• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.3
• /
• pp.549-559
• /
• 2023

As mobile edge computing (MEC) is gaining attention as a core technology of 5G networks, edge AI technology of 5G network environment based on mobile user data is recently being used in various fields. However, as in traditional AI security, there is a possibility of adversarial interference of standard 5G network functions within the core network responsible for edge AI core functions. In addition, research on data poisoning attacks that can occur in the MEC environment of standalone mode defined in 5G standards by 3GPP is currently insufficient compared to existing LTE networks. In this study, we explore the threat model for the MEC environment using NWDAF, a network function that is responsible for the core function of edge AI in 5G, and propose a feature selection method to improve the performance of detecting data poisoning attacks for Leaf NWDAF as some proof of concept. Through the proposed methodology, we achieved a maximum detection rate of 94.9% for Slowloris attack-based data poisoning attacks in NWDAF.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.3043-3067
• /
• 2022

With the mobile traffic in the network increases exponentially, multi-access edge computing (MEC) develops rapidly. MEC servers are deployed geo-distribution, which serve many mobile terminals locally to improve users' QoE (Quality of Experience). When the cache space of a MEC server is full, how to replace the cached videos is an important problem. The problem is also called the cache replacement problem, which becomes more complex due to the dynamic video popularity and the varied video sizes. Therefore, we proposed a new cache replacement scheme based on local video popularity and video size to solve the cache replacement problem of MEC servers. First, we built a local video popularity model, which is composed of a popularity rise model and a popularity attenuation model. Furthermore, the popularity attenuation model incorporates a frequency-dependent attenuation model and a frequency-independent attenuation model. Second, we formulated a utility based on local video popularity and video size. Moreover, the weights of local video popularity and video size were quantitatively analyzed by using the information entropy. Finally, we conducted extensive simulation experiments based on the proposed scheme and some compared schemes. The simulation results showed that our proposed scheme performs better than the compared schemes in terms of hit rate, average delay, and server load under different network configurations.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.1-6
• /
• 2021

For competitive hotel business, the hotel must have a sound prediction capability to balance the demand and supply of hospitality products. To have a sound prediction capability in the hotel, it should be prepared to be equipped with a new technology such as quantum computing. The quantum computing is a brand new cutting-edge technology. It will change hotel business and even the whole world too. Therefore, we study the impact of quantum computing on supply chain management (SCM) and hotel performance. Toward the goal we have developed the research model including six constructs: quantum (computing) prediction, communication, supplier relationship, service quality, non-financial performance, and financial performance. The result of the study shows a significant influence of quantum (computing) prediction on hotel performance through the mediating role of SCM in the hotel. Quantum prediction is highly significant in enhancing the SCM in the hotel. However, the direct effect between the quantum prediction and hotel performance is not significant. The finding indicates that hotels which would install the quantum computing technology and utilize the quantum prediction could hugely benefit from the performance improvement.

• The Journal of Asian Finance, Economics and Business
• /
• v.6 no.1
• /
• pp.217-229
• /
• 2019

The purpose of the research is to verify how task characteristics for business and technology characteristics, economic feasibility, technology readiness, organizational factors, environmental factors of cloud computing affect the performance of cloud computing adoption through Fit and Viability. The research aims to verify the relationship among the success factors for adopting cloud computing based on the Fit-Viability model. Respondents who work for IT companies which is using cloud computing in South Korea were chosen. The data was analyzed by the structural equating model. As a result, Task characteristics and Technology characteristics affected Fit in a positive manner, while Technology readiness, Organizational factors and Environmental factors also positively impacted Viability. Fit and Viability both affected the successful adoption of cloud equally. In particular, Environmental factors were proven to have the biggest impacts on Viability, and affected highly indirect impact on the Performance of cloud computing adoption through Viability. Entering the era of the fourth industrial revolution, corporations have established digital transformation strategies to secure a competitive edge while growing continuously, and are also carrying out various digital transformation initiatives. For the success of adoption of foundational technologies, they need to understand not only the decision-making factors of adopting cloud computing, but also the success factors of adopting cloud computing.

• Journal of information and communication convergence engineering
• /
• v.21 no.4
• /
• pp.329-336
• /
• 2023

This paper introduces an edge AI-based scene-specific object detection system for long-term traffic management, focusing on analyzing congestion and movement via cameras. It aims to balance fast processing and accuracy in traffic flow data analysis using edge computing. We adapt the YOLOv5 model, with four heads, to a scene-specific model that utilizes the fixed camera's scene-specific properties. This model selectively detects objects based on scale by blocking nodes, ensuring only objects of certain sizes are identified. A decision module then selects the most suitable object detector for each scene, enhancing inference speed without significant accuracy loss, as demonstrated in our experiments.

• Convergence Security Journal
• /
• v.21 no.3
• /
• pp.25-30
• /
• 2021

Due to the current COVID-19 pandemic, companies and institutions are introducing virtual desktop technology, one of the logical network separation technologies, to establish a safe working environment in a situation where remote work is provided. With the introduction of virtual desktop technology, companies and institutions can operate the network separation environment more safely and effectively, and can access the business network quickly and safely to increase work efficiency and productivity. However, when introducing virtual desktop technology, there is a cost problem of high-spec server, storage, and license, and it is necessary to supplement in terms of operation and management. As a countermeasure to this, companies and institutions are shifting to cloud computing-based technology, virtual desktop service (DaaS, Desktop as a Service). However, in the virtual desktop service, which is a cloud computing-based technology, the shared responsibility model is responsible for user access control and data security. In this paper, based on the shared responsibility model in the virtual desktop service environment, we propose a cloud-based hardware security module (Cloud HSM) and edge-DRM proxy as an improvement method for user access control and data security.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.141-151
• /
• 2022

In recent years, the industry and research communities have focused on developing autonomous crack inspection approaches, which mainly include image acquisition and crack detection. In these approaches, mobile devices such as cameras, drones or smartphones are utilized as sensing platforms to acquire structural images, and the deep learning (DL)-based methods are being developed as important crack detection approaches. However, the process of image acquisition and collection is time-consuming, which delays the inspection. Also, the present mobile devices such as smartphones can be not only a sensing platform but also a computing platform that can be embedded with deep neural networks (DNNs) to conduct on-site crack detection. Due to the limited computing resources of mobile devices, the size of the DNNs should be reduced to improve the computational efficiency. In this study, an architecture called pruned crack recognition network (PCR-Net) was developed for the detection of structural cracks. A dataset containing 11000 images was established based on the raw images from bridge inspections. A pruning method was introduced to reduce the size of the base architecture for the optimization of the model size. Comparative studies were conducted with image processing techniques (IPTs) and other DNNs for the evaluation of the performance of the proposed PCR-Net. Furthermore, a modularly designed framework that integrated the PCR-Net was developed to realize a DL-based crack detection application for smartphones. Finally, on-site crack detection experiments were carried out to validate the performance of the developed system of smartphone-based detection of structural cracks.

• Journal of Internet Computing and Services
• /
• v.17 no.3
• /
• pp.25-32
• /
• 2016

In this paper, we propose a method that is tracking an object in real time using particle filter and the observation model with edge. First of all, the proposed method defines the object to be tracked in the initial frame. Then, it generates the edge observation model for the object to be tracked and a set of particles. It calculates the weight by comparing the average of the middle distance in eight-way of particle filter edge model with that in edge observation model, and then updates the weight with the calculated value. After resampling particles using the updated weights, it estimates the current location of the tracked object. Finally, this paper demonstrates the performance of the stable tracking through comparison with the existing method by using a number of experimental data.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.5
• /
• pp.1009-1019
• /
• 2017

Mobile Edge Computing(MEC) is a emerging technology to cope with mobile traffic explosion and to provide a variety of services having specific requirements by means of running some functions at mobile edge nodes directly. For instance, caching function can be executed in order to offload mobile traffics, and safety services using real time video analytics can be delivered to users. So far, a myriad of methods and architectures for personalized service recommendation have been proposed, but there is no study on the subject which takes unique characteristics of mobile edge computing into account. To provide personalized services, acquiring users' context is of great significance. If the conventional personalized service model, which is server-side oriented, is applied to the mobile edge computing scheme, it may cause context isolation and privacy issues more severely. There are some advantages at mobile edge node with respect to context acquisition. Another notable characteristic at MEC scheme is that interaction between users and applications is very dynamic due to temporal relation. This paper proposes the local service recommendation platform architecture which encompasses these characteristics, and also discusses the personalized service recommendation mechanism to be able to mitigate context isolation problem and privacy issues.

• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.256-259
• /
• 2022

BaaS(Blockchain as a Service)는 블록체인의 사용이 어렵다는 단점을 유연한 자원운용이 가능하고 뛰어난 접근성의 특징을 가진 클라우드와 접목하여 쉽게 블록체인을 구축하고 사용할 수 있도록 해주는 클라우드 서비스이다. BaaS 의 등장으로 블록체인의 접근성은 큰 범위로 증가하였으며 다양한 도메인에 활용되고 있다. 하지만 클라우드 기반 서비스이기 때문에 클라우드 서비스의 문제점인 보안 이슈가 제기되었다. 본 논문에서는 BaaS 에 ZKP(Zero-Knowledge Proof)와 엣지 컴퓨팅 기술을 활용하여 보안성을 제공할 수 있는 새로운 BaaS 모델인 EBaaS 를 제안한다. EBaaS 는 엣지 컴퓨팅 기술을 적용하여 클라우드 서비스 공급업체에 대한 데이터 종속성을 제거하고 블록체인의 고가용성을 제공할 수 있으며 ZKP 를 활용하여 내부적으로 민감한 데이터에 대한 보안성도 제공할 수 있다.