DL Random Beamforming

國立臺灣大學電機資訊學院電信工程學研究所
博士論文
Graduate Institute of Communication Engineering
College of Electrical Engineering and Computer Science
National Taiwan University
Doctoral Dissertation
多輸入多輸出天線廣播通道中降低迴授量
Feedback Reduction in MIMO Broadcast Channel
李晉豪
Jin-Hao Li
指導教授：蘇炫榮博士
Advisor: Hsuan-Jung Su, Ph.D.
中華民國 102 年 10 月
October, 2013
　　　　國立臺灣大學博士學位論文
　　　　口試委員會客定書
多天線輸入輸出廣播通道下，降低回授量
　　Ｆｅｅｄｂａｃｋ　Ｒｅｄｕｃｔｉｏｎ　ｉｎ　ＭＩＭＯ　Ｂｒｏａｄｃａｓｔ
　　　　Ｃｈａｎｎｅｌ
　　
　　本論文係李晉豪君（學號Ｄ
９７９
４２０
２４）在國立臺灣大學電信工
程學研究所完成之博士學位論文，於民國１　０２
年９月１
　８日承下列考
試委員審查通過及口試及格，特此證明
口試委員：
（指導教授）
二至侈逐＋
所　　　　長（簽名）
致謝
時間過的很快，不知不覺已經在實驗室待了五個年頭，在此，感謝在這不算
短的日子裡，曾經一同奮鬥，一起經歷研究，計畫，與幫助我的人，在此分享謝
謝你們，也想與你們分享我的喜悅。
首先，謝謝一路上支持我的家人，沒有他們一直以來的幫助，支持，讓我可以
全心全力的專注於學業上，無需為金錢上煩惱，也謝謝家人帶給我的家庭溫暖，
讓我在遇到研究瓶頸時，有一個溫暖的避風港，一個可以放鬆的家。
謝謝我的指導教授，蘇炫榮老師，在研究上給予我許多的幫助，給予我很大的
空間，可以自由的發揮，選擇自己喜愛的研究題目，並且在遇到困難時，提供許
多的建議來解決問題，此外，也謝謝老師提供一個那麼棒的研究環境，提供最好
的設備，給予最大的自由，讓大家可以很放鬆的在實驗室，一起討論，努力，很
榮幸可以加入老師的實驗室。
在這五年的時間，很開心有一群熱心的學長提供意見，一起做計畫，謝謝林秉
勳學長，對於研究的方向與內容，提供許多的想法，謝謝廖偉舜學長，一起為了
計畫努力，也給予我許多的幫助，第一次一起去舊金山，開車去 outlet 的時光，
一輩子都難忘，謝謝不承認來自交大的李崇丕學長，一起吐研究的苦水，謝謝已
經畢業很天才的徐博，跟你討論，一起去土耳其伊斯坦堡報告，承蒙您照顧了。
謝謝冠文學弟，麻煩你一肩扛起了計畫的重責，太辛苦拉。還有有一群很會搞笑
的碩班學弟妹，一起修課，一起為計畫努力，一起嘴炮聊天，跟你們在一起，都
覺得自己還很年輕，帶給我許多的歡樂。
最後，要謝謝我的老婆，在這最後的一年，陪我經歷了面試，找工作，陪我等
待論文的審核，經歷了許多的困難，謝謝你在我身邊。礙於篇幅，還有需多沒列
出來的朋友，謝謝你們這些日子的鼓勵與支持，謝謝。
李晉豪
10 月/2013
博理館 508 室
摘要
在多使用者多傳輸天線傳輸多輸入天線下行廣播通道中，資料的傳輸速率可以透
過基地台的增加天線數量與排程機制來大量的增加。當使用者落在基地台的廣播範
圍，所有的使用者都收到來自於基地台的相同的資料。針對於多使用者多傳輸天線
與多接收天線系統，基地台具有 Mt 根傳輸天線，K 總共有 K 個使用者在此基地台涵
蓋範圍內且每個使用者具有 Mr 根接收天線，所有的多工增益 Mt 可以透過空間多重存
取技術或是迫零束波成形技術來達到。此外，當使用者數目遠超過傳送天線數目
K>>Mt，由於多使用者分散增益，速率總和會呈現 MtloglogK 的成長趨勢。然而，上
述的結果都來自於傳輸端具有所有的通道消息，因此，通道迴授量會隨著使用者與
天線數量線性的增加。另一方面，透過選擇適當的調變與編碼來達到可靠的傳輸，
通道消息是非常重要的，因此，對網路系統而言，迴授量是非常大的。
此論文主要的貢獻在於考慮不同的系統應用，例如，多使用者多傳輸天線多接收
天線廣播通道，與多點廣播通道，與不同系統的排程演算法，提出降低迴授量的策
略來降低無線資源的浪費。透過分析，降低迴授量的方法，透過推導的多個閥值來
執行，數個閥值是藉由序量統計的觀念配合排成演算法來得到。模擬結果可知，在
廣播通道中，迴授量可以大量的被降低並且幾乎完美的速率總和都可以被達到。對
於多點廣播通道，迴授量也可以大量的被降低並且解碼成功的使用者的數量幾乎與
傳輸端具有完美的通道消息的情況是一樣的。此外，新的排程方法配合多閥值技術
也被提出，在異直性質瑞雷衰減通道中，不同使用者被基地台挑選到的的公平性，
迴授的量，與系統的速率和的關係也被討論。
關鍵字 – 降低迴授；多用戶分散增益；束波形成技術
Abstract
In multiuser multiple-input multiple-output (MIMO) downlink broadcast
channel, data rate can be tremendously increased by adding antennas and

by scheduling algorithm at the base station (BS). When users are located
in the broadcast area, the users receive the same data which is broadcasted
from the BS. For the multiuser MIMO system with Mt transmit antennas
and K users equipped with Mr antennas, the full multiplexing gain Mt can
be achieved by using space-division multiple access schemes or zero-forcing
beamforming. Moreover, in a large user regime K >> Mt , the sum rate
grows like Mt log log K due to multiuser diversity. However, all these results
are based on the assumptions of full channel state information (CSI) at the
transmitter, thus, the feedback load linearly increases with number of users
and number of antennas. On the other hand, CSI feedback is also very
important in terms of proper selection of modulation and coding for achieving
reliable transmission. Thus the aggregate feedback load in a network is large.
The main contributions of thesis are proposing feedback reduction strate-

gies considering different applications, such as multiuser MIMO broadcast
channel and multicast channel, and scheduling algorithms to avoid waste of
radio resource used for CSI feedback. Through analysis, we derive the feed-
back reduction methods which use multiple thresholds which are obtained
through the concept of order statistics to exploit the scheduling mechanism.
Simulations show that the feedback load can be reduced dramatically and
almost full sum rate performance can be achieved in broadcast channel. For
multicast scenario, the feedback load can be reduced dramatically and num-
ber of decoded user is almost the same as the full CSI case. Moreover, a new
scheduling method with multi-threshold model is proposed and the relations

among the fairness across users, the feedback load and the sum rate in het-
erogeneous Rayleigh fading channel are also investigated.
Key Words -Feedback reduction; Multi-user diversity; Beamforming.
2
Contents
1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Overview of Thesis . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 Reduction Method in MIMO Broadcast Channel using Or-
thogonal Random Beamforming Approach 10
2.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2 The Multi-threshold Feedback Scheme . . . . . . . . . . . . . 19
2.2.1 Derivation of Multiple Thresholds . . . . . . . . . . . . 20
2.2.2 Sum Rate Loss Analysis and Minimum Number of Re-
gions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.2.3 Multiuser Diversity Using the Multi-threshold Scheme . 32
2.3 Bit Allocation and Feedback Load Analysis . . . . . . . . . . . 34
2.3.1 Optimal Bit Allocation with Given Thresholds . . . . . 35
2.3.2 Fast Bit Allocation Method . . . . . . . . . . . . . . . 37
2.3.3 Complexity Comparison . . . . . . . . . . . . . . . . . 42
2.3.4 Feedback Load Analysis . . . . . . . . . . . . . . . . . 42
2.4 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . . . 43
i
2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3 Reduction Method for Multiuser MIMO Zero-forcing Beam-

forming using Semi-orthogonal Scheduling Algorithm in Broad-
cast Channel 53
3.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

3.2 Sum Rate Analysis Using SUS Algorithm . . . . . . . . . . . . 55
3.2.1 User Selection . . . . . . . . . . . . . . . . . . . . . . . 55
3.2.2 CQI Metric . . . . . . . . . . . . . . . . . . . . . . . . 57

3.2.3 Sum Rate Analysis . . . . . . . . . . . . . . . . . . . . 59
3.3 Feedback Reduction Strategy . . . . . . . . . . . . . . . . . . 60
3.3.1 Derivation of Multiple Thresholds . . . . . . . . . . . . 60
3.4 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . . . 62
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4 Feedback Reduction in MBMS System 68
4.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.2 Threshold Design Using Order Statistics . . . . . . . . . . . . 72

4.3 Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.1 Feedback Load . . . . . . . . . . . . . . . . . . . . . . 74
4.3.2 Performance Loss Analysis . . . . . . . . . . . . . . . . 75
4.3.3 Determination of Threshold . . . . . . . . . . . . . . . 78
4.4 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . 79
4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
ii
5 Feedback Policies for Heterogeneous Rayleigh Fading Chan-
nel with Finite Feedback 85
5.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

5.2 Proposed Algorithm Using Partial Derivative . . . . . . . . . . 87
5.2.1 Greedy Algorithm . . . . . . . . . . . . . . . . . . . . . 87
5.2.2 Partial Derivative Algorithm . . . . . . . . . . . . . . . 88
5.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6 Sum Rate and Fairness Tradeoff with Feedback Reduction in
Multiuser Heterogeneous Rayleigh Fading Channel 92
6.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.1.1 Transmission and Feedback Procedure . . . . . . . . . 94
6.2 Design of Multiple Thresholds . . . . . . . . . . . . . . . . . . 95
6.2.1 Proposed Scheduler . . . . . . . . . . . . . . . . . . . . 97
6.3 Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.1 Sum Rate Analysis . . . . . . . . . . . . . . . . . . . . 98
6.3.2 Fairness Analysis . . . . . . . . . . . . . . . . . . . . . 99

6.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7 Conclusions 107
A Proof of Theorem 1 109
B Proof of Theorem 2 111
iii
C Proof of Theorem 3 113
D Proof of corollary 115
E Proof of Lemma 2 117
F Proof of Lemma 3 118
Bibliography 120
iv
List of Figures
2.1 Multiuser MIMO downlink system model. . . . . . . . . . . . 16
2.2 Multi-threshold feedback model. . . . . . . . . . . . . . . . . . 21

2.3 The probability (2.9) that user k’s SNR on the m-th beam
direction is ranked the p-th among all the users. Mr = Mt ,
K = 20 users. . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Sum rate loss versus the number of users. △RP (K) = 0.25
bps/Hz is the tolerable sum rate loss. ρ = 10 dB, Mt = Mr = 4. 29
2.5 Sum rate performance of the multi-threshold scheme with dif-
ferent numbers of regions. ρ = 10 dB, Mt = Mr = 4. . . . . . . 31
2.6 Sum rate comparison between the mathematical lower bound
and the simulation result, ρ = 10 dB, Mt = Mr = 4. . . . . . . 36
2.7 Sum rate performance comparison for different bit allocation

methods with N = 4 regions, ρ = 10 dB, Mt = Mr = 4. . . . . 41
2.8 Sum rate performance comparison for different feedback schemes.
ρ = 10 dB, Mt = Mr = 4. . . . . . . . . . . . . . . . . . . . . . 47
2.9 Feedback load comparison for different feedback schemes. ρ =
10 dB, Mt = Mr = 4. . . . . . . . . . . . . . . . . . . . . . . . 48
2.10 Sum rate as a function of the feedback load. BQ,A = BQ,B =
BQ = BQ,D = 5, ρ = 10 dB, Mt = Mr = 4. . . . . . . . . . . . 49
v
2.11 Sum rate as a function of the feedback load. BQ,A = BQ,B =
BQ = BQ,D = 5, ρ = 10 dB, Mt = 2, Mr = 4. . . . . . . . . . . 50
2.12 Sum rate as a function of the feedback load for different trans-
mission powers (ρ). BQ,A = BQ,B = BQ = BQ,D = 5, Mt =
4, Mr = 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.1 System model of zero-forcing beamforming using SUS algorithm. 56

3.2 Comparison of simulated sum rate and approximated sum rate
with different CQI metrics, P =10W, BD =8 bits. . . . . . . . . 65

3.3 Comparison of sum rate for different feedback types Mt =4,
P =10W, BD =8 bits, BQ =2 bits and ε=0.4 . . . . . . . . . . . 66
3.4 Comparison of sum rate as a function of feedback load for
different feedback types, Mt =4, P =10W, BD =8 bits, BQ =2
bits, ε=0.4 and K = 100 ∼ 500. . . . . . . . . . . . . . . . . . 67
4.1 MBMS system model . . . . . . . . . . . . . . . . . . . . . . . 70
4.2 Increment of number of successfully decoding users increasing
threshold from rth,K−n to rth,K−(n+1) . . . . . . . . . . . . . . . 78
4.3 Number of successfully decoding users of proposed scheme. . . 81

4.4 Feedback load of proposed scheme using different threshold. . 82
4.5 Number of successfully decoding users vs. feedback load. . . . 83
5.1 Rate comparison for different algorithm. ρ=10dB. . . . . . . . 90
6.1 Sum rate comparison between different number of feedback

regions (P= 10W). . . . . . . . . . . . . . . . . . . . . . . . . 103
vi
6.2 Fairness comparison between different number of feedback re-
gions (P = 10W). . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.3 Sum rate vs. feedback load and fairness vs. feedback load
with different number of regions (P = 10W ). . . . . . . . . . 105
vii
List of Tables
2.1 The greedy algorithm for bit allocation. . . . . . . . . . . . . . 37
viii
List of Abbreviations
AWGN Additive White Gaussian Noise
BS Base Station
CDF Cumulative Distribution Function
CSI Channel State Information
CQI Channel Quality Indicator
CDI Channel Direction Information
DPC Dirty Paper Coding
E-MBMS Enhanced Multimedia Broadcast Multicast Service
i.i.d Independent Identically Distribution
LTE Long Term Evolution
MIMO Multiple Input Multiple Output
MISO Multiple Input Single Output
MCS Modulation and Coding
ix
MBS Multicast Broadcast Service
MBMS Multimedia Broadcast Multicast Service
ORB Orthogonal Random Beamforming
PDF Probability Distribution Function
RVQ Random Vector Quantization
SDMA Spatial Division Multiple Access
SNR Signal-to-Noise Ratio
SINR Signal-to-Interference-Noise Ratio
SUS Semi-orthogonal Scheduling
x
Chapter 1
Introduction
1.1 Background
In the recent years, the wireless communication has been confronted
the demand for the high data rate and the quality of service (Qos). The
applications include the voice transmission, file transfer, steaming multimedia
and video telephony, each with different Qos constraint. Due to different Qos
requirements for different services, the network considers the tolerable delay,
the traffic load of the system, system fairness and channel conditions of users
to allocate the resources to the users. Through the scheduler, the multiuser
diversity is exploited through the better utilization of the spectrum [1].
In addition to multiuser diversity gain, multiple-input multiple-output
(MIMO) technologies can provide spatial diversity in wireless fading chan-

nels to improve the communication quality, increase the spectrum efficiency
and to enhance the capability of interference suppression. As a result, the sys-
tems including 3GPP Long Term Evolution (LTE) [2], High Speed Downlink
Packet Access (HSPA) and Worldwide Interoperability for Microwave Access
(WiMAX) [3] are applying the multiple antennas to meet the requirement of
1
Qos.
Recently, there has been a popular interest in the role of multiple an-
tennas for the multiuser network settings, especially in broadcast channel
and multicast channel scenarios. The MIMO broadcast channel scenario

models the network that the base station (BS) sends different data to dif-
ferent users sharing the same resource. Hence there is inter-user interference
environment, and MIMO and scheduling techniques are used to reduce inter-
ference and attain the diversity and/or multiplexing gains. Both MIMO and
scheduling techniques rely on channel state information (CSI) feedback. For
the multicast channel scenario, all users under the cell coverage receive the
same data which is sent from the BS. In this case, CSI feedback is important
for rate adaptation through selection of modulation and coding schemes.
For the downlink broadcast channel employing multiple antennas, it has
been shown recently that dirty paper coding (DPC) [4] achieves the capac-
ity [5]. However, this capacity achieving scheme is difficult to derive and
has a high encoding/decoding complexity. Thus, several works resorted to
the more practical (but suboptimal) space division multiple access (SDMA)
based designs. For example, zero-forcing beamforming (ZF-BF) was shown

in [6] to achieve the optimal sum rate growth. However, both the DPC and
the ZF-BF schemes require perfect channel state information (CSI) feedback
from the users to the BS to achieve the optimal performance. The need for
CSI feedback results in a significant burden on uplink capacity.
On the other hand, because of the increasing demand of multimedia ap-
plications, multimedia broadcast multicast service (MBMS) grows rapidly in
recent years. In MBMS systems, the users receive the same data broadcasted
2
by the BS on the common radio resources, and hence the efficiency of radio
resources is highly improved and the quality of multimedia service is then
increased. Different from the MIMO broadcast channel scenario, the main
purpose of MBMS system is to provide the reliable transmission instead of
achieving high data rate. Due to the remarkable enhancement on efficiency
of radio resources and multimedia quality, MBMS has become one of the
important features in next generation communication standards, such as the
Enhanced-MBMS (E-MBMS) in 3GPP LTE [7] and the Multicast Broadcast

Service (MBS) in WiMAX [3]. However, to enable MBMS functions, all the
users in the MBMS network have to feedback its’ current channel condition
to the BS, so that the BS can select a proper modulation and coding scheme
(MCS) for starting the broadcast transmission to all the MBMS users. For
this sake, the feedback load will be increased dramatically with the num-
ber of the users in the MBMS network and hence the spectrum efficiency is
degraded due to the accordingly increased feedback bandwidth. Therefore,
for the reason mentioned above, an effective feedback reduction method is
significant for MBMS systems.
The unrealistic assumption of perfect CSI at transmitter, the performance

gap between the sum rate performance with full CSI at the transmitter and
the sum rate performance without CSI and worse utilization of feedback
bandwidth for scenarios of the broadcast channel and multicast channel mo-
tivate the research of the scheme with partial CSI at the transmitter. Partial
CSI or limited feedback refers to any possible of form of incomplete informa-
tion of channel such as quantized channel quality indicator (CQI), quantized
channel direction information (CDI) and statistics CSI.
3
The dominant role of CSI at the transmitter in multiuser MIMO sys-
tems can be identified by the asymptotic sum rate growth under different
assumptions on CSI at the transmitter. When the full CSI is obtained at

the transmitter, the scaling law of sum rate of MIMO broadcast channel is
Mt log log(KMr ) for fixed number of receive antennas Mr , number of trans-
mit antennas Mt , large number of users K and transmit power P [8]. When
the transmitter is lack the CSI, the scaling law of sum rate of MIMO broad-
cast channel is zero for fixed Mr , Mt , P and large K. There is no multiuser

gain can be exploited since the transmitter has no knowledge of CSI. For the
limited feedback model [9], each user feeds back B bits to represent the quan-
tized CDI and unquantized CQI. When the low-complexity semi-orthogonal
user selection method [10] and ZF precoder are used, the multiuser diversity
Mt log log K is achieved. When B = log Mt , this scheme becomes to the
random beamforming scheme [11]. In this scheme, each user needs to feed
back the best unquantized SINR among all directions and the corresponding
beam index. The full multiuser diversity is also achieved. In [12], only one
bit feedback per user for the ORB scheme exhibits the same growth rate as
the ORB scheme with perfect SINR. However, the throughput may degrade
a lot due to large quantization error. Note that the above results are based
on the assumption that the receiver has the perfect CSI. The feedback load
of limited feedback schemes increases with number of users dramatically.
There are other previous works that sought to reduce the feedback load at
the scheduling stage. In [13], a threshold was set according to the scheduling
outage probability such that a user does not need to feedback when its CSI is
below the threshold. This method reduces the system feedback load without
4
affecting the scheduling performance much. In [14], multiple thresholds are
set, and the scheduler utilizes a polling process to select the best feedback
threshold from these thresholds to further reduce the aggregate feedback

load. The drawback of this scheme is the large delay incurred by the polling
process. In [15], another scheme is proposed to reduce the feedback load of
the ZF-BF systems through two-stage feedback. In the first stage, each user
feeds back the coarsely quantized version of its CSI, thus the BS has some
information to determine which users to schedule. The BS then broadcasts

to the scheduled users and asks them to feedback finer CSI to achieve good
ZF-BF performance in the second stage. The drawback of this scheme is also
the delay incurred by the two-stage feedback process.
Reliable communication over the wireless environment is a great challenge
since the time varying channel may cause the unexpected transmission error.
Automatic Repeat-reQuest (ARQ) is a popular error recovery mechanism. If
the transmitter does not receive an acknowledgment before the timeout, the
transmitter will retransmit the data until the transmitter receives an acknowl-
edgment or exceeds a predefined number of retransmissions. However, when
many receivers are located in a cell, this ARQ mechanism can cause a large
number of ARQ feedback messages (ACKs or NACKs) to be sent to the trans-
mitter at the same time, which is referred to the feedback implosion problem.
The previous works investigate this problem as shown in [16, 17]. However,
these articles only focus on the ACK/NACK feedback instead of CQI feed-
back. Through the selection MCS and resource allocation for MBMS, the
reliable communication can be also achieved. The scalable coding method
is proposed in [18] to manage the radio resources. In [19], the resources are
5
dynamically allocated to MBMS according to the channel knowledge of users
to maximize the system rate while guaranteeing the system coverage. With
CSI of users, dynamic resource allocation is feasible strategy to make full use
of frequency-selective attenuation to achieve higher performance. In [20], an
analytical model for the coverage and capacity estimation of MBMS over a
Single Frequency Network (MBSFN) and investigation of the best cell radius
and the best MCS to reach a certain spectral efficiency target are provided.
Four different approaches for the efficient selection of the appropriate MCS
in MBSFN transmissions are presented in [21]. Due to the nature of time
varying channels, the MCS has to be adjusted to the weakest user of the
multicasting group [19]. So, each user needs to feedback the CSI to the BS.
However, in this scenario, the overhead of feedback information is a heavy
drawback to the achievement of good network performance. Another feed-
back reduction method is proposed in [22]. In this method, the user makes
a decision whether it reports the CQI or not according to the recent frame
error rate (FER). However, the user can not meet the Qos constraint when
the feedback signaling time is large.
It is clear that the feedback load of multiuser MIMO downlink systems can
be reduced if the scheduling mechanism is taken into consideration. However,
most existing works only exploit the scheduling mechanism to moderate the
amount of feedback, but not incorporate the properties of scheduling into the
CSI quantization design. In view of this, in this thesis we propose to reduce
the feedback load by incorporating the scheduling mechanism in both the
determination of the amount to feedback and the CSI quantization. Note
that the required information for the proposed scheme to operate, such as
6
the signal-to-interference-noise ratio (SINR) or signal-to-noise ratio (SNR)
statistics and the number of users, are usually known at the BS. Thus, in
practice, the BS can compute the region thresholds and broadcast to the users
periodically, or broadcast the parameters of the SINR/SNR statistics and the
number of users periodically to the users to let them derive the thresholds.
In practical systems where the number of users is large, the threshold values
will be insensitive to the number of users, and may not have to be updated
frequently. Note that the scheme in [13] also has its threshold dependent on
the channel statistics and the number of users, if a certain scheduling outage
probability is desired. Thus periodic update of the threshold is also necessary
for the scheme in [13].
Conventional multiuser diversity maximizes the system throughout but
results in an unfair scheduling of the system resource across users. Propor-
tional fair scheduling achieves strict fairness among different users but this
fairness comes at the cost of a significant sum rate penalty. The perfor-
mance evaluations for different schedulers such as round robin, proportional
fair scheduling and greedy SNR scheduling are discussed in [23]. In [24], a
hybrid multiuser scheduler combines the greedy SNR scheduling and nor-
malized SNR scheduling. The BS divides all active users into n groups. The
BS selects one user per group based on either the best SNR or normalized
SNR criterion and then the BS chooses one user form n selected users based
on normalized SNR or best SNR criterion. Hence, the flexible tradeoff be-
tween the sum rate and the fairness among users can be adjusted through the
number of groups n. However, the concept of feedback load reduction does
not take into account in above scheduling methods. Thus, a new scheduler
7
is motivated and proposed to jointly consider the feedback load reduction,
the fairness among users and the sum rate to achieve efficient and flexible
system.
1.2 Overview of Thesis
We first present the design of feedback reduction method in MIMO broad-

cast channel using ORB approach in Chapter 2. The multiple thresholds are
used to divide the range of SINR into multiple regions to reduce the feed-
back load. The minimum number of regions required to meet the sum rate
loss requirement is investigated. Moreover, assigning quantization bits into
multiple regions is introduced in Chapter 2.3. Because the performance of
ORB scheme is limited by the inter-user interference, thus the ZF-BF using
semi-orthogonal user selection (SUS) algorithm is considered in Chapter 3.
However, the exact CSI is unknown either the BS nor users, the approximated
sum rate is formulated. Then, the multiple thresholds are derived to avoid
wasting of the feedback resources according to the SUS algorithm. Because
of the increasing demand of multimedia applications, MBMS grows rapidly
in recent years. The application of the feedback reduction in MBMS sys-

tem is discussed in Chapter 4. The number of successfully decoding users in
the system is considered as the performance metric. Applying the proposed
feedback scheme, the number of successfully decoding users is investigated
and the constant feedback load regardless of number of users is achieved. In
Chapter 5, we consider the optimization problem with objective that maxi-
mizes the sum rate under finite feedback load constraint in the heterogeneous
Rayleigh fading channel. Each user has its own threshold which is obtained
8
by the proposed low complexity derivative algorithm. Simulations show that
the sum rate performance of proposed algorithm is close to the sum rate
performance using greedy algorithm.

The relations among the fairness, the feedback load and the sum rate
in heterogeneous channel are discussed in Chapter 6. In the heterogeneous
fading channel, the users are grouped in to multiple groups. The users’
SINRs/SNRs in each group can be viewed as identical distribution to simplify
the system model. A new scheduling method with multi-threshold model

is proposed and the flexible fairness can be achieved through changing the
number of thresholds in the multi-threshold feedback scheme in Chapter
6.2.1. Through simulations, the flexible tradeoff among the fairness among
users, the sum rate and the feedback load can be achieved.
1.3 Notations
Vectors and matrices are denoted by boldface lower case and capital let-
ters, respectively. E{·} refers to the expected values of a random variable.
XT (xT ) stands for the transpose of matrix X (vector x), and X∗ (x∗ ) stands
for the conjugate transpose of matrix X (vector x). Moreover, X† denotes the
pseudo-inverse X∗ (XX∗ )−1 . The function ⌈x⌉ represents the smallest integer
≥ x. ⌊x⌋ denotes the largest integer equal to or less than x. log and ln are
the logarithms with base 2 and e, respectively. The function f (n) = O(g(n))

f (n)
denotes that lim ≤ α, where α is a positive constant independent of
n→∞ g(n)
n.
9
Chapter 2
Reduction Method in MIMO

Broadcast Channel using
Orthogonal Random
Beamforming Approach
For MIMO downlink channel in [25,26], a model was proposed to analyze
the sum rate loss due to imperfect (quantized) CSI. In the system considered
there, each user quantizes the channel vector to one of the 2BD quantization
vectors and feeds back the codebook index using BD bits to the BS to capture
the spatial direction and magnitude of the channel. If we consider that

each user only has BD = log Mt bits to quantize the CDI, where Mt is the
number of transmitted antennas. The opportunistic random beamforming

(ORB) scheme can be viewed as a low-feedback scheme for CDI. However,
the feedback load including CQI and CDI is increased with the number of
users. In this chapter, the ORB scheme [11] is considered to further reduce
feedback load.
There are other previous works that sought to reduce the feedback load at
the scheduling stage. In [13], a threshold was set according to the scheduling
10
outage probability such that a user does not need to feedback when its CSI is
below the threshold. This method reduces the system feedback load without
affecting the scheduling performance much. In [14], multiple thresholds are

set, and the scheduler utilizes a polling process to select the best feedback
threshold from these thresholds to further reduce the aggregate feedback
load. The drawback of this scheme is the large delay incurred by the polling
process. In [15], another scheme is proposed to reduce the feedback load of
the ZF-BF systems through two-stage feedback. In the first stage, each user
feeds back the coarsely quantized version of its CSI, thus the BS has some
information to determine which users to schedule. The BS then broadcasts
to the scheduled users and asks them to feedback finer CSI to achieve good
ZF-BF performance in the second stage. The drawback of this scheme is also
the delay incurred by the two-stage feedback process.
In the ORB scheme, the BS transmits through orthogonal beamforming
vectors to the users, and each user only needs to feedback its received SINR or
SNR on different orthogonal beamforming vectors for the purpose of schedul-
ing. However, the aggregated feedback load increases linearly with respect
to the number of users. In order to efficiently utilize the feedback resources,

The proposed scheme divides the range of CSI into multiple regions accord-
ing to the order statistics of the received SNR or SINR which reflect the
properties of scheduling. Each region corresponds to a range of SNR/SINR,
and is quantized with a specific number of bits (usually, more quantization
bits for regions with higher SNR/SINR) to further assist scheduling and link
adaptation. The CSI feedback thus consists of two parts: one indicating the
region that the received SNR/SINR falls in, and the other being the quan-
11
tized result of that region. When a user’s SNR/SINR is in the lowest region
(that is, below the lowest threshold), the user doest not feedback at all. This
way, the users opportunistically reduce the feedback load.

For a given number of regions, we derive a tight upper bound for the
sum rate loss of the proposed scheme as compared to systems with perfect
CSI feedback from all users. Then, for any given tolerable sum rate loss of
the system, the minimum number of regions required is derived. For exam-
ple, the proposed scheme with four regions is good enough to keep the sum
rate loss smaller than 0.25 bps/Hz when the number of antennas at both the
transmitter and the receivers is four and the number of users is less than 500.
In addition, the aggregate feedback load and the multiuser diversity using
the proposed scheme are also investigated. Our theoretical analysis shows
that, in contrast to the existing feedback schemes whose aggregate feedback
loads increase with the number of users, with a given number of regions,
the proposed scheme has a constant aggregate feedback load regardless of
the number of users. Moreover, while keeping the aggregate feedback load
constant, the proposed scheme almost achieves the optimal asymptotic sum
rate scaling with respect to the number of users (that is, the multiuser di-
versity). Specifically, given the number of regions, the proposed scheme can
achieve a constant portion of the optimal sum rate achievable only by the
system in which all the users always feedback, and the sum rate loss decreases
exponentially to zero as the number of regions increases. This is a strong
evidence that the proposed scheme opportunistically exploits the scheduling
gain (equivalently, the multiuser diversity) for the feedback design, unlike
the schemes in [25,26] which do not take advantage of scheduling and require
12
all users to feedback, and the scheme in [13] which does not exploit order
statistics and has the aggregate feedback load increasing with the number of
users. Through simulation, we verify the analytical results, and demonstrate

that the proposed scheme can reduce the feedback load and utilize the limited
feedback bandwidth more effectively than the existing feedback methods. A
fast bit allocation method that assigns different numbers of quantization bits
to different regions is also discussed. The simulation results show that the
sum rate performances with the complex optimal bit allocation method and
with the fast algorithm are almost the same.
Note that the required information for the proposed scheme to operate,
such as the SNR/SINR statistics and the number of users, are usually known
at the BS. Thus, in practice, the BS can compute the region thresholds
and broadcast to the users periodically, or broadcast the parameters of the
SNR/SINR statistics and the number of users periodically to the users to let
them derive the thresholds. As will be shown later, the threshold values are
proportional to ln K, where K is the number of users considered in schedul-
ing. In practical systems where the number of users is large, the threshold
values will be insensitive to the number of users, and may not have to be
updated frequently. Note that the scheme in [13] also has its threshold depen-
dent on the channel statistics and the number of users, if a certain scheduling
outage probability is desired. Thus periodic update of the threshold is also
necessary for the scheme in [13].
This chapter is organized as follows. Section 2.1 describes the system
model and briefs the order statistics. Section 2.2 introduces the proposed
feedback scheme and analyzes its sum rate loss and multiuser diversity. In
13
Section 2.3, the bit allocation problem is discussed along with the feedback
load analysis. We then give the simulation results in Section 2.4 and conclude
the chapter in Section 2.5.
2.1 System Model
The multiuser MIMO downlink system model is shown in Fig.2.1 where

the BS is equipped with Mt antennas. There are K users in the system and
each user has Mr receive antennas. We consider a full buffer traffic model,
that is, each user always has data in the buffer to transmit. According
to the ORB strategy for multiuser transmission, the BS uses a precoding
matrix W = [w1 , w2 , . . . , wMt ] to transmit Mt signals simultaneously, where
wi ∈ CMt , i = 1, 2, . . . , Mt , are random orthogonal vectors (beams) generated
from isotropic distribution, and performs maximum sum rate scheduling to
schedule a user on each beam [27]. The received signal at the k-th user can
be mathematically described as:
yk = Hk Ws + nk , (2.1)
where Hk is the Mr × Mt complex Gaussian channel matrix between the BS

and the k-th user, nk is the Mr × 1 additive white Gaussian noise (AWGN)
vector at the k-th user. The entries of Hk and nk are assumed to be in-
dependent identically distributed (i.i.d.) complex Gaussian with zero mean
and unit variance. In addition, the channel matrices for different users are
assumed to be independent.
Note that in this work we consider only identical channel distributions
for the users for the simplicity of demonstrating the idea. This assumption
14
can also be used to model the users’ instantaneous SNRs normalized by
their respective average SNRs for systems with different channel statistics
for the users, but using the SNR based proportionally fair scheduling [28]
which schedules users based on their normalized instantaneous SNRs. The
more practical situations where the users have different channel statistics or
distributions are more intricate, and are discussed in [29].
The vector s = [s1 , s2 , . . . , sMt ]T is the Mt × 1 vector of the transmitted
signal. It is assumed that the feedback channel is error-free and delay-free.
The total transmitted power is a constant Pt so that E{s∗ s} = Pt . Under
the equal power assumption of the ORB, each beam is equally allocated with
power ρ = Pt /Mt .
When Mr < Mt , inter-beam interference is unavoidable at the receivers.
The statistical distribution of the received SINR in this situation was studied
in [11]. We define SIN Rm,k as the SINR of the m-th signal sm at the k-th
user. For Mr ≥ Mt , interference avoidance is possible. We consider zero-
forcing (ZF) receivers to remove the inter-beam interference. In this case,

the received signal after the ZF filter at the k-th user is given by
(Hk W)† yk = s + (Hk W)† nk . (2.2)
Therefore, the received SNR of the m-th signal sm at the k-th user with the
ZF receiver is given by
ρ
SN Rm,k = (2.3)
[((Hk W)∗ (H k W))
−1 ]
m
where [A]m denotes the m-th diagonal element of matrix A. According
to [30], SN Rm,k is a chi-square random variable with 2(Mr − Mt + 1) degrees
15
Feedback channel
User
1
Spatial
Scheduling User
division
Algorithm 2
multiple
.. . .. . .
.. . .. . .
(Maximum
access
sum rate)
.. . .. . .
(SDMA)
K users data Mt antennas

Base Station User
K
Mr antennas
Feedback CQI, CDI
Figure 2.1: Multiuser MIMO downlink system model.
16
of freedom. For notational simplicity, we let


 SIN Rm,k , Mr < Mt
Xm,k = .

 SN Rm,k , Mr ≥ Mt
Then the probability density function (PDF) of Xm,k for x > 0 can be sum-
marized as [11] [30]
 2Mr −1 (2Mr −1)
 −x ∑ (Mr (Mt − 1) + i − 1)!

 x2Mr −2 e ρ i

 2Mr −i−1
 i=0 (ρ) (1 + x)Mr (Mt −1)+i
fXm,k (x) = , Mr < Mt .
 (2Mr − 1)!(Mr Mt − Mr − 1)!

 −x

 eρ
 ( xρ )Mr −Mt , Mr ≥ Mt
ρ(Mr − Mt )!
(2.4)
Consequently, the cumulative distribution function (CDF) of Xm,k is given
by
FXm,k (x) =
 2Mr −1 ∫ (2M −1) −x
 ∑ x r
x2Mr −2 e ρ (Mr (Mt − 1) + i − 1)!

 i

 dx
 i=0 0 (ρ)2Mr −i−1 (1 + x)Mr (Mt −1)+i
, Mr < M t ,
 (2Mr − 1)!(Mr Mt − Mr − 1)!

 − Mt + 1, xρ )

 Γ(M r
 1− , Mr ≥ Mt
(Mr − Mt )!
(2.5)
∫∞
where Γ(a, x) = x
ta−1 e−t dt is the upper incomplete gamma function. Ac-
cording to (2.4), in the special case when the transmitter and the receiver
have the same number of antennas, Xm,k has an exponential distribution
with parameter 1/ρ.
With the maximum sum rate scheduling, on each beam direction, the
BS selects, among the users who have fed back their CSIs, the user that
17
has the best channel to transmit to. If none of the users has fed back the
CSI, the BS randomly selects one user to transmit to. Due to the symmetric
property, we drop the beam index m of Xm,k , and let Xk represent the SNR or
SINR (depending on the antenna configuration) of user k for a certain beam.
K K K
Let X(1) , X(2) , . . . , X(K) be the order statistics of i.i.d. continuous random
variables X1 , X2 , . . . , XK with the common PDF (2.4) in decreasing order,
K
i.e., X(1) ≥ X(2)
K
≥ · · · ≥ X(K)
K K
. The CDF and PDF of X(j) , respectively, are
given by [31]:
∑
K
K!{FXm,j (x)}i {1 − FXm,j (x)}K−i
FX(j)
K (x) = , −∞ < x < ∞, (2.6)
(i)!(K − i)!
i=K−(j−1)
K!fXm,j (x){FXm,j (x)}K−j {1 − FXm,j (x)}j−1

fX(j)
K (x) = , −∞ < x < ∞. (2.7)
(j − 1)!(K − j)!
With the order statistics, the sum rate using the maximum sum rate schedul-
ing algorithm can be computed. As a simple example, if every user has the
probability Pf to feedback the CSI for a particular beam direction to the
BS, and the feedback events are independent of the value of the CSI and
independent from user to user and from beam to beam, the sum rate can be
obtained by
(
∑
K
K!Pf n (1 − Pf )K−n
R(K, Pf , Mt ) = Mt E n
log(1 + X(1) )
n!(K − n)!
n=1
)
+ (1 − Pf )K log(1 + Xk ) , (2.8)
where the first term accounts for the rate using the maximum sum rate
scheduling algorithm when at least one user has fed back the CSI, and the
second term is the rate when no user has fed back to the BS and the BS
randomly schedules one user k on a ceratin beam.
18
2.2 The Multi-threshold Feedback Scheme
In the remainder of this chapter, for conciseness of the presentation, we

will abuse the notation and let SN Rm,k represent not only the SNR in the
Mr ≥ Mt case but also the SINR in the Mr < Mt case. For the scheme
in [13], if the SNR of a user is greater than the outage threshold, the user
feeds back BQ bits to represent the received SNR. Otherwise it does not
feedback. The threshold is derived according to a pre-determined scheduling
outage probability (where “scheduling outage” refers to the situation when
none of the users feeds back), but not directly related to the scheduling
mechanism. Since the maximum sum rate scheduler selects users according
to their SNR orders, it is more meaningful to set the threshold according to
the order statistics of the received SNR.
The basic idea of our proposed scheme is to let a user compare its received
SNR with the thresholds derived from the order statistics. The user can
thus guess its most possible rank among all the users, and, if its rank is
high enough to make its chance to be scheduled high, it feeds back its SNR.
Otherwise the user does not feedback in order to save the reverse link resource
and avoid interfering the other users’ reverse link transmission. Note that
there might be errors in the statistical inference by the individual users about
their SNR ranks. These errors may result in the situation where the users who
actually have high SNRs do not feedback, and the BS does not have proper
users to select from. To make up for the sum rate loss due to this situation, we
allow the users with several (guessed) ranks to feedback. Therefore, for each
beam direction, a set of N thresholds Rth = {rth,1 , rth,2 , . . . , rth,N } is set (see
19
Fig. 2.2) according to the order statistics of the received SNR. Let rth,0 = ∞.
For the SNR region i bounded by the adjacent thresholds as [rth,i , rth,i−1 ), bi
additional quantization bits are used to help the BS differentiate users whose
SNRs fall in that region, and also make better link adaptation if a user in
that region is scheduled. According to the importance of the SNR regions
to the sum rate, bi , i = 1, . . . , N , (to be optimized later) are usually in non-
increasing order. When the received SNR is higher than rth,N , the user feeds
back its rank and the additional quantization bits. Otherwise the user does
not feedback at all.
2.2.1 Derivation of Multiple Thresholds
When user k has on its m-th beam SN Rm,k , Xk = snrm,k (snrm,k is a
value) and the users have i.i.d. SNR distributions, the probability that user
k’s SNR on the m-th beam direction is ranked the p-th among all the users
is
(
(K − 1)!
P {Xk = X(p)
K
|Xk = snrm,k } =
(K − p)!(p − 1)!
)
×{FXm,k (snrm,k )}K−p {1 − FXm,k (snrm,k )}p−1 (2.9)
and satisfies
∑
K
P {Xk = X(p)
K
|Xk = snrm,k } = 1, (2.10)
p=1
where FXm,k (x) is defined in (2.5).

For example, the probability that user k has on its m-th beam SN Rm,k =
20
Region Region Region
N 2 1
0 bN
.. . ... b2 b1
SNR
rth,N rth,N-1 rth,2 rth,1
Figure 2.2: Multi-threshold feedback model.
snrm,k which is the highest SNR among all the users on the m-th beam is
P {X(1)
K
= snrm,k }
P {Xk = K
X(1) |Xk = snrm,k } =
P {Xk = snrm,k }
( )K−1
= FXm,k (snrm,k ) . (2.11)
With SN Rm,k = snrm,k , user k can infer its most possible rank (i.e., the
rank with the highest probability (2.9)) among the users on the m-th beam,
denoted rank(snrm,k ), as
rank(snrm,k ) = arg max P {Xk = X(p)

K
|Xk = snrm,k }. (2.12)
p=1,...,K
From (2.11), it is clear that when snrm,k is sufficiently high, user k’s most
possible rank rank(snrm,k ) will be 1. It will be verified later that as snrm,k
decreases, rank(snrm,k ) will gradually increase.
Fig. 2.3 shows the rank probabilities (2.9) for different ranks over a wide
range of snrm,k for a system with Mr = Mt and K = 20 users. It can be
seen that as as snrm,k decreases, user k’s most possible rank rank(snrm,k )
increases, and each value of rank(snrm,k ) occurs in a continuous region of
snrm,k . To this end, we define the regions in Fig. 2.2 according to the most
possible rank.
21
Mr=Mt, K=20
1
0.9 p=1
0.8 p=2
|Xk=snrm,k)
p=3
0.7
p=4
0.6
p=5
0.5 p=6
(p)
P(Xk=XK
0.4 p=7
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8
snrm,k
Figure 2.3: The probability (2.9) that user k’s SNR on the m-th beam direc-
tion is ranked the p-th among all the users. Mr = Mt , K = 20 users.
22
Definition 1. The SNR region j denoted [rth,j , rth,j−1 ) satisfies that ∀snrm,k ∈
[rth,j , rth,j−1 ), rank(snrm,k ) = j, where rank(snrm,k ) is defined in (2.12).
Thus the number of regions N must be no larger than the number of
users K. By observing Fig. 2.3, the thresholds rth,j , j = 1, 2, . . . , N , can be

determined as the point where the probabilities (2.9) with p = j and p = j +1
equal. In the following, the thresholds are derived.
Due to the symmetric assumption that the users suffer i.i.d. Rayleigh
fading processes, also due to the ORB, the same set of thresholds applies to
all users and all beam directions. By Definition 1, rank(snrm,k ) = j when
snrm,k ∈ [rth,j , rth,j−1 ), and rank(snrm,k ) = j+1 when snrm,k ∈ [rth,j+1 , rth,j ).
Since Xk with its distribution given in (2.5) is a continuous random variable,
the probability P {Xk = X(p)

K
|Xk = snrm,k } in (2.9) is a continuous function
of snrm,k for p = 1, 2, . . . , K. Using (2.9) and (2.12), we have
P {Xk = X(j)
K
|Xk = rth,j } = P {Xk = X(j+1)
K
|Xk = rth,j } (2.13)
FXm,k (rth,j ) 1−FXm,k (rth,j )
⇔ K−j
= j
⇒ FXm,k (rth,j ) = 1 − j
K
, j = 1, 2, . . . , N. (2.14)
This result shows that rth,1 > rth,2 > · · · > rth,N , which, together with Defi-
nition 1, verifies that as snrm,k decreases, rank(snrm,k ) gradually increases.
When Mr = Mt , we have FXm,k (x) = 1 − e− ρ , and from (2.14)

x
( )
K
rth,j = ρ ln . (2.15)
j
It can be observed that the value of rth,j is insensitive to the number of users
K when K is large, as in practical systems. If FXm,k (x) does not have a

closed-form expression (for example, in the Mr < Mt case in (2.5)), (2.14)
23
can still be used to numerically search for the thresholds. From (2.14) and
the nondecreasing property of a CDF, we have
j
FXm,k (x) ≥ 1 − , when x > rth,j ,
K
j
FXm,k (x) ≤ 1 − , when x < rth,j , (2.16)
K
or equivalently (see the example in Fig. 2.3),
P {Xk = X(j)
K
|Xk = x} ≥ P {Xk = X(j+1)
K
|Xk = x}, when x > rth,j ,
P {Xk = X(j)
K
|Xk = x} ≤ P {Xk = X(j+1)
K
|Xk = x}, when x < rth,j . (2.17)
Thus each threshold can be efficiently searched by bisection. In addition,
each threshold rth,j found can serve as the initial upper boundary (and zero
is an obvious lower boundary) for the search of the next threshold rth,j+1 .
From (2.14), it can be obtained that
1
P {Xk ∈ [rth,j , rth,j−1 )} = , j = 1, 2, . . . , N. (2.18)
K
In other words, the probability for a user to infer itself as being ranked the
1
j-th place on a certain beam direction is K
, for all j = 1, 2, . . . , N , with
N ≤ K. This is very intuitive because each user has the same probability
1
K
of being ranked the j-th place, j = 1, 2, . . . , K, due to the symmetric
assumption of the users’ SNR distributions. Therefore, the probability Pf
for a user to feedback is
N
Pf = 1 − FXm,k (rth,N ) = , (2.19)
K
which decreases with the number of users.
Note that the thresholds can be computed off-line as long as the number
of users and the channel statistics are known. In practice when the network
24
operating conditions vary with time, the values of the thresholds can be
updated periodically by the BS according to the system configuration and
channel statistics, and broadcasted to the users. Or, the users can compute
the threshold values by themselves after receiving the information of network
configuration and channel statistics broadcasted by the BS.
With the thresholds set, if a user finds its SNR on a beam lower than
rth,N , then no feedback is sent for that beam. Otherwise, the user feeds
back BR = ⌈log2 (N )⌉ bits to indicate its most possible rank on that beam.
In order to account for the situation where there are more than one users
reporting to have the same rank, each region j is further quantized with bj
bits which are also fed back together with the “rank” bits.
2.2.2 Sum Rate Loss Analysis and Minimum Number

of Regions
We now consider the sum rate loss incurred by the reduction of feedback.
The sum rate loss of the proposed scheme can be attributed to two sources:
one resulting from the situation when none of the users feeds back, and the
other being the feedback quantization error. In general, when the number
of regions increases, which results in higher feedback load, the sum rate loss
decreases. Thus, for a given tolerable sum rate loss, we should apply the
minimum number of regions required in order to minimize the feedback load.

On the other hand, we also need to consider the number of quantization bits
bj for each region. Optimal joint design of the number of regions N and the
number of quantization bits for each region is quite difficult. Thus, we will
resolve the feedback design issue by a more tractable two-stage approach.
25
First, we will analyze the sum rate loss incurred when none of the users
feeds back, and determine the minimum number of regions N required to
meet the given tolerable sum rate loss of the system without considering the
quantization in each region.
Then, in Section 2.3, quantization of the regions will be considered by
allocating bits for quantizing the regions to maximize the sum rate.
Definition 2. A rate loss event occurs when all users’ SNRs on a certain
beam is smaller than the lowest threshold rth,N . (Note: In [13], the rate loss
event is called the scheduling outage event.)
When a rate loss event occurs, the BS will randomly schedule one user.
The probability of the rate loss event, denoted PL (K), is

( )K
( )K N
PL (K) = P {X(1)
K
< rth,N } = FXm,k (rth,N ) = 1− . (2.20)
K
Without loss of generality, assume that user k is selected by the BS in a rate

loss event. The sum rate loss compared to the case when the users always
feed back is
△R(K, N, Mt )
{ }
= Mt E log(1 + X(1)
K
) − log(1 + Xk ) | X(1)
K
< rth,N PL (K) (2.21)
{ ( ( K )) }
≤ Mt E log 1 + X(1) − Xk | X(1) K
< rth,N PL (K)
( {( K ) })
≤ Mt log 1 + E X(1) − Xk | X(1)
K
< rth,N PL (K)
, △RU (K, N, Mt ), (2.22)
where the inequalities are due to the convexity of the rate function and
Jensen’s inequality. By using (2.5), (2.14), and considering the Mr ≥ Mt
26
case as an example,
{( K ) K }
E X(1) − Xk |X(1) < rth,N
∫ rth,N ∫ rth,N
K−1
xKfXm,k (x)FXm,k (x) dx xfXm,k (x)dx
= 0 ( )K − 0 ( ) (2.23)
1− K N 1− K
N
∫ rth,N [ Mr ∑
−Mt +1 ]K−1
Mr −Mt +1 −x/ρ −x/ρ (x/ρ)m
K[(Mr − Mt )!] K−2
x e 1−e dx
0 m=0
m!
= ( )K
ρMr −Mt +1 1 − N
K
[ Mr ∑
−Mt +1 ]
−rth,N /ρ (rth,N )k
ρ(Mr − Mt + 1) 1 − e
ρk k!
− ( ) k=0
.
1− N
K
Using (2.22) and (2.23), the minimum number of regions required for a given
tolerable sum rate loss △RP (K) can be approximated by the minimum N
( ) ∆RP (K)
that satisfies E (X(1) − Xk ) | X(1) < rth,N < 2 Mt PL (K) − 1.
K K
The following theorem gives the relationship between the sum rate loss
and the number of users.
Theorem 1. Given Mt and N , the sum rate loss △R(K, N, Mt ) for the
multi-threshold feedback scheme increases with the number of users K.
Proof. In Appendix A.
This theorem implies that the number of thresholds should be increased

as the number of users increases in order to keep the sum rate loss smaller
than the tolerance of the network. For the Mt = Mr = 4 example, Fig. 2.4
compares the sum rate loss upper bound with the actual sum rate loss ob-
tained by simulation, and shows that the sum rate loss upper bound (2.22)
27
is tight (within 0.1 bps/Hz) when the number of regions is large. This fig-
ure also shows that the required number of thresholds is insensitive to the
number of users. Four regions are enough to keep the sum rate loss smaller
than the tolerable sum rate loss △RP (K) = 0.25 bps/Hz when the number
of users is up to 500. Using the sum rate loss upper bound to design the
system, we can see that the number of regions can be reduced to three when
the number of users is smaller than 16. When the number of users is smaller
than 5, two regions are enough.
28
0.7
∆R(K,N=3,Mt)
0.6
∆R(K,N=4,Mt)
∆RU(K,N=3,Mt)
0.5
∆RU(K,N=4,Mt)
∆RP(K)=0.25 bps/Hz)
Sum rate loss
0.4
0.3
0.2
0.1
0
0 50 100 150 200 250 300 350 400 450 500
Number of users
Figure 2.4: Sum rate loss versus the number of users. △RP (K) = 0.25
bps/Hz is the tolerable sum rate loss. ρ = 10 dB, Mt = Mr = 4.
29
Fig. 2.5 shows the simulation results of the sum rate for different numbers
of regions when the number of users increases. In this figure, the “full CSI”
case refers to the ORB system in which the users always feedback their SNRs
on all beams. It is shown that the sum rate increases with both the number
of users and the number of regions.
Under the ORB and maximum sum rate scheduling, when the number
of regions is larger than four, the sum rate achieved by the multi-threshold
scheme is very close to the sum rate with full CSI feedback (within 0.25
bps/Hz according to Fig. 2.4). An intuitive interpretation of this result is
that, on average, providing four users with the highest SNRs on each beam
to the BS suffices for the purpose of maximum sum rate scheduling.
30
16
15
14
Sum rate
13
Full CSI
12
N=1
N=2
N=3
11
N=4
N=5
N=6
10
0 20 40 60 80 100 120 140 160 180 200
Number of users
Figure 2.5: Sum rate performance of the multi-threshold scheme with differ-
ent numbers of regions. ρ = 10 dB, Mt = Mr = 4.
31
2.2.3 Multiuser Diversity Using the Multi-threshold
Scheme
In this section, we characterize the proposed multi-threshold scheme’s

asymptotic sum rate scaling with respect to the number of users, that is, the
multiuser diversity. The sum rate using this scheme can be expressed as
{ }
R(K, N, Mt ) = Mt P {rth,N ≤ X(1)
K
< ∞}E log(1 + X(1) K
) | rth,N ≤ X(1)
K
<∞
{ }
+ Mt P {0 ≤ X(1)
K
< rth,N }E log(1 + Xk ) | 0 ≤ X(1)
K
< rth,N .
(2.24)
When the number of users is large, this sum rate exhibits the following
property.
Theorem 2. Let Mt , ρ, N be given, and the lowest threshold be set as rth,N =
FX−1m,k (1 − N
K
). For both Mr ≥ Mt and Mr < Mt , the lower bound of achiev-
able sum rate of the multi-threshold feedback scheme, denoted RL (K, N, Mt ),
satisfies
RL (K, N, Mt )
lim = 1 − e−N .
K→∞ Mt log log K
Proof. In Appendix B.
When Mt = Mr , the sum rate of the multi-threshold feedback scheme can

be more precisely characterized by the following theorem.
Theorem 3. Let Mt = Mr , ρ, N be given, and the lowest threshold rth,N =

FX−1m,k (1 − N
K
) = ρ ln(K/N ). The achievable sum rate R(K, N, Mt ) of the
multi-threshold feedback scheme satisfies
R(K, N, Mt )
lim = 1 − e−N .
32
Proof. In Appendix C.
The above two theorems show that with given Mt , ρ and N , when the
number of users K is large, the multi-threshold feedback scheme can achieve a
sum rate which scales like (1−e−N )Mt log log(K). In other words, this scheme
can asymptotically achieve a constant portion (1 − e−N ) of the optimal sum
rate Mt log log(K) achievable with full CSI feedback. The remaining portion,
i.e., the sum rate loss, decreases exponentially to zero as the number of
regions N increases. In fact, e−N equals the probability of the rate loss event
PL (K), given in (2.20), when the number of users is large. This result can
be observed from Fig. 2.5 where it is shown that the sum rate loss is already
very small when the number of regions is four.
The asymptotic upper bound of the sum rate loss, e−N Mt log log(K), also
offers an intuitive explanation of why N = 4 regions are good enough for a
wide range of the number of users as shown in Fig. 2.4. Note that asymptot-
ically the sum rate loss decreases exponentially with N and increases with
log log(K). According to Fig. 2.4, as the minimum required N increases from
3 to 4 (when K becomes larger than 16), the amount of sum rate loss this
reduces is quite large. This sum rate loss reduction will not be depleted until
K is very large.
33
2.3 Bit Allocation and Feedback Load Anal-
ysis
In this section, we consider practical quantization and feeding back the
CSI values with finite numbers of bits. Assume that the users use BR bits to
represent the region information, and additional bj bits to quantize the SNR
when it falls in region j. On a given beam, whenever there are other users
feeding back the same rank indication and the same additional quantized bits
as the user who actually has the highest SNR, the BS will randomly schedule
one of them. As a result, the lowest possible rate due to this ambiguity
in scheduling will be the rate derived from the lower boundary of the SNR
quantization region in question. Let B = (bN , bN −1 , . . . , b1 ) be the vector
of the numbers of bits for quantizing the SNR in regions N, N − 1, . . . , 1,
respectively. We assume the optimal nonuniform quantization [32] [33] for
each region. The sum rate Rq (B) with both rank and SNR quantization
feedback can be lower bounded by

∑N ∑2bj ∫ rth,j,t+1
Rq (B) > Mt log (1 + rth,j,t ) fX(1)
K (x)dx, (2.25)
j=1 t=1 rth,j,t
where rth,j,1 , rth,j,2 , . . . , rth,j,2bj +1 are the quantization levels in rank region
j, with rth,j,1 = rth,j , rth,j,2bj +1 = rth,j−1 . In Fig. 2.6, the analytical lower
bound of the sum rate (2.25) is compared to the simulated sum rate for a
few numbers of SNR quantization bits. For each number of quantization
bits, the bit allocation to the regions is obtained by exhaustive search to
maximize the sum rate. Also shown in this figure is the “full CSI” case in
which all users always feedback perfect (unquantized) SNRs of all beams. For
the cases tested, the analytical lower bound of the sum rate almost matches
34
the simulated sum rate. Thus the lower bound (2.25) is very tight, and can
be used to analytically allocate available SNR quantization bits to quantize
each region to achieve the maximum sum rate, as follows.

1
Because all users have the same probability K
of inferring itself as being
ranked the jth, j = 1, 2, . . . , N , on each beam direction, the expected number

∑ bj
of feedback bits required in addition to the rank bits is N j=1 K . Define BQ =
∑N
j=1 bj which is proportional to the average number of SNR quantization
bits per beam per user (and equal to the average number of SNR quantization
bits per beam for the entire system, as computed in (2.37)). Constrained on
BQ , the bit allocation problem of the N regions for a given beam m can be
formulated as
max Rq (B)
B
∑
N
s.t. bj = BQ , bj ∈ Z+ . (2.26)
j=1
2.3.1 Optimal Bit Allocation with Given Thresholds
The optimization problem (2.26) can be solved by the greedy algorithm,

which is to assign one bit at a time to the region that will result in the
maximum sum rate, because adding one more bit to any of the regions will
increase the average feedback load by the same amount.
For the s-th single bit assigning iteration, the sum rate difference between
using bl,s−1 bits and bl,s−1 + 1 bits for region l can be expressed as
△Rl (s) = Rbl l,s−1 +1 − Rbl l,s−1 , l = 1, 2, . . . , N, (2.27)
where bl,s−1 is the number of bits for quantizing region l, resulting from the
35
16
15
14
Sum Rate
13
Full CSI
Lower bound of sum rate B=(0,0,0,3)
12
Simulation B=(0,0,0,3)
11 Simulation B=(0,0,0,4)
Simulation B=(0,0,1,4)
10
9
0 10 20 30 40 50 60 70 80 90 100
Number of users
Figure 2.6: Sum rate comparison between the mathematical lower bound and
the simulation result, ρ = 10 dB, Mt = Mr = 4.
36
(s − 1)th bit assigning iteration. Rm
l
is the sum rate of region l using m
quantization bits, and can be approximated by
∑2m ∫ rth,l,j+1
Rm ≈ Mt
l
log (1 + rth,l,j ) fX(1)
K (x)dx (2.28)
j=1 rth,l,j
as in the sum rate lower bound (2.25) which has been shown to be very tight
in Fig. 2.6. The region which gives the maximum sum rate increment with
one additional bit will be assigned one more bit at the s-th iteration. The
algorithm iterates until all the available quantization bits are allocated, i.e.,
when s = BQ . The greedy algorithm is summarized in Table 2.1 where the
sum rate increment (2.27) is approximated using (2.28).
Table 2.1: The greedy algorithm for bit allocation.

Initialize bl = 0, l = 1, 2, . . . , N.
For (s = 1 to BQ )
l = arg max △ Rt (s)
t=1,2,...,N
bl = bl + 1
End
2.3.2 Fast Bit Allocation Method
Despite that the sum rate lower bound (2.25) is analytical, it is difficult
to compute. We alternatively consider minimizing the mean square quanti-

zation error as a suboptimal but simple solution. The conditional PDF of
the SNR in region j for a given beam m is
fXm,k (x)
fXm,k (x | rth,j ≤ x < rth,j−1 ) =
P (rth,j ≤ x < rth,j−1 )
= KfXm,k (x), rth,j ≤ x < rth,j−1 .(2.29)
The SNR variance in region j can be expressed as

( ∫ rth,j−1 ) ( ∫ rth,j−1 )2
2
2
σx,j = x fXm,k (x)dx − xfXm,k (x)dx .
rth,j rth,j
37
Thus, the variance of the quantization error using bj bits can be bounded
by [34]
2
σx,j
2
σe,j ≤ ϵ 2bj ,
2
(2.30)
2
where the constant ϵ is source dependent. For example, ϵ = 1.0 for uniformly
distributed sources and ϵ = 1.4 for Gaussian sources. In our case, the SNR
PDFs for different regions are different, thus the ϵ that gives the tightest
bound (2.30) will be different for different regions. In order to simplify the
computation, we set the same ϵ for all regions such that the upper bound
(2.30) is always valid. Note that this simplification is reasonable when K
is large because in that case, the probability of each region (2.18) becomes
very small, which implies that each region interval is very small, and the
SNR distribution within each region is almost uniform. We further relax the
constraint for the number of quantization bits in (2.26) from being a positive
integer to being a positive real number. Then, a new bit allocation problem
based on minimizing the upper bound of the variance of the quantization

error can be formulated as
∑
N 2
σx,j
min
B=(bN ,bN −1 ,...,b1 )
j=1
22bj

 ∑N

 bj = BQ

 j=1
s.t. 0 ≤ bj ≤ BQ , j = 1, 2, . . . , N , (2.31)




 bj ∈ R+
where in the objective function, the same constant ϵ and same probability
1/K for all regions are dropped for conciseness without changing the problem.
Since the optimization problem (2.31) is convex, we can apply the Karush-
38
Kuhn-Tucker (KKT) conditions [35] to solve it. To simplify the expression
in (2.31), we let
∑
N ∑
N
L(B, λ, ν1 , . . . , νN , δ1 , . . . , δN ) = f0 (B) + λf1 (B) + νj hj (B) + δj qj (B)
j=1 j=1
(2.32)
where

 ∑

 f0 (B) = N 2
j=1 σx,j 2
−2bj



 f1 (B) = ∑N bj − BQ

j=1

 hj (B) = −bj





 q (B) = b − B .
j j Q
Since f0 , f1 , hj , qj are differentiable, the KKT conditions for this problem are


 ∂L(B, λ, ν1 , . . . , νN , δ1 , · · · , δN )

 = 0, j = 1, 2, . . . , N,

 ∂bj

 ∑N

 b = BQ

 j=1 j
 λ ̸= 0 (2.33)





 νj hj (B) = 0, δj qj (B) = 0, j = 1, 2, . . . , N,




 νj ≥ 0, δj ≥ 0, j = 1, 2, . . . , N.
∂L(B, λ, ν1 , · · · νN , δ1 , · · · δN )
From = 0, we have
∂bj
νj = (−2 ln 2)2−2bj σx,j

2
+ λ + δj . (2.34)
Substitute (2.34) into the fourth condition in (2.33). By considering the

{νj = 0, δj = 0, 0 < bj < BQ }, {νj = 0, δj > 0, bj = BQ } and {νj > 0, δj =
0, bj = 0} cases separately, and defining W as
( N )
1 ∑
W = Tj − V , (2.35)
N j=1
39
( )
2
ln (ln 4)σx,j BQ ln 4
where Tj = and V = , bj can be obtained by
ln 10 ln 10




 0, W > Tj

(Tj − W ) ln 10
bj = , Tj − V < W < T j (2.36)

 ln 4


 BQ , W < Tj − V.
The obtained bj ’s are then rounded to be nonnegative integers.
40
16
15
14
Sum rate
13
12
Full CSI
11 Proposed Scheme C, BQ=3, greedy algorithm
Proposed Scheme C, BQ=5, greedy algorithm
10 Proposed Scheme C, BQ=3, fast bit allocation
Proposed Scheme C, BQ=5, fast bit allocation
9
0 10 20 30 40 50 60 70 80 90 100
Number of users
Figure 2.7: Sum rate performance comparison for different bit allocation
methods with N = 4 regions, ρ = 10 dB, Mt = Mr = 4.
Fig. 2.7 compares the simulated sum rate performance using the greedy
and the fast bit allocation methods. It is shown that the sum rate perfor-
mances for these two bit allocation methods are visually indistinguishable.
Thus, the fast bit allocation method is preferred for all practical purposes.
For the cases considered, the bit allocations obtained by these two methods
are the same as the exhaustively searched bit allocation shown previously
in Fig. 2.6. The general trend is that more bits are allocated to higher
41
SNR regions. When BQ is small, the optimal bit allocation has the form
B = (0, 0, . . . , BQ ). This is reasonable because given the same probability
of a user’s SNR falling in each region, the region [rth,1 , ∞) is the thin tail of
the PDF and has a wider range. For this wide range of SNR which results in
higher rate than the other (lower) regions, it is more important to differenti-
ate the users so the user with the highest rate is more likely to be selected.
Thus, more bits should be allocated to this region.
2.3.3 Complexity Comparison
Let L be the number of quantization levels for approximating the integration
operation in (2.28), the complexity of the greedy algorithm is O(L(2(N −
1)BQ + 2BQ − 2)). On the other hand, the fast bit allocation algorithm only
needs to calculate the variance in each region, and the bit allocation vector is
obtained by (2.36). Thus, the complexity of the fast bit allocation algorithm
is O(LN ) which is lower.
2.3.4 Feedback Load Analysis
Let BR be the number of feedback bits carrying the rank information and BQ
be defined in (2.26). For the multi-threshold feedback scheme, the average

number of feedback bits for the network when the number of users is K can
be expressed as
∑N { }
1
F b = KMt (BR + bj ) = Mt (N BR + BQ ) (2.37)
j=1
K
which does not increase with the number of users, and is a constant when the
number of transmission beams Mt and the number of regions N are fixed.
42
This is in contrast to the conventional feedback schemes whose total feedback
loads for the network increase with the number of users.
2.4 Numerical Results
In this section, we compare different feedback schemes in terms of the

sum rate and feedback load performance using simulation. The transmitter
is equipped with Mt antennas and there are K users each having Mr antennas.
For the conventional feedback scheme, named Scheme A, each user always
feeds back to the BS the SNR values of the Mt beams. A reduced feedback
scheme was proposed in [11] [36] where each user only feeds back its largest
SNR value among all beams and the corresponding beam index. We refer
to this scheme as Scheme B. The feedback loads of both Scheme A and
Scheme B increase with the number of users. The multi-threshold scheme
we propose is referred to as the Proposed Scheme C. The single threshold
feedback scheme proposed in [13] will be called Scheme D. For that scheme,
each user feeds back the SNR value of a beam direction when the SNR
is greater than the threshold. In [13], the threshold is determined by the
scheduling outage probability Pout which is the probability that none of the
users feeds back. In the performance comparison, we additionally introduce a
slightly modified Scheme D based on the design philosophy proposed in this
paper by setting the threshold as rth,N of the Proposed Scheme C, such
that the Pout of this scheme equals the probability of rate loss event PL of the
Proposed Scheme C in (2.20). Thus, in the comparison, we will consider
( )
N K
Scheme D with constant Pout = 10−1 , 10−4 , and Pout = PL (K) = 1 − K .
In the simulation, Scheme A and Scheme B use BQ,A and BQ,B bits, re-
43
spectively, to quantize their SNR values in the region [0, ∞). The “full CSI”
case is Scheme A with infinite quantization bits (i.e., the SNR feedbacks
∑
are unquantized). The Proposed Scheme C has BQ = N j=1 bj bits allo-
cated to N = 4 regions using the fast bit allocation method in Section 2.3.2.
The number of regions is chosen to guarantee the sum rate loss upper bound
in (2.22) smaller than the system tolerable sum rate loss △RP (K) = 0.25
bps/Hz. For Scheme D, BQ,D bits are used to optimally quantize the region
[threshold, ∞) where the threshold depends on the scheduling outage prob-

ability Pout . Note that for the Proposed Scheme C and Scheme D, the
threshold values increase with number of users. As a result, the SNR ranges
of the regions change. Thus the bit allocation and quantization of each region
(single region for Scheme D) also depend on the number of users.
Fig. 2.8 compares the sum rates of different feedback schemes as the
number of users increases. The numbers of SNR quantization bits defined
above for different schemes are set as five. Note that for different schemes,
the relationships between the number of SNR quantization bits and the total
feedback load are different. Therefore, Fig. 2.8 is shown only to illustrate
the performance difference between similar schemes. With the same number
of SNR quantization bits, Scheme A’s total feedback load is roughly four
times that of Scheme B. Thus it is reasonable that Scheme A’s sum rate is
higher than that of Scheme B, with the sum rate difference getting smaller
as the number of users increases. This is because when the number of users is
large, feeding back only the largest SNR among all beams is good enough for
the purpose of scheduling. For Scheme D, setting the threshold such that
Pout = 10−4 results in higher sum rate compared to setting the threshold as
44
rth,4 when the number of users is large. This is because the scheduling outage
probability of the latter increases with the number of users, and is higher
than that of the former when the number of users is large. With BQ = 5, the
Proposed Scheme C’s average number of feedback bits is Mt (N BR +BQ ) =
52 which is less than Mt N BQ,D = 80 of Scheme D using the threshold rth,4
(i.e., Pout of Scheme D = PL of the Proposed Scheme C). Thus the

Proposed Scheme C’s sum rate is lower than that of Scheme D.
Fig. 2.9 shows the average total feedback loads of the cases considered
in Fig. 2.8, and confirms the above discussion on the numbers of feedback
bits for similar schemes. For example, the feedback loads of Scheme A
and Scheme B grow linearly with the number of users, and the slope of
Scheme A is four times that of Scheme B because Scheme A’s users feed-
back the SNR of every beam. The Proposed Scheme C and Scheme D
with threshold rth,4 have constant feedback loads as discussed in Section 2.3.4.
On the other hand, Scheme D with constant scheduling outage probability
(Pout = 10−1 , 10−4 ) has its feedback load increasing with the number of users,
but saturating when the number of users is high. This is because when Pout
1/K
is fixed, Scheme D’s threshold is −ρ ln(1 − Pout ). When the number of
( )
1/K
users is large, Scheme D’s feedback load is limK→∞ Mt BQ K 1 − Pout =
Mt BQ ln(1/Pout ). Thus the feedback load behaviors of the three Scheme Ds
are similar when the number of users is large.
For fair comparison between the feedback schemes, the results of sum rate
and feedback load are combined to show the sum rate as a function of the
feedback load when Mr = Mt = 4 in Fig. 2.10. As shown in Fig. 2.10, for
Scheme A and Scheme B, the feedback load has to be increased if higher
45
sum rate is desired. The Proposed Scheme C and Scheme D with rth,4 as
the threshold (Pout = PL ), which is based on the same design philosophy as
the Proposed Scheme C, have much lower and fixed feedback loads as their
sum rates grow like (1−e−N )Mt log log(K). It can be seen that, to achieve the
same sum rate, the Proposed Scheme C requires lower feedback load than
Scheme D. Note that, based on the design philosophy in [13], Scheme D
with constant scheduling outage probability has its feedback load increasing
with the sum rate, like Scheme A and Scheme B, for smaller number of
users. In fact, if the scheduling outage probability is set to zero, Scheme D
will become exactly the same as Scheme A. When the number of users is
large, the feedback load of Scheme D saturates, making it behave more like
Proposed Scheme C. When Scheme D’s Pout is large, its sum rate loss is
also large.
Fig. 2.11 shows the sum rate as a function of the feedback load when
Mr = 4 > Mt = 2. Similar trends as in Fig. 2.10 are observed. Since only
Mt = 2 beams are transmitted simultaneously, the sum rate in Fig. 2.11 is
smaller than the sum rate in Fig. 2.10, and the feedback load in Fig. 2.11 is
half the feedback load in Fig. 2.10.

Fig. 2.12 shows the sum rate as a function of the feedback load under
different transmission powers with Mt = 4 > Mr = 1. Since the inter-beam
interference can not be avoided, the sum rate is significantly degraded as
compared to the Mr = Mt = 4 case. Increasing the transmission power
improves the sum rate, but not much, because this system is interference
limited.
46
16
15
14
Sum rate
13
Full CSI
Scheme A BQ,A=5
Scheme B BQ,B=5
12
Proposed Scheme C, BQ=5 (N=4)
Scheme D BQ,D=5, Pout=PL
11 Scheme D BQ,D=5, Pout=10-4
Scheme D BQ,D=5, Pout=10-1

10
10 20 30 40 50 60 70 80 90 100
Number of users
Figure 2.8: Sum rate performance comparison for different feedback schemes.
ρ = 10 dB, Mt = Mr = 4.
47
2000
1800 Scheme A, BQ,A=5

Scheme B, BQ,B=5
1600
Proposed Scheme C, BQ=5 (N=4)
Number of feedback bits
1400
Scheme D, BQ,D=5, Pout=PL
1200 Scheme D, BQ,D=5, Pout=10-4
1000 Scheme D, BQ,D=5, Pout=10-1
800
600
400
200
0
10 20 30 40 50 60 70 80 90 100
Number of users
Figure 2.9: Feedback load comparison for different feedback schemes. ρ = 10
dB, Mt = Mr = 4.
48
15
14.5
14
13.5
13
Scheme A, fast algorithm
Sum rate
12.5
Scheme B, fast algorithm
Proposed Scheme C (N=4), fast algorithm
12 Scheme D, Pout=PL, fast algorithm
11.5 Scheme D, Pout=10-4, fast algorithm
11 Scheme D, Pout=10-1, fast algorithm
10.5
10
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Figure 2.10: Sum rate as a function of the feedback load. BQ,A = BQ,B =
BQ = BQ,D = 5, ρ = 10 dB, Mt = Mr = 4.
49
11.5
11
10.5
Scheme A, fast algorithm

Sum rate
10 Scheme B, fast algorithm

Proposed Scheme C (N=4), fast algorithm
Scheme D, Pout=PL, fast algorithm
9.5
Scheme D, Pout=10-4, fast algorithm

9 Scheme D, Pout=10-1, fast algorithm
8.5
8
0 100 200 300 400 500 600 700 800 900 1000
Figure 2.11: Sum rate as a function of the feedback load. BQ,A = BQ,B =
BQ = BQ,D = 5, ρ = 10 dB, Mt = 2, Mr = 4.
50
10
9
Scheme A, ρ=10
Scheme B, ρ=10
8 Proposed Scheme C (N=4), ρ=10
Scheme D, Pout=PL, ρ=10
7 Scheme D, Pout=10-4, ρ=10
Sum rate
Scheme D, Pout=10-1, ρ=10

6
Scheme A, ρ=20
Scheme B, ρ=20
5 Proposed Scheme C (N=4), ρ=20
Scheme D, Pout=PL, ρ=20
4 Scheme D, Pout=10-4, ρ=20
Scheme D, Pout=10-1, ρ=20

3
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Figure 2.12: Sum rate as a function of the feedback load for different trans-
mission powers (ρ). BQ,A = BQ,B = BQ = BQ,D = 5, Mt = 4, Mr = 1.
51
2.5 Summary
In this chapter, we proposed a multi-threshold feedback scheme for the sched-

uled multiuser MIMO downlink channel to opportunistically reduce the ag-
gregate feedback load. The minimum number of thresholds required for a

given tolerable sum rate loss was found. The multiuser diversity using the
proposed multi-threshold scheme was also investigated. Finally, the opti-
mal bit allocation and a fast bit allocation algorithm for the multi-threshold
scheme were discussed. Analytical and simulation results showed that the
proposed multi-threshold feedback scheme can reduce the feedback load and
utilize the limited feedback bandwidth more effectively than the existing
feedback methods. In particular, while keeping the aggregate feedback load
of the entire system constant regardless of the number of users, the proposed
scheme almost achieves the optimal asymptotic sum rate scaling with respect
to the number of users (i.e., the multiuser diversity). Its sum rate loss com-
pared to the system with full CSI feedback decreases exponentially to zero
as the number of thresholds increases.
52
Chapter 3
Reduction Method for

Multiuser MIMO Zero-forcing
Beamforming using
Semi-orthogonal Scheduling
Algorithm in Broadcast
Channel
The performance of low complexity ORB scheme discussed in previous
chapter is limited by the inter-user interference. Although the interference
can be canceled by the multiple received antennas. However, due to limited
size of the mobile device, the user equipped with multiple received antennas is
not a realistic assumption. Another way to reduce interference is through the
user selection method that selects the users with minimum inter-user inter-
ference [37]. But, the high computational complexity is required to find the
optimal scheduled users. In this chapter, the ZF-BF using semi-orthogonal
user selection (SUS) algorithm is considered to improve the performance.
Each user needs to feedback the channel direction information (CDI) and
53
channel quality indicator (CQI) to the BS and the scheduler applies the SUS
algorithm to select users that satisfy the semi-orthogonal constraint to each
other. Obviously, aggregated feedback load increases with number of users.

Thus, an efficient feedback method is required to reduce the feedback load.
Besides, the actual SINR is difficult to be obtained when the SUS algorithm
is used. This is due to the fact that the BS does not have the actual CDI
between users and the BS and the users other than the scheduled users do
not have the information of the beamforming vectors of the scheduled users.
A lower bound of the SINR metric was derived in [26] as an indicator to
represent the received signal quality. However, the sum rate derived using
the SINR lower bound does not tightly match the actual sum rate when the
semi-orthogonal constraint becomes large.
In this chapter, an approximate sum rate is derived that matches the
actual sum rate even when the semi-orthogonal parameter ϵ of the SUS algo-
rithm is large. Moreover, a novel multi-threshold feedback scheme is proposed
to reduce the feedback load. These thresholds are derived through the car-
dinality of the candidate user sets of SUS algorithm. With these thresholds,
only when the user’s estimated SINR is greater than the smallest threshold,
the user feedbacks CDI and CQI to the BS. Through simulations, the pro-
posed scheme not only exploits full multiuser diversity but also efficiently
utilizes the feedback resources.
This chapter is organized as follows. Section 3.1 describes the system
model. The SUS scheduling algorithm is introduced and an approximated
sum rate using SUS algorithm is derived in Section 3.2. A novel multi-
threshold feedback scheme is proposed in Section 3.3. The simulation results
54
are shown in Section 3.4 and we conclude this chapter in Section 3.5.
3.1 System Model
The multiuser MIMO downlink system includes a BS with Mt antennas
and K users each with single antenna. The received signal of the k-th user
in the scheduled user set S = {π(1), . . . , π(|S|)} can be expressed as
∑
yk = hk x + nk = hk vk sk + hk vj sj + nk , k ∈ S, (3.1)
j̸=k
∑
where x = k∈S vk sk ∈ CMt ×1 is the transmit symbol vector related to the
information symbol sk , k ∈ S via linear beamforming under the average
power constraint E[∥x∥2 ] = P . The entries of the channel vector hk ∈ C1×Mt
and the noise nk are i.i.d. complex Gaussian random variables with zero
mean and unit variance. The received SINR of user k is
(P/|S|)|hk vk |2
SIN Rk = ∑ , k ∈ S. (3.2)
1 + (P/|S|) k̸=j |hk vj |2
Thus, the sum rate of the scheduled user set S is given by
∑
R(S) = log(1 + SIN Rk ). (3.3)
k∈S
3.2 Sum Rate Analysis Using SUS Algorithm

3.2.1 User Selection
We assume that each user has perfect channel knowledge of hk and quan-
tizes the direction of normalized channel h̃k = hk /∥hk ∥ to a vector ĥk . The
quantization vector hk⋆ is chosen from the codebook Wk = {wk,1 , . . . , wk,2BD }
55
CQI,CDI CDI
User
1
User 1 User (1)
User 2 User (2)

User
User ZFBF

2
Selection

User K User (|S|)

Base Station
User
M antennas K
CQI,CDI CDI
Figure 3.1: System model of zero-forcing beamforming using SUS algorithm.
such that wk,j ⋆ = ĥk⋆ , where
j ⋆ = arg max |wk,j h̃∗k |. (3.4)

1≤j≤2BD
The codebook design is based on random vector quantization scheme (RVQ)

[38], and each Wk for user k is known both at the BS and user k a prior.
The user feeds back the SINR and index of codebook perfectly and instan-
taneously to the BS. At the BS, the scheduled users are selected through
the SUS scheduling algorithm under semi-orthogonal constraint ϵ which is
described in [9]. The BS selects the first scheduled user π(1) from the initial
user set B0 = {1, 2, . . . , K} as
π(1) = arg max SIN Rk . (3.5)

k∈B0
56
After selecting number of (i − 1) scheduled users, the i-th scheduled user π(i)
selected from the candidate user set Bi = {l|1 ≤ l ≤ K , l ∈
/ {π(1), . . . , π(i −
1)}, : |ĥl ĥ∗π(j) | ≤ ϵ, 1 ≤ j ≤ (i − 1)} according to
π(i) = arg max SIN Rk , (3.6)

k∈Bi
where ϵ is a design parameter that indicates the maximum spatial correlation

allowed between quantized channels. When ZF-BF is applied at the BS, the
beamformer vπ(i) is obtained by normalizing the i-th column of V which can
be obtained by pseudo inverse of Ĥ(S)
V(S) = [vπ(1) , . . . , vπ(|S|) ] = Ĥ(S)∗ (Ĥ(S)Ĥ(S)∗ )−1 , (3.7)
where Ĥ(S) = [ĥTπ(1) , . . . , ĥTπ(|S|) ]T .
3.2.2 CQI Metric
Because the SINR in (3.2) is unknown either at the BS nor the users, the
expected SINR conditioned on h̃k and ĥk is used as a CQI metric to estimate
SINR [9] [39]. Let θk be the angle between hk and ĥk . Then the h̃k can be
decomposed into
h̃k = cos(θk )ĥk + sin(θk )gk . (3.8)
Note that the gk and vj ∈ V(S) in (3.7) are unit vector on the (Mt − 1)
dimensional hyperplane orthogonal to ĥk , j ̸= k. Since vj is solely deter-
mined by ĥj , j ̸= k. Thus, the inner product |gk vj |2 is Beta distributed with
parameters (1, Mt − 1) and its expected mean value is 1/(Mt − 1). For a
57
given h̃k and ĥk , the expected SINR at user k is lower bounded by
( )
ρ∥hk ∥2 |h̃k vk |2
E{SIN Rk } = E ∑
1 + ρ∥hk ∥2 sin(θk )2 j̸=k |gk vj |2
ρ∥|hk ||2 E{|h̃k vk |2 }
≥ |S|−1
1 + ρM t −1
∥hk ∥2 sin2 θk
ρ∥hk ∥2 {cos2 (θk )E(|ĥk vk |2 ) + sin2 (θk )E{|g̃k vk |2 }
= |S|−1
1 + ρM t −1
∥hk ∥2 sin2 θk
ρ∥hk ∥2 {cos2 (θk )E(|ĥk vk |2 )}
≥ |S|−1
1 + ρM t −1
∥hk ∥2 sin2 θk
( )
(Mt −1)
ρ∥hk ∥ cos (θk ) 1 − Mt ϵ
2 2
≈ |S|−1
, γk,|S| (ϵ), (3.9)
1 + ρM t −1
∥hk ∥2 sin2 θk
( )
(Mt −1)
where E(|ĥk vk |2 ) can be approximated by 1 − Mt
ϵ [40]. Note that the
estimated SINR γk,Mt (0) is chosen as the SINR metric in [9] [41]. The CDF
of γk,|S| (ϵ) is shown below

 ( Mt x )

 1− (
−
2BD e ρ(Mt −(Mt −1)ϵ)
)Mt −1




AM x
1+ M −(Mt −1)ϵ
 x ≥ (1−δ)(MMt −(M t −1)ϵ)
Fγk,|S| (ϵ) (x) 
t t
, ,
( Mt − (Mt −Mt1)ϵ
t Aδ
= )

x
−
Mt  2BD e ρ(Mt −(Mt −1)ϵ)

 1 − ( )Mt −1 + T

 AM x


1+ M −(Mt −1)ϵ
t t
, otherwise,
Mt − (Mt − 1)ϵ
(3.10)
B −x
|S|−1 −M D B
,T = (Γ(Mt −
x 1 ρ
where υ = ρ(1−δ−Aδx)
, A= Mt −1
, δ=2 t−1 [ 2 De
Γ(Mt −1) (1+Ax)Mt −1
∫∞
1, δ(Ax + 1)υ) − Γ(Mt − 1, υ))], Γ(a, x) = x
ta−1 e−t dt.
58
3.2.3 Sum Rate Analysis
The approximate cardinality of the i-th candidate user set Bi is given

by [42]


 K , i = 1,
|Bi | = (3.11)

 ⌊Kαi ⌋ , i = 2, . . . , Mt ,
∏i−1
where αi = t=1 (1 − (1 − ϵ2 )Mt −t ). αi can be viewed as the probability
that a user in the i-th candidate set satisfies the semi-orthogonal constraint.
Then, as an example, the probability that only one user among the K users
satisfies the semi-orthogonal constraint is P|S|=1 = (1 − α2 )K−1 . With the
cardinalities of the candidate sets, the probability that j users are selected
in the SUS algorithm with constraint ϵ which denoted as P|S|=j is given in
Corollary 1.
Corollary 1. According to SUS algorithm with semi-orthogonal constraint
ϵ, the approximate cardinality of the i-th candidate user set Bi is given in
(3.11). The probability that j users are scheduled at the BS can be expressed
as follows
K−j−Q(j−1) ( j−1 )
∑
K−j
∑ ∏
P|S|=j = ··· αt+1 (1 − αt+1 )
mt
m1 =0 mj−1 =0 t=1
( )
× (1 − αj+1 )K−j−Q(j)
, j ≥ 2, (3.12)
∑
j−1
where Q(j) = mt .
t=1
Proof. In Appendix D.
According to the SUS algorithm, the user π(i) is selected in the Bi based
Bi
on π(i) = arg maxj∈Bi γj,|S| (ϵ). For notational simplicity, we define γ(1),|S| (ϵ)
59
as the largest random variable among |Bi | random variables with the common
CDF in (3.10). The sum rate with quantized CDI and SINR in (3.9) can be
approximated by
∑
Mt (∑
j )
Bi
R≈ P|S|=j E log(1 + γ(1),j )
j=1 i=1
j ∫
∑
Mt ∑ ∞
= P|S|=j log(1 + x)dFγk,|S| (ϵ) (x)|Bi | dx. (3.13)
j=1 i=1 0
3.3 Feedback Reduction Strategy
Considering the maximum sum rate criterion in each candidate user set,
the users with relatively low estimated SINR are rarely selected in the can-
didate user set. Allowing these users to feed back CSI may waste feedback
resources. Therefore, a novel multiple-threshold model is proposed to divide
the range of estimated SINR into multiple regions to reduce the feedback
load.
3.3.1 Derivation of Multiple Thresholds
The cardinality of candidate user set Bi can be approximated as ⌊Kαi ⌋
in (3.11). When the semi-orthogonal constraint ϵ decreases, the cardinality

of candidate user set is decreased except of the candidate user set B1 . For
the purpose of SUS scheduling algorithm, the users in the candidate user set
Bi are required to feedback SINRs to the BS. Thus, the number of feedback
users in the candidate user set Bi can be viewed as |Bi |.
Due to the i.i.d assumption of estimated SINR, the user in the candidate
set Bi with the highest estimated SINR among |Bi | users is 1/|Bi |. When the
60
user belongs to Bi and its estimated SINR falls in the region [rth,i , ∞), the
user is more likely to be selected in the candidate set Bi . Thus, the threshold
rth,i can be set by
1
= 1 − Fγk,Mt (ϵ) (rth,i ), i = 1, 2, . . . , Mt . (3.14)
|Bi |
In general, the cardinality of i-th candidate set |Bi | is larger than cardinality
of (i + 1)-th candidate user set |Bi+1 |. Because it’s easier to find the i sched-
uled users that satisfy the semi-orthogonal constraint than i + 1 scheduled
users that satisfy the semi-orthogonal constraint. Thus, the cardinality of
candidate sets follow |B1 | ≥ |B2 | ≥ · · · ≥ |BMt |, and these thresholds de-
crease in order as rth,1 ≥ rth,2 ≥ · · · ≥ rth,Mt . These thresholds are used
to represent the quantization levels of estimated SINR. Especially to note
that, when the user’s estimated SINR is smaller than rth,Mt , the user does
not feedback any CSI to the BS. Otherwise, the estimated SINR and CDI
are required to feed back to the BS. Besides, all thresholds can be computed
off-line as long as the number of uses, the semi-orthogonal threshold ϵ and
the fading statistic are known. The value of the thresholds can be updated
periodically according to the system configuration possibly broadcasted by
the BS.
In the proposed scheme, the quantization bits for CQI are BQ = log(Mt )
bits and the CDI is quantized using the codebook with the size 2BD . From
(3.14), when the |BMt | > 1, the feedback probability of the user can be
61
expressed as
1
P {γk,Mt (ϵ) > rth,Mt } = 1 − Fγk,Mt (ϵ) (rth,Mt ) = (3.15)
|BMt |
When |BMt | ≤ 1, the γth,Mt becomes to zero. All users are required to
feedback CSI to the BS in that case. Therefore, the feedback load of proposed
multi-threshold scheme is summarized as follows


 K(BD + BQ ) , |BMt | ≤ 1
Lf bk = (3.16)

 K (BD + BQ ) = (BD +BQ ) , |BMt | > 1
|BM | t ⌊αM ⌋ t
3.4 Numerical Results
In this section, the simulated sum rate and approximated sum rate with
different CQI metrics without considering the quantization loss are discussed.
We also compare the different feedback schemes in terms of the sum rate and
feedback load. The transmitter is equipped with Mt = 4 antennas and the
semi-orthogonal constraint is set as ε = 0.4. The contents of feedback include
two parts: one is the quantized SINR metric which is represented using BQ
bits, the other is the CDI which is quantized by the codebook with size
2BD . The transmit power is 10W while the additive white Gaussian noise at
receiver is 1W. For the conventional feedback type, named Type I, no matter
what the CQI value is, the user quantizes CQI in the range [0, ∞) using BQ
bits and the CDI is quantized through BD bits. The multiple thresholds
model we proposed is referred as Type II scheme, the thresholds are used to
represent the quantization levels. Only when the user’s SINR is greater than
the threshold rth,Mt , the number of bits for quantized CQI and CDI of the
user are BQ = log(Mt ) and BD bits. The single threshold feedback scheme
62
is called Type III. When the SINR falls in the region [rth,Mt , ∞), the SINR
is quantized by BQ bits, and CDI is quantized through the codebook using
BD bits. Otherwise, no CSI is required to feedback to the BS.
Fig. 3.2 shows the (simulated) actual sum rate and the approximate sum
rate using different SINR metrics. The derived approximated sum rate is
close to the actual sum rate when the semi-orthogonal constraint becomes
large, i.e., ϵ = 0.4.
In Fig.3.3, the Type II and the Type III schemes achieve higher sum
rate than the Type I scheme. Because the design of the quantization method
for Type I scheme without considering the scheduling method, thus the
quantization levels are not fine enough to represent the CQI value. Therefore,
when the number of users increases, the sum rate performance of the Type I
scheme is limited by the inaccuracy CQI. Different from the Type I, the
sum rate performance is close to the sum rate with perfect CQI because
the quantization levels is designed according to SUS algorithm. Although
the sum rate performance in the Type III scheme increases with number of
users, the quantization levels in SINR region (rth,Mt , ∞] can not represent the
SINR well enough. Thus, there exists the sum rate loss compared with the
Type II scheme. For the multiuser diversity, obviously, the Type II and
the Type III schemes achieve the same growth rate as the sum rate with
perfect CQI even if the CQI bits are small, i.e., BQ = 2.
The feedback load of the Type I scheme increases linearly with number
of users. Thus, when the number of users is 500, the total feedback bits
are K(BD + BQ ) = 5000. The threshold based feedback method is used
for the Type II and the Type III schemes, the feedback load is a constant
63
(BD +BQ )
regardless the number of users and can be approximated as ⌊αM ⌋
≈ 522.
In order to have a fair comparison for all schemes, the sum rate as a function
of feedback load is shown in Fig.3.4. Obviously, the threshold-based feedback

method is more efficient than traditional feedback scheme. The Type I
scheme without considering the scheduling algorithm not only wastes lots of
feedback resources but also limits the sum rate performance due to inaccuracy
CQI.
3.5 Summary
A multi-user MIMO broadcast channel with ZF-BF using SUS scheduling was
considered. With the distribution of estimated SINR and an approximate
cardinality of the candidate user set in the SUS algorithm, an approximate
sum rate was derived which is close to actual sum rate. Moreover, when
the proposed feedback scheme was used, the feedback resources were utilized
well.
64
13.5
13
Approximated sum rate (ε=0.2, CQI=γk,|S|(0))
12.5
Simulated sum rate (ε=0.2)
Approximated sum rate (ε=0.4, CQI=γk,|S|(0))
12
Sum rate (bps/Hz)
Simulated sum rate (ε=0.4)

11.5 Approximated sum rate (ε=0.4, CQI=γk,|S|(ε))
ε=0.4
11
10.5
10
9.5 ε=0.2
9
100 150 200 250 300 350 400 450 500
Number of users
Figure 3.2: Comparison of simulated sum rate and approximated sum rate
with different CQI metrics, P =10W, BD =8 bits.
65
11.5
11
10.5
Sum rate (bps/Hz)
10 Full CQI (BQ=inf, BD=8,ε=0.4)

Type-I (CQI=γk,M (ε=0.4), BQ=2, BD=8,ε=0.4)
t
9.5
Type-II (CQI=γk,M (ε=0.4), BQ=2, BD=8,ε=0.4)
t
Type-III (CQI=γk,M (ε=0.4), BQ=2, BD=8,ε=0.4)

9 t
8.5
100 150 200 250 300 350 400 450 500
Number of users
Figure 3.3: Comparison of sum rate for different feedback types Mt =4,
P =10W, BD =8 bits, BQ =2 bits and ε=0.4
66
11
10.5
Type-I (CQI=γk,M (ε=0.4), BQ=2, BD=8,ε=0.4)
t
Type-II (CQI=γk,M (ε=0.4), BQ=2, BD=8,ε=0.4)

Sum rate (bps/Hz)
t
10
Type-III (CQI=γk,M (ε=0.4), BQ=2, BD=8,ε=0.4)
t
9.5
8.5
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Number of bits
Figure 3.4: Comparison of sum rate as a function of feedback load for different
feedback types, Mt =4, P =10W, BD =8 bits, BQ =2 bits, ε=0.4 and K =
100 ∼ 500.
67
Chapter 4
Feedback Reduction in MBMS

System
Because of the increasing demand of multimedia applications, multimedia
broadcast multicast service (MBMS) grows rapidly in recent years. In MBMS
systems, the MBMS users just have to receive data broadcasted on the com-
mon radio resources, and hence the efficiency of radio resources is highly im-
proved and the quality of multimedia service is then increased. Therefore, due
to the remarkable enhancement on efficiency of radio resources and multime-
dia quality, MBMS has become one of the important features in next genera-
tion communication standards, such as the Enhanced-MBMS (E-MBMS) in

3GPP Long Term Evolution (LTE) [7] and the Multicast Broadcast Service
(MBS) in Worldwide Interoperability for Microwave Access (WiMAX) [3].
However, to enable MBMS functions, all the users in the MBMS network
have to feedback its current channel condition to the BS, so that the BS can
select a proper modulation and coding scheme (MCS) for starting the broad-
cast transmission to all the MBMS users. For this sake, the feedback load
will be increased dramatically with the number of the users in the MBMS
68
network and hence the spectrum efficiency is degraded due to the accordingly
increased feedback bandwidth. Therefore, for the reason mentioned above,
an effective feedback reduction method is significant for MBMS systems.

In the literature, feedback reduction for multiuser systems is a popular
topic. For instance, in [43] a feedback reduction method is proposed that
only the users with normalized SNR being greater than a pre-determined
threshold will feedback their SNR via a dedicated feedback channel. Due to
the reason that statistically there must be some users with SNR lower than
the threshold, the feedback load is hence effectively reduced. In [44], a feed-
back reduction method in random access channel with multiple thresholds
is proposed. In this method, the range of user’s SNR is divided into multi-
ple classes with multiple thresholds. When the user’s SNR is greater than
the smallest threshold and belongs to a certain class j, the user randomly
accesses the feedback channel with probability pj . Therefore, the feedback
load can be adjusted by the probability pj .
In this chapter, a threshold-based feedback reduction method is proposed.
The idea of the method is that only when user’s SINR belongs to the region
[0, threshold), the user feeds back its SINR to the BS. The threshold is de-
rived based on the rankings of the users’ received SINRs among all users, i.e.,
the order statistics of the users. Because of the fact that the users with lager
SINRs should not affect the MCS selection of the MBMS network, the users
whose SINRs larger than the selected threshold should not feedback their
channel condition to reduce the system feedback load. From the simulation
results, it can be seen clearly that the feedback load is reduced significantly
while the performance using proposed feedback model is kept closely to the
69
Figure 4.1: MBMS system model
full CSI case.
This chapter is organized as follows. Section 4.1 describes the MBMS sys-
tem model using orthogonal random beamforming approach. The proposed
threshold for MBMS system is derived in Section 4.2. Performance analysis
of proposed feedback scheme is investigated in 4.3. Simulation results and
summary are given in 4.4 and 4.5.
4.1 System Model
In this study, a multiuser downlink MBMS multiple-input and single-

output (MISO) system is considered. The MBMS system includes one BS
with Mt antennas and K single-antenna users as shown in Fig. 4.1. The BS
utilizes ORB scheme [45] to send information to users. In the ORB scheme,
the precoding matrix at BS can be expressed as W = [w1 , w2 , . . . , wMt ],
where wi ∈ CMt ×1 , i = 1, 2, . . . , K, are beamformer vectors which are random
70
orthogonal vectors generated from isotropic distribution [45]. The received
signal yi for user i is given by
yi = h∗i Wx + ni , (4.1)
where x = [x1 , x2 , . . . , xMt ]T ∈ CMt ×1 is the multicast signal vector, hi ∈
CMt ×1 represents the channel fading vector seen by user i with i.i.d. complex
Gaussian random variables with zero mean and unit variance, ni ∈ C is
AWGN seen by user i with zero mean and variance σ 2 . It is assumed that
the feedback channel is error-free and the feedback delay time is negligible.
There is a constraint on total transmission power Pt so that Pt = E{x∗ x}. It
is also assumed that the total power is equally allocated at each beam, i.e.,
each beam is allocated with power Pt /Mt .
The beamformer wj is served for the broadcast signal xj , then the received
signal for user i can be expressed as
∑
Mt
yi,j = h∗i wj xj + h∗i wt xt + ni , j = 1, 2, . . . , Mt . (4.2)
t=1,t̸=j
The last two terms in (4.2) are interference and noise terms, and the corre-
sponding SINR γi,j can be shown as follows
|h∗i wj |2 Zj
γi,j = ∑M t = , (4.3)
Mt
ρ
+ t=1,t̸=j |h∗i wt |2 Mt
ρ
+ Wj
∑Mt
where ρ = Pt
σ2
, Zj = |h∗i wj |2 , and Wj = t=1,t̸=j |h∗i wt |2 . Since the channels
between the BS and users are all assumed to be i.i.d. Rayleigh distribu-
tion and the precoder W is an unitary matrix, the terms Zj and Wj are
independent chi-square distribution, i.e., Zj ∼ χ(2) and Wj ∼ χ(2Mt − 2)
71
respectively [11]. The PDF and CDF of γi,j , ∀i, j can be expressed as
Mt γi,j { }
e− ρ Mt
f (γi,j ) = (1 + γi,j ) + Mt − 1 , (4.4)
(1 + γi,j )Mt ρ
and Mt γi,j
e− ρ
F (γi,j ) = 1 − . (4.5)
(1 + γi,j )Mt −1
Because of the nature of MBMS, the system needs to select a user according
to the criterion as follows
i⋆j = arg min γi,j , j = 1, 2, . . . , Mt . (4.6)

i∈{1,2,...,K}
After the user i⋆j is selected, the SINR of user i⋆j can be used to choose a
proper MCS such that most of the users for the broadcast signal xj of the
MBMS network can decode the data successfully. The sum rate of MBMS
system can be written as follows
∑
Mt
R= E{log(1 + ri⋆ ,j )}. (4.7)
j=1
To achieve the purpose of MCS selection, all users in the MBMS system need
to feed back their CSIs, i.e., the SINRs of all beams, to the BS resulting in
the significant feedback load.
4.2 Threshold Design Using Order Statistics
To effectively reduce feedback load, the threshold based feedback scheme

is proposed. For a given the threshold set R = {rth,K , rth,K−1 , . . . , rth,1 , rth,0 },
where rth,K = 0 and rth,0 = ∞, and these thresholds are in decreasing order
as rth,0 ≥ rth,1 ≥ · · · ≥ rth,K .
72
Due to symmetric property, the index j is removed in (4.3) to simplify
the representation of i-th user’s SINR γi . Let γ(1) ≥ γ(2) ≥ · · · ≥ γ(K) be
order statistics of users’ SINRs γ1 , γ2 , . . . , γK , all with common PDF (4.4) in

decreasing order. When the user k’s received SINR on j-th beam is sinrk,j
, then the probability that the SINR of user k on the j-th beam is ranked
p-th among all users is
P {γk = γ(p) |γk = sinrk,j } =

( )
(K − 1)!
{F (sinrk,j )} K−p
{1 − F (sinrk,j )} (p−1)
. (4.8)
(K − p)!(p − 1)!
The conditional probability in (4.8) satisfies
∑
K
P {γk = γ(p) |γk = sinrk,j } = 1. (4.9)
p=1
Due to the symmetric assumption that users suffer i.i.d. Rayleigh fading
process, also due to ORB, the same set of thresholds applies to all users and
all beam directions. We have
P {γk = γ(l) | γk = rth,l } = P {γk = γ(l+1) | γk = rth,l },
and then the rth,l can be solved by
l
F (rth,l ) = 1 − , l = 1, 2, . . . , K. (4.10)
K
By method mentioned above, the BS can calculate the threshold set R =

{rth,K , rth,K−1 , . . . , rth,1 } off-line and broadcast the threshold set R to users
so that users can determine their feedback load by themselves and hence the
overall feedback load can be dramatically reduced. For a given threshold

rth,l ∈ R, only when the user’s SINR belongs to the region (0, rth,l ], the user
73
feeds back the SINR to the BS. Otherwise, the user no needs to feedback any
CSI to the BS. The set of feedback users is defined as Sf (rth,l ) = {j | γj ≤
rth,l , j = 1, 2, . . . , K} and the cardinality of Sf (rth,l ) can be expressed as

follows
|Sf (rth,l )| = KP {γj ≤ rth,l } = KF (rth,l ) = K − l. (4.11)
Once no one’s SINR is smaller than the threshold rth,l , no one needs to
feedback any CSI to the BS. The BS randomly selects one user among all
users. Applying the proposed feedback scheme, the sum rate can be expressed
as follows
[( ( )K )( )
l
RF (rth,l ) = Mt E 1 − log(1 + min γj )
K γj ∈Sf
( )K ( )]
l
+ log(1 + γi ) . (4.12)
K
Especially to note that when the threshold is selected as ∞. Our proposed
feedback method is equivalent to the conventional scheme. In the following,
in order to simplify the expression of the feedback load, we set l = K −n, n =
1, 2, . . . , K.
4.3 Performance Analysis

4.3.1 Feedback Load
We consider the feedback load as the number of feedback users. Thus,

the feedback load is equivalent to cardinality of the feedback user set. For
a given threshold rth,K−n , the feedback load is |Sf (rth,K−n )| = n. Obviously,
the feedback load is a constant regardless the number of users.
74
4.3.2 Performance Loss Analysis
As many users as possible to decode the desired broadcast signals cor-

rectly is the purpose of MBMS system. Therefore, we consider the number
of users that can decode the desired broadcast data successfully as the per-
formance indicator. In the proposed feedback scheme, the number of suc-

cessfully decoding users is defined as ND (rth,K−n ) when the threshold rth,K−n
is used. All users can perfectly decode the data in the conventional feedback
scheme because the threshold value is infinity. Then, the number of success-
fully decoding users for the conventional scheme is ND (∞) = K. However,
the feedback load of the conventional feedback scheme increases dramatically
with the number of users.
The proposed scheme with threshold rth,K−n , the number of successfully
decoding users is
[ ( )K ] ( )K ∑K
K −n K −n j
ND (rth,K−n ) = 1 − K+ (4.13)
K K j=1
K
( )K
n (K − 1)
=K − 1− . (4.14)
K 2
The first term in (4.13) describes that the user with the worst SINR is fed
back to the BS and is selected at the BS. In this case, all users can decode
the data correctly. Another term explains the event occurred that when no
one feedbacks the CSI to the BS, the BS randomly selects one user among
all users. Once the user that its received sinrk,j on the j-th beam is ranked
p-th among all users is randomly selected by the BS, then p users can decode
the data correctly.
75
Lemma 1. When the proposed feedback method is applied and the threshold
rth,K−n is used, the number of successfully decoding users satisfies
ND (rth,K−n ) 1
lim = 1 − n.
K→∞ K 2e
n K
Proof. Since lim (1 − ) = e−n , the lemma holds immediately.
K→∞ K
The benefit of reduction of feedback load comes at the cost of the per-
formance loss which is defined by the difference of number of successfully
decoding users between conventional scheme and the scheme with proposed
feedback method. Thus, the performance loss compared with the conven-
tional feedback scheme can be expressed by
∆ND (∞, rth,K−n ) , ND (∞) − ND (rth,K−n )

( )K
K −n (K − 1) K −1
= ≤ . (4.15)
K 2 2
Note that when no one feeds back CSI to the BS, the performance loss is at
most about fifty percent. In general, more users feed back CSI to the BS,
the number of successfully decoding users is increased. When the threshold
increases from rth,K−n to rth,K−(n+1) , the feedback load is increased from n to

n + 1. The increment of number of successfully decoding users can be shown
as follows
( )[( )K ( )K ]
K −1 K −n K − (n + 1)
∆ND (rth,K−(n+1) , rth,K−n ) = − .
2 K K
(4.16)
In Fig. 4.2, the increment of number of successfully decoding users is de-
creased dramatically when the threshold increases from rth,K−n to rth,K−(n+1) .

When number of users K is smaller than 500 and n is greater than five, almost
76
all users can decode the data correctly. This hints that when the threshold
used in the MBMS network is greater than rth,K−5 , the feedback bandwidth
is wasted.
77
60
K=10
50
K=100
K=500
∆ ND(rth,K-(n+1), rth,K-n)
40
30
20
10
0
1 2 3 4 5 6 7 8 9
n
Figure 4.2: Increment of number of successfully decoding users increasing
threshold from rth,K−n to rth,K−(n+1) .
4.3.3 Determination of Threshold
The number of successfully decoding users increases with the parameter n

and feedback load equals to n. There exists a tradeoff between the feedback
load and the number of successfully decoding users. Besides, the performance
loss ∆ND (∞, rth,K−n ) is an increasing function with respect to the number of
users K shown in lemma 2. For a given n, an intuitive explanation of lemma
2 is that the feedback load is a constant regardless the number of users

which hurts the performance of number of successfully decoding users. The
78
threshold should be adjusted with number of users K to keep the performance
loss be allowed. Thus, for a given the tolerable performance loss ∆NL (K), the
threshold should be selected to satisfy the performance loss requirement and

to reduce the feedback load as many as possible at the same time. Therefore,
the threshold rth,K−n⋆ is selected according to
( )
⋆
n = min n | ∆ND (∞, rth,K−n ) ≤ ∆NL (K) . (4.17)
n∈{1,...,K}
Lemma 2. When the threshold is selected as rth,K−n , the feedback load be-
comes to n. The partial derivative of performance loss compared with con-
ventional feedback scheme with respect to the number of users K satisfies
∂∆ND (∞, rth,K−n )

> 0.
∂K
Proof. See Appendix E.
4.4 Simulation Results
In this section, we compare different feedback feedback schemes in terms

of the number of successfully decoding users and the feedback load using
simulations. The transmitter is equipped with Mt = 4 antennas and there
are K users, each having single antenna. The transmitter power P is 20W
and the power of AWGN at the receiver is 1W. For the conventional feed-
back scheme, the user always feeds back the CSI to the BS on each beam.
The threshold based feedback scheme we proposed is referred as proposed
scheme. For a given threshold rth,K−n ,only when user’s SINR locates in the
region (0, rth,K−n ], the user feeds back the SINR to the BS. Otherwise, the
user no needs to feedback any CSI to the BS. Besides, in order to keep the
79
performance loss be allowed, the n is selected according to (4.17) to meet the
performance loss requirement. The tolerable performance loss of number of
successfully decoding users is chosen as ten percent of all users.

Fig. 4.3 shows the performance comparison between conventional feed-
back scheme and proposed scheme in terms of the number of successfully de-
coding users. Although the number of successfully decoding users increases
with the number of users, the performance gap between conventional feed-
back scheme and proposed scheme becomes large when the number of users
increases. For a given tolerable performance loss constraint ∆NL (K) = 0.1K,
the proposed scheme using the threshold rth,K−2 can meet the performance
requirement. Besides, in this case, the feedback load is two as shown in
Fig. 4.4. Not only constant feedback load is achieved but also the perfor-
mance requirement is met when the proposed feedback scheme is used.
In order to have a fair comparison, the results of the number of success-
fully decoding users and the feedback load are combined to show the number
of successfully decoding users as a function of feedback load in Fig. 6.3. When
our proposed feedback scheme is used, the feedback load can be reduced dra-
matically and most of users can decode the data correctly.
4.5 Summary
In this paper, we proposed a novel and effective feedback reduction method

for MBMS network. The method utilized the rank of user’s SINR to find the
thresholds. The performance loss compared with conventional scheme was
also investigated. The simulation results showed that the proposed method
indeed effectively reduces feedback load while satisfying the performance re-
80
100
Full CSI
90
Simulated ND(rth,K-1)
Number of successfully decoding users
80 Analytical ND(rth,K-1)
70
Analytical ND(rth,K-2)
60
50 Analytical ND(rth,K-3)
K-∆NL(K)=0.9K
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Number of users
Figure 4.3: Number of successfully decoding users of proposed scheme.
81
100
90
Full CSI
80 Proposed scheme using rth,K-1
Proposed scheme using rth,K-2
70
Feedback load
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Number of users
Figure 4.4: Feedback load of proposed scheme using different threshold.
82
100
Full CSI
90
Number of successfully decoding users
80 Proposed scheme using rth,K-2
70
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Feedback load
Figure 4.5: Number of successfully decoding users vs. feedback load.
83
quirement.
84
Chapter 5
Feedback Policies for

Heterogeneous Rayleigh Fading
Channel with Finite Feedback
In previous chapters, we focus on designing the thresholds according to
the order statistics of users received SINRs/SNRs. In this chapter, we con-
sider the optimization problem with objective that maximizes the sum rate
under finite feedback load constraint in the heterogeneous Rayleigh fading
channel. The optimality of the homogenous threshold feedback policy in the
two-user case with limited feedback under the Rayleigh fading channel was
discussed [46]. Surprisingly, it was shown that homogenous threshold is not

always rate optimal even if users experience homogeneous fading statistics.
In addition, the general feedback polices of the homogeneous fading channel

using orthogonal random beamforming was investigated in [47].
We investigate the feedback policies for the heterogeneous Rayleigh fading
channel under finite feedback constraints in this chapter. The algorithm we
proposed assigns feedback probabilities for the users according to the partial
derivatives of the sum rate function with respect to thresholds of the users.
85
Simulation results show that our proposed low complexity algorithm can
almost achieve the performance of the greedy algorithm.
5.1 System Model
We consider the multiuser downlink transmission model which includes a

BS with single antenna and K users, each having single antenna. We assume
that the users suffer from Rayleigh fading channel, and thus the received
SNR Xj of user j is exponentially distributed with parameter λj . The CDF

and PDF of Xj are given by
Fj (x) = 1 − e−λj x , x > 0,
fj (x) = λj e−λj x , x > 0. (5.1)
The user j feeds back its SNR value to the BS when the SNR value is greater
than its pre-determined threshold τj which is broadcasted from the BS. Then
the BS performs the maximum sum rate algorithm to schedule the users.
Without loss of the generality, we assume τ1 ≥ τ2 ≥ · · · ≥ τK > 0 and
τ0 = ∞, and the system sum rate of K users can be expressed as follows

∫ ∞ (∏
K )
R(τ1 , . . . , τK ) = log(1 + x)d Fi (x)
τ1 i=1
∫ τ1 (∏
K )
+ F1 (τ1 ) log(1 + x)d Fi (x) + · · ·
τ2 i=2
∏
K−1 ∫ τK−1
+ Fj (τj ) log(1 + x)dFK (x)
j=1 τK
K (∏
∑ i−1 )∫ τi−1 (∏
K )
= Fj (τj ) log(1 + x)d Fj (x) , (5.2)
i=1 j=1 τi j=i
86
∏
0
where we define Fj (τj ) = 1. We set a total feedback load C as a constraint,
j=1
and formulate the problem as
max R(τ1 , . . . , τK )
(τ1 ,...,τK )
∑
K K (
∑ )
s.t. Pj = 1 − Fj (τj ) ≤ C. (5.3)
j=1 j=1
5.2 Proposed Algorithm Using Partial Deriva-

tive
To the best of the authors’ knowledge, a closed-form solution to the op-
timization problem (5.3) does not exist. Thus we turn to developing subop-
timal algorithms that assign feedback probabilities to users. Note that the
relation between τj and the feedback probability for user j is Pj = 1 − Fj (τj ).
5.2.1 Greedy Algorithm
We first define a positive increment of feedback probability, denoted δp , as

the unit for assigning the probabilities. Initially the feedback probability of
each user is set to zero. At the l-th assignment step, the sum rate increment
for user j can be expressed as
∆RIl (j) = R(Pl−1 + Vj ) − R(Pl−1 ), (5.4)
where Pl−1 = [P1l−1 , P2l−1 , . . . , PKl−1 ] with element Pml−1 being the feedback
probability for user m at the (l − 1)-th assignment step, and Vj = [0, . . . ,
δp , . . . , 0] which has only one non-zero element δp at the j-th location. The
probability increment δp is assigned to user j ⋆ at the l-th assignment step
87
according to
j⋆ = max ∆RIl (j). (5.5)
j=1,2,...,K.
The probability assignment is repeated until the feedback load constraint is
met.
5.2.2 Partial Derivative Algorithm
In this method, initially the feedback probability of each user is set to

one. Then the users’ feedback probabilities will be reduced until the feedback
load constraint is met.
Lemma 3. The partial derivatives of the sum rate with respect to the thresh-
olds satisfy
(
∂R(τ1 , . . . , τK ) ∏
K−1
= − log(1 + τK )fj (τj ) Fm (τm ) − fj (τj )
∂τj m=1,m̸=j
{ K−1
∑ ∏ m ∫ τm
∏K })
F v (x)
× Ft (τt ) v=m+1
dx < 0.
m=j t=1,t̸=j τm+1 (1 + x) ln 2
Proof. See Appendix F.
Whenever a threshold is increased, the corresponding feedback probability
is reduced which hurts the scheduling and the sum rate. Based on the result
of Lemma 3, the sum rate loss due to decrease of the feedback probability of
the l-th assignment can be approximated by
∆RLl (j) = R(Pl−1 − Vj ) − R(Pl−1 )

( )
∂R(Pl−1 − Vj )
≃ δp . (5.6)
∂τj
88
Then, at each assignment step, user k ⋆ will be selected for reducing the
feedback probability by δp if
k⋆ = max ∆RLl (j). (5.7)

j=1,2,...,K.
∑K
The algorithm iterates until the condition j=1 Pj ≤ C is met. However, the
computational burden of integration in Lemma 3 is significant. Therefore,
we replace the integration by an upper bound as shown below
∫ ∏K ( )
τm
Fv (x) ∏K
1 + τm
v=m+1
dx ≤ Fv (τm ) log .
τm+1 (1 + x) ln 2 v=m+1
1 + τm+1
Through simulations, we compare the sum rate performance of the pro-
posed algorithm with that of the greedy algorithm under the finite feedback
constraint. The fading channel coefficients for users are assumed to be in-
dependent and complex Gaussian distributed with zero mean and variances
σj2 , where σj2 are drawn uniformly from the interval [0, 1] to model heteroge-
neous channels. Thus, the parameter of random variable Xj can be written

σN2
as λj = P σj2
, where the transmit power P is 10W, while the additive white
2
Gaussian noise power at the receivers σN is 1W. The increment of probability
δp is set as 0.1 and the feedback constraint C is configured as 1.4. In Fig. 5.1,
the proposed low complexity algorithm achieves almost the same rate as the
greedy algorithm when the number of users is smaller than 10. The sum
rates of the low complexity algorithm and the partial derivative algorithm
are almost the same.
89
3.8
Greedy algorithm
3.6 Partial deravative algorithm: low complexity
Partial deravative algorithm
3.4
Sum rate (bps/Hz)
3.2
2.8
2.6
0 5 10 15 20
Number of users
Figure 5.1: Rate comparison for different algorithm. ρ=10dB.
90
5.4 Summary
In this chapter, we investigated the feedback polices for multiuser down-

link systems under a finite feedback constraint. The proposed low complexity
algorithm assigns the feedback probabilities to the users in a heterogenous

Rayleigh fading channel. Simulation showed that the sum rate using the
partial derivative algorithm is close to that of the greedy algorithm.
91
Chapter 6
Sum Rate and Fairness Tradeoff

with Feedback Reduction in
Multiuser Heterogeneous
Rayleigh Fading Channel
Multiuser diversity can significantly improve system throughput when the
users suffer channel fluctuations [48]. However, the feedback load increases
linearly with the number of users and the average system fairness is damaged
to achieve high sum rate.
The performance evaluations for different schedulers such as round robin,

proportional fair scheduling and greedy SNR scheduling are discussed in [23].
The hybrid scheduling is proposed in [24] to achieve the flexible tradeoff be-
tween the sum rate and the fairness among all users. In [24], a hybrid mul-
tiuser scheduler combines the greedy SNR scheduling and normalized SNR
scheduling. The users are grouped into n groups and the grouping method
can be done through sector by sector or ring by ring. One user is selected in
each group based on either the ”best SNR” or ”normalized SNR” criterion
and then the scheduled user is chosen from n selected users based on ”nor-
92
malized SNR” or ”best SNR” criterion. Thus, the flexible tradeoff between
the sum rate and the fairness among all users can be achieved through the
parameter n. However, the concept of feedback load reduction does not take
into account in above scheduling methods.
In this chapter, the relations among the sum rate, average system fairness
and the feedback load are established through the multiple thresholds and
proposed scheduling algorithm. The range of SNR is divided into N regions
according to the multiple thresholds. Only when user’s SNR is greater than
the smallest threshold, the user feeds back SNR to the BS. Thus, the reduc-
tion of feedback load is achieved and amount of feedback load is related to
the parameter N . The proposed scheduling includes two steps: a region is
selected randomly among N regions, and then the scheduled user with the
highest SNR is chosen from that region which is selected in previous step.
Through simulations, it is shown that the proposed scheduling with feedback
reduction scheme can achieve the flexible tradeoff among the sum rate, the
fairness among all users and the feedback load.
This chapter is organized as follows. Section 6.1 describes the multi-
user system in heterogeneous Rayleigh fading channel. The transmission and

feedback procedure is described in Section 6.1.1. Relations among the system
fairness, the sum rate and the feedback load using the proposed scheduler
are investigated in 6.3. Simulation results and summary are given in 6.4 and
6.5.
93
6.1 System Model
The multiuser downlink system included the BS with single antenna and
K users is considered. Each user is equipped with single antenna. The
received signal for the user i is given by

√
yi = P hi xi + ni , (6.1)
where xi is the transmitted signal, and hi represents channel fading coefficient

which is assumed to be i.i.d complex Gaussian random variable with zero
mean and variance σi2 . The noise for user i, denoted ni , is modeled as i.i.d
complex Gaussian distributed with zero mean and unit variance. We assume
that the total transmit power at the BS is P . The received SNR for user i,
denoted Yi , is exponential distributed with parameter λi and the PDF of Yi
is shown as follows
fYi (y) = λi e−λi y , y ≥ 0, (6.2)
where λi = 1/(P σi2 ).
6.1.1 Transmission and Feedback Procedure
The transmission and feedback procedure are described as follows. The

transmission model including the training mode and the data transmission
mode is proposed. In the training mode, the BS broadcasts the reference sig-
nal to let the users estimate the received SNRs. After receiving the SNRs that
fed back from users, the BS calculates the threshold set rth,1 , rth,2 , . . . , rth,N
with decreasing order as {rth,1 ≥ rth,2 ≥ · · · ≥ rth,N } and broadcasts these
thresholds periodically. In the data transmission mode, the data transmis-

sion is performed and the users’ received SNRs are still feeding back to the
94
BS to enhance the accuracy of received SNRs at the BS. Especially to note
that only when the user’s received SNR is greater than the pre-determined
threshold, the user feedbacks CSI to the BS. Otherwise, the user no needs to
feed back any CSI to the BS.
6.2 Design of Multiple Thresholds
To derive the thresholds in the training mode, the statistical mean of the
random variables Y1 , Y2 , . . . , YK are ranked with decreasing order as 1

λ(K)
≥
1
≥ · · · ≥ λ(1)
λ(K−1)
1
and uniformly divided into N groups with size L =
{ } { }
1 1 1 1 1 1
K/N as λ(K) , λ(K−1) , . . . , λ(K−L+1) ,. . ., λ(L) , λ(L−1) , . . . , λ(1) . To simplify
the notation, we let random variable Ymn represent the m-th statistic in the
n-th group. Then, Ymn is exponentially distributed with statistical mean

1
λ(K−(n−1)L−(m−1))
. The random variables in group n are {Y1n , Y2n , . . . , YLn } and
n
let Y(1) ≥ Y(2)
n
≥ · · · ≥ Y(L)
n
be the order statistics of that group. For a
measured instantaneous SNR yi at user i, we say that the rank of yi in group

n
n is d if Y(1) ≥ Y(2)
n
≥ · · · ≥ yi ≥ Y(d+1)
n
≥ · · · ≥ Y(L)
n
.
| {z } | {z }
(d − 1) variables (L − d) variables
The threshold of each group is designed according to the probability of
a user’s measured instantaneous SNR being a particular rank in that group.
i
Let Pm,n (r) denote the probability that user m is ranked n among the L
95
users in group i when its instantaneous SNR is r.
i
Pm,n (r) = P {Ymi = Y(n) i
|Ymi = r}
∑
= P {Yti1 ≥ Yti2 ≥ · · · ≥ Ytin−1 ≥ r ≥ Ytin+1 · · · ≥ YtiL }
S
| {z } | {z }
(n − 1) variables (L − n) variables
{ n−1
∑∏
1
= (n−1)!(L−n)! exp (−λ(K−(i−1)L−(ta −1)) r)
S a=1 }
∏
L
× (1 − exp−λ(K−L(i−1)−(tb −1)) r ) ,
b=n+1
∑
where S denotes the summation over all permutations (t1 , . . . , tn−1 , tn+1 ,
. . . , tL ) of (1, 2, . . . , m − 1, m + 1, . . . , L) in the group i. For example, when
the instantaneous SNR of user t in group one is infinity, the rank of user t
1
among the L users in group one is one with probability one, i.e., Pt,1 (∞) =
P {Yt1 = Y(1)
1
|Yt1 = ∞} = 1. In order to reduce computational complexity,
we approximate by assuming that the random variables Ymi , m = 1, 2, . . . , L
in group i have the same exponential distribution with mean µi obtained by

∑L
λ(K−(i−1)L−(m−1))
µi = m=1 . (6.3)
L
i
Then, Pm,n (r) is the same for all users in the group i, and can be obtained
by
L−n
i (L − 1)!e(−µi (n−1)r) (1 − e(−µi r) )
Pm,n (r) = . (6.4)
(n − 1)!(L − n)!
The most probable rank of user m in group i when its instantaneous SNR is
r can be obtained by
n⋆ = arg max i
Pm,n (r). (6.5)
n∈{1,2,...,L}
i
Let Qin be defined such that Pm,n i
(Qin ) = Pm,n+1 (Qin ). It can be seen that
[ )
when the instantaneous SNR of user m falls in the range Qin , Qin−1 , the
96
most probable rank of user m in group i is n. To reduce the feedback load
, we let only the users who are most likely to be rank one in each group to
feedback. With this approximation, the closed-form solution of the threshold

rth,i can be derived as
1
rth,i = Qi1 = ln L, i = 1, 2, . . . , N. (6.6)
µi
6.2.1 Proposed Scheduler
In order to achieve the flexible tradeoff among the sum rate, the average
system fairness and the feedback load, the proposed scheduling method in
the data transmission mode is described as follows:
1. The BS uses multiple thresholds rth,1 , rth,2 , . . . , rth,N to divide the range
of SNR into N regions. Note that these thresholds are broadcasted by
the BS in the training mode. Only when the user’s received SNR is
greater than the smallest threshold rth,N , the user feeds back the SNR
to the BS. Otherwise, the user no needs to feedback any CIS to the BS.
After the BS collecting the CSI, the candidate user group i, denoted
as Gi = {j | Yj ∈ [rth,i+1 , rth,i ), j = 1, 2, . . . , K} can be obtained at
the BS. Then the BS randomly selects a candidate user group with
non-zero cardinality.
2. After the BS selecting a candidate group, then the BS selects a user

with the highest SNR among the users in that group.
3. When no one feeds back SNR to the BS, the BS randomly selects one
user among K users.
97
6.3 Performance Analysis
6.3.1 Sum Rate Analysis
In this section, we investigate the sum rate performance using the pro-
posed scheduling method with multi-threshold feedback scheme. We de-

fine the set of SNRs for the candidate user set Gl as Sl = {Yj | Yj ∈
[rth,l+1 , rth,l ), j = 1, 2, . . . , K} with the cardinality |Sl | and let Y(1)
Sl
be the
largest order statistic in the set Sl . We also define the indicator function as
follows:


 1, x>0
I(x) = (6.7)

 0, otherwise
Thus, the actual number of regions that the BS can randomly select is Q =
∑N ∑N
l=1 I(|Sl |) ≤ N and the number of feedback users is m = l=1 |Sl |.
The sum rate is composed of two parts, one is the sum rate obtained in
the case when the number of candidate user groups is greater than or equal
to one, the other is the sum rate contributed by a user that is randomly
98
selected at the BS. The sum rate of proposed scheme is shown below
[ K (
∑ ∑ K−m ∏ P (Ytm+j < rth,N ) ([
R= P (Yt1 ∈ S1 | Yt1 ≥ rth,N )
m=1 G j=1
m!(K − m)!
× P (Yt2 ∈ S1 | Yt2 ≥ rth,N ) × · · · × P (Ytm ∈ S1 | Ytm ≥ rth,N )

] [
× log(1 + Y(1)
S1
) + P (Yt1 ∈ S1 | Yt1 ≥ rth,N )P (Yt2 ∈ S2 | Yt2 ≥ rth,N )
1
× P (Yt3 ∈ S1 | Yt3 ≥ rth,N ) × · · · × P (Ytm ∈ S1 | Ytm ≥ rth,N )
2
[ ]] [
× log(1 + Y(1) S1 S2
) + log(1 + Y(1) ) + · · · + P (Yt1 ∈ SN | Yt1 ≥ rth,N )
)
])
× · · · × P (Ytm ∈ SN | Ytm ≥ rth,N ) log(1 + Y(1)
SN
)
∏ (1 − e−λi rth,N ) ∑ [ ]
K K
+ E log(1 + Yq | Yq < rth,N ) , (6.8)
i=1
K q=1
∑
where G denotes the summation over all permutations (t1 , t2 , . . . , tm , . . . ,
tK ) of (1, 2, . . . , m, . . . , K). The sum rate of selected user when its SNR
smaller than the pre-determined threshold is

[ ] ∫ rth,N log(1 + y)λq e−λq y dy
E log(1 + Yq | Yq < rth,N ) = 0
.
(1 − e−λq rth,N )
6.3.2 Fairness Analysis
The fairness is another performance metric of the system. In [24], the

− log(Pi )
self-fairness defined as Fi = log(K)
is used to measure the fairness of user
i where Pi is the scheduled probability of user i. Thus, the average system
∑
fairness can be formulated as F = K i=1 Pi Fi . Obviously, in order to obtain
the average system fairness, the access probability of user i, Pi should be

investigated first. The Pi is composed of two parts: one is the probability
that the user i is scheduled according to its feedback and the other is the
99
probability that the user i is selected randomly at the BS. Then, the Pi can
be expressed as follows:
[ K−1 ( (
∑ 1 ∑ m−1 ∏
Pi = P (Ytm+j < rth,N )
m=1
(m − 1)!(K − m)! ′ i=1
G
[
P (Yt1 ∈ S1 | Yt1 ≥ rth,N )P (Yt2 ∈ S1 | Yt2 ≥ rth,N ) × · · ·
[ 1
× P (Ytm−1 ∈ S1 | Ytm−1 ≥ rth,N ) P (Ym ∈ S1 | Ym ≥ rth,N )
|S1 |
∑N
] ] [
+ P (Ym ∈ Sl | Ym ≥ rth,N ) + P (Yt1 ∈ S1 | Yt1 ≥ rth,N )
l=2
× P (Yt2 ∈ S2 | Yt2 ≥ rth,N ) × · · · × P (Ytm−1 ∈ S1 | Ytm−1 ≥ rth,N )

[ 1 1
× P (Ym ∈ S1 | Ym ≥ rth,N ) + P (Ym ∈ S2 | Ym ≥ rth,N )
|S1 | 2
∑N
]] [
+ P (Ym ∈ Sl | Ym ≥ rth,N ) + · · · + P (Yt1 ∈ SN | Yt1 ≥ rth,N )
l=3
× P (Yt2 ∈ SN | Yt2 ≥ rth,N ) × · · · × P (Ytm−1 ∈ SN | Ytm−1 ≥ rth,N )

))]
∑ 1 ]]
−1
[N
× P (Ym ∈ Sl | Ym ≥ rth,N ) + P (Ym ∈ SN | Ym ≥ rth,N )
l=1
|SN |
∏
K
(1 − e−λi rth,N )
+ , (6.9)
i=1
K
∑
where denotes the summation over all permutations (t1 , . . . ,
′
G
, tm−1 , tm+1 , . . . , tK ) of (1, 2, . . . , m − 1, m + 1, . . . , K).
When a user’s instantaneous SNR is smaller than the smallest threshold
rth,N , the user no needs to feed back any CSI to the BS. Otherwise, the user
feeds back CSI to the BS with probability e−λk rth,N , k = 1, 2, . . . , K. The
100
average total feedback load can be expressed as
K (
∑ )
−λk rth,N
Fb = e (6.10)
k=1
From the equation (6.10), the average system feedback load can be reduced
through the threshold γN .
In this simulation, we show the sum rate, the average fairness of the
system and the feedback load for the proposed scheduler with multi-threshold
model. The fading channel coefficients for users are assumed to be i.i.d.
complex Gaussian distribution with zero and variance σi2 , where σi2 are drawn
uniformly from the interval [0, 1] to model heterogeneous channels. The
transmit power P is 10W, while the additive white Gaussian noise power
2
at the receivers σN is 1W. The proposed scheduling scheme with feedback
reduction method is compared with the greedy SNR scheduling, proportional
fair scheduling (normalized SNR) and the scheduling using random selection
at the BS. The parameter N is changed from 1 to K to achieve the flexible
tradeoff among the average system fairness, the sum rate and the feedback
load. When number of region N equals to one, our scheme becomes to
the scheme with single threshold using maximum SNR scheduling algorithm.
Once the number of regions becomes to K, the threshold rth,N turns to zero.
Thus, our scheme in that case is identical with the conventional feedback
scheme which means no feedback load can be reduced. The full fairness can
be achieved but the gain of multiuser diversity is damaged.
In Fig. 6.1, when the number of regions N increases, then the sum rate
101
performance is decreased. Because the BS randomly selects a region among
N regions first, the probability that the scheduled user has lower SNR is
increased. When the number of regions is equal to one, the sum rate perfor-
mance of our proposed scheme closes to the greedy SNR algorithm. The sum
rate performance using random selection at the BS is also matched with our
proposed scheme when the number of regions is K. The random selection
phase in our proposed scheduling scheme selecting one region among N = K
regions just likes random selection of one user among K users. In that case,
the cardinality of candidate user set of our proposed scheme is too small to
obtain the multiuser diversity gain.
As shown in Fig. 6.2, the average system fairness increases with the num-
ber of regions. When the number of regions increases, the number of candi-
date user sets is increased. The more users with lower SNR covered in the
candidate user sets, the more average fairness is obtained. Therefore, the
flexible tradeoff between the sum rate and the fairness among users can be
achieved in our proposed scheme.
In Fig. 6.3, we plot sum rate as the function of the feedback load as an
indication of the efficiency. The proposed scheme with N = 1 can achieve

the highest sum rate and reduce the feedback load dramatically at the same
time. In the other hand, the relation between the fairness and the feedback
load in our proposed scheme can be observed. In order to obtain the more
fairness, the more users are required to feed back CSI to the BS.
102
5.5
4.5
Sum rate (bps/Hz)
4
Max SNR scheduling
Normalized random selection
3.5 Random selection
Proposed scheme (N=K)
3 Proposed scheme (N=4)
Proposed scheme (N=3)
2.5
1.5
10 20 30 40 50 60 70 80 90 100
Number of users
Figure 6.1: Sum rate comparison between different number of feedback re-
gions (P= 10W).
103
1
0.9
0.8
Average system fairness
0.7
0.6
Max SNR scheduling

0.5
Normalized random selection
Random selection
0.4
0.3
0.2 Proposed scheme (N=2)
0.1
0
0 10 20 30 40 50 60 70 80 90 100
Number of users
Figure 6.2: Fairness comparison between different number of feedback regions

(P = 10W).
104
Sum rate (bps/Hz)
6

5
4 Proposed scheme (N=3)
3 Proposed scheme (N=K)
1
0 10 20 30 40 50 60 70 80 90 100
Feedback load
1
System fairness
0.95
0.9

0.7
0.65
0 10 20 30 40 50 60 70 80 90 100
Feedback load
Figure 6.3: Sum rate vs. feedback load and fairness vs. feedback load with
different number of regions (P = 10W ).
6.5 Summary
In this chapter, we proposed a scheduling scheme with multi-threshold

model to balance the sum rate and the system fairness. The analysis of sum
rate, average system fairness and the feedback load were given. According to
the multi-threshold feedback scheme, the number of regions N could be used
to adjust the system requirement, i.e., sum rate, system fairness, feedback
load. The simulations showed that the flexible tradeoff among sum rate, the
105
fairness among users and the feedback load is achieved through the number
of regions N .
106
Chapter 7
Conclusions
The feedback reduction methods considering the different scheduling al-
gorithms are proposed for different transmission scheme such as ORB scheme,
ZF-BF using SUS algorithm and MBMS system using ORB approach. In the
ORB scheme with proposed feedback scheme, the sum rate growth is 1−e−N ,
almost achieves the full multiuser diversity, where N is the minimum number
of regions to satisfy the pre-determine performance requirement. Moreover,
the feedback load is a constant regardless the number of users. For the ZF-
BF using SUS algorithm scheme, the threshold is designed according to the
scheduling algorithm and thus the feedback method is more efficient than
the existing methods. However, the sum rate performance is not the desired
purpose of MBMS system. Thus, the number of successfully decoding users
with proposed feedback scheme is investigated. Simulation shows that the
number of successfully decoding users is almost the same with traditional
feedback scheme while the feedback load keeps in a small constant. For the
heterogeneous scenario, each user has its own threshold to reduce feedback
load. For a given feedback load constraint, the threshold is derived through
the proposed low complexity derivative algorithm. Simulations show that the
107
sum rate performance using proposed algorithm is close to the performance
using the greedy algorithm.
We also discussed the relation among the fairness across users, the feed-
back load and the sum rate in the heterogeneous Rayleigh fading channel. A
new scheduling method with multi-threshold feedback scheme is proposed to
achieve the flexible fairness of the system. The fairness is increased when the
number of regions of feedback scheme increases. Thus, the system fairness
can be changed through the number of regions to satisfy the requirement of

system.
108
Appendix A
Proof of Theorem 1
Proof. In the proposed multi-threshold feedback scheme, N ≤ K. When
K = N , the users always feedback and there is no sum rate loss. Therefore,
∂△R(K,N,Mt )
we will prove this theorem by showing that > 0 for the nontriv-
(
∂K
ial K > N case. We first define f1 (Xk , K) , E log(1 + X(1) K

) − log(1 +
)
Xk ) | X(1)
K
< rth,N to simplify the expressions. The sum rate loss in (2.21)
can be expressed
{ }
△R(K, N, Mt ) = Mt E log(1 + X(1)
K
) − log(1 + Xk ) | X(1)
K
< rth,N PL (K)
= Mt f1 (Xk , K)PL (K), (A.1)
where PL (K) is the probability of rate loss event defined in (2.20). The
partial derivative of △R(K, N, Mt ) with respect to K is

[ ]
∂△R(K, N, Mt ) ∂f1 (Xk , K) ∂PL (K)
= Mt PL (K) + f1 (Xk , K) (A.2)
.
∂K ∂K ∂K
109
According to the inequality ln(x) ≥ 1 − 1/x, we have
( )K (
∂PL (K) ∂ N N )( N )K N ( N )K−1
= 1− = ln 1 − 1− + 1−
∂K ∂K K K K K K
( ) [ ( ) ( ) ]
N N K−1 K N N
= 1− 1− ln 1 − +1
K K N K K
[ ]
N( N )K−1 ( K )( 1 )
≥ 1− −1 1− + 1 = 0. (A.3)
K K N 1− K
N
On the other hand, using the Leibnitz rule, we have

( ∫ rth,N ( )K−1
∂f1 (Xk , K) ∂ log(1 + x)KfXm,k (x) FXm,k (x)
= ( )K dx
∂K ∂K 0 FXm,k (rth,N )
∫ rth,N )
log(1 + x)fXm,k (x)
− dx
0 FXm,k (rth,N )
[( ∫ ( )K−1
rth,N
log(1 + x)fXm,k (x){ FXm,k (x) }
= ( )K dx
0 FXm,k (rth,N )
∫ rth,N ( )K−1 ( )
log(1 + x)fXm,k (x){K FXm,k (x) ln FXm,k (x) }
+ ( )K dx
0 FXm,k (rth,N )
( )K−1 ( ))
log(1 + rth,N )KfXm,k (rth,N ) FXm,k (rth,N ) ∂rth,N
+ ( )K
FXm,k (rth,N ) ∂K
( ( ) )]
log(1 + rth,N )fXm,k (rth,N ) ∂rth,N
− > 0, (A.4)
FXm,k (rth,N ) ∂K
∂rth,N
where ∂K
> 0 can be observed from (2.14), and fXm,k (x) and FXm,k (x) are,
respectively, the PDF and CDF of Xm,k . Since f1 (Xk , K) > 0, PL (K) > 0,
∂PL (K)
∂K
≥ 0 and ∂f1 (Xk ,K)
∂K
> 0, we conclude that ∂△R(K,N,Mt )
∂K
> 0.
110
Appendix B
Proof of Theorem 2
Proof. The sum rate can be lower bounded by
{ }
R(K, N, Mt ) ≥ Mt P {rth,N ≤ X(1)
K
< ∞}E log(1 + X(1)
K
) | rth,N ≤ X(1)
K
<∞
≥ Mt (1 − PL (K)) log(1 + rth,N ) , RL (K, N, Mt ), (B.1)
where PL (K) is the probability of the rate loss event defined in (2.20). Be-
cause the distribution of Xm,k is different for Mr ≥ Mt and Mr < Mt , we dis-
cuss these two cases separately. For Mt > Mr , since rth,N has a similar form as
N
the variable un in the proof of [11, Theorem 4] (that is, FXm,x (rth,N ) = 1− K ),
we can follow the steps in the proof of [11, Theorem 4] to compute the lower
bound of rth,N by using the lower bound of fXm,k (x) for x > 1. Therefore,
L
when the number of users is large, the lower bound of rth,N , denoted as rth,N ,
can be expressed as
( )
L
rth,N = ρ ln(K/N ) − ρ Mr (Mt + 1) − 3 ln ln(K/N ) + O(ln ln ln(K/N )).
(B.2)
111
Then,
RL (K, N, Mt ) (1 − PL (K)) Mt log(1 + rth,N )

lim = lim
K→∞ Mt log log K K→∞ Mt log log K
(1 − PL (K)) Mt log(1 + rth,N
L
)
≥ lim
( ( )K ) L
N log(1 + rth,N )
= lim 1 − 1 −
K→∞ K log log K
= 1 − e−N . (B.3)
For Mr ≥ Mt , rth,N can be found using the asymptotic expression of the in-
complete Gamma function [49]. For large x, the incomplete Gamma function
can be approximated as follows
Γ(m, x) ≈ xm−1 e−x (1 + O(1/x)). (B.4)
When K is large, rth,N can be solved by

rth,N
Γ(Mr − Mt + 1, ρ
)
1 − FXm,k (rth,N ) =
(Mr − Mt )!
(r )Mr −Mt −rth,N N
th,N
= e ρ (1 + O(ρ/rth,N )) = . (B.5)
ρ K
We have
rth,N
= ln(K/N ) + (Mr − Mt ) ln ln(K/N ) + O(ln ln ln(K/N )). (B.6)
ρ
log(1 + rth,N )
Then lim = 1 and we have the same result as (B.3).
K→∞ log log K
112
Appendix C
Proof of Theorem 3
Proof. With the property of the sum rate lower bound already shown in
Theorem 2, we now focus on the sum rate upper bound. Using Jensen’s
inequality, the sum rate R(K, N, Mt ) can be upper bounded by

( { K })
R(K, N, Mt ) ≤ Mt (1 − PL (K)) log 1 + E X(1) | rth,N ≤ X(1)
K
<∞
( )
+ Mt PL (K) log 1 + E {Xk } . (C.1)
According to [50], for i.i.d. random variables X1 , X2 , . . . , XK having the same
CDF FX (x),
∫ ∫
1 { K} 1
FX−1 (u)du ≤ E X(1) ≤K FX−1 (u)du.
1
0 1− K
{ }
Thus, E X(1) | rth,N ≤ X(1) < ∞ can be upper bounded by
K K
{ K } { K }
E X(1) | rth,N ≤ X(1)
K
< ∞ ≤ E X(1) | rth,N ≤ Xk < ∞, k = 1, 2, . . . , K
∫ 1 ∫ 1 ( )
−1
≤K FX|X≥rth,N (u)du = K −ρ ln (1 − u)e−rth,N /ρ du
1 1
1− K 1− K
∫ 1 ( ) ( )
=K FX−1m,k (u) + rth,N du = ρ ln(K) + 1 + rth,N , (C.2)
1
1− K
113
where FX|X≥rth,N (x) is the conditional CDF of X, which is distributed like
FXm,k (x) defined in (2.5) with Mr = Mt , given that X ≥ rth,N . That is,
e−x/ρ
FX|X≥rth,N (x) = 1 − .
e−rth,N /ρ
Substituting (C.2) and E {Xk } = ρ into (C.1), an upper bound of the sum
rate RU (K, N, Mt ) can be found as
(( )
R(K, N, Mt ) ≤ Mt 1 − PL (K) log(1 + ρ ln K + ρ + rth,N )
)
+ PL (K) log(1 + ρ)
, RU (K, N, Mt ).
For the upper bound RU (K, N, Mt ), we have

( )
RU (K, N, Mt ) (1 − PL (K)) log (1 + ρ ln K + ρ + rth,N )
lim = lim
K→∞ Mt log log K K→∞ log log K
PL (K) log(1 + ρ)
+ lim
K→∞ log log K
( ( )K )
N
= lim 1 − 1 −
K→∞ K
( )
log 1 + 2ρ ln(K) + ρ − ln N
× lim
K→∞ log log K
= 1 − e−N .
114
Appendix D
Proof of corollary
Proof. With the cardinalities of the candidate sets, the probability that j
users are selected in the SUS algorithm with constraint ϵ is denoted by P|S|=j .
Only one scheduled user selected at the BS can be interpreted that the other
(K−1) users not satisfy the semi-orthogonal constraint with the selected user.
The first user is selected from B1 = {1, 2, . . . , K} according to maximum
SINR criterion with α1 = 1. The probability that only one scheduled user at
the BS can be expressed by
P|S|=1 = (1 − α2 )K−1 . (D.1)
For the case of only two scheduled users, the second user is selected from
/ {π(1)}, : |ĥl ĥ∗π(1) | ≤ ϵ} according to maximum
B2 = {l|1 ≤ l ≤ K , l ∈
SINR criterion. Due to only two scheduled users, one belongs to B1 and the
other one belongs to B2 . We assume that m users not satisfy the constraint
with π(1) with probability (1 − α2 ) and number of (K − 2 − m) users can
not satisfy the constraint with π(1) and π(2) with probability (1 − α3 ). In
other words, the cardinality of B3 is zero. Thus,
∑
K−2
P|S|=2 = α2 (1 − α2 )m (1 − α3 )K−2−m . (D.2)
m=0
115
Similarity, the probability that only j users are selected at the BS is
K−j−Q(j−1) ( j−1 )
∑
K−j
∑ ∏
P|S|=j = ··· αt+1 (1 − αt+1 )
mt
m1 =0 mj−1 =0 t=1
( )
× (1 − αj+1 )K−j−Q(j)
, j ≥ 2, (D.3)
∑
j−1
where Q(j) = mt .
t=1
116
Appendix E
Proof of Lemma 2
Proof.
∂∆ND (∞, rth,K−n )

=
∂K [(
( )K )K ( )
1 K −n K −1 K −n K −n
+ ln
2 K 2 K K
( )K−1 ( ) ]
K −n K −n 1
−K −
K K2 K
( )K−1 [
K K −n 1
=
2(K − n)(K − 1) K K −1
( ) ]
K −n n
+ ln +
K K −n
According to the inequality ln(x) ≥ 1 − 1/x, we have

( )K−1
∂∆ND (∞, rth,K−n ) K K −n
≥
∂K 2(K − n)(K − 1) K
[ ( ) ]
1 K n
× + 1− + >0
K −1 K −n K −n
117
Appendix F
Proof of Lemma 3
Without losing the generality, we assume that the thresholds of the system
with the decreasing order as τ1 ≥ τ2 ≥ . . . ≥ τK , then
∂R(τ1 , . . . , τK )
< 0, ∀j
∂τj
Proof.
[ j−1 ) ∫ τj−1
∂ (∏ (∏ )
K
∂R(τ1 , . . . , τK )
= Fm (τm ) log2 (1 + x)d Fm (x)
∂τj ∂τj m=1 τj m=j
(∏
j )∫ τj ( ∏
K )
+ Fm (τm ) log2 (1 + x)d Fm (x)
m=1 τj+1 m=j+1
(∏
j+1 )∫ τj+1 ( ∏
K )
+ Fm (τm ) log2 (1 + x)d Fm (x) + · · ·
m=1 τj+2 m=j+2
∏
K−1 ∫ τK−1 ]
+ Fm (τm ) log2 (1 + x)dFK (x)
m=1 τK
118
According to Leibniz’s rule for differentiation under the integral form and
integration by parts, we have
[{
∂R(τ1 , . . . , τK ) (∏
j−1 ) ( ∏
K )
= − Fm (τm ) log2 (1 + τj )fj (τj ) Fm (τj )
∂τj m=1 m=j+1
(∏ j ) ( ∏K ) }
− Fm (τm ) log2 (1 + τj )d Fm (τj )
m=1 m=j+1
{( ∏
j−1 )[ ( ∏
K )
+ Fm (τm )fj (τj ) log2 (1 + τj ) Fm (τj )
m=1 m=j+1
( ∏
K )] ∫ τj
∏K }
Fm (x)
m=j+1
− log2 (1 + τj+1 ) Fm (τj+1 ) − dx
m=j+1 τj+1 (1 + x) ln 2
{ ∏
j+1 [ ∏
K
+ fj (τj ) Fm (τm ) log2 (1 + τj+1 ) Fm (τj+1 )
m=1,m̸=j m=j+2
] ∫ ∏K }
∏
K τj+1
m=j+2 Fm (x)
− log2 (1 + τj+2 ) Fm (τj+2 ) − dx + · · ·
m=j+2 τj+2 (1 + x) ln 2
{ ∏
K−1 [ ]
+ fj (τj ) Fm (τm ) log2 (1 + τK−1 )FK (τK−1 ) − log2 (1 + τK )FK (τK )
m=1,m̸=j
∫ τK−1 }]
FK (x)
− dx
τK (1 + x) ln 2
(
∏
K−1
= − log2 (1 + τK )fj (τj ) Fm (τm )
m=1,m̸=j
{ K−1
∑ ∏ m ∫ τm
∏K })
Fv (x)
− fj (τj ) Ft (τt ) v=m+1
dx <0
m=j t=1,t̸=j τm+1 (1 + x) ln 2
119
Bibliography
[1] R. Knopp and P. Humblet, “Information capacity and power control in
single cell multiuser communications,” IEEE Int. Conf. on Commun.,

June 1995.
[2] 3GPP, “Long term evolution, evolved universal terrestrial radio access
(e-utra); physical layer; general description,” TS 36.201, vol. v1.0.0,
March 2007.
[3] Air Interface for Fixed Broadband Wireless Access Systems. IEEE Std.
802.16-2004, 2004.
[4] M. Costa, “Writing on dirty paper,” IEEE Trans. on Inf. Theory, vol. 29,
no. 3, pp. 439–441, May 1983.
[5] H. Weingarten, Y. Steinberg, and S. Shamai, “The capacity region of

the Gaussian multiple-input multiple-output broadcast channel,” IEEE
Trans. on Inf. Theory, vol. 52, no. 9, pp. 3936–3964, Sept. 2006.
[6] T. Yoo and A. Goldsmith, “On the optimality of multiantenna broad-

cast scheduling using zero-forcing beamforming,” IEEE J. Select. Areas
Commun., vol. 24, no. 3, pp. 528–541, Mar. 2006.
120
[7] Evolved Universal Terrestrial Radio Access (E-UTRA); LTE Physical
Layer-General Description (Release 8). 3GPP TSG RAN TS 36.201
Version 8.3.0, Sep. 2009.
[8] M. Sharif and B. Hassibi, “A comparison of time-sharing, dpc, and
beamforming for mimo broadcast channel with many users,” vol. 55,
no. 1, pp. 11–15, Jan. 2007.
[9] T. Yoo, N. Jindal, and A. Goldsmith, “Multi-antenna downlink chan-
nels with limited feedback and user selection,” IEEE J. Select. Areas
Commun., vol. 25, no. 7, pp. 1478–1491, Sept. 2007.
[10] T. Yoo and A. Goldsmith, “On the optimality of multiantenna broadcast
scheduling using zero-forcing beamforming,” vol. 24, pp. 528–541, Mar.
2006.
[11] M. Sharif and B. Hassibi, “On the capacity of MIMO broadcast channel
with partial side information,” IEEE Trans. on Inf. Theory, vol. 51,
no. 2, pp. 506–522, Feb. 2005.
[12] J. Diaz, O. Simeone, and Y. Bar-Ness, “Sum-rate of mimo broadcast

channels with one bit feedback,” June 2006.
[13] D. Gesbert and M. S. Alouini, “How much feedback is multi-user diver-

sity really worth?” IEEE Int. Conf. on Commun., pp. 234–238, June
2004.
121
[14] V. Hassel, M. S. Alouini, D. Gesbert, and G. E. Oien, “Exploiting mul-
tiuser diversity using multiple feedback thresholds,” IEEE VTC, vol. 2,
pp. 1302–1306, May 2005.
[15] R. Zakhour and D. Gesbert, “A two-stage approach to feedback design
in multi-user MIMO channels with limited channel state information,”

IEEE PIMRC, vol. 24, pp. 1–5, Sept. 2007.
[16] C. Liang, N. Hock, and Z. Liren, “Feedback suppression in reliable mul-
ticast protocol,” IEEE Int. Conf. on Commun., vol. 2, p. 1436V1439,

June 2000.
[17] M. Anastasopoulos and P. Cottis, “High altitude platform networks:
a feedback suppression algorithm for reliable multicast/broadcast ser-
vices,” IEEE Trans. on Wireless Communications, vol. 8, no. 4, pp.
1639–1643, April 2009.
[18] S. Yu, M. Peng, and W. Wang, “Adaptive traffic multiplexing for mbms
in the umts system based on the scalable coding,” Communications and
Networking in China, pp. 1–5, Oct. 2006.
[19] S. Lu, Y. Cai, L. Zhang, J. Li, P. Skov, C. Wang, and Z. He, “Channel-
aware frequency domain packet scheduling for mbms in lte,” IEEE VTC,
pp. 1–5, April 2009.
[20] L. Rong, O. B. Haddada, and S. Elayoubi, “Analytical analysis of the

coverage of a mbsfn ofdma network,” IEEE Globecom, pp. 1–5, April
2008.
122
[21] A. Alexiou, C. Bouras, V. Kokkinos, A. Papazois, and G. Tsichritzis,
“Efficient mcs selection for mbsfn transmissions over lte networks,” IFIP
Wireless Days, pp. 1–5, April 2010.
[22] S. Sohn, H.-S. Kim, and Y. Y. Kim, “A novel multicast scheme for
feedback-based multicast services over wireless networks,” EURASIP

Journal on Wireless Communications and Networking, Feb. 2013.
[23] T. B. Sorensen and M. R. Pons, “Performance evaluation of proportional
fair scheduling algorithm with measured channels,” IEEE VTC, vol. 4,

pp. 2580–2585, Sept. 2005.
[24] L. Yang, M. Kang, and M.-S. Alouini, “On the capacity-fairness tradeoff
in multiuser diversity system,” IEEE Trans. on Vehicular Technology,
vol. 56, no. 4, pp. 1901–1907, July 2005.
[25] N. Jindal, “MIMO broadcast channels with finite-rate feedback,” IEEE
Trans. on Inf. Theory, vol. 52, no. 11, pp. 5045–5060, Nov. 2006.
[26] T. Yoo, N. Jindal, and A. Goldsmith, “Multi-antenna downlink chan-

nels with limited feedback and user selection,” IEEE J. Select. Areas
Commun., vol. 25, no. 7, pp. 1478–1491, Sept. 2007.
[27] B. Hassibi and T. L. Marzetta, “Multiple-antennas and isotropically

random unitary inputs: The received signal density in closed form,”
IEEE Trans. on Inf. Theory, vol. 48, no. 6, pp. 1473–1484, June 2002.
123
[28] L. Yang, M. Kang, and M.-S. Alouini, “On the capacity-fairness tradeoff
in multiuser diversity systems,” IEEE Trans. on Vehicular Technology,
vol. 56, no. 4, pp. 1901–1907, July 2007.
[29] J.-H. Li and H.-J. Su, “Feedback reduction for MIMO broadcast chan-
nel with heterogeneous fading,” IEEE Symposium on Computers and

Communications (ISCC), June 2011.
[30] D. A. Gore, R. W. Heath, and A. J. Paulraj, “Transmit selection in
spatial multiplexing systems,” IEEE Commun. Lett., vol. 6, pp. 491–

493, Nov. 2002.
[31] S. Ghahramani, Fundamentals of probability with stochastic processes,
3rd ed. Prentice Hall, 2005.
[32] D. K. Sharma, “Design of absolutely optimal quantizers for a wide class
of distortion measures,” IEEE Trans. on Inf. Theory, vol. 24, no. 6, pp.
693–702, Nov. 1978.
[33] C. Anton-Haro, “Optimal quantization schemes for orthogonal random

beamforming-a cross-layer approach,” ICASSP, vol. 3, pp. 637–640,
April 2007.
[34] H.-M. Hang and J.-J. Chen, “Source model for transform video coder
and its application−Part I: Fundamental theory,” IEEE Trans. Circuits
Syst. Video Technol., vol. 7, no. 2, pp. 287–298, April 1997.
[35] S. Boyd and L. Vandenberghe, Convex Optimization. Cambridge Uni-

versity Press, 2004.
124
[36] M. Pugh and B. D. Rao, “Reduce feedback schemes using random beam-
forming in MIMO broadcast channels,” IEEE Trans. on Signal Process-
ing, vol. 58, no. 3, pp. 1821–1832, Mar. 2010.
[37] T. Kang and H. Kim, “Optimal beam subset and user selection for
orthogonal random beamforming,” IEEE Commun. Lett., vol. 12, no. 9,

pp. 636–638, 2008.
[38] N. Jindal, “MIMO broadcast channels with finite-rate feedback,” IEEE
Trans. on Inf. Theory, vol. 52, no. 11, pp. 5045–5060, Nov. 2006.
[39] M. Trivellato, F. Boccardi, and F. Tosato, “User selection schemes
for MIMO broadcast channels with limited feedback,” IEEE VTC, pp.
2089–2093, April 2007.
[40] K. Ko, S. Jung, and J. Lee, “Hybrid MU-MISO scheduling with limited
feedback using hierarchical codebooks,” IEEE Trans. on Communica-
tions, pp. 1–13, 2012.
[41] M. Kountouris, “Multiuser multi-antenna sysytems with limited feed-

back,” Ph. D dissertation, Jan. 2008.
[42] Y. Shao and J. Yuan, “A lower bound to the sum-rate of mimo broadcast
channels with limited-rate feedback,” IEEE Int. Conf. on Commun., pp.
3674–3678, May 2008.
[43] G. U. Hwang and F. Ishizaki, “Design of a fair scheduler exploiting mul-

tiuser diversity with feedback information reduction,” IEEE Commun.
Lett., vol. 12, no. 2, pp. 124–126, Feb. 2008.
125
[44] J. So, “Opportunistic feedback with multiple classes in wireless sys-
tems,” IEEE Commun. Lett., vol. 13, no. 5, pp. 384–386, June 2009.
[45] T. L. Marzetta and B. M. Hochwald, “Capacity of a mobile multiple-

antenna communication link in Rayleigh flat fading,” IEEE Trans. on
Inf. Theory, vol. 45, pp. 139–157, 1999.
[46] H. Inaltekin, T. Samarasinghe, and J. S. Evans, “Rate optimal limited

feedback policies for the MIMO downlink,” International Symposium on
Modeling and Optimization in Mobile, Ad Hoc and Wireless Networks

(WiOpt), May 2011.
[47] T. Samarasinghe, H. Inaltekin, and J. S. Evans, “Optimal
selective feedback policies for opportunistic beamforming,”
http://arxiv.org/abs/1107.4822, 2011.
[48] P. Viswanath, D. Tse, and R. Laroia, “Opportunistic beamforming using
dumb antennas,” IEEE Trans. on Inf. Theory, vol. 48, no. 6, pp. 1277–
1294, Jun. 2002.
[49] I. Gradshteyn and I. Ryzhik, Table of Integrals, Series, and Products,

7th ed. U.K.:Academic, 2007.
[50] O. Gascuel and G. Caraux, “Bounds on expectations of order statistics

via extremal dependences,” Statist. Probab. Lett., pp. 143–148, Sept.
1992.
126

DL Random Beamforming

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DL Random Beamforming

Uploaded by

Copyright:

Available Formats

國立臺灣大學電機資訊學院電信工程學研究所

Feedback Reduction in MIMO Broadcast Channel

In multiuser multiple-input multiple-output (MIMO) downlink broadcast

channel, data rate can be tremendously increased by adding antennas and

be achieved by using space-division multiple access schemes or zero-forcing

beamforming. Moreover, in a large user regime K >> Mt , the sum rate

important in terms of proper selection of modulation and coding for achieving

reliable transmission. Thus the aggregate feedback load in a network is large.

The main contributions of thesis are proposing feedback reduction strate-

scheduling method with multi-threshold model is proposed and the relations

2 Reduction Method in MIMO Broadcast Channel using Or-

thogonal Random Beamforming Approach 10

2.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.2 The Multi-threshold Feedback Scheme . . . . . . . . . . . . . 19

2.2.1 Derivation of Multiple Thresholds . . . . . . . . . . . . 20

2.2.2 Sum Rate Loss Analysis and Minimum Number of Re-

3 Reduction Method for Multiuser MIMO Zero-forcing Beam-

3.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

3.2.2 CQI Metric . . . . . . . . . . . . . . . . . . . . . . . . 57

3.3 Feedback Reduction Strategy . . . . . . . . . . . . . . . . . . 60

3.3.1 Derivation of Multiple Thresholds . . . . . . . . . . . . 60

3.3.2 Feedback Load Analysis . . . . . . . . . . . . . . . . . 61

3.4 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . . . 62

4 Feedback Reduction in MBMS System 68

4.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

4.2 Threshold Design Using Order Statistics . . . . . . . . . . . . 72

5.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

6 Sum Rate and Fairness Tradeoﬀ with Feedback Reduction in

Multiuser Heterogeneous Rayleigh Fading Channel 92

6.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

6.1.1 Transmission and Feedback Procedure . . . . . . . . . 94

6.2 Design of Multiple Thresholds . . . . . . . . . . . . . . . . . . 95

6.2.1 Proposed Scheduler . . . . . . . . . . . . . . . . . . . . 97

6.3 Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . 98

6.3.1 Sum Rate Analysis . . . . . . . . . . . . . . . . . . . . 98

6.3.2 Fairness Analysis . . . . . . . . . . . . . . . . . . . . . 99

6.3.3 Feedback Load Analysis . . . . . . . . . . . . . . . . . 100

A Proof of Theorem 1 109

B Proof of Theorem 2 111

D Proof of corollary 115

E Proof of Lemma 2 117

F Proof of Lemma 3 118

2.1 Multiuser MIMO downlink system model. . . . . . . . . . . . 16

2.2 Multi-threshold feedback model. . . . . . . . . . . . . . . . . . 21

direction is ranked the p-th among all the users. Mr = Mt ,

bps/Hz is the tolerable sum rate loss. ρ = 10 dB, Mt = Mr = 4. 29

2.5 Sum rate performance of the multi-threshold scheme with dif-

ferent numbers of regions. ρ = 10 dB, Mt = Mr = 4. . . . . . . 31

2.6 Sum rate comparison between the mathematical lower bound

and the simulation result, ρ = 10 dB, Mt = Mr = 4. . . . . . . 36

2.7 Sum rate performance comparison for diﬀerent bit allocation

3.1 System model of zero-forcing beamforming using SUS algorithm. 56

with diﬀerent CQI metrics, P =10W, BD =8 bits. . . . . . . . . 65

P =10W, BD =8 bits, BQ =2 bits and ε=0.4 . . . . . . . . . . . 66

3.4 Comparison of sum rate as a function of feedback load for

diﬀerent feedback types, Mt =4, P =10W, BD =8 bits, BQ =2

bits, ε=0.4 and K = 100 ∼ 500. . . . . . . . . . . . . . . . . . 67

4.1 MBMS system model . . . . . . . . . . . . . . . . . . . . . . . 70

4.2 Increment of number of successfully decoding users increasing

threshold from rth,K−n to rth,K−(n+1) . . . . . . . . . . . . . . . 78

4.3 Number of successfully decoding users of proposed scheme. . . 81

5.1 Rate comparison for diﬀerent algorithm. ρ=10dB. . . . . . . . 90

6.1 Sum rate comparison between diﬀerent number of feedback