\xpatchcmd
Proof.
UV-Plane Beam Mapping for Non-Terrestrial Networks in 3GPP System-Level Simulations
Abstract

Due to the high altitudes and large beam sizes of satellites, the curvature of the Earth’s surface can impact system-level performance. To consider this, 3GPP introduces the UV-plane beam mapping for system-level simulations of non-terrestrial networks (NTNs). This paper aims to provide a comprehensive understanding of how beams and user equipments (UEs) are placed on the UV-plane and subsequently mapped to the Earth’s surface. We present a general process of projecting UEs on the UV-plane onto the Earth’s surface. This process could offer a useful guideline for beam and UE deployment when evaluating the system-level performance of NTNs.

Index terms — Satellite communication systems, system-level simulations, non-terrestrial networks, UV-plane, beam mapping.

I Introduction

System-level simulation is an important process for evaluating system performance, such as user equipment (UE) throughput, total cell throughput, and cell-edge throughput. A stochastic geometry-based approach was widely used to evaluate the system performance in a very effective way and to check the impacts of various system parameters on the performance [1]. Various system-level simulations have been conducted using stochastic geometry in terrestrial networks (TNs) [2, 3, 4, 5] and non-terrestrial networks (NTNs) [6, 7, 8, 9, 10].

Unlike the stochastic model-based system performance evaluation, 3GPP system-level simulations assume a deterministic hexagonal grid for cell or beam layout and consider physical, medium access control, and transport layer protocols to estimate more realistic system performance. To evaluate the system performance of physical layer techniques, 3GPP proposed the cluster-based fast fading channel model in [11], which also includes a detailed methodology for calibrating the channel model in specific scenarios. A guideline to establish satellite channels was provided in [12] for NTNs. In [13], the NTN system environments and parameters for link- and system-level performance evaluation were proposed in thirty different study cases assuming single satellite scenarios.

As conventional TNs have considered a flat deployment surface, UEs are distributed within hexagonal beams or cells on the 2D plane. However, in NTNs, due to the high altitudes and large beam sizes of satellites, the curvature of the Earth’s surface might affect the system-level performance, so that this should be taken into account in the NTN beam layout. Therefore, in [13], the UV-plane is defined in the satellite reference frame, and hexagonal beams are placed on that plane. This paper aims to provide a comprehensive understanding of how beams and UEs are placed on the UV-plane and then mapped to the Earth’s surface. And also we hope to offer useful guidelines for beam and UE deployment in NTN system-level simulations.

Refer to caption
Figure 1: Principle of the UV-plane beam mapping for the 3dB beamwidth θ3dBsubscript𝜃3dB\theta_{\mathrm{3dB}}italic_θ start_POSTSUBSCRIPT 3 roman_d roman_B end_POSTSUBSCRIPT where PssubscriptPs\mathrm{P}_{\mathrm{s}}roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT is the satellite position, and D𝐷Ditalic_D is the beam radius in the UV-plane. The central beam center is at nadir point.

II 3GPP NTN Beam Layout

For NTN system-level simulations, 3GPP has established the following beam layout assumptions for a single satellite simulation [13].

  • Hexagonal mapping of the beam boresight directions on the UV-plane is assumed.

  • The 3dB beamwidth is used to determine the beam diameter and adjacent beam spacing (ABS).

  • A 19-beam layout that incorporates a wrap-around mechanism: 18 beams around the central beam, surrounded by two distinct interfering tiers.

Based on these assumptions, we discuss the principle of the UV-plane beam mapping hereafter. As shown in Fig. 1, the UV-plane is defined as the plane that includes the position of a satellite and is perpendicular to the Nadir direction, i.e., the direction toward the Earth’s center from the satellite, where the U- and V-axes are the orthogonal basis spanning this plane. The beam layout projected onto the UV-plane is determined by a unit sphere and the antenna pattern of each beam. In Fig. 1, the purple region shows the center beam area spanned in the UV-plane considering the 3dB beamwidth θ3dBsubscript𝜃3dB\theta_{\mathrm{3dB}}italic_θ start_POSTSUBSCRIPT 3 roman_d roman_B end_POSTSUBSCRIPT. Thus, the beam radius, denoted by D𝐷Ditalic_D, is readily calculated using the angle θ3dBsubscript𝜃3dB\theta_{\mathrm{3dB}}italic_θ start_POSTSUBSCRIPT 3 roman_d roman_B end_POSTSUBSCRIPT and the characteristics of the unit sphere as

D=sinθ3dB2.𝐷subscript𝜃3dB2\displaystyle D=\sin\frac{\theta_{\mathrm{3dB}}}{2}.italic_D = roman_sin divide start_ARG italic_θ start_POSTSUBSCRIPT 3 roman_d roman_B end_POSTSUBSCRIPT end_ARG start_ARG 2 end_ARG . (1)

The relationship between the beam radius D𝐷Ditalic_D and the ABS is shown in Fig. 2. Due to the properties of a regular hexagon, GFMGFM\angle\text{GFM}∠ GFM is equal to 30 degrees. By using FG¯=D=sinθ3dB2¯FG𝐷subscript𝜃3dB2\overline{\text{FG}}=D=\sin\frac{\theta_{\mathrm{3dB}}}{2}over¯ start_ARG FG end_ARG = italic_D = roman_sin divide start_ARG italic_θ start_POSTSUBSCRIPT 3 roman_d roman_B end_POSTSUBSCRIPT end_ARG start_ARG 2 end_ARG, the ABS, which is the same as FH¯¯FH\overline{\text{FH}}over¯ start_ARG FH end_ARG, is given by

ABS=2×FM¯=2×32D=3sinθ3dB2.ABS2¯FM232𝐷3subscript𝜃3dB2\displaystyle\text{ABS}=2\times\overline{\text{FM}}=2\times\frac{\sqrt{3}}{2}D% =\sqrt{3}\sin\frac{\theta_{\mathrm{3dB}}}{2}.ABS = 2 × over¯ start_ARG FM end_ARG = 2 × divide start_ARG square-root start_ARG 3 end_ARG end_ARG start_ARG 2 end_ARG italic_D = square-root start_ARG 3 end_ARG roman_sin divide start_ARG italic_θ start_POSTSUBSCRIPT 3 roman_d roman_B end_POSTSUBSCRIPT end_ARG start_ARG 2 end_ARG . (2)

From (2), we can calculate ABSs for several scenarios as in Table I, which aligns with the results in [13]. Once the beam layout is configured using the ABS derived in (2), a pre-defined number of UEs are randomly distributed within each hexagonal beam on the UV-plane for system-level performance evaluation. Each UE on the UV-plane should then be mapped to a position on the Earth’s surface by utilizing the definition and characteristics of the UV-plane. In the next section, we present a general process of mapping UEs from the UV-plane to positions on the Earth’s surface.

Refer to caption
Figure 2: Relationship between the beam radius D𝐷Ditalic_D and the ABS where F and H are beam centers, and G is the intersection point of the three hexagons’ vertices.
TABLE I: Beamwidth and ABS for the scenarios in [13]
Set Scenarios Beamwidth [deg] ABS
Set-1 GEO S-band 0.4011 0.0061
GEO Ka-band 0.1765 0.0027
LEO S-band 4.4127 0.0667
LEO Ka-band 1.7647 0.0267
Set-2 GEO S-band 0.7353 0.0111
GEO Ka-band 0.4412 0.0067
LEO S-band 8.832 0.1334
LEO Ka-band 4.4127 0.0667
Refer to caption
Figure 3: Description of the relationship between the UE position in the UV-plane PuuvsuperscriptsubscriptPuuv\mathrm{P}_{\mathrm{u}}^{\mathrm{uv}}roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT and the actual UE position PusubscriptPu\mathrm{P}_{\mathrm{u}}roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT where OO\mathrm{O}roman_O is the Earth’s center.
Refer to caption
(a) UV-plane, FRF=1FRF1\text{FRF}=1FRF = 1
Refer to caption
(b) Earth’s surface (top view), FRF=1FRF1\text{FRF}=1FRF = 1
Refer to caption
(c) UV-plane, FRF=3FRF3\text{FRF}=3FRF = 3
Refer to caption
(d) Earth’s surface (top view), FRF=3FRF3\text{FRF}=3FRF = 3
Figure 4: Beam layout (black solid lines) and UE locations (colored dots) on both the UV-plane and the Earth’s surface for FRF=1FRF1\text{FRF}=1FRF = 1 and FRF=3FRF3\text{FRF}=3FRF = 3. A LEO satellite with Set-1 parameters operating in the S-band is considered where the center beam elevation of 70 degrees. The red stars represent the satellite positions in the UV-plane, and the yellow hexagons are beams for collecting statistics.

III Projection from UV-plane to Earth’s surface

Let rEsubscript𝑟Er_{\mathrm{E}}italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT be the Earth’s radius and a𝑎aitalic_a denote the satellite altitude. Without loss of generality, we set the satellite position in the Cartesian coordinates as

Ps=[0,0,rE+a]TsubscriptPssuperscript00subscript𝑟E𝑎T\displaystyle\mathrm{P}_{\mathrm{s}}=[0,0,r_{\mathrm{E}}+a]^{\mathrm{T}}roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT = [ 0 , 0 , italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT + italic_a ] start_POSTSUPERSCRIPT roman_T end_POSTSUPERSCRIPT (3)

and focus on one UE whose position on the UV-plane is denoted by PuuvsuperscriptsubscriptPuuv\mathrm{P}_{\mathrm{u}}^{\mathrm{uv}}roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT, and that on the Earth surface by PusubscriptPu\mathrm{P}_{\mathrm{u}}roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT. The important geometry is shown in Fig. 3, and the process of placing the UE on the Earth’s surface is described as follows.

  1. 1.

    Calculate the angle ω𝜔\omegaitalic_ω between OPs¯¯subscriptOPs\overline{\mathrm{O}\mathrm{P}_{\mathrm{s}}}over¯ start_ARG roman_OP start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT end_ARG (the line between the Earth’s center and the satellite) and PsPu¯¯subscriptPssubscriptPu\overline{\mathrm{P}_{\mathrm{s}}\mathrm{P}_{\mathrm{u}}}over¯ start_ARG roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT end_ARG (the line between the satellite and the actual UE position on the Earth’s surface). Let duuvsuperscriptsubscript𝑑uuvd_{\mathrm{u}}^{\mathrm{uv}}italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT denote the distance between the satellite and the UE position in the UV-plane, i.e.,

    duuv=PsPuuv.superscriptsubscript𝑑uuvdelimited-∥∥subscriptPssuperscriptsubscriptPuuv\displaystyle d_{\mathrm{u}}^{\mathrm{uv}}=\lVert\mathrm{P}_{\mathrm{s}}-% \mathrm{P}_{\mathrm{u}}^{\mathrm{uv}}\rVert.italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT = ∥ roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT - roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT ∥ . (4)

    Then, the angle ω𝜔\omegaitalic_ω is obtained by the characteristics of the unit sphere as

    ω=arcsinduuv.𝜔superscriptsubscript𝑑uuv\displaystyle\omega=\arcsin d_{\mathrm{u}}^{\mathrm{uv}}.italic_ω = roman_arcsin italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT . (5)
  2. 2.

    Calculate the zenith of departure (ZOD) angle of the line-of-sight (LOS) path using the angle ω𝜔\omegaitalic_ω as

    θLOS,ZOD=πω.subscript𝜃LOSZOD𝜋𝜔\displaystyle\theta_{\mathrm{LOS,ZOD}}=\pi-\omega.italic_θ start_POSTSUBSCRIPT roman_LOS , roman_ZOD end_POSTSUBSCRIPT = italic_π - italic_ω . (6)
  3. 3.

    Calculate the azimuth of departure (AOD) angle of the LOS path as

    θLOS,AOD=PuuvPs𝐮^subscript𝜃LOSAODsuperscriptsubscriptPuuvsubscriptPs^𝐮\displaystyle\theta_{\mathrm{LOS,AOD}}=\angle\mathrm{P}_{\mathrm{u}}^{\mathrm{% uv}}\mathrm{P}_{\mathrm{s}}\hat{\mathbf{u}}italic_θ start_POSTSUBSCRIPT roman_LOS , roman_AOD end_POSTSUBSCRIPT = ∠ roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_uv end_POSTSUPERSCRIPT roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT over^ start_ARG bold_u end_ARG (7)

    where 𝐮^^𝐮\hat{\mathbf{u}}over^ start_ARG bold_u end_ARG is the unit vector toward the U-axis.

  4. 4.

    Using the geometry of the blue triangle in Fig. 3, calculate the elevation angle as

    α=arccos((rE+a)sinθLOS,ZODrE).𝛼subscript𝑟E𝑎subscript𝜃LOSZODsubscript𝑟E\displaystyle\alpha=\arccos\left(\frac{(r_{\mathrm{E}}+a)\sin\theta_{\mathrm{% LOS,ZOD}}}{r_{\mathrm{E}}}\right).italic_α = roman_arccos ( divide start_ARG ( italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT + italic_a ) roman_sin italic_θ start_POSTSUBSCRIPT roman_LOS , roman_ZOD end_POSTSUBSCRIPT end_ARG start_ARG italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT end_ARG ) . (8)
  5. 5.

    Calculate the slant range as

    du=PsPu¯=rEsinα+rE2sin2α+a2+2rEa.subscript𝑑u¯subscriptPssubscriptPusubscript𝑟E𝛼superscriptsubscript𝑟E2superscript2𝛼superscript𝑎22subscript𝑟E𝑎\displaystyle d_{\mathrm{u}}=\overline{\mathrm{P}_{\mathrm{s}}\mathrm{P}_{% \mathrm{u}}}=-r_{\mathrm{E}}\sin\alpha+\sqrt{r_{\mathrm{E}}^{2}\sin^{2}\alpha+% a^{2}+2r_{\mathrm{E}}a}.italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT = over¯ start_ARG roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT end_ARG = - italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT roman_sin italic_α + square-root start_ARG italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT roman_sin start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_α + italic_a start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + 2 italic_r start_POSTSUBSCRIPT roman_E end_POSTSUBSCRIPT italic_a end_ARG . (9)
  6. 6.

    Obtain the UE projection vector PsPusubscriptPssubscriptPu\overrightarrow{\mathrm{P}_{\mathrm{s}}\mathrm{P}_{\mathrm{u}}}over→ start_ARG roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT end_ARG by converting the spherical coordinates with the ZOD and AOD angles into the Cartesian coordinates with a radius of dusubscript𝑑ud_{\mathrm{u}}italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT, i.e.,

    ΔP=[dusin(θLOS,ZOD)cos(θLOS,AOD)dusin(θLOS,ZOD)sin(θLOS,AOD)ducos(θLOS,ZOD)].ΔPmatrixsubscript𝑑usubscript𝜃LOSZODsubscript𝜃LOSAODsubscript𝑑usubscript𝜃LOSZODsubscript𝜃LOSAODsubscript𝑑usubscript𝜃LOSZOD\displaystyle\Delta\mathrm{P}=\begin{bmatrix}d_{\mathrm{u}}\sin(\theta_{% \mathrm{LOS,ZOD}})\cos(\theta_{\mathrm{LOS,AOD}})\\ d_{\mathrm{u}}\sin(\theta_{\mathrm{LOS,ZOD}})\sin(\theta_{\mathrm{LOS,AOD}})\\ d_{\mathrm{u}}\cos(\theta_{\mathrm{LOS,ZOD}})\end{bmatrix}.roman_Δ roman_P = [ start_ARG start_ROW start_CELL italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT roman_sin ( italic_θ start_POSTSUBSCRIPT roman_LOS , roman_ZOD end_POSTSUBSCRIPT ) roman_cos ( italic_θ start_POSTSUBSCRIPT roman_LOS , roman_AOD end_POSTSUBSCRIPT ) end_CELL end_ROW start_ROW start_CELL italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT roman_sin ( italic_θ start_POSTSUBSCRIPT roman_LOS , roman_ZOD end_POSTSUBSCRIPT ) roman_sin ( italic_θ start_POSTSUBSCRIPT roman_LOS , roman_AOD end_POSTSUBSCRIPT ) end_CELL end_ROW start_ROW start_CELL italic_d start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT roman_cos ( italic_θ start_POSTSUBSCRIPT roman_LOS , roman_ZOD end_POSTSUBSCRIPT ) end_CELL end_ROW end_ARG ] . (10)
  7. 7.

    Calculate the UE position as Pu=Ps+ΔPsubscriptPusubscriptPsΔP\mathrm{P}_{\mathrm{u}}=\mathrm{P}_{\mathrm{s}}+\Delta\mathrm{P}roman_P start_POSTSUBSCRIPT roman_u end_POSTSUBSCRIPT = roman_P start_POSTSUBSCRIPT roman_s end_POSTSUBSCRIPT + roman_Δ roman_P.

From the above process, we can project all UEs in the UV-plane onto their actual positions on the Earth’s surface. With these actual positions, we can proceed with the system-level simulations, including large- and small-scale channel generation and reference signal received power-based UE attachment.

Refer to caption
(a) FRF=1FRF1\text{FRF}=1FRF = 1
Refer to caption
(b) FRF=3FRF3\text{FRF}=3FRF = 3
Figure 5: Distributions of satellite-UE distances.

IV Simulation Results

In this section, we simulate the UE projection process presented in the previous section. In our scenario, a LEO satellite is located at an altitude of 1200 km with the center beam elevation of 70 degrees. We assume the satellite configuration is based on Set-1 and the S-band, which includes a 3dB beamwidth of 4.4127 degrees and an ABS of 0.0667 in the UV-plane.

Fig. 4 shows the beam layout and UE locations in both the UV-plane and the Earth’s surface for the scenario of a LEO satellite with Set-1 parameters operating in the S-band, and the center beam elevation θcsubscript𝜃c\theta_{\mathrm{c}}italic_θ start_POSTSUBSCRIPT roman_c end_POSTSUBSCRIPT of 70 degrees. The numbers of beams for FRF=1FRF1\text{FRF}=1FRF = 1 and FRF=3FRF3\text{FRF}=3FRF = 3 are 61 and 127, respectively, and ten UEs are randomly dropped within each beam. For system-level simulations, the inner-19 beams (yellow hexagons) are used to collect UE performance statistics, while the outer beams only generate interference to the inner beams. The center of the center beam is located at (sinθc,0)=(0.2878,0)subscript𝜃c00.28780(\sin\theta_{\mathrm{c}},0)=(0.2878,0)( roman_sin italic_θ start_POSTSUBSCRIPT roman_c end_POSTSUBSCRIPT , 0 ) = ( 0.2878 , 0 ) in the UV-plane. As expected, the shape of the beam layout in the UV-plane is distorted when being projected onto the Earth’s surface due to the curvature of the Earth’s surface. It can be seen that the actual beam size on the Earth’s surface is highly dependent on the number of beams per satellite. When a satellite generates more beams, the edge beams have higher elevation angle, resulting in a more distorted beam shape.

Fig. 5 illustrates the distributions of distances between the satellite and UEs. The UEs in the beams closer to the satellite exhibit the shorter distances due to their higher elevation angles. For example, for FRF=1FRF1\text{FRF}=1FRF = 1, the UEs in the beam 50 have the shortest distances, while those in the beam 38 have much larger distances. For FRF=3FRF3\text{FRF}=3FRF = 3, the edge beams are more distorted due to the longer distance from the center beam. Thus, distances are spread over a larger range, e.g., from 1200 km (for UEs in the beam 76 or 77) to 1771 km (for UEs in the beam 92) compared to the case of FRF=1FRF1\text{FRF}=1FRF = 1.

V Conclusions

This paper considered the methodology of UV-plane beam mapping used in 3GPP NTN system-level simulations. First, we described the definition of the UV-plane and the relationship between the 3dB beamwidth and beam size in the plane. Next, we presented a general process for placing beams and UEs on the Earth’s surface. This process could provide a valuable guideline for beam and UE deployment when evaluating the system-level performance of NTNs.

Acknowledgment

This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No.2021-0-00847, Development of 3D Spatial Satellite Communications Technology).

References

  • [1] F. Baccelli and B. Blaszczyszyn, Stochastic geometry and wireless networks: Volume I theory.   Found. Trends in Networking, 2009.
  • [2] M. Haenggi, J. G. Andrews, F. Baccelli, O. Dousse, and M. Franceschetti, “Stochastic geometry and random graphs for the analysis and design of wireless networks,” IEEE J. Sel. Areas Commun., vol. 27, no. 7, pp. 1029–1046, 2009.
  • [3] S. Singh, H. S. Dhillon, and J. G. Andrews, “Offloading in heterogeneous networks: Modeling, analysis, and design insights,” IEEE Trans. Wireless Commun., vol. 12, no. 5, pp. 2484–2497, 2013.
  • [4] S. Singh and J. G. Andrews, “Joint resource partitioning and offloading in heterogeneous cellular networks,” IEEE Trans. Wireless Commun., vol. 13, no. 2, pp. 888–901, 2014.
  • [5] J. G. Andrews, T. Bai, M. N. Kulkarni, A. Alkhateeb, A. K. Gupta, and R. W. Heath, “Modeling and analyzing millimeter wave cellular systems,” IEEE Trans. Commun., vol. 65, no. 1, pp. 403–430, 2017.
  • [6] N. Okati, T. Riihonen, D. Korpi, I. Angervuori, and R. Wichman, “Downlink coverage and rate analysis of low Earth orbit satellite constellations using stochastic geometry,” IEEE Trans. Commun., vol. 68, no. 8, pp. 5120–5134, 2020.
  • [7] D.-H. Jung, J.-G. Ryu, W.-J. Byun, and J. Choi, “Performance analysis of satellite communication system under the shadowed-rician fading: A stochastic geometry approach,” IEEE Trans. Commun., vol. 70, no. 4, pp. 2707–2721, Apr. 2022.
  • [8] D.-H. Jung, G. Im, J.-G. Ryu, S. Park, H. Yu, and J. Choi, “Satellite clustering for non-terrestrial networks: Concept, architectures, and applications,” IEEE Veh. Technol. Mag., vol. 18, no. 3, pp. 29–37, Sep. 2023.
  • [9] D.-H. Jung, H. Nam, J. Choi, and D. J. Love, “Modeling and analysis of GEO satellite networks,” 2023, arXiv:2312.15924.
  • [10] M. Lee, S. Kim, M. Kim, D.-H. Jung, and J. Choi, “Analyzing downlink coverage in clustered low Earth orbit satellite constellations: A stochastic geometry approach,” 2024, arXiv:2402.16307.
  • [11] 3GPP TR 38.901 v14.3.0, “Study on channel model for frequencies from 0.5 to 100 GHz,” Jan. 2018.
  • [12] 3GPP TR 38.811 v15.4.0, “Study on NR to support non-terrestrial networks,” Sep. 2020.
  • [13] 3GPP TR 38.821 v16.0.0, “Solutions for NR to support non-terrestrial networks (NTN),” Dec. 2019.