Co-existence of Terrestrial and Non-Terrestrial Networks in S-band

TL;DR

The paper addresses the challenge of co-existing terrestrial TN and non-terrestrial NTN networks in overlapping S-band frequencies by developing a stochastic-geometry framework to derive exact expressions for the TN coverage probability $P_c(T)$ and average rate $ar{C}$ under two interference scenarios: Case I with NTN DL as the aggressor and Case II with NTN UL as the aggressor. It models TN BSs with a binomial point process within a circular region and NTN components with LEO geometry, leveraging Nakagami-$m$ fading for serving links and Laplace transforms for interference. The authors present closed-form analytical expressions for both cases and validate them through Monte Carlo simulations, revealing that the relative performance of the two co-existence modes depends on network parameters such as NTN altitude, ISD, load, and isolation distance. The results offer design insights and guidelines for dynamic spectrum sharing between TN and NTN, including minimum isolation distances and case switching to optimize network performance. Overall, the work provides analytical tools to inform practical spectrum-sharing strategies for integrated TN-NTN 6G networks.

Abstract

Co-existence of terrestrial and non-terrestrial networks (NTN) is foreseen as an important component to fulfill the global coverage promised for sixth-generation (6G) of cellular networks. Due to ever rising spectrum demand, using dedicated frequency bands for terrestrial network (TN) and NTN may not be feasible. As a result, certain S-band frequency bands allocated by radio regulations to NTN networks are overlapping with those already utilized by cellular TN, leading to significant performance degradation due to the potential co-channel interference. Early simulation-based studies on different co-existence scenarios failed to offer a comprehensive and insightful understanding of these networks' overall performance. Besides, the complexity of a brute force performance evaluation increases exponentially with the number of nodes and their possible combinations in the network. In this paper, we utilize stochastic geometry to analytically derive the performance of TN-NTN integrated networks in terms of the probability of coverage and average achievable data rate for two co-existence scenarios. From the numerical results, it can be observed that, depending on the network parameters, TN and NTN users' distributions, and traffic load, one co-existence case may outperform the other, resulting in optimal performance of the integrated network. The analytical results presented herein pave the way for designing state-of-the-art methods for spectrum sharing between TN and NTN and optimizing the integrated network performance.

Figures (9)

• Figure 1: Co-existence scenarios from TR38.863 where NTN DL (a) or NTN UL (b) interferes with TN DL.
• Figure 2: Geometry of the system model used for derivation of \ref{['eq:CDF_Rn_ntn']} in co-existence Case II. The shaded area is the area obtained from the intersection of the disc centered at TN UE with radius ${\mathcal{R}}_n$ and the outer annulus defining the outer border for NTN UEs.
• Figure 3: Verification of serving distance distribution given in \ref{['eq:CDF_R0']} for urban (ISD = 0.75 km) and rural (ISD = 7.5 km) areas.
• Figure 4: Verification of the interfering distance distribution, for cases I (a), given as in \ref{['eq:Rn_R0_CDF']}, and II (b) given as in \ref{['eq:Rn_R0_CDF']}.
• Figure 5: Coverage probability of Case I for 100% load (a), and 25% load (b), for different LEO satellite altitudes. The simulations, shown by lines, verify the theoretical results, depicted by markers, given in Theorem \ref{['theorem:coverage senario3']}.

Theorems & Definitions (14)

• Theorem 1
• proof
• Lemma 1
• proof
• Lemma 2
• proof
• Theorem 2
• proof
• Theorem 3
• proof