A thesis submitted in partial fulfillment of the requirement of Nazarbayev University for the degree of Doctor of Philosophy

(1)

HARVESTING AND NON-ORTHOGONAL MULTIPLE ACCESS

Sultangali Arzykulov

A thesis submitted in partial fulfillment of the requirement of Nazarbayev University for the degree of Doctor of Philosophy

April 2019

(2)

Declaration

I declare that the research contained in this thesis, unless otherwise formally indicated within the text, is the original work of the author. The thesis has not been previously submitted to this or any other university for a degree, and does not incorporate any material already submitted for a degree.

Signed:

Dated:

(3)

Abstract

The increasing demand for wireless applications is making radio spectrum scarce. Mean- time, studies show that the assigned spectrum is not thoroughly utilized. The cognitive radio (CR) technology is proposed as a feasible key technology to solve issues related to the spectrum scarcity. CR can improve the spectrum utilization by reusing the unused spectrum occupied by licensed users. Introduction of CR networks produces two kinds of interference: interference from the CR network (secondary network) to the primary network (PN) and the interference among secondary users. All unwanted interference should be adequately managed in order not to jeopardize the performance of the PN and at the same time improve the performance of CR systems. Interference alignment (IA) is a promising technique that can efficiently manage interference. One of the aims of this thesis is to mitigate the interference by deploying multiple antennas at both transmitter and receiver sides in order to improve the performance of CR networks.

The rapid growth of data-hungry wireless applications is forcing us to perform energy harvesting (EH) from external power sources for the next-generation of wireless communication systems. Especially, CR networks, where receiver nodes need advanced hardware to process a large amount of data, require higher energy consumption. Thus, another goal of the current thesis is to investigate simultaneous wireless information and power transfer (SWIPT) in CR networks in the presence of intra- and inter-network interference over various channel state conditions.

Firstly, a cooperative CR network is investigated over the generalα−µfading channel.

The contribution of this study is mainly described by the exact closed-form expression for the outage probability (OP) of secondary users, which clearly shows how the outage saturation paradigm appears when the interference level at primary receiver is applied.

Secondly, the proposed cooperative CR is extended by applying multiple-input multiple out (MIMO) antennas and an IA technique to deal with intra- and inter-network interference. The negative effect of interference at both primary and secondary receivers is mitigated by using precoding and interference suppression beamforming matrices. The management of interference at primary receivers allows secondary transmitters to increase the transmit power level. Moreover, the instantaneous capacity performance is assessed for the same CR system by applying two EH methods, i.e., time-switching (TS) and power-splitting (PS). Then, the optimal values of TS and PS portions are determined for different channel state information (CSI) scenarios. In addition, the effect of imperfect CSI on bit error rate and capacity performance is provided.

Finally, we jointly study a cooperative CR and non-orthogonal multiple access (NOMA), where we derive closed-form expressions for the OP of NOMA secondary destination users for detect-and-forward and amplify-and-forward relaying techniques. Furthermore, power allocation factors for different distances of secondary NOMA users are found to satisfy OP fairness for all users. In addition, the proposed CR-NOMA network model is further studied with enabled SWIPT technology.

(4)

Acknowledgments

First and above all, I thank the Almighty for giving me the strength and patience to work through all these years.

My greatest appreciations go to my supervisor - Professor Theodoros A. TSIFTSIS. Thank him for introducing me into the research realm of wireless communications, the patient guidance, encouragement and advice he has provided throughout my Ph.D. study.

He has taught me the methodology to carry out the research and to present the research works as clearly as possible. It was a great privilege and honor to work and study under his guidance.

My heartfelt thanks also go to my supervisor - Associate Professor Behrouz MA-

HAM, co-supervisor - Assistant ProfessorRefik C. KIZILIRMAK and external supervisor - Assistant ProfessorMohamed M. ABDALLAHwho cared about my work so much and responded to my questions and queries so promptly. I feel extremely lucky to have them as supervisors.

I want to express my deep gratitude to Assistant ProfessorGalymzhan NAURYZBAYEV

for all his help from the beginning of my studies, for generously giving his time to offer me valuable comments toward improving my work and for being a good friend.

I would also like to take this opportunity to thank ProfessorMohamed-Slim ALOUINI

and Associate Professor Mohammad S. HASHMI - my viva examiners, for their very helpful comments and suggestions.

I am also very grateful to Professor Mohammad S. OBAIDAT, who has given me valuable support and who had a word of encouragement during my Ph.D. studies.

Some special words of gratitude go to Director of Ph.D. Studies Associate Professor Luis R. Rojas-SOLÓRZANOfor his suggestions throughout my research and dealing with the bureaucracies of doing a Ph.D.

I would like to thank all my Ph.D. colleagues, with whom I have shared moments of deep anxiety but also of big excitement. Their presence was very important in a process that is often felt as tremendously solitaire. I would also like to thank all my friends for accepting nothing less than excellence from me.

I am extremely grateful to my parents for their love, prayers, caring and sacrifices for educating and preparing me for my future. I am very much thankful to my wife and my sons for their love, understanding and continuing support to complete this research work. Also, I express my thanks to my sisters and brothers for their support and valuable prayers. To you, I owe the deepest and most sincere thanks.

Finally, my thanks go to all the people who have supported me to complete the research work directly or indirectly.

Sultangali ARZYKULOV

(5)

Preface

Mr. Sultangali ARZYKULOV received the B.Sc. (Hons.) degree in radio engineering, electronics and telecommunications from Kazakh National Technical University named after K. I. Satpayev, Almaty, Kazakhstan, in June 2010. He received M.Sc. degree in communication engineering from theUniversity of Manchesterin September2013. From 2010 to2011, he worked as a network engineer in JSC “Kazakhtelecom”, and then as a teaching assistant at Nazarbayev Universityfrom 2014to 2018. Since September2015, he has been pursuing his Ph.D. degree in Science, Engineering and Technology with a research focus on cognitive radio, multi-user MIMO systems, interference mitigation, wireless energy harvesting, non-orthogonal multiple access and millimeter wave communications

Main Supervisors

The topic of the current thesis was proposed by Prof. Theodoros A. TSIFTSIS, who was the main supervisor from September2015to March2018. He left Nazarbayev University in March2018and his position as the main supervisor was taken by Dr.Behrouz MAHAM

who was also in the initial team of the Ph.D. supervisory committee. This change was made for typical reasons due to the Ph.D. Program regulations ofNazarbayev University.

(6)

List of Figures

2.1 The radio spectrum allocations in the USA [1]. . . 9

2.2 Cognitive radio network. . . 10

2.3 The Cognitive radio cycle. . . 11

2.4 Cognitive radio paradigms. . . 12

2.5 The MIMO network withNT transmit andNRreceive antennas. . . 13

2.6 A transceiver with energy harvesting module. . . 16

2.7 An example of a power-domain NOMA for two users.. . . 18

2.8 An illustration of CR-NOMA [2]. . . 20

3.1 The basic cooperative relaying schemes: amplify-and-forward and decode- and-forward.. . . 22

3.2 A system model for cooperative underlay CR network consisting of a secondary source, relay and destination. . . 25

3.3 The OP versus SNR performance for various ITC over the Weibull distribution. . . 31

3.4 The OP versus SNR performance for different ITC over Rayleigh and Negative Exponential distributions. . . 32

3.5 The SER versus SNR performance over the Weibull distribution when I = 20dB. . . 32

4.1 Time-switching and power-splitting energy harvesting modes. . . 35

4.2 The IA- and EH-based CRN withLPUs and one SN sharing the spectrum simultaneously. . . 38

4.3 The time frame structure of the TSR. . . 42

4.4 Time frame structure of the PSR. . . 46

4.5 Average capacity versus the EH TS and PS coefficients forDin TSR and PSR protocols at20dB, respectively. . . 51

4.6 Average capacity versus SNR of P R_j and D operating in the TSR and PSR protocols under different CSI scenarios: a) Perfect CSI versus CSI mismatch (κ = 0, ψ = 0.001). b) CSI mismatch: (κ = 1.5, ψ = 15) versus(κ= 1, ψ= 10). . . 52

(11)

4.7 BER versus SNR for the PSR protocol under different CSI scenarios: a) BER performance for perfect CSI and SNR-independent CSI mismatch cases((0, 0.001)and(0, 0.005)). b) BER performance for SNR-dependent CSI mismatch cases((0.75, 10), (1, 10)and(1.5, 15)). . . 54 4.8 BER versus CSI mismatch parametersκandψofDat20dB in the PSR

mode. . . 55 4.9 The OP of the PN and the SN with different values of data rate threshold

at25dB with the perfect CSI case: a) The OP of the system in the PSR protocol. b) The OP of the system in the TSR protocol. . . 56 4.10 Average harvested power versus distance for the TSR and PSR protocols

at20dB: a) The TSR protocol. b) The PSR protocol. . . 57 5.1 A system model of the downlink dual-hop underlay CR-NOMA network

with transmit power constraint at the secondary source. . . 61 5.2 The OP ofx₁andx₂versus SNR performance whend₁ ={1.5d,2d,3d,5d},

α₁ = 0.8andα₂ = 0.2. . . 67 5.3 The OP of x₁ and x₂ versus the SNR threshold at 20 dB when d₁ =

{1.5d,2d,3d,5d},α₁ = 0.8andα₂ = 0.2. . . 68 5.4 The OP ofx₁andx₂versus PA factors at20dB whend₁ ={1.5d,2d,3d,5d}. 68 5.5 The OP ofx₁andx₂versus the transmit SNR performance with OF-based

PA factors whend₁={2d,3d,5d}. . . 69 5.6 The OP versus SNR threshold with OF-based PA factors at20dB, when

d₁ ={2d,3d,5d}. . . 70 5.7 The OP versus the transmit SNR performance for NOMA and OMA users

withα₁ = 0.8,d₁ = 2dandu= 4.7dB. . . 70 5.8 The OP of D₁ and D₂ versus the transmit SNR when d₁ = {2d,4d},

α₁ = 0.8,α₂ = 0.2andτ = 0. . . 72 5.9 The OP ofD₁andD₂versus the transmit SNR whend₁ = 3d, α₁ = 0.8,

α₂ = 0.2andτ = 0. . . 72 5.10 The OP ofD₁andD₂versus the PA factors at20dB whend₁ ={1.5d,2d,4d}

andτ = 0. . . 73 5.11 The OP ofD1andD2versus the transmit SNR with OF-based PA factors

whend₁ ={1.5d,2d,4d}andτ = 0. . . 74 5.12 The OP ofD₁andD₂versus the predefined SNR threshold with OF-based

PA factors at20dB whend₁ ={1.5d,2d,4d}andτ = 0. . . 74 5.13 A dual-hop underlay dual-hop CR-NOMA network. . . 76 5.14 The OP versus the transmit SNR forx₁ andx₂ whend₂ ={2 m, 4 m, 6

m},a₁ =b₁ = 0.2anda₂ =b₂ = 0.8. . . 79 5.15 The OP versusa₁ forx₁ andx₂ at 30 dB whend_SR ={d, 2d}. . . 80

(12)

5.16 The optimized OP versus the transmit SNR forx₁andx₂whend₂={2d, 4d, 6d}. . . 80 5.17 A downlink underlay dual-hop CR-NOMA network. . . 82 5.18 The OP of messages x1 and x2 versus the transmit SNR when d1 =

{1.5d,2d,3d},α₁ = 0.8,α₂ = 0.2,ν = 0andσ²_˜

hι = 0. . . 87 5.19 The OP of messagesx₁ andx₂ versus the SNR threshold at 20dB when

d₁ ={1.5d,2d,3d},α₁ = 0.8,α₂ = 0.2,ν= 0andσ²_˜

hι = 0. . . 88 5.20 The OP of messages x₁ and x₂ versus PA factors at 20 dB when d₁ =

{1.5d,2d,3d},ν = 0andσ_˜²

hι = 0. . . 89 5.21 The OP of messagesx₁ and x₂ versus the transmit SNR with OF-based

PA factors whend₁={1.5d,2d,3d},ν= 0andσ²_˜_h

ι = 0. . . 90 5.22 The OP of messagesx1 andx2 versus the SNR threshold with OF-based

PA factors at20dB whend1 ={1.5d,2d,3d},ν = 0andσ_˜_h²

ι = 0. . . 90 5.23 The OP of messages x₁ andx₂ versus the transmit SNR for NOMA and

OMA scenarios withα₁ = 0.8,d₁ = 2d,ν = 0andσ_˜²

hι = 0. . . 91 5.24 The OP of messages x₁ and x₂ versus the transmit SNR with imperfect

CSI and PN’s interference whenα₁ = 0.8andd₁ = 2d. . . 92 5.25 The average throughput versus the transmit SNR for NOMA messages

withα₁ = 0.8andd₁ = 2d. . . 92 5.26 A system model of the downlink dual-hop underlay CR-NOMA network

with the AF relaying. . . 93 5.27 The OP of Users1 and2versus the transmit SNR whend1 = {2d,4d},

α₁ = 0.8,α₂ = 0.2andτ = 0. . . 96 6.1 A downlink dual-hop underlay CR-NOMA network. . . 99 6.2 The time frame structure of the PSR protocol. . . 100 6.3 The OP of x₁ and x₂ versus the ransmit SNR threshold at 25dB when

d₁ = 3d,α₁ = 0.8andα₂ = 0.2. . . 102 6.4 The OP ofx₁ andx₂ versus PA factors at20dB whend₁ ={1.5d,3d,5d}. 103 6.5 The OP ofx₁andx₂versus the transmit SNR performance with OF-based

PA factors whend1={1.5d,3d,4d}. . . 104 6.6 Harvested power versus theS-Rdistance performance with variousθvalues.104

(13)

List of Tables

5.1 OF-based PA factors for differentd₁for the DF mode.. . . 69

5.2 The observed OF-based PA factors for the AF mode. . . 73

5.3 The observed OF-based PA factors (Kusers). . . 89

6.1 OF-based PA factors for differentd1with EH. . . 103

(14)

List of Abbreviations

1G First Generation

3GPP Third Generation Partnership Project 4G Fourth Generation

5G Fifth Generation AF Amplify-and-Forward

AWGN Additive White Gaussian Noise

BF Beamforming

BPSK Binary Phase Shift Keying

CDF Cumulative Distribution Function CDMA Code Division Multiple Access CF Compress-and-Forward

CR Cognitive Radio

CRN Cognitive Relaying Network CSI Channel State Information DF Decode-and-Forward DoF Degrees of Freedom DSA Dynamic Spectrum Access EH Energy Harvesting

FCC Federal Communications Commission FDMA Frequency-Division Multiple Access i.i.d. Independent and Identically Distributed

i.n.i.d. Independent But Not Necessarily Identically Distributed IA Interference Alignment

ID Information Decoding IoT Internet of Things IT Information Transfer

ITC Interference Temperature Constraint LDS Low Density Spreading

LPMA Lattice Partition Multiple Access LTE Long Term Evolution

MA Multiple Access

MGF Moment Generating Function

(15)

MIMO Multiple-Input Multiple-Output MPA Message Passing Algorithm

MUST Multiple User Superposition Transmission NOMA Non-Orthogonal Multiple Access

Ofcom Office of Communications

OFDMA Orthogonal Frequency Division Multiple Access OMA Orthogonal Multiple Access

OF Outage Fairness OP Outage Probability PA Power Allocation

PDF Probability Density Function PDMA Pattern Division Multiple Access

PN Primary Network

PS Power-Splitting

PSR Power-Splitting Relaying

PU Primary User

QoS Quality of Service

RF Radio Frequency

RV Random Variable

SCMA Sparse Code Multiple Access SER Symbol Error Rate

SIC Successive Interference Cancellation SINR Signal-to-Interference-plus-Noise Ratio SN Secondary Network

SNR Signal-to-Noise Ratio SR Selective Relaying

SU Secondary User

SWIPT Simultaneous Wireless Information and Power Transfer TDMA Time-Division Multiple Access

TS Time-Switching

TSR Time-Switching Relaying WBAN Wireless Body Area Network WEH Wireless Energy Harvesting WiFi Wireless Fidelity

WiMax Worldwide Interoperability for Microwave Access WPT Wireless Power Transfer

(16)

List of Symbols

(·)^[ι] Time slot index

(·)^† Hermitian transpose of a matrix

CN(a, b) Complex normal distribution withamean andbvariance f_X(x) Probability density function of random variableX F_X(x) Cumulative distribution function of random variableX Γ(·) Gamma function

Γ(a, b) Upper incomplete gamma function γ_inc(a, b) Lower incomplete gamma function E(·) Statistical expectation operator h_ji Channel gain between nodesj andi

P_(·) Transmit power at the secondary transmitter

P¯_(·) Maximum allowed transmit power at the secondary transmitter I Interference temperature constraint

σ² Noise power at the receive node ξ Predefined SNR threshold

K₁(·) Modified Bessel function of the second kind of order 1 R_th Predefined rate threshold

min(a, b) Function that finds minimum betweenaandb M_γ(−g) Moment generating function

x Lower bold case denotes vector X Upper bold case denotes matrix

H MIMO channel

I Identity matrix tr(X) Trace of a matrix

|x| Absolute value ofx kxk Euclidean norm ofx V(·) Variance operator

C^N×M Space of complexN ×M matrices

span(X) Vector space spanned by the column vectors eig(X) Eigenvectors ofX

d_m,n Distance between nodesmandn

(17)

τ Path-loss coefficient V Precoding BF matrix

U Interference suppression matrix f Number of data streams

T_j Primary transmitterj

n AWGN noise

ρ Nominal transmit SNR α Time fraction of the TS θ Power fraction of the PS Q(x) Gaussian-Qfunction

α(·) NOMA power allocation factor at the source β(·) NOMA power allocation factor at the relay D Secondary Destination

G Amplification factor at the relay P R Primary receiver

S Secondary Source

R Secondary Relay

(18)

DEDICATED WITH EXTREME AFFECTION AND GRATITUDE TO

my parents Mr. Ussenbatyr ARZYKULOVand Mrs. Zhantore ARZYKULOVA

my brothers NURZHANand RAKHYMZHAN

my sisters NURGUL, MARZHAN and YERKEZHAN

my wife DINARA and my sons DAMIRand ARSEN

(19)

Chapter 1

Introduction

1.1 Modern Wireless Systems

Wireless communication is the technology that has certainly changed the lives of hu- man beings. Nowadays, many applications establish their broadcasting using wireless communication concept. The range of those applications varies from highly commer- cialized satellite and mobile communication networks to amateur radio communications, from regularly practiced Wi-Fi systems to rarely used deep-space networks, from fully infrastructure-based radio/television networks to non-infrastructure ad-hoc systems. The number of wireless applications will be increased dramatically in the future due to the high demand for the wireless medium.

1.2 Motivation

The radio frequency (RF) spectrum, an irreplaceable carrier underlying the wireless communication system, is conditionally assigned to a particular wireless application for independent use by regulatory authorities, such as the Federal Communications Commission (FCC) in the United States, the Office of Communications (Ofcom) in the UK and Min- istry of Information and Communication of the Republic of Kazakhstan. Nowadays, the radio spectrum is becoming increasingly scarce due to the fact that more and more devices become wireless and occupy more RF spectrum. Meantime, studies show that some spectrum bands are left without being fully utilized [1]. The research works on the mea- surement of spectrum usage showed that up to 85% of the spectrum in some frequency bands below3GHz are not utilized [3]. The imbalance between the likely spectrum deficit and possible inadequate spectrum utilization inspired the innovative paradigm shift in access to the spectrum by permitting dynamic spectrum sharing and reuse, which is called as dynamic spectrum access (DSA) [4]. The most prominent candidate technology for the DSA is Cognitive Radio (CR), which is able to sense the environment and adapt its operating parameters dynamically. Hence, CR can coexist with existing systems (primary

(20)

systems) autonomously in a non-intrusive fashion [1]. Doing so, CR is a promising technology that can significantly improve the spectrum usage by reusing the spectrum that belongs to primary systems [5].

One of the major impairments in wireless communications is interference, which appears due to the broadcast and superposition nature of wireless medium. In the meaning of CR systems, interference problems are an extremely serious issue due to the following two main aspects. On the one hand, CR adheres to the fundamental premise of not cre- ating any harmful interference to the primary system. On the other hand, the efficiency of CR can be limited by interference coming from either primary or other secondary CR nodes (secondary users). Hence, the problems associated with interference in CR networks deserve a thorough and extensive study, which is one of the main focuses of the thesis.

The eventual fate of mobile communication is expected to be completely different from what we experience today. Ultra top-notch videos and better wide-screen resolutions in mobile devices are forcing us either to look for better sustainable power sources or to perform energy harvesting (EH) from external power sources for the next-generation of wireless communication systems. Especially, CR networks, where receiver nodes need advanced hardware to process a large amount of information data, require higher energy consumption. Thus, another focus of the thesis is to investigate EH in CR networks in the presence of intra- and inter-network interference.

The fast increase in the demand for the mobile Internet has propelled 1000-fold data traffic boost by 2020 for the fifth generation (5G) mobile communication networks. In addition, 5G networks need to maintain massive connectivity of devices to meet the requirement for low-cost devices, low latency, and various service types due to the rapid development of the Internet of Things (IoT). Henceforth, such explosive data traffic de- livers even more key challenges to efficiently handle the wireless spectrum. So far, non- orthogonal multiple access (NOMA) has been proposed as a potential candidate technology for 5G [6], which not only efficiently utilize the wireless spectrum as CR but also expected to increase the system throughput and support massive connectivity. Therefore, to further improve the performance of5G networks, the application of NOMA to CR networks seems to be very crucial. Therefore, one more aim of the thesis is to examine a synergy between these two advanced communication techniques considering embedded EH techniques in the presence of interference.

(21)

1.3 Key Contributions

The main contributions of the current thesis are briefly summarized as follows:

• We model a cooperative underlay CR system where a primary receiver applies interference temperature model to secondary transmit nodes. The end-to-end closed- form expressions for the outage performance of secondary users are provided over the generalizedα−µfading distribution, which includes various other distributions, i.e., Rayleigh, Gamma, Exponential, Nakagami-m, etc. The impact of interference temperature on the outage performance and symbol error rate of the secondary network is investigated.

• Further, the proposed cooperative underlay CR is studied by applying multiple antennas and interference managing techniques to deal with intra- and inter-network interference. We apply precoding and interference suppression beamforming matrices to mitigate the negative effect of interference at primary and secondary receivers. The management of interference at primary receivers allows secondary transmitters to increase the transmit power level. Closed-form expressions for the outage performance of primary and secondary networks are determined.

• The performance of CR relaying networks is investigated by applying EH at the secondary relay node. EH methods are studied under imperfect channel conditions.

Moreover, the instantaneous capacity performance is assessed for both methods.

In addition, some useful insights into how the channel state information quality impacts on capacity performance are also provided.

• The synergy of cooperative underlay CR and NOMA is studied in terms of the outage performance of NOMA users by applying relaying protocols over Rayleigh and Nakagami-m fading channels. We find power allocation factors for NOMA secondary users to guarantee outage fairness among NOMA users. Moreover, considering the effect of the imperfect CSI on the outage performance, we derive a general closed-form solution for the outage probability when the number of secondary NOMA nodes is extended toK users. Finally, EH-enabled CR-NOMA is investigated in terms of the outage and EH performance.

(22)

1.4 List of Publications

The work presented in this thesis has led to the following publications:

Published and Accepted Papers

Journal Papers

• S. Arzykulov, G. Nauryzbayev, T. A. Tsiftsis, and M. Abdallah, “On the Perfor- mance of Wireless Powered Cognitive Relay Network with Interference Align- ment,” IEEE Transactions on Communications, vol. 66, no. 9, pp. 3825-3836, September 2018.

• S. Arzykulov, T. A. Tsiftsis, G. Nauryzbayev, and M. Abdallah, “Outage Perfor- mance of Cooperative Underlay CR-NOMA with Imperfect CSI,” IEEE Commu- nications Letters, vol. 23, no. 1, pp. 176-179, January 2019.

• S. Arzykulov, G. Nauryzbayev, and T. A. Tsiftsis, “Underlay Cognitive Relaying System overα−µFading Channels,” IEEE Communications Letters, vol. 21, no.

1, pp. 216-219, January 2017.

Conference Papers

• S. Arzykulov, T.A. Tsiftsis, G. Nauryzbayev, and M. Abdallah, “Outage Perfor- mance of Underlay CR-NOMA Networks with Detect-and-Forward Relaying,”2018 IEEE Global Communications Conference (GLOBECOM), Abu Dhabi, United Arab Emirates, 2018, pp. 1-6.

• S. Arzykulov, G. Nauryzbayev, T.A. Tsiftsis, and M. Abdallah, “On the Capacity of Wireless Powered Cognitive Relay Network with Interference Alignment,”GLOBE- COM 2017 - 2017 IEEE Global Communications Conference, Singapore, 2017, pp.

1-6.

• S. Arzykulov, G. Nauryzbayev, T.A. Tsiftsis, and M. Abdallah, “On the BER performance of Wireless Powered Cognitive Relay Network with Interference Align- ment,” in 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Montreal, QC, 2017, pp. 1-5.

• S. Arzykulov, T.A. Tsiftsis, G. Nauryzbayev, and M. Abdallah, “Outage Perfor- mance of Underlay CR-NOMA Networks,”2018 10th International Conference on Wireless Communications and Signal Processing (WCSP), Hangzhou, 2018, pp.

1-6.

(23)

• G. Nauryzbayev, S. Arzykulov, T.A. Tsiftsis, and M. Abdallah, “Performance of Cooperative Underlay CR-NOMA Networks over Nakagami-m Channels,” 2018 IEEE International Conference on Communications Workshops (ICC Workshops), Kansas City, MO, 2018, pp. 1-6.

• G. Nauryzbayev, S. Arzykulov, E. Alsusa, and T. A. Tsiftsis, “An Alignment-based Interference Cancellation Scheme for Network-MIMO Systems,” 2016 10th Inter- national Conference on Signal Processing and Communication Systems (ICSPCS), Gold Coast, QLD, 2016, pp. 1-5.

• G. Nauryzbayev, S. Arzykulov, E. Alsusa, and T. A. Tsiftsis, “A Closed-form Solu- tion to Implement Interference Alignment and Cancellation Scheme for the MIMO Three-user X-channel Model,”2016 10th International Conference on Signal Pro- cessing and Communication Systems (ICSPCS), Gold Coast, QLD, 2016, pp. 1-6.

Submitted Papers

Journal Papers

• S. Arzykulov, G. Nauryzbayev, T. A. Tsiftsis, B. Maham, and M. Abdallah, “On the Outage Performance of Relaying CR-NOMA,” IEEE Transactions on Cognitive Communications and Networking, April 2019.

• S. Arzykulov, G. Nauryzbayev, T. A. Tsiftsis, and B. Maham, “Outage of AF-based Underlay CR-NOMA Networks,” IEEE Transactions on Vehicular Technology, January 2019.

• S. Arzykulov, G. Nauryzbayev, T. A. Tsiftsis, and B. Maham, “New Results on the Outage for CR-NOMA Relaying Systems,” IEEE Access, March 2019.

Conference Papers

• S. Arzykulov, G. Nauryzbayev, T.A. Tsiftsis, B. Maham, and K. M. Rabie, “Wire- less Powered Cognitive Relay Networks: Outage Analysis,” IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, Istanbul, Turkey, September 2019.

• G. Nauryzbayev, S. Arzykulov, and K. M. Rabie, “Underlay CR-NOMA Relay- ing Networks over Generalized Fading Channels,” IEEE Global Communications Conference (Globecom), Waikoloa, HI, USA, December 2019.

• N. Miridakis, S. Arzykulov, T. A. Tsiftsis, G. Yang, and G. Nauryzbayev, “Green CR-NOMA: A New Interweave Energy Harvesting Transmission Scheme for Sec- ondary Access,” 16th International Symposium on Wireless Communication Sys- tems, Oulu, Finland, August 2019.

(24)

• A. U. Makarfi, K. M. Rabie, R. Kharel, O. S. Badarneh, G. Nauryzbayev, X. Li, S. Arzykulov, and O. Kaiwartya, “Performance Analysis of SWIPT Networks over Composite Fading Channels,” IEEE Global Communications Conference (Globe- com), Waikoloa, HI, USA, December 2019.

1.5 Thesis Organization

The thesis consists of seven chapters. Chapter 1is the introduction, which provides an overview of the research, motivations and contributions. The following chapters are orga- nized as the next.

Chapter 2 presents a literature review on the relevant background for the research work presented in this thesis. We first provide an introduction to the concepts of the CR with its main three paradigms. Then, some review on using multiple antennas in CR networks are provided. Moreover, different types of interference involved in CR networks are analyzed, and an interference managing technique named interference alignment is proposed as a novel approach to deal with interference. In the end, we review works where the improvement in the performance of the CR networks by applying wireless energy harvesting and NOMA techniques is shown.

Chapter3starts with some introduction of relaying networks in the CR systems. Then, a system model of a dual-hop CR with interference temperature constraints at a primary receiver is presented. The end-to-end outage probability (OP) for the secondary network is derived for Rayleigh and generalα−µfading channels. Moreover, evaluation of the symbol error rate is derived for the proposed system model. Furthermore, in the numer- ical part, the correctness of the derived analytical results are validated through extensive Monte Carlo simulations.

In Chapter 4, we investigate a dual-hop decode-and-forward wireless powered cognitive relaying network with interference alignment over Rayleigh fading channels. We assume a secondary relay node to be energy-constrained which needs to harvest energy from both information and interference signals. By applying beamforming matrices to primary and secondary networks, the performance metrics such as the OP, capacity and bit error rate are studied under perfect and imperfect channel state information scenarios for both power-splitting relaying and time-switching relaying protocols. Moreover, the Chapter studies the optimal network performance by calculating the optimal energy harvesting time-switching and power-splitting coefficients. Finally, closed-form expressions for the OP of primary and secondary users are derived which are validated through Monte Carlo simulations.

Chapter 5 focuses on a downlink dual-hop CR-NOMA model with an imposed interference constraint at the primary receiver. The chapter considers several different

(25)

scenarios for the proposed system model. First, it proposes a system model with two NOMA secondary destination users, where only one secondary transmit node is limited by maximum transmit power. We derive the OP for the considered system model applying decode-and-forward (DF) and amplify-and-forward (AF) relaying methods over Rayleigh and Nakagami-m fading channels. Then, by considering the effect of imperfect CSI on the system performance, we derive a general closed-form solution for the OP when the number of secondary NOMA users is extended toK users. In addition, the OP for the scenario where both source and relay are restricted by the maximum transmit power is studied. Furthermore, in order to obtain fairness between NOMA users, the power allocation factors are optimized depending on the distance between secondary users. Finally, the obtained OP performance for cooperative NOMA compared to that for conventional cooperative OMA to show the supremacy of the former.

Chapter 6 studies the OP of the dual-hop underlay CR network consisted of a secondary relay node with energy harvesting and two NOMA secondary destination users.

Moreover, the optimal power allocation factors are found for different distances of NOMA users in order to satisfy OP fairness for both secondary destination users. Additionally, the proposed CR-NOMA is compared with conventional CR multiple access in terms of the OP.

Finally, Chapter7concludes the thesis and discusses related future works.

(26)

Chapter 2

Literature Review

2.1 Radio Spectrum

In recent years, wireless communication networks have obtained significant popularity due to its convenient untethered connectivity and mobile access. However, the radio spectrum in wireless communication networks is a naturally limited resource. Traditionally, spectrum regulators adopt the fixed spectrum access technique in order to support various wireless services and applications, where each spectrum band is assigned to one or several dedicated users. As a result of doing this, only the delegated, a.k.a. licensed, users get access to communicate over the allotted spectrum band, while other unlicensed users do not have rights to use that spectrum, even if the spectrum is not used by licensed users.

The recent fast-growing demand on wireless communications systems has resulted in the fact that most of the available spectrum bands have fully been occupied, which leads to the spectrum scarcity issue. Moreover, the increase of the different wireless services, i.e., voice, web, multimedia, etc., has triggered the overcrowding of spectrum usage, which has consequently brought forth the low quality of service (QoS) for wireless network users [5]. Fig. 2.1 shows the radio spectrum allocation in the USA, where it illustrates that the radio spectrum has been entirely reserved by different wireless applications. From the figure, one can judge that the spectrum is too limited to support additional wireless applications. However, due to the high demand for wireless communications, the spectrum needs to be further utilized by newly emerging wireless networks. For example, over the last three decades, the need for extra bandwidth (radio spectrum) of cellular communication systems has increased exponentially while evolving from the voice-oriented first generation (1G) to the multimedia-rich fourth generation (4G). Despite the widely observed spectrum deficiency, spectrum measures reveal a surprising fact on the spectrum utilization. Thus, the recent studies have affirmed that some licensed spectrum bands, namely, amateur radio, television bands, etc., are mostly not been utilized leaving large spectral holes [7,8] due to the fixed spectrum allocation and the independent usage of the spectrum band. Hence, more dynamic and flexible spectrum utilization methods are required to avoid the scarce problem in the wireless networks.

A thesis submitted in partial fulfillment of the requirement of Nazarbayev University for the degree of Doctor of Philosophy

HARVESTING AND NON-ORTHOGONAL MULTIPLE ACCESS

Sultangali Arzykulov