불확실한 채널 상태 정보를 갖는 다중 셀 대규모 분산형 안테나 시스템에서 에너지 효율적인 자원 할당
Energy-Efficient Resource Allocation in Multicell Large Scale Distributed Antenna System with Imperfect CSI
- 주제(키워드) Massive MIMO
- 발행기관 아주대학교
- 지도교수 윤원식
- 발행년도 2016
- 학위수여년월 2016. 8
- 학위명 석사
- 학과 및 전공 일반대학원 전자공학과
- 실제URI http://www.dcollection.net/handler/ajou/000000022725
- 본문언어 영어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
Distributed antenna system is a technology that is capable of increasing both the energy efficiency and spectral efficiency of a cellular system by using adaptive power allocation, flexible antenna configuration and dynamic rate adjustment. The distributed antenna system is able to mitigate the effects of large scale fading and path loss by distributing the large number of antennas geographically. The basic idea of a distributed antenna system is that the antennas of the base station, which are known as remote radio heads are distributed geographically in a cell. The RRHs are connected to a single baseband processing unit through high speed backhaul like optical fibers or microwave repeaters. The digital data messages are formed at this baseband processing unit and finally transmitted as radio signals from each remote radio head. In this thesis, we study the energy-efficient resource allocation for the downlink of a multi-cell large scale distributed antenna system. A non-convex fractional optimization problem is formed which maximizes the objective function of energy efficiency, which guarantees a minimum QoS and considers total power consumption including the circuit power and the maximum allowed transmit power per RRH. By using the Dinkelbach method, the fractional optimization problem is converted into subtractive form. Next, an iterative algorithm is proposed where antenna and user selection and power allocation are done in separate steps. Norm based joint antenna and user selection algorithm using tabu search is first proposed to select optimum antennas and users for transmission. Next, lagrangian decomposition method is used to solve the sub-optimal power control. Simulation results validate that proposed scheme is always able to achieve its optimal value in limited iterations and that increasing the number of antennas or users doesn’t always necessarily result in higher EE. Besides, the proposed iterative scheme is always able to achieve a better performance than existing norm based algorithm and simplified CRAN based energy efficient power allocation scheme (S-CEEPA), no matter how many antennas or users are selected.
more목차
Chapter 1. Introduction 1
1.1 Background and Motivation 1
1.2 Contributions 2
1.3 Overview 4
Chapter 2. Massive MIMO Systems 5
2.1 Overview of Massive MIMO systems 5
2.2 Large Scale Distributed Antenna Systems 10
2.3 Pilot Contamination 11
2.4 System Model 13
2.5 Uplink Channel Estimation 14
2.6 Downlink Precoding 16
Chapter 3. Resource Allocation 18
3.1 Problem Formulation 18
3.2 Proposed Solution 20
3.2.1 Joint Antenna and User Selection Algorithm 21
3.2.2 Power allocation 25
Chapter 4. Simulation Results 31
4.1 Proposed JASUS vs conventional JASUS 32
4.2 Overall EE vs iterations 33
4.3 Overall EE vs antenna selection number 34
4.4 Overall EE vs user selection number 35
4.5 Overall EE vs SE 36
Chapter 5. Conclusion 38
Reference 40
