Title Page
Abstract
Contents
Chapter 1. Introduction 20
1.1. Background 20
1.2. Motivations 24
1.2.1. Towards 6G: fundamentals and requirements of NTN 24
1.2.2. 3GPP standards of NR to support non-terrestrial networks 26
1.2.3. Transonic speed mobility support techniques for 6G-non-terrestrial networks 28
1.3. Dissertation outline 30
Chapter 2. 5G-NR Physical Layer-based Solutions to Support High Mobility in 6G Non-Terrestrial Networks 32
2.1. Overview of non-terrestrial networks 34
2.1.1. NTN scenario 37
2.1.2. NTN use cases 39
2.1.3. NTN challenges and solutions 40
2.2. Overview of NR-NTN channel models 42
2.3. 5G-NR link level performance analysis under 3GPP NR-NTN channel models 49
2.3.1. Proposed 5G-NR-based solutions for high mobility 49
2.3.2. Link-level simulations for performance evaluation of the proposed solutions 57
2.4. Demodulation reference signal (DM-RS) based channel estimation for NTN to support high mobility 72
2.4.1. Link-level simulator 72
2.4.2. Design elements for physical layer 74
2.4.3. DM-RS in 5G-NR 77
2.4.4. Proposed front-loaded DM-RS design for high mobility 79
2.4.5. Simulation results and discussion 85
Chapter 3. Enhanced 6G Non-Terrestrial Network Link Performance using Deep Learning-based Channel Estimation and Doppler Compensation Techniques 95
3.1. Deep-learning for channel estimation 98
3.1.1. Conventional channel estimation techniques 98
3.1.2. Deep-learning-based channel estimation approaches 99
3.2. Doppler pre-compensation and compensation techniques 102
3.2.1. Doppler pre-compensation techniques 103
3.2.2. Doppler compensation techniques 104
3.3. Proposed deep-learning assisted channel estimation 104
3.3.1. Deep learning model architecture 107
3.3.2. Training and hyperparameter selection 112
3.4. Proposed doppler pre-compensation methodology 114
3.4.1. Design of doppler compensation techniques 114
3.4.2. Integration with 6G-non-terrestrial networks 118
3.5. Performance evaluation and validation 120
3.5.1. Simulation setup 121
3.5.2. System model 122
3.5.3. Channel models 124
3.5.4. Deep learning channel estimation 126
3.5.5. Doppler pre-compensation and compensation 127
3.6. Simulation results and discussion 129
Chapter 4. 6G Non-Terrestrial Networks Downlink Link Budget Analysis for High Altitude Adaptation 139
4.1. 6G-NTN scenario for link budget analysis 143
4.1.1. Potential Frequencies for 6G-NTN Communications 148
4.1.2. 6G Frame Structure for NTN Communications 153
4.1.3. 6G-NTN link analysis techniques 159
4.1.4. Link design and optimization 160
4.1.5. Link budget evaluation method 164
4.2. Link budget parameters 168
4.2.1. Target bit rate and link margin 170
4.2.2. Transmit power and noise figure 172
4.3. 6G-NTN link analysis 177
Chapter 5. Conclusions 186
References 189
Table 2.1. Link-level simulation parameters. 56
Table 2.2. 5G-NR LLS simulation parameters for NTN. 84
Table 2.3. Maximum spectral efficiency and required SNR for various MCS levels. 91
Table 2.4. Maximum spectral efficiency and required SNR for front-loaded DM-RS additional positions under SISO and MIMO. 91
Table 3.1. Link-level simulation parameters for 6G-NTN. 119
Table 3.2. Training parameters for CNN based channel estimator model. 122
Table 3.3. Analysis of channel estimation under DL. 133
Table 4.1. Conventional versus emerging technologies for 6G-NTN link analysis. 146
Table 4.2. Possible Differences Between 5G-NR And 6G Waveform Parameters for NTN Communication 157
Table 4.3. Link Budget Simulation Properties. 174
Table 4.4. 6G-NTN Service Link Parameters. 175
Table 4.5. 6G-NTN DL Link Properties. 176
Table 4.6. NTN Downlink link margin analysis. 181
Figure 2.1. Non-terrestrial networks scenario and use cases. 37
Figure 2.2. 5G-NR frame structure with flexible numerology. 51
Figure 2.3. Link-level simulation model of the 5G-NR physical layer to support high mobility. 52
Figure 2.4. 5G-NR sub-frame with DM-RS symbol configuration to support high mobility. 55
Figure 2.5. BER performance under various MCS levels for SCS 15 kHz at stationary UE: (a) QPSK, (b) 16-QAM, (c) 64-QAM, and (d) 256-QAM. 59
Figure 2.6. Throughput performance under various MCS levels for SCS 15 kHz at stationary UE: (a) QPSK, (b) 16-QAM, (c) 64-QAM, and (d) 256-QAM. 62
Figure 2.7. Link-level performance under pilot-aided perfect and DM-RS-based practical channel estimation at 500 km/h for TDL-A (NLOS): (a) BER, (b) Throughput, (c) Spectral Efficiency; 63
Figure 2.8. Link-level performance under various 5G-NR numerologies at 500 km/h: (a) BER, (b) Throughput, (c) Spectral Efficiency. 65
Figure 2.9. Link-level performance under various MIMO schemes at 500 km/h: (a) BER, (b) Throughput, (c) Spectral Efficiency. 67
Figure 2.10. Link-level performance of front-loaded DM-RS based channel estimation with additional DM-RS symbol for TDL-A (NLOS): (a) BER, (b) Throughput, (c) Spectral Efficiency. 69
Figure 2.11. Link-level performance of front-loaded DM-RS based channel estimation with additional DM-RS symbol for TDL-E (LOS): (d) BER, (e) Throughput, (f) Spectral Efficiency. 70
Figure 2.12. Block diagram of 5G-NR link-level simulator. 73
Figure 2.13. Front-loaded DM-RS configuration with additional DM-RS symbols to support high mobility. 83
Figure 2.14. Front-loaded DM-RS versus non-front-loaded design performance for high mobility: (a) BER and (b) Spectral Efficiency. 86
Figure 2.15. LLS performance of front-loaded DM-RS for various MCS Levels (a) BER and (b) Spectral Efficiency. 87
Figure 2.16. DM-RS additional position performance under SISO and MIMO: (a) BER and (b) Spectral Efficiency. 88
Figure 2.17. Performance of first front-loaded DM-RS symbol location (I0 = 2 and 3): (a) BER and (b) Spectral Efficiency. 90
Figure 3.1. NTN-aided downlink transmission in TN network under Doppler pre-compensation technique. 97
Figure 3.2. System model of proposed Deep learning-based CNN channel estimator for 6G-NTN. 101
Figure 3.3. Training process for data generation of CNN model for DMRS channel estimation. 123
Figure 3.4. Channel estimation under training data of the CNN model for 612 subcarriers, 14 OFDM symbols and MIMO [612x14 x8] under 64-QAM, 120... 132
Figure 3.5. Mobility and altitude analysis under Doppler pre-compensation: (a) BER, (b) Throughput, and (c) Spectral Efficiency. 135
Figure 3.6. NR-NTN PDSCH LLS performance under Doppler pre-compensation and Rx Doppler compensation: (a) BER, (b) Throughput, and (c) Spectral Efficiency. 137
Figure 4.1. NTN scenario for link budget analysis. 144
Figure 4.2. Architecture of 6G-NTN downlink communication. 145
Figure 4.3. Adaptive 6G frame structure with OFDM waveform for NTN downlink communication. 154
Figure 4.4. Non-terrestrial transmitter-to-receiver link with typical signal loss and noise sources. 165
Figure 4.5. Link budget analysis of 6G-NTN downlink communication. 183