Earticle

본 논문은 자율주행차량 환경에서 V2X(Vehicle-to-Everything) 통신의 성능을 최적화하고, 높은 이동성과 네트워크 불안정성이 특징인 VANET(Vehicular Ad Hoc Network) 환경에서 효과적인 지능형 교통 시스템(ITS)을 구현하기 위한 SDVN(Software-Defined Vehicular Network) 기반 설계 방법을 제안한다. 기존의 VANET은 분산형 구조로 인해 패킷 손실, 지연 증가, 네트워크 단절 등 여러 한계가 존재하며, 이를 해결하기 위해 본 연구는 SDN 아키텍처를 차량 통신망에 통합하였다. SDN의 중앙집중 제어 특성을 활용하여 네트워크 토폴로지를 실시간으로 관리하고, 트래픽 상황에 따라 최적의 경로를 동적으로 설정할 수 있도록 하였다. 시뮬레이션은 VEINS, SUMO, OMNeT++, Mininet-WiFi, Ryu 컨트롤러를 통합하여 구축되었으며, 다양한 차량 밀도와 도시 교차로 기반 시나리오에서 VANET 대비 SDVN의 성능을 정량적으로 비교하였다. 실험 결과, SDVN 기반 ITS는 평균 지연 시간에서 최대 25% 감소, 패킷 전달률(PDR) 15% 증가, 네트워크 처리량(Throughput) 약 30% 향상을 나타냈으며, 중앙제어로 인한 제어 메시지 오버헤드는 소폭 증가하였다. 본 연구는 향후 자율주행 기술의 확산과 더불어 차량 통신 인프라의 고도화를 위한 기반 기술로 활용될 수 있을 것이다.

This paper presents an in-depth optimization strategy for Vehicle-to-Everything (V2X) communication tailored for autonomous vehicle networks and proposes a robust Intelligent Transportation System (ITS) architecture built upon Software-Defined Vehicular Networking (SDVN) within highly dynamic VANET environments. While traditional Vehicular Ad Hoc Networks (VANETs) offer decentralized communication among mobile nodes, they suffer from inherent limitations such as frequent topology changes, high latency, and unreliable data delivery. The proposed architecture integrates real-time mobility simulation (SUMO), discrete event network modeling (OMNeT++), vehicular communication (VEINS), wireless emulation (Mininet-WiFi), and programmable flow control via the Ryu SDN controller. Various urban traffic scenarios with differing vehicle densities (low, medium, high) were simulated, and the performance of the SDVN-based ITS was benchmarked against a conventional VANET model. The SDVN model demonstrated a reduction of up to 25% in latency, a 15% improvement in PDR, and up to 30% higher throughput, while incurring only a moderate increase in control overhead due to SDN signaling. These findings validate the feasibility and efficiency of SDVN-enhanced ITS frameworks for next-generation smart transportation infrastructures and offer valuable insights for large-scale deployment in future autonomous mobility ecosystems.

One of the latest emerging technologies is artificial intelligence, which makes the machine mimic human behaviour. The most important component used to detect cyber attacks or malicious activities is the intrusion detection system (IDS). Artificial intelligence plays a vital role in detecting intrusions and widely considered as the better way in adapting and building IDS. In modern days, neural network algorithms are emerging as a new artificial intelligence technique that can be applied to real-time problems. The proposed system is to detect a classification of botnet attack which poses a serious threat to financial sectors and banking services. The proposed system is created by applying artificial intelligence on a realistic cyber defence dataset (CSE-CIC-IDS2018), the latest IDS Dataset in 2018 by Canadian Institute for Cybersecurity (CIC) on AWS (Amazon Web Services). The proposed system of Artificial Neural Networks provides an outstanding performance of Accuracy score is 99.97% and an average area under ROC(Receiver Operator Characteristic) curve is 0.999 and an average False Positive rate is a mere value of 0.03. The proposed system of Artificial Intelligence-based Intrusion detection of botnet attack classification is powerful, more accurate and precise. The novel proposed system can be applied to conventional network traffic analysis, cyber-physical system traffic analysis and also can be applied to the real-time network traffic data analysis.

This paper investigates the Contention Window (CW) optimization problem in multi-agent scenarios, where the fully cooperative among mobile stations is considered. A partially observable environment is employed to model and analyze the CW optimization problem, and Smart Exponential-Threshold-Linear with Deep Q-learning Network (SETL-DQN) Multi-Agent (MA) algorithm is proposed to obtain the optimal system throughput through the CW Threshold optimization. In the determined scenarios, SETL-DQN(MA) can effectively cope with the mutual interaction among mobile stations. The simulation results show that our proposed method is superior from both static and dynamic scenarios and has the highest optimum packet transmission efficiency.

Intrusion detection acts as a vital function in providing information security, and additionally the key technology is to precisely classify diverse attacks. Intrusion detection system (IDS) is identified as an important security issue within the cloud network environment. In this paper, IDS is given based on an innovative optimized custom RC-NN (Recurrent Convolutional Neural Network) which is proposed for intrusion detection along with the Ant Lion optimization algorithm. By this method, CNN (Convolutional Neural Network) is made hybrid with LSTM (Long Short Term Memory). Thus, all the attacks identified with the network layer of cloud are classified efficiently. The experimental results shown below describe the presentation of the IDS classification model with high accuracy, thus improving the detection rate or error rate. The optimized custom RC-NN-IDS model thus achieved an improved classification accuracy of 94% and also a decreased error rate of 0.0012. Additionally true positive rate, true negative rate and precision are considered as performance metrics. The proposed approach is evaluated using the DARPA IDS evaluation Data Sets and CSE-CIC-IDS2018 dataset and is compared with some existing approaches.

This paper investigates QoS-aware computation offloading issues for mobile edge computing in the 6G network. To minimize the end-to-end delay, we harness the Information-Centric Network (ICN) to ensure resource-constrained mobile user offloading computation-sensitive tasks in a distributed manner. Then, a two-stage approach based on a Multi-Agent Reinforcement Learning (MARL) algorithm entwined with optimization-embedding offloading ratio is proposed to enhance server selection for load balancing. Numeral results demonstrate that, with reference to a workshop-scale scenario, the proposed method can achieve outperformed performance in reducing delay and balancing loads on edge servers than the other four baseline schemes.

신체영역무선통신망(WBAN)에서 클러스터 헤드(CH) 선출 및 최적경로의 라우팅은 에너지 효율 향상과 네트워크 노드 운영수명 연장을 위해 해결해야 할 이슈이다. 이러한 연구를 위해 본 논문에서는 BKOA알고리즘과 그리드 기반 멀티홉 라우팅 프레임워크를 결합한 하이브리드 BKOA-GRID를 제안한다. 시뮬레이션 수행결과 제안된 BKOA-GRID는 PSO, LEACH, EEUC 등 기존 알고리즘보다 노드생존율 90%, 잔류에너지 지속성은 총 에너지의 약 60%를 유지하여 높은 에너지 효율을 보였다.

Cluster head(CH) election and optimal path routing in a Wireless Body Area Network(WBAN) are issues that must be addressed to improve energy efficiency and extend the operating life of network nodes. To address these issues, this paper proposes a hybrid BKOA-GRID (Black Kite Optimization Algorithm-GRID) framework, which integrates the Black Kite Optimization Algorithm with a grid-based multi-hop routing structure. Simulation results demonstrate that the proposed BKOA-GRID exhibits superior energy efficiency compared to existing algorithms such as PSO, LEACH, and EEUC, maintaining a node survival rate of 90% and preserving approximately 60% of the total residual energy.

Confusion network (CN) is an aligned and compact representation form of word lattice that stores median results of decoding procedure and connects the acoustic decoding and succeeding process steps. The quality of confusion network is very important for multi‐pass decoding based recognition method. Three new quality evaluation measures are proposed for confusion network in this paper. The first is called confusion network word error rate (CWER), which gives a lower bound of recognition word error rate. The second is used to measure the size of confusion set, called confusion word density (CWD). The third is word confusion probability (WCP), which can measure average distinguishability between words in a confusion set. Based on the proposed measures, we present a new method of quality optimization for confusion network, called as confusion probability based pruning algorithm. The experiments, carried out on a large vocabulary Chinese continuous speech recognition system, demonstrate that the new method leads to a significant reduction in CWD and WCP without an increase of CWER.

본 연구는 폐쇄루프 공급망을 최적화하기 위한 녹색 폐쇄루프 공급망 네트워크(GCSN)를 제안한다. GCSN은 물류 유통과정에 투입되는 익스프레스 박스의 재활용에 중점을 두고 연구하였다. 이를 통해 물류공급망 작업의 표준화 와 전자상거래 업체의 증가에 따른 익스프레스 박스의 사용의 가속화와 전자상거래 업계의 익스프레스 상자의 재활용 을 통한 자원 낭비를 최소화하는 녹색공급망을 추진한다. GCSN은 유통흐름과 자원의 투입요소는 수학식으로 표현하 며 LINGO를 사용하여 구현된다. 녹생 즉, 환경 보호의 관점에서 전자상거래 산업의 지속가능한 발전을 견인하는 유통 의 과정이다. 또한 자원 부족에 따른 원자재 가격 상승에 대응하고 전자상거래 업체들의 친환경적으로 익스프레스 상 자를 투입하는 전략의 필요성을 부각시켰다. 이와 같은 이유로 본 연구는 전자상거래 업체에서 사용하는 재활용이 가 능한 익스프레스상자를 연구대상으로 하며, 공급망 네트워크에서 시설 위치와 할당문제도 통합적으로 고찰한다. 마지 막으로 본 연구의 결론과 시사점을 도출하였다.

This paper proposes a green closed-loop supply chain network (GCSN) for optimizing closed-loop supply chains. The GCSN focuses on the application of the recycling express box in logistics circulation, accelerates the standardization of logistics operations and the use of express packaging in e-commerce companies, and promotes the reduction and greening of recycling express box in the e-commerce industry. The GCSN is represented as a mathematical formulation and implemented using LINGO. Greening, environmental protection, and wisdom are the general trends for promoting the growth of the e-commerce industry. Meanwhile, the price of raw materials has increased owing to a shortage of resources, which emphasizes the need for e-commerce enterprises to develop green packaging. Therefore, this study considers the shared circular packaging launched by e-commerce enterprises as the research object, and integrates the problem of facility positioning and path planning in the logistics system. The conclusion summarizes the significance of this study.

Sequential decision-making in dynamic, heterogeneous environments is often hindered by multivariate and correlated outcomes. This study introduces a unified Copula-CNN-LSTM Deep Q-Network (DQN) framework for multi-stage individualized policy learning in pseudo temporal settings. Motivated by the need for agents that account for inter-outcome dependencies, we extend static covariates from benchmark datasets (Boston Housing and Wine Quality Red) into pseudo-temporal sequences to emulate state transitions. Multivariate rewards with controlled correlations (ρ = 0.5 and ρ = −0.5) are standardized via an empirical copula transformation to assess policy robustness under varying dependency structures. The DQN agent optimizes policies using experience replay and temporal discounting of state-action reward trajectories. The framework demonstrates stable convergence in average rewards across both datasets under positive and negative correlation structures. Analysis of the resulting dynamic conditional average treatment effects (CATEs) across outcome dimensions highlights the model’s ability to discern heterogeneous treatment impacts. Furthermore, learned policy matrices and dynamic Directed Acyclic Graphs (DAGs) reveal interpretable temporal dependencies, with edge structures reflecting the complex multivariate nature of the optimal policy. Overall, the proposed framework effectively captures inter-temporal dependencies and adapts to correlated rewards, providing a scalable and interpretable solution for sequential decision making in complex environments.

SAGD공법(증기배유공법)은 중유나 비튜멘의 생산에 효과적이지만 증기 생성을 위해 다량의 담수와 천연가스를 사용하는 한계를 가지고 있다. 천연가스 사용이 생산 비용의 큰 부분을 차지하므로 적절한 압력의 증기주입은 SAGD공법의 경제성에 매우 중요하다. 이 연구에서는 SAGD공법의 에너지 효율을 향상시키고 다수의 증기 주입압력 시나리오의 신속한 평가를 위하여 인공신경망을 이용한 증기주입압력 최적화 모델을 개발하였다. 개발한 인공신경망 모델로 SAGD공법의 생산기간별 증기주입압력에 따른 누적 증기-오일비와 ES-SAGD공법의 생산이윤을 예측하였다. 개발한 모델을 대상 저류층에 적용한 결과, 저류층 시뮬레이션 결과와 개발 모델의 예측값이 상관계수 0.9 이상, 예측 오차 5.5% 이하로 인공신경망 모델의 예측 성능을 확인하였다. 생산기간동안 증기주입압력을 조절한 SAGD공법의 최적 시나리오는 일정 압력으로 증기를 주입하는 기존의 경우에 비해 26% 낮아진 누적 증기-오일비를 나타내었고, ES-SAGD도 34% 향상된 생산이윤을 얻었다.

SAGD has been widely used to extract oil from heavy oil and bitumen reservoir as in-situ recoverable technology. Effective steam injection was important because SAGD depended on the combustion of natural gas to generate steam of which cost was dominant. Numerical approach based on ANN (Artificial Neural Network) scheme was developed to obtain an optimized schedule of variable injection pressure in SAGD. ANN model predicts cSOR in SAGD while it does a production benefit in ES-SAGD. The model was verified by showing accurate prediction error, less than 5.5% compared with typical reservoir simulator. Optimized schedule decreased 26% of cSOR in SAGD while improved an economical value up to 34% in ES-SAGD rather than that of conventional approach, constant pressure scheduling.

미래 전장의 전쟁수행 역량은 AICBMS(AI, Cloud, Bigdata, Mobile, Security)라 일컫는 4차 산업혁명의 차세대 기술을 적 용하여 혁신적인 국방력을 확보할 수 있는가에 달려 있다 해도 과언이 아니다. 또한, 미래의 군 작전환경은 네트워크를 기반 으로 모든 무기체계가 하나의 통합된 정보통신망 내에서 실시간으로 전장정보를 상호공유하며 작전을 수행하게 되는 네트워 크 중심전(NCW)으로 급변하고, 유·무인 복합전투체계 운용범위로 확대되고 있다. 특히, 초고속, 초연결성을 책임지는 통신 네 트워크는 여러 전투 요소를 연결하고 정보의 원활한 유통을 위해 높은 생존성과 다계층(국방 모바일, 위성, M/W, 유선) 네트 워크 기반의 전력 운용의 효율성을 요구한다. 이러한 관점에서 본 연구는 제원이 고정된 기존의 단일매체, 단일경로 전송과는 달리, 가용한 통신 유무선 인프라 다매체를 동시 사용하여 통신량 폭주시 부하분산과 RNN(Recurrent Neural Networks) 알고 리즘을 이용한 인공지능 기반의 전송기술로 다매체다중경로(MMMP-Multi-Media Multi-Path) 적응적 네트워크 기술 개발하는 것이다.

The performance capability of the future battlefield depends on whether the next-generation technology of the Fourth Industrial Revolution, called ABCMS (AI, Bigdata, Cloud, Mobile, Security), can be applied to secure innovative defense ca pabilities It is no exaggeration to say. In addition, the future military operation environment is rapidly changing into a net work-oriented war (NCW) in which all weapon systems mutually share battlefield information and operate in real-time wi thin a single integrated information and communication network based on the network and is expanding to the scope of o peration of the manned and unmanned complex combat system. In particular, communication networks responsible for hig h-speed and hyperconnectivity require high viability and efficiency in power operation based on multi-tier (defense mobile, satellite, M/W, wired) networks for the connection of multiple combat elements and smooth distribution of information. Fro m this point of view, this study is different from conventional single-media, single-path transmission with fixed specificati ons, It is an artificial intelligence-based transmission technology using RNN (Recurrent Neural Networks) algorithm and lo ad distribution during traffic congestion using available communication wired and wireless infrastructure multimedia simult aneously and It is the development of MMMP-Multi-Media Multi-Path adaptive network technology.

Researches and developments on BEMS, which are the cutting edge technology for energy saving in buildings, are performed world-widely through sustainable management in various conditions. However, there are many obstacles to adapt the system in existing buildings because it needs not only the extra administrator who manages the energy, but also needs highly expensive equipments, which are designed for newly built buildings, to install. Therefore, there are numerous limits exist when applying the BEMS in established buildings. The purpose of this study is to evaluate the possibility of energy saving according to user's participation through related case study and to propose PEMS(Portable Environment Monitoring System) that could overcome the weakness of previous system. In addition, this study will assess the optimization of WSN(Wireless Sensor Networks), which is the essential technology to develop PEMS. Lastly, this study has performed field measurements as well as basic tests to evaluate the adaptability of PEMS in buildings. The result of this study is as follows ; 1) Energy saving of 10~20% in buildings can be achieved through energy feedback. 2) As the RF output power decreases from 0dbm to -25dbm, maximum reception distance of sensor node was reduced as well. 3) Maximum reception distance can be different depending on the RF output power differences and different types of materials. 4) As RF output power increases, the number of relay node reduced, therefore, the WSN showed more stability.

LiDAR는 자율 주행뿐만 아니라 다양한 산업 현장에 적용되어 대상의 크기와 거리를 측정 하는 데 사용되고 있다. 이에 더하여 이 센서는 반사된 빛의 양을 바탕으로 반사 강도 영상 또한 제공한다. 이는 측정 대상의 형상에 대한 정보를 제공하여 센서 데이터 처리에 긍정적인 효과를 일으킨다. LiDAR는 고해상도가 될수록 높은 성능을 보장하지만 이는 센서 비용의 증 가를 야기하는데, 이 점은 반사 강도 영상에도 해당된다. 높은 해상도의 반사 강도 영상을 취득 하기 위해서는 고가의 장비 사용이 필수적이다. 따라서 본 연구에서는 저해상도의 반사 강도 영상을 고해상도의 영상으로 개선하는 인공지능을 개발하였다. 이를 위해서 본 연구에서는 최 적의 초해상화 신경망 모델을 위한 파라미터 분석을 수행하였다. 또한, 초해상화 알고리즘을 2,500여 장의 반사 강도 영상에 적용하여 훈련과 검증을 하였다. 결과적으로 반사 강도 영상의 해상도를 향상시켰다. 바라건대 본 연구의 결과가 향후 자율 주행 분야에 적용되어 주행환경 인식과 장애물 탐지 성능 향상에 기여할 수 있기를 기대하는 바이다.

LiDAR is used in autonomous driving and various industrial fields to measure the size and distance of an object. In addition, the sensor also provides intensity images based on the amount of reflected light. This has a positive effect on sensor data processing by providing information on the shape of the object. LiDAR guarantees higher performance as the resolution increases but at an increased cost. These conditions also apply to LiDAR intensity images. Expensive equipment is essential to acquire high-resolution LiDAR intensity images. This study developed artificial intelligence to improve low-resolution LiDAR intensity images into high-resolution ones. Therefore, this study performed parameter analysis for the optimal super-resolution neural network model. The super-resolution algorithm was trained and verified using 2,500 LiDAR intensity images. As a result, the resolution of the intensity images were improved. These results can be applied to the autonomous driving field and help improve driving environment recognition and obstacle detection performance

산업사회에서 유통문제는 점차 비중이 커지고 있다 . 유통 네트워크를 운영하는 개인이나 기업 은 지점들 간의 유통 문제가 빈번히 발생한다. 더구나 고유가로부터 야기된 운송비 상승뿐만 아 니라, 교통량 증가로 인한 운송 장애로 물류비용이 점차로 증가하는 추세이다. 이러한 물류 환경 에서 물류 네트워크 정보를 어떻게 사용하느냐에 따라 유통에 소요되는 비용과 시간은 달라질 수 있다. 본 연구에서는 이러한 개념을 바탕으로 유통 네트워크 정보에 MAX-Flow 알고리즘을 적 용하여 유통량을 극대화하며, 극대화된 유통 네트워크에서 유통 관리 기능을 제공함으로써 물류 시스템의 효율성을 향상시킨다. 본 연구에서는 물류 시스템의 운영에 있어서 유통 네트워크의 최 적화로 물류 관리 비용을 절감하고 실제 산업현장의 여러 특성에 알맞게 적용할 수 있는 기반을 연구한다.

Network Flow has been playing important role in the modern industrial society. No matter what people or company run the network, both of them can’t avoid circulating goods among the many branches. But in practical situations, not only the price rising in network increase the transportation costs, but the huge traffic flow volumes increase the transportation costs. Given to such a network environment, how to flow goods in the network is very important. In this paper, the MAX-Flow algorithm will be applied to network flow in order to maximize the network flow volumes. As far as the functions of network are correctly provided, the optimized network system always can make the flow process efficiently.

국내 교통은 날로 증가하는 차량으로 인해 도로의 상습정체, 대기오염 발생 등의 다양한 교통문제가 발생되고 있다. 이러한 문제의 해결을 위해 지자체는 지능형교통체계(ITS : Intelligent Transport System), 첨단교통관리시스템(ATMS : Advanced Traffic Management Systems) 등의 시스템 구축을 통해 교통관리를 시행하고자 했으나 인프라 중심의 교통시스템 구축만으로 는 만성적인 교통문제 해결에 효과가 미비하여 기존 시설물에 운영관리 기능을 강화한 시스템 고도화가 필요한 시점이다. 도 시부 내 교통류는 임의의 시간대별로 특성 차량군이 형성되어 다양한 교통패턴이 존재하며, 이러한 상황별 교통패턴을 처리할 수 있는 지자체 네트워크 차원의 상황별 신호운영 설계가 필요하다고 판단된다. 따라서, 본 연구에서는 기존의 획일적인 신호 운영의 문제점을 개선하기 위해 단기적 교통상황 예측 데이터의 교통패턴을 기반으로 Frame Signal을 설정한 뒤 네트워크 차 원의 신호최적화를 통한 상황별 신호제어 운영방안을 목적으로 연구를 진행하고자 한다.

An increasing number of vehicles is causing various traffic problems such as chronic congestion of highways and air pollution. Local governments have been managing traffic by constructing systems such as Intelligent Transport Systems (ITS) and Advanced Traffic Management Systems (ATMS) to relieve such problems, but construction of an infrastructure-based traffic system is insufficient in resolving chronic traffic problems. A more sophisticated system with enhanced operational management capabilities added to the existing facilities is necessary at this point. As traffic patterns of the urban traffic flow is time-specific due to the different vehicle populations throughout the time of the day, a local network-wide signal operation plan that can manage such situation-specific traffic patterns is deemed to be necessary. Therefore, this study is conducted for the purpose of establishment of a plan for contextual signal control management through signal optimization at the network level after setting the Frame Signal in accordance to the traffic patterns gathered from the short-term traffic forecast data as a means to mitigate the problems with existing standardized signal operations.