http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
다중접근범주를 지원하는 IEEE 802.11 기반 네트워크의 모형화와 성능 분석
이계민,강영경 한국자료분석학회 2007 Journal of the Korean Data Analysis Society Vol.9 No.1
We model a network under an IEEE 802.11-based protocol for multiple access categories and carry out the performance analysis. Currently, 802.11 is the most widely deployed wireless LAN standard. The MAC layer of 802.11 is endowed with flexibility to provide a differentiated service for various types of packets through a diffrentiated backoff procedure. We present a Markov model for the state of each station under a finite load traffic condition and derive an numerical algorithm to find the transmission probability and the throughput. Most of the restrictive assumptions in the previous works are loosened in the way that each station can send an arbitrary type of packets and the length of a packet can be different depending on the access category. The proposed model is validated through simulation under various system settings. 무선랜 표준으로 가장 많이 사용되고 있는 IEEE 802.11을 확장하여 다중접근범주를 지원하는 개선된 메카니즘을 제안하고, 네트워크의 모형화와 성능분석을 실시한다. 제안된 프로토콜은 패킷의 다양한 유형별로 차별화된 서비스를 제공하기 위하여, 802.11의 매체접근제어층에 유연성을 부여하여 패킷의 접근 범주에 따른 차별화된 백오프 절차를 통하여 다중접근범주를 지원한다. 유한 트래픽 조건 아래서 네트워크의 각 스테이션의 상태를 마코프 모형으로 분석하고 상태별 극한 확률과 전송처리율을 구하는 수치해석적 알고리즘을 유도한다. 기존 연구에서의 제한적인 가정들을 완화하여 각 스테이션은 임의의 유형의 패킷 전송을 시도할 수 있고, 유형별 패킹의 길이도 다르다고 가정한다. 마지막으로 다양한 시스템 설정 하에서 시뮬레이션을 통한 모형 검증과 성능 평가를 실시한다.
A Network Model Induced by Accumulated Wealth
이계민,김광일 한국물리학회 2006 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.49 No.4
In this paper, we propose a growing network model designated as the network induced by ac- cumulated wealth: This network is composed by three essential characteristic ingredients: wealth which is the motive for each node to participate in the network and which serves as the criterion by which the node gains the newly added link at each time step, the nodes which produce and accumulate wealth, and the links which work as chances for both nodes located at the end of each link to produce new wealth with given success probabilities of their own. We clarify the relation between wealth distribution and degree distribution of the network, which simultaneously explains how wealth induces the network structure and how the network structure serves for wealth accu- mulation. Furthermore, we prove the scale-free behavior of the degree distribution of the network. We also prove that the wealth distribution is doubly scale-free in the sense that the graph of the distribution consists of two power-law segments and a simple joiner between the segments, which is a universal phenomenon in several network models using the wealth-governing equation in the Bouchaud-M¶ezard model.
Equilibrium Point Analysis of A Multiple Access Protocol with a Code Reservation Scheme
이계민,강영경,전종우 한국자료분석학회 2004 Journal of the Korean Data Analysis Society Vol.6 No.6
We propose a multiple access protocol for a CDMA-based wireless communication system in which the terminals contend for the spreading codes and reserve the codes for multiple talk spurts. Equilibrium Point Analysis(EPA) is employed for the analysis of the proposed model and the recursive formulas for the system parameters are derived. We also present some numerical results corroborated by a simulated example.
Equilibrium Point Analysis of the Performance of IEEE 802.11 DCF under Poisson Packet Arrivals
이계민,강영경,최창원 한국자료분석학회 2005 Journal of the Korean Data Analysis Society Vol.7 No.5
The IEEE 802.11 standard has been widely deployed for wireless local area networks(WLANs) in recent years. The distributed coordination function (DCF) of 802.11 coordinates transmissions onto the shared communication channel. Under the DCF each station transmits packets depending on the channel state and its transmission packet queue. We propose a discrete time Markov model for 802.11 WLANs when packets arrive according to the Poisson process. We derive the average number of packets transmitted during a session and the channel throughput of the proposed model. Validation of the model is provided by numerical results which is corroborated by simulated examples.
Stratification in a Network Model of Information Exchange
이계민,김성곤,김광일 한국물리학회 2014 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.64 No.11
We propose a network model in which a node exchanges information with its neighbors and rewiresits link from a neighbor to the neighbor’s neighbor when it is given a greater chance to get a newitem of information. In simulations, the degree distribution evolves briefly from an initial uniformassignment to be scale-free in the tail, and then freezes with a heavier tail, which corresponds tothe monopolization of links by a few hubs. After freezing, the class of nodes in the middle of thedegree distribution shrinks, and information levels become stratified. Stratification is controlled bythe rewiring rate and the number of links per node. We suggest that increasing the connectednessof nodes is the best way to avoid stratification.
Creating Level Set Trees Using One-Class Support Vector Machines
이계민 한국정보과학회 2015 정보과학회논문지 Vol.42 No.1
A level set tree provides a useful representation of a multidimensional density function. Visualizing the data structure as a tree offers many advantages for data analysis and clustering. In this paper, we present a level set tree estimation algorithm for use with a set of data points. The proposed algorithm creates a level set tree from a family of level sets estimated over a whole range of levels from zero to infinity. Instead of estimating density function then thresholding, we directly estimate the density level sets using one-class support vector machines (OC-SVMs). The level set estimation is facilitated by the OC-SVM solution path algorithm. We demonstrate the proposed level set tree algorithm on benchmark data sets.
Analysis of IEEE 802.11 DCF in Non-saturated Conditions
이계민,강영경,최창원 한국자료분석학회 2005 Journal of the Korean Data Analysis Society Vol.7 No.4
The IEEE 802.11 standard defines the physical layer and media access control(MAC) layer for a wireless LAN and this standard is largely being deployed. The primary MAC technique of 802.11 is a random access scheme based on carrier sense multiple access with collision avoidance(CSMA/CA) protocol and is called distributed coordination function(DCF). We propose a Markov model for the behavior of a given station governed by DCF under non-saturated traffic conditions and evaluate the channel throughput. We validate the accuracy of the proposed model through simulation.