PaperlayerView exact revision on arXiv

Source-linked AI summary

Upper Mid-Band Spectrum for 6G: Vision, Opportunity and Challenges

Ahmad Bazzi, Roberto Bomfin, Marco Mezzavilla, Sundeep Rangan, Theodore Rappaport, Marwa Chafii

arXiv:2502.17914v3eess.SP

TL;DR

6G needs additional spectrum for rising data traffic, AI, sensing, and gigabit-per-second services, but FR3 channel evidence and deployment guidance remain incomplete. The paper reviews measured FR3 propagation, evaluates spectrum agility and coverage–rate tradeoffs, and examines mMIMO, ISAC, coexistence, and standardization. It identifies frequency agility as important for maintaining suitable rate–coverage tradeoffs while documenting channel-modeling, RF, power, and deployment challenges.

  • Problem

    6G requires new spectrum beyond crowded sub-6 GHz bands, while FR3 channel models and deployment approaches must account for its physical properties and incumbent coexistence.

  • Method

    The paper synthesizes FR3 channel measurements and analyzes spectrum agility, mMIMO, ISAC multiband sensing, commercial deployment challenges, coexistence, and 6G standardization.

  • Results

    At rates up to 200 Mbps, 7 GHz dominates, while above 200 Mbps the optimal band shifts to 18 GHz, reaching up to 575 Mbps and shifting again at 850 Mbps.

  • Takeaways & Limitations

    Dynamically hopping among FR3 frequencies can maintain favorable rate–coverage tradeoffs, while realistic antenna gains are necessary for meaningful cross-frequency comparisons.

Abstract

from arXiv · show

Driven by the pursuit of gigabit-per-second data speeds for future 6G mobile networks, in addition to the support of sensing and artificial intelligence applications, the industry is expanding beyond crowded sub-6 GHz bands with innovative new spectrum allocations. In this paper, we chart a compelling vision for 6G within the frequency range 3 (FR3) spectrum, i.e. $7.125$-$24.25$ $\GHz$, by delving into its key enablers and addressing the multifaceted challenges that lie ahead for these new frequency bands. Here we highlight the physical properties of this never-before used spectrum for cellular by reviewing recent channel measurements for outdoor and indoor environments, including path loss, delay and angular spreads, and material penetration loss, all which offer insights that underpin future 5G/6G wireless communication designs. Building on the fundamental knowledge of the channel properties, we explore FR3 spectrum agility strategies that balance coverage and capacity tradeoffs, while examining coexistence with incumbent systems, such as satellites, radio astronomy, and earth exploration. Moreover, we discuss the potential of massive multiple-input multiple-output technologies, challenges for commercial deployment, and potential solutions for FR3, including multiband sensing for FR3 integrated sensing and communications. Finally, we outline 6G standardization features that are likely to emerge from 3GPP radio frame innovations and open radio access network developments.

I. INTRODUCTION

FR3, spanning 7.125–24.25 GHz, is presented as a “Golden Band” for 6G because it balances coverage and bandwidth while requiring advance allocation and coexistence planning.

  • By 2034, global mobile data traffic is expected to grow five- to nine-fold, with AI accounting for one-third of traffic.
  • FR3 deployment requires spectrum allocation years ahead of the 6G launch so incumbents, technology readiness, and nationwide business cases can be addressed.
  • FR3 spans 7.125–24.25 GHz and is positioned between FR1 and FR2 as an upper-midband option for 6G.
  • The spectrum is being evaluated for cellular expansion while accommodating satellite communications, radio astronomy, earth exploration, and other incumbent services.

B. Improvement items by ITU, 3GPP & Organization

ITU, 3GPP, and related organizations are shaping 6G requirements and FR3 channel-modeling work to support substantially higher data rates and additional capabilities.

  • IMT-2030 targets a peak data rate of up to 200 Gbps, ten times the IMT-2020 target.
  • 3GPP began FR3 channel-modeling work in Release 19 in 2024, while the ITU defined improvements and additional IMT-2030 capabilities.

A. FR3 Channel Modeling

FR3 channel modeling remains constrained by sparse measurements, especially in the 7–24 GHz range, where 3GPP has relied on frequency interpolation to estimate channel parameters.

  • TR 38.901 spans 0.5–100 GHz but was developed without many field measurements across that range and without FR3 measurements.
  • More accurate FR3 channel models are needed to replace estimates based on sparse measurements from selected frequency bands.
  • Because 5G modeling emphasized frequencies below 6 GHz and above 24 GHz, 3GPP used frequency interpolation to estimate channels in the 7–24 GHz band.

B. FR3 pathloss exponents (PLEs) over different environments

Measured FR3 path-loss exponents indicate favorable propagation in several LoS and NLoS settings, although coverage comparisons must account for frequency-dependent antenna aperture and array gain.

  • Indoor hotspot: 1.32 is the measured omnidirectional LoS PLE at 16.95 GHz in indoor hotspot conditions, compared with 1.34 at 6.75 GHz and 1.2 at 28 GHz.
  • Urban microcell: 1.85 is the omnidirectional LoS PLE at 16.95 GHz in UMi, compared with 1.79 at 6.75 GHz and 2.02 at 28 GHz.
  • Coverage interpretation: FR3 can offer improved NLoS coverage relative to higher-frequency mmWave bands, but equal-aperture antenna gains can offset higher-frequency channel loss.
  • Urban microcell: 2.2 dB stronger signal per decade of distance follows from the UMi LoS PLE comparison between 1.79 at 6.75 GHz and 2.02–2.1 at 28 GHz.

C. FR3 delay spreads (DSs), angular spreads (ASs) & pene-

FR3 channels exhibit delay and angular spreads between lower-frequency and mmWave regimes, while material penetration remains frequency- and material-dependent. These properties shape signaling, sensing, beamforming, and coverage decisions.

  • FR3 delay spreads (DSs): 12.7-37.7 ns RMS delay spreads were measured across indoor factory, urban microcell, and indoor hotspot environments.At 6.75 GHz and 16.95 GHz in indoor factory line-of-sight conditions, RMS delay spreads were 14 ns and 12.7 ns, respectively.
  • FR3 angular spreads (ASs): FR3 angular spreads narrow at 16.95 GHz relative to 6.75 GHz, indicating fewer and more focused multipath components.The narrower spread can benefit spatial multiplexing and beamforming by reducing interference between signal paths.
  • FR3 penetration loss: 16.95 GHz has higher material penetration loss than lower frequencies, with losses depending on material type and polarization.Low-emissivity glass and concrete are identified as challenging materials.
  • FR3 penetration loss: FR3 can penetrate materials more effectively than higher-frequency mmWave bands, despite limitations for low-emissivity glass and concrete.The measurements therefore expose a frequency-dependent penetration trade-off rather than uniformly improved penetration.

D. FR3 frequency-dependent features and losses

FR3 propagation varies substantially with frequency, including indoor path loss, rain attenuation, and foliage loss. These measured dependencies are important for updating channel models used in future wireless designs.

  • Frequency-dependent path loss: The frequency-weighted path loss exponent model is better suited than the close-in model for representing measured indoor frequency dependence.The CIF model extends the physics-based close-in reference by incorporating frequency dependence into the path-loss slope.
  • Rain attenuation and foliage loss: 1.16 dB/km is the specific attenuation at 24 GHz under heavy rain, compared with 0.04 dB/km at 7 GHz.The passage reports this frequency difference for a rain rate of 8 mm/hr.
  • Rain attenuation and foliage loss: 49.18 dB foliage loss at 24 GHz over 100 m exceeds the 34.66 dB measured at 7 GHz over the same distance.The reported difference is 14.52 dB at 100 m.
  • Implications for channel modeling: Real-world FR3 measurements are needed because TR 38.901 has largely ignored frequency-dependent temporal and spatial channel characteristics between 6 and 100 GHz.The measurements are intended to inform a future version of the 3GPP channel model.

III. ITU ENHANCEMENTS VIA FR3

FR3 supports ITU IMT-2030 objectives through wider bandwidth, multiband operation, sensing, spectrum agility, and massive MIMO. Its use requires balancing rate, coverage, propagation, and deployment constraints across frequencies.

  • Peak data rates & bandwidth requirements: At least 400 MHz-wide spectrum blocks are proposed for 6G, with three FR3 carriers providing up to 1.2 GHz of non-contiguous bandwidth.The passage connects this bandwidth scale with the 50-200 Gbps peak-rate vision.
  • Peak data rates & bandwidth requirements: 16 antennas can provide 192 Gbps, while additional 400 MHz carrier bandwidth can also achieve 192 Gbps.Combining increased spatial streams and carrier bandwidth is reported to reach 256 Gbps.
  • Location accuracy: A 4f_ref scaling of carrier frequency and bandwidth yields a delay-estimation CRB below 10^-3 at 17 dB SNR.The reported result corresponds to sub-10 cm location accuracy and meets the IMT-2030 target.
  • Coverage-rate tradeoffs & Spectrum agility: At rates up to 200 Mbps, 7 GHz dominates coverage, sustaining 200 Mbps for 30% of the time under the stated bandwidth allocation.The allocations are 100, 200, 300, and 400 MHz at 7, 14, 18, and 24 GHz, respectively.
  • Coverage-rate tradeoffs & Spectrum agility: Beyond 200 Mbps, the optimal band shifts to 18 GHz up to 575 Mbps, with another shift at 850 Mbps.These shifts motivate dynamically hopping among FR3 frequencies to maintain rate-coverage tradeoffs.

IV. CHALLENGES FOR COMMERCIAL DEPLOYMENT

FR3 commercial deployment is constrained by fragmented and dynamic spectrum, higher-frequency phase noise, coexistence requirements, and RF-front-end limitations. These challenges increase processing demands and require broadband, linear, and tunable hardware.

  • Non-contiguous bandwidth and dynamic spectrum: Non-contiguous carrier aggregation requires higher-resolution ADCs and DACs plus sophisticated baseband algorithms, increasing processing complexity and power consumption.
  • Non-contiguous bandwidth and dynamic spectrum: Dynamic spectrum access must support coexistence with satellite, radar, and other wireless services operating in FR3.
  • Phase noise: Higher-frequency phase noise and sampling jitter degrade clock SNR and signal performance across FR3.
  • Spectrum coexistence: Lower FR3 bands overlap fixed-link and defense allocations, while upper FR3 intersects Ku-band satellite services, requiring accurate incumbent sensing and multidimensional sharing policies.
  • RF frontend, impairments and linearity: Commercial-scale broadband PAs, mixers, and tunable filters remain unavailable, while RF front ends must control leakage, compensate band-dependent impairments, and preserve linearity.

V. OPEN QUESTIONS & POTENTIAL SOLUTIONS

Open questions span how FR3 measurements should reshape beam management, sensing, localization, spectrum agility, and coexistence. Proposed directions combine multiband operation, adaptive protocols, AI-controlled radio access, and new propagation-aware designs.

  • Channel-aware beam management: FR3 angular-spread measurements imply revised beam management, while precise beam alignment is needed for ISAC target detection, localization, and tracking.
  • Joint sensing and communication design: Hybrid beamforming can reduce power consumption, but trades away beamforming and selection gains and can degrade communication SINR, spectral efficiency, and sensing accuracy.
  • Coherence time and frequency allocation: FR3’s lower frequencies suit more dynamic environments, whereas higher frequencies suit less mobile environments because coherence times differ across the band.
  • Spectrum agility: Frequency hopping can improve spectral efficiency under severe NLoS blockage, but evaluations should use practical aperture-limited antenna gains.At 28 GHz, typical additional link-budget gains are about 26 dB at the BS and 13 dB at the UE, versus 10 dB and 6 dB at 7 GHz.
  • Incumbent coexistence: Terrestrial interference and malicious actions threaten satellite services, motivating incumbent sensing and real-time interference-nulling beams for coexistence.
  • FR3 localization: RSS-only localization may miss the IMT-2030 1–10 cm target, while angle-based methods are sensitive to dense-urban multipath; joint multiband time- and angle-based processing may be needed.
  • New propagation traits: Diffuse scattering creates dense multipath components whose statistics can help identify and classify targets, while FR3 supports hybrid near-field and far-field beamforming as Fraunhofer distance varies.
  • 3GPP and O-RAN innovations: Multi-band aggregation, AI-configured slots, sparse multi-band pilots, scalable numerology, and agile timing advance support efficient operation across fragmented FR3 carriers.

VI. CONCLUSIONS

The paper presents FR3 as a promising 6G upper-mid-band resource while identifying unresolved propagation, physical-layer, and regulatory issues. It connects spectrum agility, mMIMO, multiband sensing, and intelligent RAN standardization to FR3 deployment.

  • The paper outlines FR3’s upper-mid-band potential, challenges, and required standardization changes for 6G.
  • FR3 offers advantages in capacity, coverage, and spectral efficiency, while its channel characteristics affect beam management.
  • Frequency hopping is highlighted to mitigate blockages, maximize spectral efficiency, and support incumbent coexistence.
  • The paper examines mMIMO design challenges, particularly for uplink, and performance bounds for FR3 ISAC multiband sensing.
  • Proposed 6G standardization features include 3GPP radio-frame modifications and O-RAN RIC-driven xApps and rApps for intelligent spectrum management.
  • Open questions remain regarding FR3 propagation characteristics, PHY design, and regulatory considerations.
Loading 2502.17914v3…