• Journal of the Korea Institute of Military Science and Technology
• /
• v.21 no.4
• /
• pp.497-507
• /
• 2018

Tactical networks are evolving into an All-IP based network for network centric warfare(NCW). Owing to the flexibility of IP based network, various military data applications including real-time and multi-media services are being integrated in tactical networks. Because each application has diverse Quality-of-service(QoS) requirements, it is crucial to develop a QoS provisioning method for guaranteeing QoS requirements efficiently. Conventionally, differentiated services(DiffServ) have been used to provide a different level of QoS for traffic flows. However, DiffServ is not designed to guarantee a specific requirement of QoS such as delay, loss, and bandwidth. Therefore, it is not suitable for military applications with a tight bound of QoS requirements. In this paper, we propose a multi-layer network virtualization scheme that allocates traffic flows having different QoS requirements to multiple virtual networks, which are constructed to support different QoS policies such as virtual network functions(VNFs), routing, queueing/active queue management(AQM), and physical layer policy. The experiment results indicate that the proposed scheme achieves lower delays and losses through multiple virtual networks having differentiated QoS policies in comparison with conventional networks.

• Journal of the Korean Geographical Society
• /
• v.43 no.4
• /
• pp.526-543
• /
• 2008

Based on the social network analysis(SNA), this study examines the structural and spatial characteristics of the actor networks of the manufacturing industries for the elderly. In the field of economic geography, former researches on network have mainly focused on the network governance. However, this study focused on the social network analysis. Centrality indexes are used to analyze the topological structure of actor networks of firms and organizations. In order to investigate the spatial structure of actor networks, not only the regional distribution of actors but also the correlation between centrality index and distance are analyzed. Network matrixes among actors are transformed to network matrixes among regions using block modeling method to reveal the spatial characteristics of the actor networks. In spite of the importance of the Capital Region, networks in the non-Capital Region like Chungnam and Pusan were showed high network density. This suggested that some kinds of policy project operating in the non-Capital Region had the influence on this network in the initial stage of industry.

• The Journal of Korea Robotics Society
• /
• v.14 no.1
• /
• pp.40-49
• /
• 2019

Reinforcement learning has been applied to various problems in robotics. However, it was still hard to train complex robotic manipulation tasks since there is a few models which can be applicable to general tasks. Such general models require a lot of training episodes. In these reasons, deep neural networks which have shown to be good function approximators have not been actively used for robot manipulation task. Recently, some of these challenges are solved by a set of methods, such as Guided Policy Search, which guide or limit search directions while training of a deep neural network based policy model. These frameworks are already applied to a humanoid robot, PR2. However, in robotics, it is not trivial to adjust existing algorithms designed for one robot to another robot. In this paper, we present our implementation of Guided Policy Search to the robotic arms of the Baxter Research Robot. To meet the goals and needs of the project, we build on an existing implementation of Baxter Agent class for the Guided Policy Search algorithm code using the built-in Python interface. This work is expected to play an important role in popularizing robot manipulation reinforcement learning methods on cost-effective robot platforms.

• Journal of the Economic Geographical Society of Korea
• /
• v.9 no.1
• /
• pp.97-122
• /
• 2006

The purpose of this paper is to analyze production and innovation networks of services in the long-live area of Gangwon and Jeju Provinces and to suggest a policy direction for regional development of rural areas where have been neglected in the knowledge-based information society. Four counties in the Honam Region, the long-live belt of Korea and two cities (Jeonju, Gwangju) are surveyed for the purpose of comparison with the Gangwon and Jeju areas. Production and innovation networks of research and supporting activities and tourist services are analyzed based on intensive interview surveys of the regions. The result of the analysis suggests that the innovation networks among the economic actors have considerable impacts on the innovation processes of the service activities and the service functions in the rural area are somewhat developed with local industry after the practice of local autonomy. The processes of innovation networks are progressed differently by the hierarchy of the regions as well as by the function of services such as research and supporting activities and tourist services. The direction of the rural development in the knowledge-based information society seems to be intensifying the networks among the innovative actors and developing virtual innovation networks for the development of rural innovation systems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.4B
• /
• pp.174-183
• /
• 2008

In this paper we design and implement a sensor network management framework(SNMF) for ubiquitous sensor networks(USNs). The SNMF employs the policy-based management approach for the autonomous and energy-efficient management of USNs. Moreover, a new light-weight policy distribution protocol called TinyCOPS-PR is designed and USN PIB for low-level policy is also defined. This allows the high-level policies defined by an administrator to translate into the specific low-level policies. The low-level policies are executed on sensor nodes so it can fulfill the proper management actions. The sensor nodes that receive some policies from an administrator perform local management actions according to those policies. SNMF can therefore realize small energy consumption and bring long network lifetime. It can also manage USNs automatically with a minimum of human interference.

• Journal of Communications and Networks
• /
• v.9 no.4
• /
• pp.408-418
• /
• 2007

In this paper, we address the question of dynamic network provisioning for time-varying traffic rates, with the objective of maximizing the system throughput. We assume that the network is capable of providing bandwidth guaranteed traffic tunnels for an ingress-egress pair and present an approach that (1) updates the tunnel routes and (2) adjusts the tunnel bandwidths, in an incremental, adaptive manner, based on the variations in the incoming traffic. First, we consider a simpler scenario where tunnel routes are fixed, and present an approach for adjusting the tunnel bandwidths dynamically. We show, through simulations, that our dynamic bandwidth assignment algorithm significantly outperforms the optimal static bandwidth provisioning policy, and yields a performance close to that of the optimal dynamic bandwidth provisioning policy. We also propose an adaptive route update algorithm, which can be used in conjunction with our dynamic bandwidth assignment policy, and leads to further improvement in the overall system performance.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.192-198
• /
• 2024

This paper investigates the application of multi-agent deep reinforcement learning in the fighting game Samurai Shodown using Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C) algorithms. Initially, agents are trained separately for 200,000 timesteps using Convolutional Neural Network (CNN) and Multi-Layer Perceptron (MLP) with LSTM networks. PPO demonstrates superior performance early on with stable policy updates, while A2C shows better adaptation and higher rewards over extended training periods, culminating in A2C outperforming PPO after 1,000,000 timesteps. These findings highlight PPO's effectiveness for short-term training and A2C's advantages in long-term learning scenarios, emphasizing the importance of algorithm selection based on training duration and task complexity. The code can be found in this link https://github.com/Lexer04/Samurai-Shodown-with-Reinforcement-Learning-PPO.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.10
• /
• pp.2376-2394
• /
• 2013

In order to ensure both of the whole system capacity and users QoS requirements in heterogeneous wireless networks, admission control mechanism should be well designed. In this paper, Multi-agent Q-learning based Admission Control Mechanism (MQACM) is proposed to handle new and handoff call access problems appropriately. MQACM obtains the optimal decision policy by using an improved form of single-agent Q-learning method, Multi-agent Q-learning (MQ) method. MQ method is creatively introduced to solve the admission control problem in heterogeneous wireless networks in this paper. In addition, different priorities are allocated to multiple services aiming to make MQACM perform even well in congested network scenarios. It can be observed from both analysis and simulation results that our proposed method not only outperforms existing schemes with enhanced call blocking probability and handoff dropping probability performance, but also has better network universality and stability than other schemes.

• Journal of Digital Convergence
• /
• v.11 no.5
• /
• pp.313-318
• /
• 2013

QoS in wireless multimedia sensor networks is an important issue to enhance streaming media service. This paper proposes a new 2-layer based QoS evaluation scheme for enhancing the streaming media QoS of wireless multimedia sensor networks. The proposed scheme performs the fuzzy relevance to control the streaming between application layer and network layer, and it performs 2-layer mapping process to enhance the transmission reliability and throughput. The simulation results show that the proposed scheme achieves improved performance in packet control ratio, transmission reliability, and delay overhead ratio compared with those of other existing schemes.

• Journal of Communications and Networks
• /
• v.12 no.5
• /
• pp.499-509
• /
• 2010

A plethora of transmission control protocol (TCP) congestion control algorithms have been devoted to achieving the ultimate goal of high link utilization and fair bandwidth sharing in high bandwidth-delay product (HBDP) networks. We present a new insight into the TCP congestion control problem; in particular an end-to-end delay-based approach for an HBDP network. Our main focus is to design an end-to-end mechanism that can achieve the goal without the assistance of any network feedback. Without a router's aid in notifying the network load factor of a bottleneck link, we utilize goodput and throughput values in order to estimate the load factor. The obtained load factor affects the congestion window adjustment. The new protocol, which is called TCP-goodput and throughput (GT), adopts the carefully designed inversely-proportional increase multiplicative decrease window control policy. Our protocol is stable and efficient regardless of the link capacity, the number of flows, and the round-trip delay. Simulation results show that TCP-GT achieves high utilization, good fairness, small standing queue size, and no packet loss in an HBDP environment.