• The KIPS Transactions:PartD
• /
• v.19D no.2
• /
• pp.127-132
• /
• 2012

In the Seoul metropolitan bus network, it may be necessary for a bus passenger to pick up a parcel, which has been purchased through e-commerce, at his or her convenient bus stop on the way to home or office. The flow-capturing location-allocation model can be applied to select pickup points for such bus stops so that they maximize the captured passenger flows, where each passenger flow represents an origin-destination (O-D) pair of a passenger trip. In this paper, we propose a fast heuristic algorithm to select pickup points using a large O-D matrix, which has been extracted from five million transportation card transactions. The experimental results demonstrate the bus stops chosen as pickup points in terms of passenger flow and capture ratio, and illustrate the spatial distribution of the top 20 pickup points on a map.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.7B
• /
• pp.1001-1008
• /
• 2010

In this paper, we propose the dynamic spectrum allocation (DSA) algorithm using spectrum sharing method considering the long-term priority between NOs and service classes for the maritime communication system environment where a ship locates at either near shore (or land) or off-shore. It was shown that the proposed algorithm using spectrum sharing with priorities could deliver better satisfaction ratio (SR) than the fixed allocation schemes, in the context of provision of required bandwidth (or spectrum) for each users. Therefore, we conclude that the proposed DSA with priorities could apply to the maritime communication environment and exploit the under-used (or unused, idle) spectrum of terrestrial communication networks.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.7 s.361
• /
• pp.81-87
• /
• 2007

While conventional frequency reuse techniques for cellular communication systems divide frequency resources into multiple regions and each mobile is statically assigned to a certain frequency region, frequency reuse techniques for cellular OFDMA communication systems can be regarded as dynamic scheduling problems of finding best-fitted subcarriers for each packet transmission. Unlike conventional frequency reuse techniques allocating mutually exclusive frequency resources to adjacent cells, this paper proposes the use of a frequency reuse technique with irregular frequency allocation patterns assigned statically based on the cell numbers. This paper shows that the use of irregular frequency patterns can allow efficient interference avoidance and high data throughputs comparable to those with carefully planned frequency patterns.

• Electronics and Telecommunications Trends
• /
• v.28 no.3
• /
• pp.95-105
• /
• 2013

현재의 인터넷은 라우팅과 어드레싱에 대한 심각한 확장성 문제를 가지고 있다. 이러한 확장성의 가장 큰 원인은 멀티호밍, 트래픽 엔지니어링, 집적되지 못하는 주소 할당 등이며, 이로 인하여 백본 지역의 라우팅 테이블 크기가 기하급수적으로 증가하는 현상을 보이고 있다. 또한 현재의 IP 주소는 호스트의 식별자와 위치자의 의미를 함께 사용하기 때문에 호스트의 이동성 및 멀티호밍을 지원하는 데 한계점이 있다. 이러한 현재 인터넷의 문제점 해결 방안으로 식별자와 위치자 분리 구조가 연구되고 있다. 본고에서는 식별자와 위치자 분리 구조에서 필수적으로 요구되는 식별자와 위치자 간의 매핑 시스템에 관한 연구들을 소개하고 각각의 장단점을 분석한다. 본고에서는 현재 인터넷 기반의 대표적 식별자와 위치자 분리 구조 중 IETF(Internet Engineering Task Force)에서 잘 알려진 LISP(Locator Identifier Separation Protocol), HIP(Host Identity Protocol), ILNP(Identifier Locator Network Protocol)에서 제안하는 매핑 시스템들과 미래인터넷 기반의 대표적 식별자와 위치자 분리 구조 중 미국의 MobilityFirst와 한국의 MOFI(Mobile-Oriented Future Internet)에서 제안하는 매핑 시스템들을 중점적으로 소개한다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.7
• /
• pp.156-161
• /
• 2020

In an environment that operates multiple UAVs, the use of a decentralized task allocation algorithm has more robustness from a single failure of UAV on the mission because there is no central command center. In addition, UAVs have situational awareness and redistribute tasks among themselves, which can expand the mission range. The use of multiple UAVs in a mission has increased as the agent hardware has decreased in size and cost. The decentralized mission-planning algorithm has the advantages of a larger mission range and robustness to a single failure during the mission. This paper extended the type of mission the uses CBBA, which is the most well-known decentralized task allocation algorithm, to the point mission and en-route mission. This will describe the real mission situation that has the purpose of surveillance. A Monte-Carlo simulation was conducted in the case of multiple agents in the task-rich environment, and the global rewards of each case were compared.

• Journal of Broadcast Engineering
• /
• v.14 no.2
• /
• pp.178-188
• /
• 2009

In this paper, we propose the method which provides efficient sub-channel allocation for handoff and new call supporting multi-media service in AMC-based OFDMA system. Firstly, we apply the multi-band method which provides different AMC method according to the location of user terminals. Also, in OFDMA system environment that a base station has a lot of sub-channels, we adopt the sub-channel allocation scheme that provides a higher priority to handoff call and real-time service about handoff and new calls with multi-meida service. The simulation results show that the proposed scheme plays a role in increasing the number of new and handoff calls meeting the required blocking rate.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.12A
• /
• pp.1191-1199
• /
• 2008

We propose the Staggered-zone Resource Allocation (SRA) in order to relax throughput decrease problems by the co-channel interference of the cell boundary users at the cellular OFDMA system using frequency reuse factor K=1 and analyze the throughput improvement. The proposed algorithm allocates the resources to the users in compliance with resource allocation rule which is planned in order to minimize co-channel interference between cells without any additional information. The resource allocation method in the SRA lines up the users in pathloss order as descending series, and then allocates from pre-determined resource allocation region where decides differently in each cell. This algorithm prevents the co-channel interferences of the cell boundary user to be caused by using same resource simultaneously and equalizes interference to the users in the cell.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.9A
• /
• pp.888-897
• /
• 2007

In this paper, we propose an uplink dynamic resource allocation algorithm to increase sector throughput and fairness among users in 802.16e OFDMA TDD system. In uplink, we address the difference between uplink and downlink channel state information in 802.16e OFDMA TDD system. The simulation results show that not only an increment of 10% of sector throughput but higher level of fairness is achieved by round-robin using the FLR and the rate / margin adaptive inner closed-loop power control algorithm. The FLR algorithm determines the number of sub-channels to be allocated to the user according to the user's position. Also, we get 31.8% more sector throughput compared with the round-robin using FLR by FASA algorithm using uplink channel state information. User selection, sub-channel allocation, power allocation algorithms and simulation methodology are mentioned in Part I.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.77-84
• /
• 2022

In this paper, we propose a Label Reassignment Strategy to improve the performance of an object detection algorithm. Our approach involves two stages: an inference stage and an assignment stage. In the inference stage, we perform multi-scale inference with predefined scale sizes on a trained model and re-infer masked images to obtain robust classification results. In the assignment stage, we calculate the IoU between bounding boxes to remove duplicates. We also check box and class occurrence between the detection result and annotation label to re-assign the dominant class type. We trained the YOLOX-L model with the re-annotated dataset to validate our strategy. The model achieved a 3.9% improvement in mAP and 3x better performance on AP_S compared to the model trained with the original dataset. Our results demonstrate that the proposed Label Reassignment Strategy can effectively improve the performance of an object detection model.

• Journal of KIISE:Information Networking
• /
• v.29 no.3
• /
• pp.265-275
• /
• 2002

A default server strategy is a very popular scheme for managing location and state information of mobile hosts in cellular networks. But the communication cost increases if the call requests are frequent and the distant between the default server and the client is long. Still more any connection to a mobile host cannot be established when the default server of the destination mobile host fails. These problems can be solved by replicating default server and by letting nearest replicated default server process the query request which is sent from a client. It is important to allocate replicated default servers efficiently in networks and determine the number of replicated default servers. In this paper, we suggest and evaluate a default server replication strategy to reduce communication costs and to improve service availabilities. Furthermore we propose and evaluate an optimized allocation algorithm and an optimal replication degree for replicating: dofault servers in nn grid networks and binary tree networks.