Source-linked AI summary
Rapid Fading Due to Human Blockage in Pedestrian Crowds at 5G Millimeter-Wave Frequencies
George R. MacCartney, Theodore S. Rappaport, Sundeep Rangan
TL;DR
Pedestrian blockage creates rapid fading in mmWave links, while outdoor evidence for its channel dynamics has been limited. The paper measures blockage on 73.5 GHz P2P links with multiple antenna beamwidths and models events using two- and four-state approaches. Fade durations are longer for narrower beams, and average fade attenuation increases as antenna HPBW decreases, with asymmetric decay and rise rates.
Problem
Outdoor measurements have been limited for characterizing mmWave channel dynamics and fading, despite blockage susceptibility affecting communications systems.
Method
The study conducts crowded-urban P2P measurements at 73.5 GHz using 7°, 15°, and 60° HPBW antenna pairs, then applies two-state and four-state blockage models.
Results
299.0 ms versus 260.2 ms mean fade duration was measured for 7° versus 60° HPBW antennas, while mean attenuation was 15.8 dB versus 11.5 dB, respectively.
Takeaways & Limitations
The models provide statistical descriptions of pedestrian blockage for extending mmWave channel models and simulating real-world fading in 5G system design.
Abstract
from arXiv · showhide
Rapidly fading channels caused by pedestrians in dense urban environments will have a significant impact on millimeter-wave (mmWave) communications systems that employ electrically-steerable and narrow beamwidth antenna arrays. A peer-to-peer (P2P) measurement campaign was conducted with 7-degree, 15-degree, and 60-degree half-power beamwidth (HPBW) antenna pairs at 73.5 GHz and with 1 GHz of RF null-to-null bandwidth in a heavily populated open square scenario in Brooklyn, New York, to study blockage events caused by typical pedestrian traffic. Antenna beamwidths that range approximately an order of magnitude were selected to gain knowledge of fading events for antennas with different beamwidths since antenna patterns for mmWave systems will be electronically-adjustable. Two simple modeling approaches in the literature are introduced to characterize the blockage events by either a two-state Markov model or a four-state piecewise linear modeling approach. Transition probability rates are determined from the measurements and it is shown that average fade durations with a -5 dB threshold are 299.0 ms for 7-degree HPBW antennas and 260.2 ms for 60-degree HPBW antennas. The four-state piecewise linear modeling approach shows that signal strength decay and rise times are asymmetric for blockage events and that mean signal attenuations (average fade depths) are inversely proportional to antenna HPBW, where 7-degree and 60-degree HPBW antennas resulted in mean signal fades of 15.8 dB and 11.5 dB, respectively. The models presented herein are valuable for extending statistical channel models at mmWave to accurately simulate real-world pedestrian blockage events when designing fifth-generation (5G) wireless systems.
I. INTRODUCTION
The introduction motivates studying pedestrian-induced blockage in outdoor mmWave channels because existing work has limited measurements of channel dynamics and fading. This paper addresses that gap with P2P measurements and simple statistical models in a crowded urban setting.
- MmWave signals are highly susceptible to blockage, causing rapid channel variations that affect the communications protocol stack.
- Outdoor measurements across mmWave bands have been limited for understanding channel dynamics and fading.
- Earlier studies reported substantial human-blockage attenuation and fade durations ranging from approximately 100 ms for one person to 300 ms or more for large groups.
- Prior studies linked fading effects to antenna beamwidth and bandwidth but did not provide precise statistical characterization of blockage dynamics for simulations.
- This work studies rapid pedestrian fading through P2P measurements across a busy walkway in a downtown open square and applies simple statistical modeling techniques.
II. MEASUREMENT SETUP AND HARDWARE
The campaign measured pedestrian blockage on a fixed 73.5 GHz P2P link in a crowded Brooklyn open square using antenna pairs spanning 7° to 60° HPBW. Repeated 135-second observations enabled comparison across beamwidths.
- The measurement site was a high-foot-traffic open square in downtown Brooklyn where pedestrians, strollers, and bicyclists crossed between the transmitter and receiver.
- The transmitter and receiver were positioned on opposite sides of the walkway at 1.5 m height, with a 14.2 m separation and traffic moving mainly perpendicular to the link.
- Three tests used matched 7°, 15°, and 60° azimuth and elevation HPBW horn-antenna pairs to examine beamwidth-dependent fading.
- Each test recorded a 135-second free-running window containing 40,800 PDP snapshots at approximately 3.3 ms spacing.
- The channel sounder operated at 73.5 GHz with 1 GHz RF null-to-null bandwidth and 2 ns multipath-component time resolution.
III. MODELING METHODOLOGY
The paper uses two simple blockage-event representations: a two-state model distinguishing shadowed from unshadowed conditions, and a four-state model that resolves signal decay and recovery.
- The two-state approach models blockage events with unshadowed and shadowed states.
- The four-state approach separates unshadowed, decaying, shadowed, and rising signal-level states.
A. Two-State Blockage Modeling
The paper models pedestrian blockage using two-state and four-state approaches. The four-state formulation separates unshadowed, decaying, shadowed, and rising signal regions and parameterizes their timing and attenuation.
- Two-state approach: Two-state modeling classifies a link as unshadowed or shadowed using a predefined threshold, typically between 0 dB and -10 dB.Transition probabilities describe movement between the two states.
- Four-state approach: Four-state modeling adds decaying and rising signal regions between unshadowed and shadowed states.This represents the blockage waveform as a sequence of distinct signal-level regions.
- Four-state approach: The four-state model defines SE(t) using mean attenuation, decay and rise rates, and fade duration over the blockage event.The decay and rise rates are determined from decay time, rise time, and either a predefined threshold or SEmean.
- Threshold selection: Threshold selection changes measured decay and rise times and therefore changes the corresponding signal-strength rates.Different applications may require thresholds matched to system requirements and fade-margin limitations.
- Four-state approach: The four-state Markov model assigns transitions from unshadowed to decaying, decaying to shadowed, shadowed to rising, and rising to unshadowed.These transitions correspond to the four regions used in the piecewise linear representation.
A. Two-State Blockage Model
The two-state model uses threshold-based state transitions and measured transition rates to characterize pedestrian blockage. The results show longer fades for narrower beams, with fade-duration distributions closely matching empirical observations.
- Measurement analysis: The comparison uses a -3 dB threshold from the unshadowed 0 dB attenuation level to define blockage observations across antenna pairs.The transition probability rate is computed as λ = p/T from transition probability p and sampling interval T.
- Transition rates: Approximately 0.20 trans/sec is the transition rate into a shadowed state for all three antenna HPBW pairs and the combined measurements.This rate implies approximately 5 seconds between shadowed states.
- Transition rates: The 7° and 60° antenna pairs have transition-rate-based blockage durations of 298 ms and 260 ms, respectively.The paper attributes longer narrow-beam fades to less energy spread around obstructions and a narrower receiving view of diffracted energy.
- Fade durations: 299.0 ms, 267.4 ms, and 260.2 ms are the mean fade durations for 7°, 15°, and 60° TX/RX HPBW pairs, respectively.Mean fade duration decreases as antenna beamwidth increases.
- Fade durations: The fitted fade-duration CDFs have goodness-of-fit values of at least 0.95, indicating a good match to empirical data.Figure 8 presents CDF curves for each antenna set and for the combined observations.
- System significance: A 200–300 ms blockage can severely degrade throughput or cause an outage when packet transmission times are at the microsecond level.The paper identifies access-point diversity and beam switching around obstacles as possible reliability measures.
B. Four-State Blockage Model
The four-state model represents pedestrian blockage as unshadowed, decaying, shadowed, and rising signal states, then fits distributions to event dynamics across antenna beamwidths. Wider beams generally reduce attenuation and fade duration, while decay and recovery rates are asymmetric.
- Four-state model: The model uses unshadowed, decaying, shadowed, and rising signal-strength states, with a 0 dB threshold defining the event boundaries.Fade duration, decay rate, mean attenuation, and rise rate are estimated for each blockage event.
- Distribution fits: The fitted CDFs characterize decay rate, rise rate, mean attenuation, and fade duration for 7°, 15°, and 60° HPBW antenna pairs.The fits use empirical blockage-event data and corresponding parameters in Table IV.
- Decay and rise rates: 0.15 dB/ms was the median decay rate for 60° HPBW antennas, compared with 0.13 dB/ms and 0.12 dB/ms for 7° and 15° antennas.Two unusually rapid 15° events skewed the mean, making median rates more representative of the trend.
- Decay and rise rates: 0.13 dB/ms and 0.14 dB/ms were the median rise rates for 15° and 60° antennas, versus 0.10 dB/ms for 7° antennas.Decay and rise rates were asymmetric, with the precise reason left unresolved.
- Mean attenuation: 15.8 dB, 12.4 dB, and 11.5 dB were the average attenuations for 7°, 15°, and 60° HPBW antennas, respectively.The inverse relationship reflects wider antennas capturing more diffracted energy around pedestrians.
- Fade duration: 321.3 ms and 378.0 ms were the mean fade durations for 60° and 7° HPBW antennas, respectively, showing shorter fades with wider beams.Wider viewing angles capture more diffracted energy around pedestrian blockers.
- Applications: The transition rates and fitted CDF parameters support simulations of blockage events and the design of PHY/MAC protocols and beam-management algorithms.The paper also connects these durations to microsecond-scale beam switching or diversity strategies for millimeter-wave systems.
V. CONCLUSION
The measurements characterize pedestrian blockage at mmWave using two complementary models, showing beamwidth-dependent fade duration and attenuation, asymmetric signal transitions, and simulation relevance for system design.
- The two-state Markov model indicates that shadowing events last longer with narrowbeam antennas than with widebeam antennas.Conditional transition probability rates from the measurements can be used to simulate random blockage events in mmWave system and network simulators.
- The four-state model finds asymmetric signal-strength decay and rise rates during blockage events.The asymmetry is attributed to nonsymmetrical blockers, groups of pedestrians walking together, or nonperpendicular crossings of the line-of-sight path.
- Mean signal fades increase as antenna HPBW decreases: 11.5 dB for 60-degree, 12.4 dB for 15-degree, and 15.8 dB for 7-degree antenna pairs.A simple model for mean signal attenuation as a function of antenna HPBW is introduced from this observation.
- These pedestrian blockage measurements are relevant to designing electrically steerable mmWave systems that must maintain connectivity and throughput during rapid fading.The conclusion identifies future study of microsecond-speed beam switching to locate reflectors and scatterers when the dominant path is obstructed.