Sharing Spectrum and Services in the 7-24 GHz Upper Midband

TL;DR

This work investigates spectrum and service sharing in the upper midband (7–24 GHz, FR-3) for 6G by combining ray-tracing in realistic urban environments with a policy/technology lens. It compares FR-1, FR-2, and FR-3 in terms of coverage and throughput, showing that FR-3 offers substantial bandwidth with comparable outdoor coverage and potential for high-rate, location-based services. The authors analyze incumbent coexistence across mobile, fixed, radiolocation, satellite, and scientific services, using realistic interference analyses and digital-twin concepts to identify critical interferers and mitigation strategies. They propose an Open RAN–based network architecture to support dynamic spectrum and service sharing, including telemetry-driven orchestration, spectrum slicing, and adaptable RF frontends. The work highlights specific candidate bands (7.125–8.5, 10–10.5, 12.2–13.25, 18.8–20.2 GHz) as promising for 6G FR-3 deployments and emphasizes the role of programmable networks and interference management in enabling practical coexistence with incumbents.

Abstract

The upper midband, spanning 7 to 24 GHz, strikes a good balance between large bandwidths and favorable propagation environments for future 6th Generation (6G) networks. Wireless networks in the upper midband, however, will need to share the spectrum and safely coexist with a variety of incumbents, ranging from radiolocation to fixed satellite services, as well as Earth exploration and sensing. In this paper, we take the first step toward understanding the potential and challenges associated with cellular systems between 7 and 24 GHz. Our focus is on the enabling technologies and policies for coexistence with established incumbents. We consider dynamic spectrum sharing solutions enabled by programmable and adaptive cellular networks, but also the possibility of leveraging the cellular infrastructure for incumbent services. Our comprehensive analysis employs ray tracing and examines real-world urban scenarios to evaluate throughput, coverage tradeoffs, and the potential impact on incumbent services. Our findings highlight the advantages of FR-3 over FR-2 and FR-1 in terms of coverage and bandwidth, respectively. We conclude by discussing a network architecture based on Open RAN, aimed at enabling dynamic spectrum and service sharing.

Figures (5)

• Figure 1: Coverage and throughput for networks deployed at different frequencies in FR-1, FR-2, and FR-3, together with coverage maps for the area considered in the city of Boston. The red dots represent the base station locations.
• Figure 2: Current spectrum allocations by the , the , and the Region 2.
• Figure 3: Candidate bands by fcctac2023preliminary, ATIS atis2023, and 38820, and allocation changes discussed during 23 (overlapping regions indicate different resolutions are under consideration for those frequencies).
• Figure 4: Interference to a antenna (red cross) from the (dots) in the neighboring area. The heatmaps report the produced by each in the worst (top) and average (bottom) case, over 500 Monte Carlo iterations. The color and size of the markers on the map represent the average generated by the . The green dots represent the whose does not exceed the $-10$ dB threshold even in the worst case, at any frequency.
• Figure 5: Extension of the Open RAN architecture for FR-3 spectrum and services sharing.