Source-linked AI summary
6G Wireless Channel Measurements and Models: Trends and Challenges
C. -X. Wang, J. Huang, H. Wang, X. Gao, X. -H. You, Y. Hao
TL;DR
Future 6G networks require channel knowledge spanning diverse frequencies, scenarios, and applications because wireless channels underpin system design, network optimization, and performance evaluation. This paper presents a comprehensive survey of measurements, characteristics, and models, then identifies open challenges including general modeling, IRS-based channels, and AI-enabled approaches.
Problem
6G channels span heterogeneous frequency bands, integrated space-air-ground-sea networks, and diverse mobility and application scenarios, creating an open issue for a general standard channel model framework.
Method
The paper surveys 6G channel measurements, characteristics, and models across all frequency bands, scenarios, and application types, and discusses future research challenges.
Results
The survey covers mmWave, THz, optical wireless, satellite, UAV, maritime, underwater acoustic, HST, V2V, ultra-massive MIMO, OAM, and industry IoT channels.
Takeaways & Limitations
More measurements are needed for emerging frequency bands and scenarios, while ray tracing and GBSM can serve as common modeling approaches within their supported scopes.
Abstract
from arXiv · showhide
In this article, we first present our vision on the application scenarios, performance metrics, and potential key technologies of the sixth generation (6G) wireless communication networks. Then, 6G wireless channel measurements, characteristics, and models are comprehensively surveyed for all frequency bands and all scenarios, focusing on millimeter wave (mmWave), terahertz (THz), and optical wireless communication channels under all spectrums, satellite, unmanned aerial vehicle (UAV), maritime, and underwater acoustic communication channels under global coverage scenarios, and high-speed train (HST), vehicle-to-vehicle (V2V), ultra-massive multiple-input multiple-output (MIMO), orbital angular momentum (OAM), and industry Internet of things (IoT) communication channels under full application scenarios. Future research challenges on 6G channel measurements, a general standard 6G channel model framework, channel measurements and models for intelligent reflection surface (IRS) based 6G technologies, and artificial intelligence (AI) enabled channel measurements and models are also given.
I. INTRODUCTION
6G is motivated by requirements beyond 5G and is envisioned to extend broadband, IoT, intelligence, coverage, spectrum use, and application scenarios. The paper therefore surveys 6G channel measurements, characteristics, and models across frequencies and scenarios, and identifies future modeling and measurement challenges.
- 6G is expected to enhance mobile broadband, expand IoT coverage, increase network intelligence, and support long-distance, high-mobility, and extremely-low-power communications.
- 6G targets 1–10 Tbps peak data rate, more than 1 Gbps/m2 area traffic capacity, 3–5 times higher spectrum efficiency, about 10 times higher energy efficiency, and 10–100 times higher connection density than 5G.
- The proposed paradigm shifts are global coverage, all spectrums, full applications, and strong or endogenous security.
- 6G enabling technologies increase sum capacity through greater bandwidth, signal power, spatial/time/frequency channels, and coverage, while reducing interference and noise.
- The paper comprehensively surveys channels from mmWave, THz, and optical wireless bands; space-air-ground-sea scenarios; and HST, V2V, ultra-massive MIMO, OAM, and industry IoT applications.It also discusses future challenges involving measurements, a general standard model framework, IRS-based technologies, and AI-enabled channel modeling.
II. 6G CHANNEL MEASUREMENTS AND CHARACTERISTICS
The survey organizes heterogeneous 6G channel measurements and characteristics across all spectrums, global coverage scenarios, and full application scenarios. It highlights established and emerging measurement areas, including mmWave and THz channels.
- 6G channels span multiple frequency bands and scenarios, with channel sounders and characteristics differing substantially across channel types.
- The survey groups channel measurements and characteristics under all spectrums, global coverage scenarios, and full application scenarios.
- MmWave generally covers 30–300 GHz, whereas THz denotes 0.1–10 THz; both offer large bandwidth but also face path loss, blockage, and atmospheric absorption.
- MmWave measurements remain needed for MIMO, high-dynamics, and outdoor environments, while channel characteristics above 300 GHz remain unclear and require extensive future measurements.
2) Optical wireless channel:
Optical wireless channels use infrared, visible-light, and ultraviolet spectra across diverse environments and exhibit propagation and device characteristics distinct from traditional wireless frequency bands.
- Optical wireless bands comprise infrared wavelengths of 780–106 nm, visible light of 380–780 nm, and ultraviolet wavelengths of 10–380 nm.
- Optical wireless channels support indoor, outdoor, underground, and underwater communications, with directed LOS, non-directed LOS, non-directed NLOS, and tracked scenarios.
- Their distinctive characteristics include material-dependent scattering, nonlinear transmitter/receiver photoelectric behavior, and background-noise effects.
- Unlike traditional frequency bands, optical wireless channels have no multipath fading, Doppler effect, or bandwidth regulation.
- Measured optical wireless parameters include CIR/CTF, path loss, shadowing fading, and RMS delay spread.
2) UAV channel:
UAV channels support aerial communication scenarios and are characterized by three-dimensional deployment, high mobility, non-stationarity, and airframe shadowing. Measurements commonly distinguish air-to-air and air-to-ground links.
- UAV channels exhibit 3D deployment, high mobility, spatial and temporal non-stationarity, and airframe shadowing.
- UAV channels are generally classified as air-to-air and air-to-ground channels.
- UAVs can reduce channel-measurement costs compared with small or medium-sized manned aircraft.
3) Maritime channel:
Maritime channels include air-to-sea and near-sea-surface links, while underwater acoustic channels face low-frequency, high-loss, multipath, dispersion, time variation, and Doppler effects. HST and V2V channels extend the survey to high-mobility scenarios, with measurements and models addressing their changing propagation conditions.
- Maritime channel: Maritime channels comprise UAV-to-ship air-to-sea links and ship-to-ship or ship-to-land near-sea-surface links.
- Underwater acoustic channel: Underwater acoustic channels operate at low frequencies with high transmission loss because of ambient ocean noise.
- Underwater acoustic channel: Underwater acoustic propagation is affected by refraction, reflection, and scattering, producing multipath, time-frequency dispersion, time variation, and Doppler effects.
- HST channel: Ultra-HST speeds exceeding 500 km/h create frequent fast handover and large Doppler spread, motivating mmWave/THz and massive MIMO technologies.
- HST channel: Preliminary HST measurements cover open space, hilly terrain, viaducts, tunnels, cuttings, stations, and intra-wagon environments.
- V2V channel: MmWave V2V measurements span 28, 38, 60, 73, and 77 GHz across highway, urban, open-area, campus, and parking environments, but use single antennas at both ends.
2) Ultra-massive MIMO channel:
Ultra-massive MIMO uses thousands of antennas to improve communication-system efficiency and throughput, while producing spherical wavefronts, spatial non-stationarity, and channel hardening. OAM multiplexing can increase spectrum efficiency, but divergence, misalignment, reflection, and limited measurements constrain its current feasibility.
- Ultra-massive MIMO channel: Ultra-massive MIMO uses thousands of antennas to improve spectral efficiency, energy efficiency, throughput, robustness, and degrees of freedom.
- Ultra-massive MIMO channel: Its large arrays produce spherical wavefronts, spatial non-stationarity, and channel hardening, validated by prior sub-6 GHz and mmWave measurements.
- OAM channel: OAM multiplexing uses orthogonal modes to transmit different signals and potentially improve spectrum efficiency.
- OAM channel: OAM beam divergence and misalignment reduce transmission distance, while reflection destroys mode orthogonality and degrades non-line-of-sight performance.
- Industry IoT channel: Industry IoT channels exhibit varied path loss, random fluctuations, non-line-of-sight propagation, many scatterers, and multi-mobility.
- Industry IoT channel: Only a few industry IoT measurements exist, mainly below 6 GHz, although mmWave measurements are promising for future high-rate massive connections.
III. 6G CHANNEL MODELS FOR ALL FREQUENCY BANDS AND ALL SCENARIOS
The survey classifies 6G channel models into deterministic and stochastic families across frequency bands and scenarios. It describes ray-based, geometry-based, distributional, Markov, and related models for mmWave, THz, optical, satellite, UAV, maritime, and underwater channels.
- 6G channel models are broadly classified as deterministic or stochastic models.
- Deterministic models include measurement-based and ray-tracing approaches, with map-based and point-cloud models as simplified ray-tracing variants.
- MmWave and THz: MmWave models include ray tracing, map-based, point-cloud, quasi-deterministic, SV, propagation-graph, and GBSM approaches.
- Optical wireless: Optical wireless models include recursive, iterative, DUSTIN, ceiling-bounce, geometry-based deterministic, GBSM, and non-GBSM models.
- Satellite: Satellite models commonly describe received-signal amplitude distributions or classify channel conditions as good, moderate, and bad using Markov chains.
- UAV, maritime, and underwater: UAV models include ray tracing, analytical two-ray, RS-GBSM, IS-GBSM, non-GBSM, and Markov models, while maritime and underwater models use ray tracing, two-ray, three-way, GBSM, TWDP, Rayleigh, Ricean, and log-normal distributions.
C. 6G channel models for full application scenarios
Full-application 6G channel models must represent mobility, non-stationarity, large-array propagation, OAM behavior, and industry IoT path-loss conditions. The surveyed approaches combine stochastic, deterministic, dynamic, Markov, propagation-graph, and scenario-specific models, while OAM propagation modeling remains open.
- HST/V2V channel: HST and V2V models must account for high mobility and non-stationarity.
- HST/V2V channel: Available HST/V2V models include ray tracing, GBSM, QuaDRiGa-based, dynamic, Markov, and propagation-graph models.
- Ultra-massive MIMO channel: Ultra-massive MIMO models represent spherical wavefronts through individual-antenna propagation distances and non-stationarity through visible regions and cluster birth-death processes.
- OAM channel: OAM research emphasizes wave generation, detection, antenna design, and feasibility analysis, while channel modeling for OAM propagation remains an open issue.
- Industry IoT channel: Industry IoT path-loss models include free-space, single-slope, 3GPP scenario-specific, industry indoor, and overall models.
- Industry IoT channel: The overall industry IoT path-loss model incorporates line-of-sight and non-line-of-sight conditions to describe fluctuating channel status.
D. Comparison of channel modeling methods for different frequency bands and scenarios
Ray tracing and GBSM provide broadly applicable deterministic and stochastic modeling approaches, while specialized models address particular frequency bands or scenarios. Some combinations, including OAM and industry IoT channels, still require further study.
- Ray tracing can model most 6G channel types, but higher-THz and optical bands require further investigation because relevant material properties are lacking.
- Ray tracing is not applicable to satellite channels because of their long distances and wide coverage areas.
- GBSM has the widest generality and acceptable insights for specific frequency bands and scenarios.
- BDCM converts the underlying channel into the angle/beam domain for specific frequency-band or scenario insights.
- OAM channels, industry IoT channels, and combined frequency-band and scenario cases need further study.
IV. FUTURE RESEARCH CHALLENGES
Future work must improve 6G channel measurements and develop a general model framework spanning heterogeneous frequencies, integrated network scenarios, mobility patterns, and emerging technologies. IRS channel validation and modeling are also open research issues.
- High-performance channel sounders are needed to measure diverse 6G channels quickly and efficiently.Existing mmWave sounders include VNA-based, commercial off-the-shelf, and custom-designed systems; THz sounders are mostly VNA-based.
- Existing standardized channel models focus on terrestrial networks and frequencies up to mmWave bands.
- A general 6G channel model framework must integrate radio-frequency and optical bands, terrestrial and space-air-sea scenarios, and varied 2D and 3D mobility.
- Evaluating 6G channel models requires considering accuracy, complexity, and generality alongside relationships among frequency bands, scenarios, and system parameters.
- IRS channel measurements and modeling remain open issues needed to validate intelligent and reconfigurable wireless environments.
D. AI enabled channel measurements and models
Rapidly growing 6G measurement datasets exceed traditional processing capacity, motivating AI and ML methods for channel measurement and modeling. These methods support tasks such as MPC clustering, scenario classification, channel prediction, and model construction.
- Growing frequency bands, scenarios, and antenna counts make 6G measurement data too large for traditional processing methods.
- AI and ML can support MPC clustering, scenario classification, and channel prediction through clustering, classification, and regression algorithms.
- ANN, CNN, and GAN algorithms can be applied to wireless channel modeling.
- The paper identifies AI-enabled channel measurements and models as part of its future challenges for 6G.