검색 상세

Energy-Aware Access Point Management for Green WLAN

GREEN WLAN을위한 에너지 인식 액세스 포인트 관리

초록/요약

Over the past few decades, wireless communication has experienced phenomenal growth and has now have become fundamental to daily activities. However, this unprecedented growth comes at a price: due to the always-on usage model, these standards are responsible for a large amount of energy consumption. Optimizing the energy consumption of access points (APs) has become a new challenge for the research community, governments and industries in order to reduce CO2 emissions and operational energy costs. In this context, wireless local area networks (WLANs) that consist of a high-density of hundreds to thousands of APs are being deployed rapidly in corporate offices and universities to satisfy user demands for high bandwidth, mobility, and reliability. Moreover, these networked APs are provisioned for busy or rush hour loads, which typically exceed their average utilization by a wide margin. In addition, these margins are rarely reached and when reached, last only for a short period of time. Thousands of WLANs worldwide compound this problem, as they remain idle for long periods of time, raising serious concerns about energy losses. In response to this compelling problem, this thesis presents a set of contributions that address the challenge of increasing energy efficiency in Wi-Fi networks. In particular, we introduce novel energy efficient algorithms for dynamically powering off certain APs by exploiting the knowledge of the distance between the User Equipment’s (UEs) and servicing APs while retaining the best user experiences. The network design was mainly evaluated based on the benefits of a centralized structure and the resulting turn off WLAN APs upshot in a network that provides adequate radio signal coverage and the required data rate capacity to serve user traffic demand in the service region. Our proposed algorithms are thoroughly evaluated by means of ns-2 simulations. The proposed solution achieves significant power reduction of the network up to 30 to 40 % compare with always on case without significant reduction of overall network throughput. In this context, wireless local area networks (WLANs) that consist of a high-density of hundreds to thousands of APs are being deployed rapidly in corporate offices and universities to satisfy user demands for high bandwidth, mobility, and reliability. Moreover, these networked APs are provisioned for busy or rush hour loads, which typically exceed their average utilization by a wide margin. In addition, these margins are rarely reached and when reached, last only for a short period of time. Thousands of WLANs worldwide compound this problem, as they remain idle for long periods of time, raising serious concerns about energy losses. In response to this compelling problem, this thesis presents a set of contributions that address the challenge of increasing energy efficiency in Wi-Fi networks. In particular, we introduce novel energy efficient algorithms for dynamically powering off certain APs by exploiting the knowledge of the distance between the User Equipment’s (UEs) and servicing APs while retaining the best user experiences. The network design was mainly evaluated based on the benefits of a centralized structure and the resulting turn off WLAN APs upshot in a network that provides adequate radio signal coverage and the required data rate capacity to serve user traffic demand in the service region. Our proposed algorithms are thoroughly evaluated by means of ns-2 simulations. The proposed solution achieves significant power reduction of the network up to 30 to 40 % compare with always on case without significant reduction of overall network throughput.

more

초록/요약

지난 수십 년 동안 무선통신은 엄청난 성장을 해왔으며, 현재 일상생활에 기본이 되고 있다. 하지만 전례 없는 성장과 함께 상시 사용 모델로 인해 표준은 많은 양의 에너지 소비에 책임이 있다. 이산화탄소 배출량 및 운영 에너지 비용 감축을 위한 Access Points (APs)의 에너지 소비 최적화는 많은 연구단체와 정부나 산업에서 새로운 과제가 되고 있다. 이런 맥락에서, 기업이나 대학에서 무선 근거리 통신망은 높은 대역폭과 이동성과 안정성에 대한 사용자의 요구를 충족하기 위해 수백에서 수천 개의 AP 로 구성되어 있다. 또한 이렇게 연결된 네트워크는 러시아워 같은 시간에는 큰 차이로 평균 사용량을 초과하여 공급된다. 또한 이러한 차이는 아주 짧은 시간 동안만 도달하거나 거의 도달하지 못하는 경우가 많다. 전세계의 많은 무선 근거리 통신망이 이 문제를 가지고 있으며, 유휴 상태로 오래 머물러 있을 때 에너지 손실문제를 초래한다. 이 문제를 해결하기 위해, 본 논문에서는 Wi-Fi 네트워크에서 에너지 효율 향상이라는 과제에 기여하고자 한다. 특히 우리는 최상의 사용자 경험을 유지하면서 사용자장비와 서비스중인 AP 의 거리에 대한 정보를 이용하여 동적으로 특정 AP 의 전원을 끄는 에너지 효율을 위한 알고리즘을 제안한다. 네트워크 설계는 주로 중앙 통제 형 구조의 장점과 적절한 무선 신호의 범위와 서비스 영역에서 사용자 트래픽 수요를 제공하기 위해 필요한 데이터 속도 용량을 제공을 바탕으로 평가 되었다. 제안한 알고리즘은 ns-2 를 사용하여 시뮬레이션 하였다. 제안한 문제해결 방법은 전체 네트워크 처리량을 대폭 감소시키지 않으면서도 기존 대비 30%까지의 상당한 전력 절감을 얻을 수 있었다.

more

목차

I. INTRODUCTION 1--
II. RELATED WORK 3--
III. BACKGROUND 6--
3.1 WLAN Overview 6--
3.1.1 Network Capacities and User Density 8--
3.1.2 Coverage 9--
3.1.3 Frequency Channel Assignment 9--
3.1.4 Structure of Service Areas 10--
3.1.5 Redundancy 11--
3.2 Principle of RSSI 11--
3.2.1 Relationships between RSSI and distance 12--
IV. PROBLEM STATEMENT 14--
4.1 High-density WLANs 14--
4.2 Traffic Intensity 14--
4.3 Case Studies 15--
V. PROPOSED ALGORITHM 17--
5.1 Network Architecture 17--
5.2 Power Saving Scheme 18--
5.2.1 Power saving algorithm 18--
5.2.2 Implementation of switch-off scheme 21--
VI. SIMULATION 23--
6.1 Simulation scenario 23--
6.2 Performance Metric Used 24--
6.3 Result and Analysis 24--
VII. CONCLUSION 31--
REFERENCES 32--
ACKNOWLEDGMENT 36

more
반출 Meta View 목록