor is calculated by using the posterior probability modeling method. Secondly, to deal with the problem that the classic evidence theory is likely to cause fusion failure when processing high-conflict data, the idea is used to allocate the importance of each evidence, which can increase the reliability of the evidence to the decision result and weaken the conflicting .The effect of evidence on decision-making results. In order to prevent the loss of the effective information of the original evidence after modifying the evidence source, a conflict allocation coefficient is introduced on the basis of fusion rules to improve the accuracy of the decision-making stage. Finally, a gas safety evaluation example analysis is carried out on the evaluation model established in this paper. The results show that the evidence theory based on improved fusion rules has improved the accuracy rate of 2.8% and 15.7% respectively compared to the evidence theory based on modified evidence sources and the classic evidence theory; multi-sensor data fusion evaluation results have improved compared with single sensor evaluation results 63.5%.
Authors: Zhenming SUN (School of Energy and Mining Engineering, China University of Mining and Technology (Beijing)), Dong Li (School of Energy and Mining Engineering, China University of Mining and Technology (Beijing)), Yunbing HOU (School of Energy and Mining Engineering, China University of Mining and Technology (Beijing)),
Hide Authors & Abstract

Show Authors & Abstract
11:25 - 11:40
Sports pose estimation based on LSTM and Attention

For the estimation of motion pose, we should extract not only spatial features, but also temporal features. We use the two-branch multi-stage CNN [1] to extract human joints as features, which not only guarantees the real-time performance, but also ensures the accuracy. For the time dimension, the extracted joint features are input into the LSTM-Attention model for training. In order to verify the effec-tiveness of our proposed method, we collected data for processing and trained with the proposed model. The experimental results show that our method has a high performance
Authors: Chuanlei Zhang (School of Computer Science and Information Engineering, Tianjin University of Science and Technology), Lixin Liu (School of Computer Science and Information Engineering, Tianjin University of Science and Technology), Qihuai Xiang (Non), Jianrong Li (School of Computer Science and Information Engineering,Tianjin University of Science and Technology), Xuefei Ren (School of Computer Science and Information Engineering,Tianjin University of Science and Technology),
Hide Authors & Abstract

Show Authors & Abstract
11:40 - 11:55
3D Online Mine Ventilation Simulation System Based on GIS

An efficient coal mine ventilation system is the guarantee of the safety produc-tion in the underground coal mine system. In the procedure of the "intelligent mine" development, the intelligent construction of ventilation system is of primary concern. With the rapid development of the advanced technology, such as the Internet of Things. This paper aims to implement the online and fast ventilation simulation to provide real-time result for the afterwards deci-sion support. Firstly, the ventilation simulation model based on circuit wind flux method and Scott-Hinsley algorithm is described. Secondly, the architec-ture of the web system is designed as well as the ventilation network graph model and simulation model is constructed. Thirdly, a prototype web system is developed based on GIS technology with ventilation model integrated at the back-end, which is called 3D VentCloud. The result demonstrated that the sys-tem is efficient in providing real-time and online ventilation simulation result, which is potential to guide the fast decision support for coal mine safety pro-duction.
Authors: Hui Liu (Peking University), Shanjun Mao (Peking University), Mei Li (Peking University), Pingyang Lyu (Peking University),
Hide Authors & Abstract

Show Authors & Abstract

Lunch Break 12:00 - 12:30

Keynote 2: Professor Sherman Shen 12:30 - 13:00

Reinforcement Learning for Resource Management in Space-Air-Ground (SAG) Integrated Vehicular Networks

Session #2 13:00 - 14:30

13:00 - 13:15
On Relay Selection for Relay-Assisted D2D Communications with Adaptive Modulation and Coding in Cellular Systems

Developing device-to-device (D2D) communications in cellular systems can improve spectrum efficiency and traffic offloading capability of the cellular systems. But only considering direct D2D communication mode may limit these benefits brought in by D2D communications, for direct D2D mode may not be available due to the long separation distance and poor link quality. Hence, relay-assisted D2D communication is presented to expand the coverage of D2D communications. One of the key points in developing relay-assisted D2D communication is to find out the optimal relay user equipment (UE) to assist data transmission between the source and destination D2D-capable UEs. In this paper, a cross-layer relay selection scheme is researched, which considers the end-to-end data rate, end-to-end transmission delay, and remaining battery time of the relay-capable UEs. Specially, we propose a method to estimate the end-to-end transmission delay for the relay-involved D2D path when adaptation modulation and coding (AMC) is taken into account. Performances of the proposed scheme are also evaluated.
Authors: Xixi Bi (Harbin Institute of Technology, Weihai), Ruofei Ma (Harbin Institute of Technology), Zhiliang Qin (Beiyang Electric Group Co. Ltd.), Gongliang Liu (Harbin Institute of Technology),
Hide Authors & Abstract

Show Authors & Abstract
13:15 - 13:30
Research on SDN Enabled by Machine Learning: An Overview

Network abstraction brings the birth of Software Defined Network (SDN). SDN is a promising network architecture that separates the control logic the network from the underlying forwarding elements. SDN gives network centralized control ability and provides developers with programmable ability. In this review, the latest advances in the field of artificial intelligence (AI) have provided SDN with learning capabilities and superior decision-making capabilities. In this study, we focus on a sub-field of artificial intelligence: machine learning (ML) and give a brief review of recent researches on introducing ML into SDN. Firstly, we introduce the backgrounds of SDN and ML. Then, we conduct a brief review on existing works about how to apply several typical ML algorithms to SDN. Finally, we give conclusion towards integrating SDN with ML.
Authors: Pu Zhao (College of Computer, National University of Defense Technology), Wentao Zhao (College of Computer, National University of Defense Technology), Qiang Liu (College of Computer, National University of Defense Technology),
Hide Authors & Abstract

Show Authors & Abstract
13:30 - 13:45
Decentralized Resource Sharing Platform for Mobile Edge Computing

Recently, the Internet of things technology is booming in the industrial field, and more and more industrial devices begin to connect to the Internet. Compared with cloud computing, edge computing can well shorten the delay time on information transmission and improve the quality of service (QoS) of task computing, which promotes the development of the industrial Internet of things to some extent. The state-of-the-art edge computing service providers are specifically designed for customized applications. In order to improve the utility of these edge nodes, we proposed a blockchain-based toll collection system for public edge-sharing in our previous work, which provides a transparent, quick and cost-efficient solution to encourage the participation of edge service providers. However, the system still exists a debatable issue since it contains a centralized proxy. In this paper, we introduce consortium blockchain to record the results of the service matching process in order to solve the issue. In addition, we propose a service matching algorithm which is based on the location of the industrial Internet of Things(IIoT) devices to select the optimal node.
Authors: Hongbo Zhang (The Chinese University of Hong Kong, Shenzhen), Sizheng Fan (The Chinese University of Hong Kong, Shenzhen), Wei Cai (The Chinese University of Hong Kong, Shenzhen),
Hide Authors & Abstract

Show Authors & Abstract
13:45 - 14:00
Mining Raw Trajectories for Network Optimization from Operating Vehicles

Improving the user peak rate in hot-spots is one of the original intention of design for 5G networks. The cell radius shall be reduced to admit less users in a single cell with the given cell peak rate, namely Hyper-Dense Networks (HDN). Therefore, the feature extraction of the node trajectories will greatly facilitate the development of optimal algorithms for radio resource management in HDN. This paper presents a data mining of the raw GPS trajectories from the urban operating vehicles in the city of Shenzhen. As the widely recognized three features of human traces, the self-similarity, hot-spots and long-tails are evaluated. Mining results show that the vehicles to serve the daily trip of human in the city always take a short travel and activate in several hot-spots, but roaming randomly. However, the vehicles to serve the goods are showing the opposite characteristics.
Authors: Lei Ning (Shenzhen Technology University), Runzhou Zhang (Shenzhen Technology University), Jing Pan (Chaincomp Technologies), Fajun Li (Shenzhen Technology University),
Hide Authors & Abstract

Show Authors & Abstract
14:00 - 14:15
An Intelligent Elevator System Based on Low Power Wireless Networks

The trend of Internet of Things (IoT) and wireless network techniques have resulted in a promising paradigm, and a more rational and intelligent elevator control system shall be considered. Many previous works are devoted to improve traffic congestion capability to increase time efficiency. However, these approaches confront the limitations of destination perception in advance or there is a steady stream of persons coming to wait the elevator that may increase the uncertainty of sensing the traffic load, since the user interface is still around elevator car. In this paper, an improved elevator system is proposed with remote calling and cloud scheduling based on low power wireless networks. It enables users to call the elevator remotely through portable devices, solves the problem of elevator invalid stop, reduces system energy consumption, and improves the service life of the elevator. It can match the running state of the elevator with the multi-user call request, shorten the time for users to take the elevator, and improve the comprehensive operation efficiency of the elevator.
Authors: Congying Yu (Shenzhen Technology University), Ruize Sun (Shenzhen Technology University), Qijun Hong (Shenzhen Technology University), Weiwen Chao (Shenzhen Technology University), Lei Ning (Shenzhen Technology University),
Hide Authors & Abstract

Show Authors & Abstract
14:15 - 14:30
Research on Time-Sensitive BBU Shaper for Supporting the Transmission of Smart Grid

Large-scale of data terminals are deployed in the smart grid, especially the power distribution network. The effectiveness of data collection or scheduling for Source-Grid-Load-Storage is affected to some extent by the accuracy of the time synchronization of the terminal equipments. Here we propose a design of BBU Shaper to deal with time synchronization error on the above issues. This article analyzes the impact of synchronization errors between terminal devices within the coverage of a single base station which provides deterministic transmission for the power IoT. For the time-sensitive service flows, when the terminal has different priorities according to the service type, the time synchronization error of the terminal will affect the stability and effectiveness of the time-sensitive service flow transmission on the network. It is mainly reflected in timestamp deviation of the data collection, disrupting the determinism of other service flows and reducing the overall network resource utilization. Based on the analysis of the above problems, we proposes an edge data shaper which can reduces the impact of air interface time synchronization errors on big data services through mechanisms such as ``ahead awaits; overtime elevates''. Evaluation shows that the Shaper can reduce the impact of time synchronization errors on the average delay of services to a certain extent.
Authors: Jiye Wang (State Grid Corporation of China, No.86 West Changan St., Beijing, China), Chenyu Zhang (Beijing University of Posts and Telecommunications), Yang Li (Big Data Center of State Grid Corporation of China), Baozhong Hao (Big Data Center of State Grid Corporation of China), Zhaoming Lu (Beijing University of Posts and Telecommunications),
Hide Authors & Abstract

Show Authors & Abstract

Coffee Break #2 14:30 - 14:40

Session #3 14:40 - 16:10

14:40 - 14:55
Using Water Monitoring to Analyze the Livability of White Shrimp

The research develops an intelligent aquaculture system to detect the water quality of a culture pond. Additionally, using Fuzzy Logic to evaluate water quality that influences the aquaculture livability. Each species requires a different environ-ment of water quality; therefore, the study utilizes an intelligent aquaculture sys-tem to detect the water quality of white shrimp ponds. After using Fuzzy Logic to analyze water quality, the result is delivered as equally divided into five levels of signals sections. The purpose of the research is to understand whether the aqua-culture environment is suitable for white shrimps by detecting the water quality; consequently, through studying the livability to understand the importance of wa-ter quality. From the experimental results, the water quality of targeted aquacul-ture ponds are all within the livability range of white shrimp; the result has shown a livability rate of 33%, which is considered high livability in marine white shrimp farming. Hence, it is concluded that water quality has a high correlation with livability. Moreover, the study demonstrates that water monitoring and water quality analysis are beneficial to monitor the aquaculture environment, which can further increase the livability of white shrimp and boost income.
Authors: Hsin-Te Wu (Department of Computer Science and Information Engineering, National Ilan University, Taiwan), Wu-Chih Hu (Department of Computer Science and Information Engineering, National Penghu University of Science and Technology), Jun-Wei Zhan (Department of Computer Science and Information Engineering, National Penghu University of Science and Technology, Taiwan), Jing-Mi Zhang (Department of Computer Science and Information Engineering, National Penghu University of Science and Technology), Fan-Hsun Tseng (Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
14:55 - 15:10
Using an Automatic Identification System (AIS) to Develop Wearable Safety Devices

The numbers of people who participate in outdoor activities are increasing; how-ever, it has a certain risk of doing outdoor activities, such as wilderness rescue and losing the force while sailing. Therefore, people performing outdoor activities usually wear relevant safety devices. Currently, most users send emergency alarms through mobile phones; yet, the penetration rate of the fourth generation of broadband cellular network technology (4G Network) is not fully equipped in some rural areas, which causes victims failed to send emergency alarms. This study uses an Automatic Identification System (AIS) to develop a wearable safe-ty device because the transmission range of AIS is wider than other types of sys-tems, and it is easier to prepare essential functions in AIS wearable devices. When a victim presses the panic button, the system will send the alarm with a Global Positioning System (GPS) signal; the system will only detect the GPS lo-cation once when pressing the button to reduce power consumption, which avoids additional power consumption while repeating the detection. The research conducted practical experiments and discovered that the method is feasible in real-world applications.
Authors: Hsin-Te Wu (Department of Computer Science and Information Engineering, National Ilan University, Taiwan), Fan-Hsun Tseng (Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
15:10 - 15:25
VR and AR Technology on Cultural Communication

The theme of this platform is centered around "the road of Hess". Starting from the beginning of the 21st century maritime silk road-QuanZhou, as the core city for design and programming. The history of silk road, will combined with panorama technology, which is one of the multimedia technology of network. In the form of Web APP, humanities and scenic sites of QuanZhou will be showed in 720° perspective, an immersive experience of 3D live virtual, to offer the real records and spread. The aim of this study is to simulate and realize the all-around real scene. In the paper, the tasks of the VR panorama fused construction include image acquisition, image registration, image smoothing, image modification by the algorithm and mathematical theory behind splicing fusion, the panoramic camera category and projection model, panoramic transformation and collection principle, the image registration of feature point and transform frequency domain. Focused on the product positioning of "maritime silk route culture", we will take the images of 18 unique scenic spots in quanzhou. The study makes an overall design and the organization of connected different scenes for the virtual scene roaming system which is based on image. In a new way, the civilization of maritime silk route is better integrated into modern technology.
Authors: Minqi Guo (School of Humanities and Social Sciences, Harbin Institute of Technology, Shenzhen, China), Hsuan-Yu Chen (Department of Radiology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan), Tsung-Chih Hsiao (College of Computer Science and Technology, Huaqiao University, Fujian, China), Zhen-Yu Wu (Department of Information Management, National Penghu University of Science and Technology, Penghu, Taiwan), Huan-Chieh Tseng (Department of Information Management, Tunghai University, Taichung, Taiwan), Yu-Tzu Chang (Department of Marketing and Logistics Management, Chaoyang University of Technology, Taichung, Taiwan), Tzer-Shyong Chen (Department of Information management, Tunghai University, Taichung, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
15:25 - 15:40
Virtual Reality, Augmented Reality and Mixed Reality on the Marketing of Film and Television Creation Industry

Virtual reality and augmented reality technology has been on the road to scientific research for many years since the end of the twentieth century, and although the scope of mixed reality application is relatively narrow compared to the first two applications, we have seen the results. Realistic contents also have exploring progress in the field of film and television creation in recent years, such as virtual reality technology can be utilized to make film clips. However, in fact, these technologies can be applied in film and television creation far more than that. I then illustrate some other aspects which these realistic contents can also be applied to in film and television creation in the future and the influence on the industry.
Authors: Rui Ge (College of Computer Science and Technology, Huaqiao University, Fujian, China), Hsuan-Yu Chen (Department of Radiology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan), Tsung-Chih Hsiao (College of Computer Science and Technology, Huaqiao University, Fujian, China), Yu-Tzu Chang (Department of Marketing and Logistics Management, Chaoyang University of Technology, Taichung, Taiwan), Zhen-Yu Wu (Department of Information Management, National Penghu University of Science and Technology, Penghu, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
15:40 - 15:55
Multi-function electric scooter assistant system for seniors

Many countries are moving towards an aging society. Many elderly people live alone, and they need to take themselves. The purpose of this article is to increase the safety when elderly people ride on a scooter alone. When an ac-cident occurs, their family can clearly know the location of the accident and can record the entire accident process. This paper uses the IFTTT (If This Then That) system. When the accident occurs, IFTTT is used to send the lo-cation to the family LINE. The family can know the location of the elderly through remotely monitor the real-time behavior of the monitored person. Therefore, they can immediately understand the situation when the accident occurs. In addition, they can also know whether the elderly is riding the scooter through the pressure sensor. It can transmit GPS location immediately and start the camera to record.
Authors: Tzer-Long Chen (Department of Finance, Providence University, Taichung, Taiwan), Chien-Yun Chang (Department of Fashion Business and Merchandising, Ling Tung University, Taichung, Taiwan), Zhen-Yu Wu (Department of Information Management, National Penghu University of Science and Technology, Penghu, Taiwan), Chia-Hui Liu (Department of Applied Mathematics, Chinese Culture University, Taipei, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
15:55 - 16:10
Design and Implementation a Smart Pillbox

This research uses a smart phone with a smart pill box to supervise and re-mind the elderly to take medicine. Considering that most of the users are el-derly people who are not familiar with the operation process of the mobile APP in the smart phone, a novel real-time transmission of electronic drug or-ders was developed in the system. It supports a simple and useful user inter-face for elderly. In the proposed system, the doctor completed the diagnosis and sent the electronic drug list from Near-field communication (NFC) to the APP of the elderly through the doctor ’s mobile phone APP at first. Then, the information such as setting the medication time and the number of days to return to the doctor are set. When it is time to take the medicine for elderly, the mobile phone APP starts the alarm to remind the user to take the medi-cine. In the proposed smart pillbox, we use the Arduino UNO development board to design, control, and add a time RTC clock module on the board to control the time. When it is time to take medicine, the servo motor will open / close the pillbox's medicine port.
Authors: Chin-Chia Hsu (Department of Business Administration, St. John’s University, New Taipei City, Taiwan), Tzer-Long Chen (Department of Finance, Providence University, Taichung, Taiwan), I-Fang Chang (Department of Finance, Providence University, Taichung, Taiwan), Zhen-Yu Wu (Department of Information Management, National Penghu University of Science and Technology, Penghu, Taiwan), Chia-Hui Liu (Department of Applied Mathematics, Chinese Culture University, Taipei, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
Day 2 16/08/2020
Room #1

08:30 - Keynote 3: Professor Tarik Taleb 08:30 - 09:00

Cross Roads for Cloud & Networks

Keynote 4: Professor Mohsen Guizani 09:00 - 09:40

6G Properties & Requirements for Complex Systems

Coffee Break #1 09:40 - 09:50

Session #4 09:50 - 12:00

09:50 - 10:05
Towards the Future Data Market: Reward Optimization in Mobile Data Subsidization

Mobile data subsidization launched by network operators is a promising business model to provide some economic insights on the evolving direction of the 4G/5G and beyond mobile data market. The scheme allows content providers to partly subsidize mobile data consumption of mobile users in exchange for displaying a certain amount of advertisements. The users are motivated to access and consume more content without being concerned about overage charges, yielding higher revenue to the data subsidization ecosystem. For each content provider, how to provide appropriate data subsidization (reward) competing with others to earn more revenue and gain higher profit naturally becomes the key concern in such a ecosystem. In this paper, we adopt a hierarchical game approach to model the reward optimization process for the content providers. We formulate an Equilibrium Programs with Equilibrium Constraints (EPEC) problem to characterize the many-to-many interactions among multiple providers and multiple users. Considering the inherent high complexities of the EPEC problem, we propose to utilize the distributed Alternating Direction Method of Multipliers (ADMM) algorithm to obtain the optimum solutions with fast-convergence and decomposition properties of ADMM.
Authors: Zehui Xiong (Alibaba-NTU Joint Research Institute, NTU), Jun Zhao (School of Computer Science and Engineering, NTU), Jiawen Kang (Energy Research Institute, NTU), Dusit Niyato (School of Computer Science and Engineering, NTU), Ruilong Deng (College of Control Science and Engineering, and School of Cyber Science and Technology, Zhejiang University), Shengli Xie (Guangdong-Hong Kong-Macao Joint Laboratory for Smart Discrete Manufacturing),
Hide Authors & Abstract

Show Authors & Abstract
10:05 - 10:20
Private Cloud in 6G Networks: A Study from the Total Cost of Ownership Perspective

Security and privacy concerns are increasingly important when massive data processing and transferring becomes a reality in the era of the Sixth Generation (6G) Networks. Under this circumstance, it becomes a trend that the enterprises tend to host their data and services on private clouds dedicated to their own use, rather than the public cloud services. However, in contrary to the well-investigated total cost of ownership (TCO) for public clouds, the analytic research on the cost of purchase and operation for private clouds is still a blank. In this work, we first review the state-of-the-art TCO literature to summarize the models, tools, and cost optimization techniques for public clouds. Based on our survey, we envision the TCO modeling and optimization for private clouds by comparing the differences of features between public and private clouds.
Authors: Yuanfang Chi (Alibaba Group, Beijing, China), Wei Dai (The Chinese University of Hong Kong, Shenzhen, China), Yuan Fan (Alibaba Group, Beijing, China), Jun Ruan (Alibaba Group, Beijing, China), Kai Hwang (The Chinese University of Hong Kong, Shenzhen), Wei Cai (The Chinese University of Hong Kong, Shenzhen),
Hide Authors & Abstract

Show Authors & Abstract
10:20 - 10:35
System-Level Performance Analysis in 3D Drone Mobile Networks

We present a system-level anaysis for drone mobile networks on a finite three-dimensional (3D) space. A performance boundary derived by deterministic random (Brownian) motion model over Nakagami-m fading interfering channels is developed. This method allows us to circumvent the extremely complex reality model and obtain the upper and lower performance bounds of actual drone mobile networks. The validity and advantages of the proposed framework are confirmed via extensive Monte-Carlo(MC) simulations. The results reveal several important trends and design guidelines for the practical deployment of drone mobile networks.
Authors: Jiayi Huang (South China University of Technology), Arman Shojaeifard (BT Labs), Jie Tang (South China University of Technology), Kai-Kit Wong (University College London),
Hide Authors & Abstract

Show Authors & Abstract
10:35 - 10:50
Joint Power Allocation and Splitting Control in SWIPT-Aided Multi-Carrier NOMA System

The combination of non-orthogonal multiple access (NOMA) and simultaneous wireless information and power transfer (SWIPT) contributes to improve the spectral efficiency (SE) and the energy efficiency (EE) at the same time. In this paper, we investigate the throughput maximization problem for the downlink multi-carrier NOMA (MC-NOMA) system with the application of power splitting (PS)-based SWIPT, in which power allocation and splitting are jointly optimized with the constraints of maximum transmit power supply as well as the minimum demand for energy harvesting (EH). To tackle the non-convex problem, a dual-layer approach is developed, in which the power allocation and splitting control are separated and the corresponding sub-problems are respectively solved through Lagrangian duality method. Simulation results validate the theoretical findings and demonstrate the superiority of the application of PS-based SWIPT to MC-NOMA over SWIPT-aided single-carrier NOMA (SC-NOMA) and SWIPT-aided orthogonal multiple access (OMA).
Authors: Jie Tang (South China University of Technology), Jingci Luo (South China University of Technology), Junhui Ou (South China University of Technology), Xiuyin Zhang (South China University of Technology), Nan Zhao (Dalian University of Technology), Daniel So (The University of Manchester), Kai-Kit Wong (University College London),
Hide Authors & Abstract

Show Authors & Abstract
10:50 - 11:05
The Development of Smart Home Disaster Prevention System

Building smart home within Internet of Things (IoT) technology can bring residents living convenience. A smart home infrastructure needs to be built with the Internet of things and wireless network. In this article, using the raspberry Pi to build a smart home disaster prevention system. The system functions proposed in this article are: 1. Real-time monitoring of home envi-ronment and safety issues, 2. Indoor temperature and humidity sensing, 3. Pot temperature and humidity sensing, 4. Disaster detection, 5. Fish tank wa-ter level detection. The proposed method helps to provide a comfortable home environment. It also reduces the damage caused by home disasters. Using the concept of home automation, we propose the smart home disaster prevention system which can reduce the pressure on residents to maintain the home environment.
Authors: Zhan-Ping Su (Department of Business Management, Chungyu University of Film and Arts, New Taipei City, Taiwan), Chin-Chia Hsu (Department of Business Administration, St. John’s University, New Taipei City, Taiwan), Zhen-Yu Wu (Department of Information Management, National Penghu University of Science and Technology, Penghu, Taiwan), Chia-Hui Liu (Department of Applied Mathematics, Chinese Culture University, Taipei, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
11:05 - 11:20
On the Security Policy and Privacy Protection in Electronic Health Information System

The integration of information technology and medical techniques could benefit expanding the time and location for medical services as well as provide better medical quality that it has become a trend in medical domain. For this reason, this study proposes to develop an integrated healthcare information system. Furthermore, based on user privacy and system information security, the key in a smart national health insurance card or a patient’s fingerprint together with a doctor’s group signature and public key are utilized for retrieving electronic medical records from the system so as to protect the confidential information in the system and guarantee patients’ privacy. This system covers other relevant functions of medical record mobility, data link, information security protection, and drug conflict avoidance. In the integrated healthcare information system, a patient’s privacy could be protected through hiding. The retrieval location of medical records is always the same that doctors from different hospitals could access to the medical records through authorization. When dealing with an urgent case, the real-time medical record retrieval could effectively enhance the recovery rate. The information in the system procedure broadly covers diagnoses of patients, insurance claims, and drug collection in order to prevent doctors from prescribing wrong medicine, avoid the troublesome of insurance claim application, and reduce patients’ problem about collecting receipts.
Authors: Hsuan-Yu Chen (Department of Radiology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan), Yao-Min Huang (Department of Management Sciences, National Chiao Tung University, Hsinchu, Taiwan), Zhen-Yu Wu (Department of Information Management, National Penghu University of Science and Technology, Penghu, Taiwan), Yin-Tzu Huang (Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract
11:20 - 11:35
Research on detection method of malicious node based on flood attack in VANET

Due to the variability and mobility of the topology of the VANET, it is vulnerable to be attacked by attackers. This paper combines the rapidity and real-time re-quirements of the detection of malicious nodes in the security of the VANET. At the same time, in order to enhance the ability to identify and detect malicious nodes, a TCP SYN (Synchronize Sequence Numbers) flooding attack detection and UDP traffic attack detection are proposed. For the TCP SYN flooding attack, RSU (Road Side Unit side unit) is used to monitor the number of SYN request packets of each node in real time and limit the number of requests for network ac-cess. For UDP traffic attack, RSU is used to monitor and analyze whether the da-ta traffic of each node in the network is abnormal. Through experimental analysis, the above two detection methods can effectively detect flooding attacks in the VANET, thus assisting the network defense mechanism to ensure network secu-rity.
Authors: yizhen xie (Beijing University of Posts and Telecommunications), yuan li (Southwest University of Science and Technology), yongjian wang (National Internet Emergency Center),
Hide Authors & Abstract

Show Authors & Abstract
11:35 - 12:00
Research on Simulation Method of 5G Location Based on Channel Modeling

It is of great significance to study 5G positioning as the new series of tech-nologies introduced by 5G can improve the positioning accuracy and achieve indoor and outdoor seamless navigation and positioning except for meeting the needs of communication. However, 5G base stations and devices are far from popular, and standards of 5G positioning has not fully formed. Thus, simulation experiment is the main research method. Accordingly, we propose a 5G positioning simulation experiment scheme and present its flow chart. The implementation ideas and the specific processes of the three main parts of the simulation experiment including scene generation, signal propagation simulation and position estimation are introduced. including. Furthermore, the positioning accuracy in the 5G simulation environment established by the scheme is verified and the influence factors of location is studied.
Authors: GUO Chi (GNSS Research Center, Wuhan University, Wuhan 430079, China), Yu Jiang (School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China), GUO fei (GNSS Research Center, Wuhan University, Wuhan 430079, China), DENG Yue (School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China), LIU nan (GNSS Research Center, Wuhan University, Wuhan 430079, China),
Hide Authors & Abstract

Show Authors & Abstract

Lunch Break 12:00 - 12:30

Session #5 12:30 - 14:30

12:30 - 12:45
Hybrid Duplex Relay Selection in Cognitive Radio Network

The existing cognitive single relay selection technology is lack of a full-duplex/half-duplex mode switching criteria, and only maximizes the system performance of secondary users in the single-duplex mode relay selection process, which result in a single relay mode and the overall performance of single primary networks and secondary networks is not optimal. To tackle these problems, the criteria for switching the hybrid duplex mode of the relay node is first derived. In the case of adopting the amplify-and-forward (AF) and decode-forward (DF) protocols, each candidate relay terminal selects the optimal duplex mode according to the signal-to-noise ratio (SNR) of the self-interference channel and the critical point of self-interference. Then propose a low interference hybrid duplex relay selection algorithm, the relay node uses the comprehensive performance loss of the primary network performance and the performance gain of the secondary network user as the indicator to optimize the relay selection, which maximizes the sacrifice value of the primary network. The simulation results show that the proposed algorithm can significantly improve the performance of the primary network system and the whole system. Furthermore, the performance of the secondary network is slightly reduced (or even improved).
Authors: Yaru Xue (Chongqing University of Posts and Telecommunications), Weina Hou (Chongqing University of Posts and Telecommunications), Zhaoyi Li (Chongqing University of Posts and Telecommunications), Shuaishuai Xia (Chongqing University of Posts and Telecommunications),
Hide Authors & Abstract

Show Authors & Abstract
12:45 - 13:00
A virtual reality simulation system for coal safety based on cloud rendering and AI technology

Coal mining, regarded as a high-risk industry, has strongly demand on Virtual Reality (VR) training environment for safety mining and emergency rescue. In order to solve the problems of the current VR platform of lack of immersive ex-perience and interest, a Browser/Client serious-game VR simulation system are designed and realized. Three key techniques are studies, which are cloud ren-dering system architecture, AI behavior tree and mining disaster animation. The system has been successfully applied in the Virtual Reality Teaching and Ex-periment Laboratory for undergraduate in China University of Mining & Tech-nology-Beijing. The research, as a new tool for miner training and disaster drill-ing, has a signification meaning of the work safety IT construction.
Authors: Zhan Jiang (Peking University), Mei Li (Peking University), Zhenming Sun (China University of Mining & Technology), Zheng Tan (Beijing Longruan Technologies), Jinchuan Chen (Beijing Longruan Technologies),
Hide Authors & Abstract

Show Authors & Abstract
13:00 - 13:15
High-discrimination multi-sensor information decision algorithm based on distance vector

In the process of sensor target recognition, attitude estimation and information decision-making, most of the current sensor information decisions require probability conversion or weight calculation of sensor data. The calculation process is complex and requires a large amount of computation. In addition, the decision result is greatly affected by the probability value. This paper proposes a multi-sensor information decision algorithm with high-discrimination based on distance vectors. At the same time, the support function, dominance function and discrimination function for the algorithm are presented. The dominance function is obtained through the normalization processing of the support matrix, and then the dominance function after normalization is sorted. The maximum value is taken as the optimal solution. The discrimination function mainly provides the basis for the evaluation of the algorithm. The simulation results show that the discrimination degree of this method in sensor information decision-making reaches more than 0.5, and the decision-making effect is good. Compared with the classic D-S evidence theory, this algorithm can effectively avoid the phenomenon that D-S evidence theory contradicts with the actual situation when making a decision. It is less affected by a single sensor and the decision result is stable. Compared with the probabilistic transformation of the initial data of the sensor in the decision-making process, it has obvious advantages.
Authors: Lingfei Zhang (Qinghai Nationalities University), Bohang Chen (Qinghai Normal University),
Hide Authors & Abstract

Show Authors & Abstract
13:15 - 13:30
Research on Monitoring and Regularity of Coal Mine Methane Extraction State

The existing gas drainage monitoring system not only can't control the gas concentration of the drainage, but also can't dynamically predict the residual gas and the drainage cycle of the coal seam according to the drainage parameters. The mining conditions of deep coal seams have become complicated by the increasing depth of mining. The measurement of the extraction status becomes more and more difficult. It is of great practical significance to effectively predict the drainage time and the gas content of coal seam. In order to increase the gas drainage concentration of coal seam under the coal mine, combined with the actual situation at the coal mine site, a gas drainage pipeline control system was designed and developed. It is mainly composed of gas sensors, microcontroller, miniature electric valves, power conversion circuits, and wireless communication modules. In view of the non-linear, multi-coupling and hysteresis characteristics of gas flow, a fuzzy control algorithm is adopted by the controller. By measuring the gas concentration and pressure of the field extraction pipeline, the microcontroller STM8 working as the control core controls opening of the pipeline valve according to the fuzzy control law .Finally the purpose of increasing the concentration of the extracted gas is achieved. Both the fuzzy control algorithm and the system circuit is designed in detail.
Authors: Chao Wu (Shandong University of Science and Technology,College of Intelligent Equipment), Yuliang Wu (Sichuan Staff University of Science and Technology), Changsheng Zhu (Shandong University of Science and Technology), Chengyuan Zhang (Shandong University of Science and Technology), Qingjun Song (Shandong University of Science and Technology),
Hide Authors & Abstract

Show Authors & Abstract
13:30 - 13:45
Modulation Recognition with Alpha-Stable Noise over Fading Channels

This paper proposes a method based on kernel density estimation(KDE)and expectation condition maximization(ECM) to realize digital modulation recognition over fading channels with non-Gaussian noise in the cognitive radio networks. A compound hypothesis test model is adopt here.The KDE method is used to estimate the probability density function of non-Gaussian noise,and the improved ECM,algorithm is used to estimate the fading channel parameters. Numerical results show that the proposed method is robust to the noise type over fading channels. Moreover, when the GSNR is 10dB, the correct recognition rate for the digital modulation recognition under non-Gaussian noise is more than 90%. Gaussian noise, and the improved ECM algorithm is used to estimate the fading channel parameters. Numerical results show that the proposed method is robust to the noise type over fading channels. Moreover, when the GSNR is 10dB, the correct recognition rate for the digital modulation recognition under non-Gaussian noise is more than 90%.
Authors: Lingfei Zhang (Qinghai Nationalities University), Mingqian Liu (Xidian University), Jun Ma (Qinghai Normal University), Chengqiao Liu (Qinghai Normal University),
Hide Authors & Abstract

Show Authors & Abstract
13:45 - 14:00
An improved Linear threshold model

Linear threshold model is one of the widely used diffusion models in influence maximization problem. It simulates influence spread by activating nodes for an iterative way. However, the way it makes activation decisions limits the convergence speed of itself. In this paper, an improvement is proposed to speed up the convergence of Linear threshold model. The improvement makes activation decisions with one step ahead considering the nodes which will be activated soon. To assist in activation decision making, the improvement introduces a new state and updates state transition rules. The experiment results verify the performance and efficiency of the improvement.
Authors: Xiaohong Zhang (Henan Polytechnic University), Nanqun He (Henan Polytechnic University), Kai Qian (Henan Polytechnic University), Wanquan Yang (Henan College of Survey and Mapping), Jianji Ren (Henan Polytechnic University),
Hide Authors & Abstract

Show Authors & Abstract
14:00 - 14:15
Performance Analysis for Caching in Multi-tier IoT Networks with Joint Transmission

The rapid growth of the number of IoT devices in the network has brought huge traffic pressure to the network. Caching at the edge has been regarded as a promising technique to solve this problem. However, how to further improve the successful transmission probability (STP) in cache-enabled multi-tier IoT networks (CMINs) is still an open issue. To this end, this paper proposes a base station (BS) joint transmission scheme in CMIN where the nearest BS that stores the requested files in each tier is selected to cooperatively serve the typical UE. Based on the proposed scheme, we derive an integral expression for the STP, and optimize the content caching strategy for a two-tier network case. The gradient projection method is used to solve the optimization problem, and a locally optimal caching strategy (LCS) is obtained. Numerical simulations show that the LCS achieves a significant gain in STP over three comparative baseline strategies.
Authors: Tianming Feng (Harbin Institute of Technology), Shuo Shi (Harbin Institute of Technology), Xuemai Gu (Harbin Institute of Technology), Zhenyu Xu (Huizhou Engineering Vocational College),
Hide Authors & Abstract

Show Authors & Abstract
14:15 - 14:30
Collaborative Mobile Edge Caching Strategy Based on Deep Reinforcement Learning

Recently, with the advent of the 5th generation mobile networks (5G) era, the emergence of mobile edge devices has accelerated. Nevertheless, the generation of massive edge data brought by massive edge devices challenges the connectivity and cache computing capabilities of the internet of things (IoT) devices. Therefore, mobile edge caching, as the key to realize efficient prefetch and cache of edge data and improve the performance of data access and storage, has attracted more and more experts and scholars’ attention. However, the complexity and heterogeneity of the devices in the edge cache scenario make it unable to meet the low latency requirements of 5G. In order to make the mobile edge caching more intelligent, based on the widely deployed macro base stations (ξBSs) and micro base stations (µBSs) in 5G scenarios, the ξBS cooperation space and µBS cooperation space is conceived in this paper. Besides, a deep reinforcement learning (DRL) algorithm with both perceptive ability and decision-making ability is used to realize collaborative edge caching. Then, we deploy federated learning (FL) to train DRL agents locally, which can not only solve the problem of resource imbalance but also realize the localization of training data. In addition, we formulated the energy consumption problem in the collaborative cache as an optimization problem. Simulation results show that the proposed scheme greatly reduces the caching cost and improves the user’s internet experience.
Authors: Jianji Ren (HeNan Polytechnic University), Tingting Hou (HeNan Polytechnic University), Shuai Zheng (HeNan Polytechnic University),
Hide Authors & Abstract

Show Authors & Abstract

Coffee Break #2 14:30 - 14:40

Session #6 14:40 - 17:00

14:40 - 14:55
MPR Selection Based on Link Transmission Quality and Mobile Similarity in OLSR for FANETs

With the development and popularization of 5G networks and UAV applications, UAV self-organizing networks have received more and more attention. Routing protocols have always been a key technology in mobile ad hoc net-works (MA-NETs), especially in Flying Ad-hoc Networks (FANETs). The rapid changes in network topology brought about by the high-speed mobility of UAV nodes make the network performance more susceptible, which poses a greater challenge to FANETs routing technology. This paper mainly focuses on the optimization problem of OLSR protocol in FANETs under planar topology, considering the link transmission quality and link stability between nodes, uses a weighted index to replace the node connection degree as the MPR set selection criterion, an MPR set selection algorithm based on mobile similarity and link transmission quality is proposed. Then in the NS2 emulator, a comparative analysis of PDR, end-to-end delay, and routing overheads are performed to verify the performance improve-ment of the proposed algorithm.
Authors: Ziheng Li (Harbin Institute of Technology), Shuo Shi (Harbin Institute of Technology), Xuemai Gu (Harbin Institute of Technology),
Hide Authors & Abstract

Show Authors & Abstract
14:55 - 15:10
Research on D2D Resource Fair Allocation Algorithm in Graph Coloring

In a heterogeneous network composed of cellular users and device-to-device (D2D) users, D2D users multiplex the spectrum resources of cellular users in heterogeneous networks, which improves the shortage of spectrum resources. But this will bring a series of interference problems, which will greatly affect the throughput of the system and the service rate of users. On the premise of ensuring the service quality and throughput of users in the system in a heterogeneous network, to improve the user service rate, a fair distribution algo-rithm of D2D resources in graph coloring is proposed. First, in a heterogeneous network system, allowing multiple D2D users to share the resources of the same cellular user at the same time can improve the utilization of spectrum resources; Secondly, the interference graph is constructed by users and the interference be-tween users in the heterogeneous network, and then the resource allocation col-ored by the D2D graph is added with a priority factor so that the fairness of user resource acquisition in the system is improved. Finally, it is verified by simula-tion, the algorithm improves the fairness of D2D users' access to resources while maintaining stable system throughput. It also reduces the system's packet loss rate and improves the user's service quality.
Authors: Yanliang Ge (Northeast Petroleum University), Qiannan Zhao (Northeast Petroleum University), Guanghua Zhang (Northeast Petroleum University), Weidang Lu (Zhejiang University of Technology),
Hide Authors & Abstract

Show Authors & Abstract
15:10 - 15:25
Anti-interference UAV Powered Cooperative MEC Network for Intelligent Agriculture

Mobile edge computing (MEC) provides computing resources for a large number of Internet-of-things (IoT) devices in intelligent agriculture, improv-ing the efficiency of agriculture. Integrating unmanned aerial vehicle (UAV) into wireless powered MEC system, the energy collection efficiency can be improved. In this paper, we proposed an anti-interference UAV powered co-operative mobile edge computing scheme for intelligent agriculture, in which IoT devices utilizes different subcarriers to transmit the information. An op-timization problem is formulated to minimize the transmit energy of the UAV through power allocation of IoT devices. Simulation results demon-strate the performance of the proposed scheme.
Authors: Chenkai Li (Zhejiang University of Technology), weidang lu (Zhejiang University of Technology), Xiaohan Xu (Zhejiang University of Technology), hong peng (Zhejiang University of Technology), Guoxing Huang (Zhejiang University of Technology),
Hide Authors & Abstract

Show Authors & Abstract
15:25 - 15:40
Fast estimation for the number of Clusters

Clustering analysis has been widely used in many areas. In many cases, the number of clusters is required to been assigned artificially, while inappropriate assignments affect analysis negatively. Many solutions have been proposed to estimate the optimal number of clusters. However, the accuracy of those solutions drop severely on overlapping data sets. To handle the accuracy problem, we propose a fast estimation solution based on the cluster centers selected in a static way. In the solution, each data point is assigned with one score calculated according to a density-distance model. The score of each data point does not change any more once it is generated. The solution takes the top k data points with the highest scores as the centers of k clusters. It utilizes the significant change of the minimal distance between cluster centers to identify the optimal number of the clusters in overlapping data sets. The experiment results verify the usefulness and effectiveness of our solution.
Authors: Xiaohong Zhang (Henan Polytechnic University), Zhenzhen He (Henan Polytechnic University), Zongpu Jia (Henan Polytechnic University), JIANJI REN (Henan Polytechnic University),
Hide Authors & Abstract

Show Authors & Abstract
15:40 - 15:55
Research on Risk Transmission Process and Immune Strategy of Mine Electric Power Information Network

The power information network is becoming more and more important in the safe and efficient production operation of the mine power system. Meanwhile, the power information network may be subject to security risks, such as malicious virus attacks, which poses challenges to mine safety pro-duction. Based on the complex network theory, this paper proposes a com-plex network model of the power information network. Aiming at the possi-ble attack risk of the power information network, the SIR epidemic model is used to analyze and research on the evolution process of the power infor-mation network risk. On this basis, two immunization strategies are pro-posed to suppress the continuous propagation of power information network security risks. The immunization process of the power information network is simulated to verify the significance of the immunization strategy in the process of power information network security risk transmission.
Authors: caoyuan ma (China University of Mining and Technology), qi chen (China University of Mining and Technology), wei chen (Jiangsu Normal University), long yan (China University of Mining and Technology), xianqi huang (China University of Mining and Technology),
Hide Authors & Abstract

Show Authors & Abstract
15:55 - 16:10
Rockburst prediction of multi-dimensional cloud model based on improved hierarchical and critic

In high terrestrial stress regions, rockburst is a major geological disaster influencing underground engineering construction significantly. How to car-ry out efficient and accurate rock burst prediction remains to be solved. Comprehensively consider the objective information of the index data and the important role of subjective evaluation and decision-making in rockburst prediction, and use the improved analytic hierarchy process and the CRITIC method based on index correlation to obtain the subjective and objective weights of each index, and obtain comprehensive weights based on the prin-ciple of minimum discriminant information. The original cloud model and the classification interval of the forecast index were modified to make up for the lack of sensitivity of the original cloud model to the average of the grade interval. A hierarchical comprehensive cloud model of each index was gener-ated through a cloud algorithm. Finally, the reliability and effectiveness of the model were verified through several sets of rockburst examples, and compared with the entropy weight-cloud model, CRITIC-cloud model and set pair analysis-multidimensional cloud model. The results show that the model can describe various uncertainties of interval-valued indicators, quick-ly and effectively determine rockburst severity.
Authors: wei yang, Xiaoyue Liu (North China University of Science and Technology, Tangshan Hebei 063200, China),
Hide Authors & Abstract

Show Authors & Abstract
16:10 - 16:25
Improvement of Online Education Based on A3C Reinforcement Learning Edge Cache

Online education is the complement and extension of campus education. Aiming at the situation that the mainstream network speed in online education scenarios is difficult to meet the requirements for smooth video playback, such as ultra-high-definition video resources, live broadcast, etc., this paper proposes an A3C-based online education resource caching mechanism. This mechanism uses A3C reinforcement learning-based edge caching to cache video content, which can meet the requirements of smooth video playback while reducing bandwidth consumption and improving network throughput. We use the Asynchronous Advantage Actor-Critic (A3C) technology of asynchronous advantage actors as a caching agent for network access. The agent can learn based on the content requested by the user and make a cache replacement decision. As the number of content requests increases, the hit rate of the cache agent gradually increases, and the training loss gradually decreases. The experimental comparison of LRU, LFU, and RND shows that this scheme can improve the cache hit rate.
Authors: Haichao Wang (Henan Polytechnic University), Tingting Hou (Henan Polytechnic University), JIANJI REN (Henan Polytechnic University),
Hide Authors & Abstract

Show Authors & Abstract
16:25 - 16:40
Adaptive Collaborative Computing In Edge Computing Environment

The rapid development of 5th generation mobile networks (5G) and Internet of Things (IoT) technologies will generate a large amount of data, the processing and analysis requirements of big data will challenge existing networks and processing platforms. As the most promising technology in 5G networks, edge computing will greatly ease the pressure on network and data processing analysis on the edge. In this paper, we consider the coordination between compute and cache resources between multi-level edge computing nodes(ENs), users under this system can offload computing tasks to ENs to improve quality of service(QoS). We aim to maximize the long-term profit on the edge, while satisfying the low-latency computing of the users, and jointly optimize the edge-side node offloading strategy and resource allocation. However, it is challenging to obtain an optimal strategy in such a dynamic and complex system. Therefore, we use double deep Q-learning(DDQN) to make decisions to solve the complex resource allocation problem on the edge and make edge have certain adaptation and cooperation. Ability to maximize long-term gains while making quick decisions. The simulation results prove the effectiveness of DDQN in maximizing revenue when allocation resources on the edge.
Authors: Jianji Ren (Henan Polytechnic University), Haichao Wang (Henan Polytechnic University), Xiaohong Zhang (Henan Polytechnic University),
Hide Authors & Abstract

Show Authors & Abstract
16:40 - 16:50
Construction of Smart Carbon Monitoring Platform for Small Cities in China Based on Internet of Things

The rapid development of the Internet of Things has promoted the construction of smart cities around the world. Research on carbon reduction path based on Internet of Things technology is an important direction for global low carbon city research. Carbon dioxide emissions in small cities are usually higher than in large and medium cities. However, due to the large difference of data environment between small cities and large and medium-sized cities, the weak hardware foundation of the Internet of Things and the high input cost, the construction of a small city smart carbon monitoring platform has not yet been carried out. This paper proposes a smart carbon monitoring platform that combines traditional carbon control methods with IoT technol-ogy. It can correct existing long-term data by using real-time data acquired by the sensing de-vice. Therefore, the dynamic monitoring and management of low-carbon development in small cities can be realized. The conclusion are summarized as follows: (1) Intelligent thermoelectric systems, industrial energy monitoring systems, and intelligent transportation systems are the three core systems of the monitoring platform. (2) The initial economic input of the monitoring platform can be reduced by setting up IoT identification devices in departments and enterprises with data foundations and selecting samples by using classification and stratified sampling.
Authors: He Zhang (School of Architecture, Tianjin University), Jianxun Zhang (School of Architecture, Tianjin University), Rui Wang (School of Architecture, Tianjin University), Qianrui Peng (School of Architecture, Tianjin University), Xuefeng Shang (School of Architecture, Tianjin University), Chang Gao (Urban Planning Institute of Tianjin University),
Hide Authors & Abstract

Show Authors & Abstract
16:50 - 17:00
Group paging mechanism with pre-backoff for machine-type communication

In order to resolve congestion of the random access channel (RAC) caused by UE’s concentrated access to the network in the process of group paging, this paper introduces the pre-backoff algorithm on the basis of studying MTC service characteristics to the group paging mechanism, and proposes the analysis model based on the group paging mechanism with pre-backoff. In the group paging mechanism with pre-backoff, when monitoring results indicate ongoing of group paging, all UE within the group, before access, will implement pre-backoff, and through pre-backoff, different pieces of UE are evenly distributed to a period of access time to alleviate collisions and conflicts resulted from concentrated access. Simulation results of three performance indexes, including the probability of conflict, probability of successful access and delay of average access, are used to analyze and verify the validity of the analysis model based on the group paging mechanism with pre-backoff.
Authors: Yong Liu (School of Telecommunication Engineering, Beijing Polytechnic), Qinghua Zhu (School of Telecommunication Engineering, Beijing Polytechnic), Jingya Zhao (School of Telecommunication Engineering, Beijing Polytechnic), Wei Han (School of Telecommunication Engineering, Beijing Polytechnic),
Hide Authors & Abstract

Show Authors & Abstract

Closing Speech 17:00 - 17:10