[GN Vol.3 No.2] Popular Articles and Research Trends

  • Writer

    관리자
  • RegDate

    2021-12-31 00:00
  • Hits

    31
  • Category

    Popular Articles and Research Trend

 Popular Articles and Research Trend 

 

by Taewoon Kim, Hallym University, South Korea

 

One effective way to identify the recent research directions and trends is to pay attention to the most popular research papers. This article reviews such most accessed and downloaded papers in Journal of Communications and Networks (JCN) and Information & Communications Technology Express (ICT Express), both of which are published by The Korean Institute of Communications and Information Sciences (KICS).

 

<JCN>


 

Paper

F. Nait-Abdesselam et al., "Towards enabling unmanned aerial vehicles as a service for heterogeneous applications,” vol. 23, no. 3, Jun. 2021.

DOI

https://doi.org/10.23919/JCN.2021.000015

Abstract

The increasing use of unmanned aerial vehicles (UAVs) in various commercial applications, such as precision agriculture and aerial remote sensing, is fast contributing to a significant growth in the UAV market. Also, it is crucial to provide continuous coverage after failures of wireless network components or additional bandwidth in high traffic situations. By introducing the concept of UAVs as a service (UaaS), we propose a novel framework, dubbed D3S, consisting of four phases: Demand, decision, deployment, and service. The main objective of this framework is to provide a realistic and streamlined approach to support the implementation of the UaaS paradigm. The technical problems involved include determining the type and number of UAVs to be deployed and their final locations (e.g., hovering or on-ground). They also include the trajectory planning, possibly several times, between charging stations and deployment locations. We present the application of the D3S framework to two case studies with the goal of providing wireless connectivity services to (i) static users after failures of wireless network components, including long-term and short-term failures, and (ii) dynamic users in wireless relaying systems.

 

 

 

Paper

Z. Wu et al., "Joint deployment and trajectory optimization in UAV-assisted vehicular edge computing networks," pp. 1-12, Sep. 2021. (early access)

DOI

https://doi.org/10.23919/JCN.2021.000026

Abstract

As the general mobile edge computing (MEC) scheme cannot adequately handle the emergency communication requirements in vehicular networks, unmanned aerial vehicle (UAV)-assisted vehicular edge computing networks (VECNs) are envisioned as the reliable and cost-efficient paradigm for the mobility and flexibility of UAVs. UAVs can perform as the temporary base stations to provide edge services for road vehicles with heavy traffic. However, it takes a long time and huge energy consumption for the UAV to fly from the stay charging station to the mission areas disorderly. In this paper, we design a pre-dispatch UAV-assisted VECNs system to cope with the demand of vehicles in multiple traffic jams. We propose an optimal UAV flight trajectory algorithm based on the traffic situation awareness. The cloud computing center (CCC) server predicts the real-time traffic conditions, and assigns UAVs to different mission areas periodically. Then, a flight trajectory optimization problem is formulated to minimize the cost of UAVs, while both the UAV flying and turning energy costs are mainly considered. In addition, we propose a deep reinforcement learning(DRL)-based energy efficiency autonomous deployment strategy, to obtain the optimal hovering position of UAV at each assigned mission area. Simulation results demonstrate that our proposed method can obtain an optimal flight path and deployment of UAV with lower energy consumption.

 

 

 

<ICT Express>

 

Paper

C.-W. Tsai et al., “An efficient parallel machine learning-based blockchain framework,” vol. 7, no. 3, pp. 300-307, Sep. 2021.

DOI

https://doi.org/10.1016/j.icte.2021.08.014

Abstract

The unlimited possibilities of machine learning have been shown in several successful reports and applications. However, how to make sure that the searched results of a machine learning system are not tampered by anyone and how to prevent the other users in the same network environment from easily getting our private data are two critical research issues when we immerse into powerful machine learning-based systems or applications. This situation is just like other modern information systems that confront security and privacy issues. The development of blockchain provides us an alternative way to address these two issues. That is why some recent studies have attempted to develop machine learning systems with blockchain technologies or to apply machine learning methods to blockchain systems. To show what the combination of blockchain and machine learning is capable of doing, in this paper, we proposed a parallel framework to find out suitable hyperparameters of deep learning in a blockchain environment by using a metaheuristic algorithm. The proposed framework also takes into account the issue of communication cost, by limiting the number of information exchanges between miners and blockchain.

 

 

 

Paper

J. Polge et al., “Permissioned blockchain frameworks in the industry: A comparison,” vol. 7, no. 2, pp. 229-233, Jun. 2021.

DOI

https://doi.org/10.1016/j.icte.2020.09.002

Abstract

Permissioned and private blockchain platforms are increasingly used in today’s industry. This paper provides a comprehensive and comparative study of the 5 major frameworks (Fabric, Ethereum, Quorum, MultiChain and R3 Corda) with regard to the community activities, performance, scalability, privacy and adoption criteria. Based on a literature review, this study shows that even if Fabric is promising, the final selection of a framework for a specific case-study is always a trade-off. Finally, lessons learnt are given for industrial practitioners and researchers.

 

 

It is indubitable that Information and Communication Technology (ICT) has published many articles on jaw-dropping advancements, such as self-driving cars, artificial intelligence, virtual/augmented reality, and block chain, in modern society. Such success has been achieved by the development of conventional core technologies in ICT as well as related applications that either turned out to be beneficial add-ons or opened new markets. In this issue, we would like to talk about utilizing unmanned aerial vehicles (UAVs) as an add-on to wireless communications and blockchain as a tool for parallel computing.

 

Wu et al. studied UAV-assisted vehicular edge computing networks (VECNs), where UAVs function as base stations (BSs). Owing to the high mobility of UAVs and their increasing resources, such as computing power and battery life, the idea of utilizing UAVs as a temporary BS has received much attention from both, industry and academia. The proposed idea consists of three main components: [C1] traffic flow forecasting, [C2] UAV flight trajectory optimization, and [C3] UAV hovering position optimization. C1 is carried out in the cloud computing layer to predict the traffic conditions so that UAVs can be pre-dispatched to the demanding area. To do so, the light-GBM algorithm is applied to the historical dataset collected in advance. Once the mission areas are chosen, C2 optimizes the UAV flight paths by using a generic algorithm, focusing on minimizing the battery use of the UAV during flight. Finally, the authors proposed the usage of  the Deep Q Network to precisely locate the dispatched UAVs at the position with the highest signal reception at the ground stations in the mission area.

 

On the other hand, Nait-Abdesselam et al. proposed a general framework called D3S, where UAVs can be utilized as a service or UaaS. The proposed D3S framework comprises four phases: demand, decision, deployment, and service; it can be used to implement UaaS to provide end-to-end connectivity to diverse applications. In the demand phase, service requests are placed with a set of high-level parameters. The decision phase then determines the types of UAVs to deploy, their numbers, deployment locations, and bandwidth for communication. The deployment phase defines the flight trajectories of selected UAVs. Finally, in the service phase, a proper coverage service to achieve end-to-end connectivity is provided. In addition, the authors introduce two case studies: UaaS for self-healing and UaaS for dynamic users, to illustrate the potential use of the D3S frameworks in real-world scenarios.

 

In addition to the use of UAVs, blockchain has gained much attention in recent years. It is well known that blockchain applications go far beyond cryptocurrency. Some promising examples are NFT marketplaces, personal identity security, and voting mechanisms. Polge et al. conducted a comparative study of five blockchain frameworks: Hyperledger Fabric, Ethereum, Quorum, MultiChain, and R3 Corda. The paper first identifies features of the five frameworks in terms of what the industry focus is, how to gain consensus, whether smart contract is used or not, whether it is open source or not, who is the supporting/governing organization, and whether it has underlying cryptocurrency or not. Then, the authors compare the five frameworks in terms of open-source community activities, industry use cases, privacy-preserving mechanisms, scalability, throughput, and latency.

 

  One interesting application of blockchain is parallel processing. Tsai et al. proposed a blockchain-based parallel framework for machine/deep learning. The basic idea behind such applications is that by using the blockchain as a shared memory, multiple computing nodes can jointly participate in a computationally heavy task, such as optimization. This is a promising concept because blockchain, by default, provides a secure and private way to exchange messages between participating nodes while guaranteeing immutability. Tsai proposed a parallel DL framework with a blockchain, called PDLKC, which operates as follows: First, the primary node sends the training data along with a set of training tasks (i.e., hyper-parameters) to the blockchain, called the knowledge chain; the computing nodes then receive what is inside the chain and randomly select one of the parameters. Thus, each node trains its own DL model using different parameters. The resulting model and parameters used are sent to the knowledge chain, and then the model with the best result is chosen to create a new task for the upcoming round. Owing to such iterative “training with different parameters and with the best” process, a high-precision DL model can be trained in a distributed and parallel manner. 

 

References 
[1] http://jcn.or.kr (JCN is technically co-sponsored by IEEE Communication Society) 
[2] https://www.journals.elsevier.com/ict-express (JCN is produced and hosted by Elsevier) 
[3] https://eng.kics.or.kr