Earticle

Recently, as data demand has increased owing to the rapidly increasing demand for wireless devices and the influence of data traffic, various technologies are being developed to support it. Among them, millimeter-wave (mmWave) frequencies with rich spectra and high data-transmission rates suffer from the problem of large path loss. Accordingly, there is a growing interest in unmanned aerial vehicles (UAVs) and reconfigurable intelligent surfaces (RISs), which can be utilized advantageously to reconstruct wireless communication environments. Therefore, this work considers a large-scale system comprising a number of users and Flying RISs, combining UAVs and RISs to increase algorithm utilization. We propose a deep neural network-based algorithm that places Flying RISs in an appropriate location so that they can support as many users as possible. Simulation results confirmed that the proposed technique could place Flying RISs in an efficient location with higher accuracy and speed in large-scale systems compared to existing techniques.

With the rapid development of the Internet and the rapid popularity of the network. Increasing network traffic also leads to frequent abnormal attacks, which seriously affects user information security. Then intrusion detection is studied as an active defense technology. In order to increase the ability of network abnormal traffic detection, a network abnormal traffic detection algorithm based on variational autoencoder is proposed in this paper. In addition to normal data, the algorithm subdivides the intrusion attack types into 4 major categories, which contain 38 specific attacks, each specific attack category as an intrusion label, which will be used as the input of the decoder together with the hidden vector to obtain the classification results more accurately and efficiently and to recover the data features in the incomplete data set. Finally, the trained Deep Belief Nets (DBN) is used for further feature extraction and classification of the data. It is verified that the proposed variational autoencoder-based supervised classification scheme is simple and effective, which is able to identify new attacks well and outperforms other classification schemes in terms of data classification.

IoT 서비스는 경량 디바이스 발전으로 그 범위가 계속 확대되고 있으며, 이에 따라 네트워크의 혼잡성은 크게 증가하고 있다. IoT 네트워크 환경에서 CoAP를 위한 기존 혼잡 제어 메커니즘은 동적 네트워크 조건에 대한 적응력이 부족하다. 본 연구에서는 IoT 서비스 환경의 네트워크 환경의 네트워크 혼잡 문제를 해결하기 위해 적응 형 신경망 기반 혼잡 제어(ANN-CoCC) 알고리즘을 제안한다. 정적 매개변수나 휴리스틱 규칙에 의존하는 기존의 혼잡 제어 방식과 달리 ANN-CoCC는 머신러닝 모델을 적용하여 실시간 네트워크 상태에 따라 재전송 시간 초과 (RTO), RTT 분산, 혼잡 윈도우와 같은 주요 전송 매개변수를 동적으로 최적화한다. 다양한 패킷 손실률과 노드 밀도에서 실험 평가를 수행한 결과, 본 연구에서 제안하는 알고리즘은 CoCoA+, BDP-CoAP, pCoCoA과 같은 기존 방식과 비교했을 때 처리량을 30% 이상 향상시키며, 지연 시간 감소, 패킷 손실을 최소화하는 것으로 확인되 었다.

The scope of IoT services continues to expand with the development of lightweight devices, and network congestion is increasing significantly. In this study, we propose an adaptive neural network-based congestion control (ANN-CoCC) algorithm to solve the network congestion problem in the network environment of IoT service environments. Unlike existing congestion control methods that rely on static parameters or heuristic rules, ANN-CoCC applies a machine learning model to dynamically optimize key transmission parameters such as retransmission timeout (RTO), RTT distribution, and congestion window according to real-time network conditions. The experimental evaluation results under various packet loss rates and node densities show that the proposed algorithm improves throughput by more than 30%, reduces delay, and minimizes packet loss compared to existing methods such as CoCoA+, BDP-CoAP, and pCoCoA.

As a new entertainment and social way, online games now have a huge and increasing user group, so it is of great significance to identify the data stream of online games. Using the excellent nonlinear fitting ability of BP neural network and the advantages of global search of genetic algorithm, the initial weights and thresholds of BP neural network are optimized, and the BP neural network model optimized by genetic algorithm is established. The muti-dimensional input information is proposed to identify online game data streams. Through the experimental simulation, it shows that the selected muti-dimensional information and the established model can be well applied to online game stream recognition.

오늘날에는 기후, 전파천문, 고성능 컴퓨팅 자원 공동 활용 등의 첨단 기술을 요구하는 그룹별 연구 활동이 강화되고 있다. 이에 따라 이와 같은 오늘날의 첨단 응용 연구를 뒷받침하기 위해 첨단 연구망 역시 첨단 응용연구를 위한 수단 으로서 점차적으로 자리매김하고 있다. 따라서 이러한 추세를 반영해서 첨단 연구망 네트워크의 관리적 측면 역시 집중 적으로 조명되어야 할 필요성이 있다. 그렇기 때문에, 본 논문에서는 제시하고자 하는 가장 큰 맥락에서의 목표가 ‘여러 첨단 응용 연구의 도구적 수단으로서의 네트워크를 안정적으로 관리할 수 있도록 하는 연구’이다.

Today, group research efforts are increasing for cutting edge technologies related to climate, radio astronomy, co -utilization of high performance computing resources and more. Accordingly, networks for cutting-edge researches a re also regarded as means for cutting-edge researches nowadays. Therefore, it is necessary for administrating cutti ng-edge network to be considered, reflecting such a current trend. In this sense, 'Research which can help to admin istrate the networks which are usually used as means for a lot researches' was set as a goal of this paper.

최근 무선 센서 네트워크에서 센서 노드의 에너지 효율성을 향상시켜 네트워크의 수명을 향상시키는 라우팅기법의 연구가 많이 수행되고 있다. 무선 센서 네트워크는 제한된 에너지 및 하드웨어 성능 때문에 배터리, 메모리 및 프로세서 등 모든 자원을 효율적으로 사용해야 하며, 이를 위해 관련 통신 프로토콜도 저-전력 소모를 달성할 수 있도록 설계하고 구현해야 한다. 센서 네트워크에 많이 사용되는 LEACH, HEED등과 같은 클러스터 기반의 라우팅 방식들은 클러스터 헤더들과 해당 멤버 노드들 간의 데이터 전달을 통하여 에너지의 효율성을 얻는다. 이 논문에서는 에너지를 효율적으로 사용할 수 있는 클러스터 라우팅 알고리즘을 제안하고 시뮬레이션을 이용하여 성능 결과를 제시한다.

Recently, a lot of researches have been done to increase the life span of network using the energy efficient sensor node in WSN. In the WSN environment, we must use limited amount of energy and hardware. Ther efore, it is necessary to design energy efficient communication protocol and use limited resources. Cluster b ased routing method such as LEACH and HEED get the energy efficient routing using data communication between cluster head and related member nodes. In this paper, we propose an energy efficient routing algori thm as well as performance result using simulation.

This study proposes a novel recommendation algorithm that reflects the results from trust/distrust network analysis as a solution to enhance prediction accuracy of recommender systems. The recommendation algorithm of our study is based on memory-based collaborative filtering (CF), which is the most popular recommendation algorithm. But, unlike conventional CF, our proposed algorithm considers not only the correlation of the rating patterns between users, but also the results from trust/distrust relationship network analysis (e.g. who are the most trusted/distrusted users?, whom are the target user trust or distrust?) when calculating the similarity between users. To validate the performance of the proposed algorithm, we applied it to a real-world dataset that contained the trust/distrust relationships among users as well as their numeric ratings on movies. As a result, we found that the proposed algorithm outperformed the conventional CF with statistical significance. Also, we found that distrust relationship was more important than trust relationship in measuring similarities between users. This implies that we need to be more careful about negative relationship rather than positive one when tracking and managing social relationships among users.

본 논문에서는 기존에 전문가에 의해서 이루어지던 국가 대기오염 측정망 데이터들의 이상 탐지 작업을 인공지능을 통해 자동화하고자 심층 신경망을 이용한 이상 탐지 모델을 제안하였다. 환경과학원에서 제공받은 기상자료 데이터의 결측치 및 이상치를 분석하여 학습데이터를 생성하였으며 비지도 학습 방식의 BeatGAN 모델에 기반하여 커널 구조 변경과 합성곱 필터층 및 전치 합성곱 필터층의 추가를 통해 새로운 모델을 제안하여 이상 탐지 성능을 높이고자 하였다. 또한 제안하는 모델의 생성적 특징을 활용하여 새로운 데이터를 생성하고 이를 학습에 사용하는 재학습 알고리즘을 구현 및 적용하여 기존 BeatGAN 모델뿐 아니라 다른 비지도 학습 모델인 Iforest, One Class SVM과 비교하였을 때 제안모델의 성능이 가장 높았음을 확인할 수 있었다. 본 연구를 통해 실제 산업현장에서 센서의 이상, 점검 등의 여러 요인으로 인해 학습 데이터가 부족한 상황에서 추가 적인 비용없이 과적합을 피하며 제안하는 모델의 이상탐지 성능을 올릴 수 있는 방법을 제시할 수 있었다.

In this paper, We propose an anomaly detection model using deep neural network to automate the identification of outliers of the national air pollution measurement network data that is previously performed by experts. We generated training data by analyzing missing values and outliers of weather data provided by the Institute of Environmental Research and based on the BeatGAN model of the unsupervised learning method, we propose a new model by changing the kernel structure, adding the convolutional filter layer and the transposed convolutional filter layer to improve anomaly detection performance. In addition, by utilizing the generative features of the proposed model to implement and apply a retraining algorithm that generates new data and uses it for training, it was confirmed that the proposed model had the highest performance compared to the original BeatGAN models and other unsupervised learning model like Iforest and One Class SVM. Through this study, it was possible to suggest a method to improve the anomaly detection performance of proposed model while avoiding overfitting without additional cost in situations where training data are insufficient due to various factors such as sensor abnormalities and inspections in actual industrial sites.

Recently, a multi facility, multi product and multi period industrial problem has been widely investigated in Supply Chain Network(SCN). One of keys issues in the current SCN research area involves minimizing both production and distribution costs. This study deals with finding an optimal solution for minimizing the total cost of production and distribution problems in supply chain network. First, we presented an integrated mathematical model that satisfies the minimum cost in the supply chain. To solve the presented mathematical model, we used a genetic algorithm with an excellent searching ability for complicated solution space. To represent the given model effectively, the matrix based real-number coding schema is used. The difference rate of the objective function value for the termination condition is applied. Computational experimental results show that the real size problems we encountered can be solved within a reasonable time.

인터넷의 경직성으로 인하여 다양한 서비스를 제공하기 어려운 문제를 해결하기 위한 가장 효과적인 방법 이 네트워크 가상화 방안이다. 공통의 물리 네트워크 자원을 공유하는 가상 네트워크를 효율적으로 생성하기 위해서 는 자원을 효율적으로 사용할 수 있는 방안이 필요하다. 이러한 한정된 물리 자원을 효율적으로 사용하기 위하여 본 논문에서는 사용중인 자원이라도 더 높은 서비스 우선순위를 갖는 가상 네트워크에 할당하는 방법을 제안한다. 높은 우선순위의 네트워크 생성 요청을 위하여 운용중인 낮은 우선순위의 가상 네트워크에서 효율적으로 대역폭을 회수하 는 알고리즘을 제안한다. 제안하는 대역폭 할당 알고리즘은 낮은 우선순위를 갖는 가상 네트워크에서 서비스 지속성 을 유지하면서 할당된 대역폭을 회수할 수 있다. 제안한 알고리즘의 실험을 통하여 약 8%정도의 가상 네트워크 생 성 수락률의 향상을 확인하였다.

Due to the ossification of the Internet, it is difficult to accommodate variety services. One of the efficient solution to this problem is network virtualization. It allows multiple parallel virtual networks to run on the shared physical infrastructure. It needs new resource allocation mechanism to share efficient physical resources. In this paper, we present efficient bandwidth allocation algorithm for virtual network request with high service priority. Our proposed algorithm can withdraw allocated bandwidth from low-level priority virtual network and maintain low-level virtual network service. We evaluated the performance of our proposed algorithm using simulation and found the improvement of approximately 8% acceptance rate.

북미시장에서 국내 가전업체의 브랜드 및 제품 인지도는 크게 성장했으며 북미 소비자들에게 국내 기업의 제품은 성능이 좋고 혁신적인 기술 제품으로 인식되고 있다. 또한 에너지 절약을 원하는 소비자가 늘어나면서 국내 가전제품의 에너지 절약 측면에서 우수성이 부각됨에 따라 시장점유율이 상승으로 이어지고 있다. 최근 스마트폰과 모바일 기기 시장 확대 및 스마트 그리드 기술 발달의 영향으로 가전제품 시장에도 스마트 열풍이 거세게 몰아치고 있는데, 국내 기업들은 가전제품과 결합된 다양한 부가기능을 통해 소비자 편의를 제공함에 따라 지속적인 제품개발 을 하고 있다. 본 연구에서는 지속적인 경쟁우위를 유지하기 위한 방안으로 국내 A사의 북미시장에서의 TV 판매 데 이터를 이용하여 북미시장에서의 가전제품 판매예측을 위한 융합모델을 개발하고자 한다. 본 연구에서는 인공신경망 과 유전자 알고리즘 기반의 융합모델을 이용한 가전제품의 판매예측을 수행하기로 한다. 추가적으로 본 연구에서는 제안한 융합모델과 기존의 예측모델과의 비교분석을 통해 제안한 융합모델의 우수성을 입증하기로 한다.

The brand and product awareness of Korean electronics companies in the North American market has grown significantly and North American consumers has been recognized as an innovative technology products good performance of Korean electronics appliances. The consumer need of energy saving has led to a rise in market share because Korean electronics appliances have the excellence in energy saving aspects. The expansion of smartphones and mobile devices and the development of smart grid technology can affect electronics market. Domestic companies are continuously develop new product to provide consumers convenient with a variety of additional features combined consumer products. This study proposes a convergence model for sales prediction of electronic appliances using sales data of A company from the North American market. We develop the convergence model for sales prediction based on based on artificial neural network and genetic algorithm. In addition, we validate the superiority of the proposed convergence model by comparing the prediction performance of traditional prediction models.

Despite their great adoption in e-commerce sites, recommender systems are still vulnerable to unscrupulous producers who try to promote their products by shilling the systems. In the past decade, network based recommendation approaches have been demonstrated to be both more efficient and of lower computational complexity than collaborative filtering methods, however as far as we know, there is rare research on the robustness of network based recommendation approaches. In this paper, we conducted a serious of experiments to examine the robustness of five typical network based recommendation algorithms. The empirical results obtained from the movielens dataset show that all the two limited knowledge shilling attacks are successful against the network based algorithms, and the bandwagon attack affects very strongly against most network based recommendation algorithms, especially the algorithms considering the preferential diffusion at the last step. One way to relieve the attack impact is to assign the algorithm a heterogeneous initial resource configuration.